summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rwxr-xr-xBuild/tools/tpm-ctan-check2
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/CATALOG12
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/History91
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/INSTALL43
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/License339
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/MANIFEST21
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/README92
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/Index16
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-deutsch.sty83
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-french.sty64
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-italian.sty63
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/README.txt131
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cwbasehy.tex2239
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cweb-hy.cls1538
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/nodoc.tex121
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/contrib/wagner/cwebzw.sty323
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/cweb-conf.pdfbin0 -> 36580 bytes
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/cweb-user.pdfbin0 -> 137862 bytes
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/cwebbase.tex2080
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/Makefile29
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.aux26
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.bbl8
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.dvibin0 -> 24308 bytes
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.idx40
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.log114
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.scn32
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.tex412
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/rcs.sty227
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/tex-it203
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.bib23
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.w327
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/Imakefile106
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/README31
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/TODO168
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cwbl-german.sty109
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweave-spec.tex403
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-conf.tex103
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-doc.sty133
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.fig78
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.latex62
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.ltx29
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb-user.tex779
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cweb.doc1396
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cwebarray.sty50
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cwebbase.doc2059
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cwebparts.doc452
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/cwebx.sty406
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.dtx347
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.ins6
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/style/cweb-doc.el22
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/Imakefile142
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/badend.w21
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/badopts.w13
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/change.ch5
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/change.w20
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/enddocbegin.w16
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/flat.w47
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/german.w33
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.ch5
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.w31
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/minimal.w12
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/modes.w16
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/newif.w39
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/nolists.w16
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/parts-code.w32
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/parts.tex47
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/refname.w21
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/report.w48
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/section.w55
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-changes.w20
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-format.w17
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-lists.w16
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/titlepage.w19
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/token.w29
-rw-r--r--Master/texmf-dist/doc/latex/cweb-latex/src/test/vbar.w30
-rw-r--r--Master/texmf-dist/source/latex/cweb-latex/Makefile79
-rw-r--r--Master/texmf-dist/tex/latex/cweb-latex/cwbl-german.sty109
-rw-r--r--Master/texmf-dist/tex/latex/cweb-latex/cweb.cls1417
-rw-r--r--Master/texmf-dist/tex/latex/cweb-latex/cweb.cls.patch36
-rw-r--r--Master/texmf-dist/tex/latex/cweb-latex/cwebarray.sty50
-rw-r--r--Master/texmf-dist/tex/latex/cweb-latex/keyvald.sty74
-rw-r--r--Master/tlpkg/tlpsrc/collection-latexextra.tlpsrc1
-rw-r--r--Master/tlpkg/tlpsrc/cweb-latex.tlpsrc2
83 files changed, 17955 insertions, 1 deletions
diff --git a/Build/tools/tpm-ctan-check b/Build/tools/tpm-ctan-check
index 6ad2c896cb7..e904dbcb559 100755
--- a/Build/tools/tpm-ctan-check
+++ b/Build/tools/tpm-ctan-check
@@ -36,7 +36,7 @@ my @WorkingTPM = qw(
crop crossreference csquotes csvtools ctable
cursolatex cursor cuisine
currvita curve curve2e curves
- custom-bib cyrillic
+ custom-bib cweb-latex cyrillic
dancers dashbox datatool
dateiliste datetime dcpic decimal delimtxt digiconfigs
dinat dingbat directory disser dk-bib dlfltxb doi doipubmed dottex
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/CATALOG b/Master/texmf-dist/doc/latex/cweb-latex/CATALOG
new file mode 100644
index 00000000000..4e8e259d023
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/CATALOG
@@ -0,0 +1,12 @@
+Name: cweb
+Version: 1.0 [29 Aug 95]
+Problems-To: schrod@iti.informatik.th-darmstadt.de
+Author: Joachim Schrod
+Address: TU Darmstadt, Alexanderstr. 10, D-64283 Darmstadt, Germany.
+ No phone calls, please.
+
+Description: LaTeX CWEB: Use LaTeX2e markup for your CWEB programs.
+ The cweb class provides hierarchic document structure (with
+ real sections, subsections, etc.), and is highly configurable.
+
+Keywords: CWEB, Literate Programming, LaTeX
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/History b/Master/texmf-dist/doc/latex/cweb-latex/History
new file mode 100644
index 00000000000..fba36fdefb6
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/History
@@ -0,0 +1,91 @@
+cweb-sty/History 30 Nov 95 -*- Indented-Text -*-
+---------------------------------------------------------------------------
+
+PL1 30 Nov 95
+
+ Make it usable with LaTeX <1994/12/01>.
+
+
+Version 1.1 20 Nov 95
+
+ Section numbers are properly reset when a higher-level section begins.
+
+ Support for German, French, and Italian CWEB documents.
+
+ Experimental version of cwebparts package, to input CWEB
+ programs in LaTeX documents. (No user documentation yet.)
+
+ Chunk numbers in margin are more often aligned to the first
+ program line. Chunk marker (paragraph sign) doesn't have a
+ depth any more.
+
+ Index doesn't produce an error with base classes report and
+ book any more.
+
+
+Version 1.0 29 Aug 95
+
+ Updated to LaTeX2e, this is now a document class.
+ User configuration via cweb.cfg and keyword-value class options,
+ the base class of this document class may be configured.
+ Improved user manual, added more concept explanations.
+
+ Realized hierarchic document structure.
+ Flat document structure is an option now.
+
+ Suppression of change hints, unchanged sections, identifier
+ index, and reference list is supported.
+
+ Moved Counter cwebSecNoEject to protected interface.
+ Chunk numbers of CWEAVE are used instead of self-computed ones.
+ Different document base sizes than 10pt may be used.
+
+
+Version 0.4 10 Aug 93
+
+ Added cwebarray.sty to the official release.
+ Added cwebzw.sty as contributed option.
+
+ Left shift operator (`<<') is now defined correctly.
+ CWEB macro and program part in one section works now.
+ Typed numbers now produce a subscript, as in the plain version.
+
+ Cross references to section numbers do not output a period
+ after the section number any more.
+
+ Updated user documentation.
+
+ Start new page on main section only if @*2 or higher (configurable).
+
+
+Version 0.3 17 Jun 93
+
+ CWEB 3.0 was released yesterday. Mentioned that in the doc.
+
+
+Version 0.2 15 Jun 93
+
+ cweb is now a style, not an option.
+
+ CWEAVE markup has changed in April '93. Documents and
+ implements the new and changed tags.
+ In particular, supports the group level construct of
+ starred sections, i.e., `@**', `@*1', etc.
+
+ Handles @( foo @> constructs.
+
+ The detection of `@.' entries is more robust now.
+
+ Works with LaTeX versions prior to 1992, too.
+
+ Allows an optional \con at the file end (was mandatory before).
+ Typesets identifier index or refinement list only when
+ they exist.
+
+ Improved the documentation, both internal and user.
+
+
+Version 0.1 09 Apr 93
+
+ First release, experimental alpha
+ still an option, not a style
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/INSTALL b/Master/texmf-dist/doc/latex/cweb-latex/INSTALL
new file mode 100644
index 00000000000..da0c453681d
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/INSTALL
@@ -0,0 +1,43 @@
+$Id: INSTALL,v 1.2 1995/11/30 15:41:27 schrod Exp $
+------------------------------------------------------------
+
+
+You're not on Unix? Jump ahead to `INSTALLATION BY HAND'.
+
+
+INSTALLATION ON UNIX
+====================
+
+I have supplied a Makefile for your convenience. You have to edit it
+to tell make where your TeX installation lives.
+
+Do you have a TDS-compliant installation?
+(E.g., teTeX, highly recommended.)
+
+ yes => Change the value of TEXMF to your TDS root directory.
+ no => Change even more variables (the ones below TEXMF).
+ Make sure all parent directories exist.
+
+
+Type
+ make
+
+Everything should be installed now.
+
+If you have a TDS installation structure and make complains that the
+LaTeX directory is not found, install LaTeX first.
+
+
+
+INSTALLATION BY HAND
+====================
+
+Copy *.cls, *.sty, and contrib/*.sty to some directory where LaTeX finds it.
+ The correct TDS directory name is tex/latex/cweb/, relative to the
+ TDS root.
+
+Copy *.dvi from the current directory to the directory
+where you keep the documentation for your installed LaTeX styles.
+ What? You don't have such a directory? Shame on you. ;-)
+ The TDS proposes one of the following directory names: doc/latex/cweb/
+ or doc/latex/styles/cweb/. I use the latter.
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/License b/Master/texmf-dist/doc/latex/cweb-latex/License
new file mode 100644
index 00000000000..a43ea2126fb
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/License
@@ -0,0 +1,339 @@
+ GNU GENERAL PUBLIC LICENSE
+ Version 2, June 1991
+
+ Copyright (C) 1989, 1991 Free Software Foundation, Inc.
+ 675 Mass Ave, Cambridge, MA 02139, USA
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+ Preamble
+
+ The licenses for most software are designed to take away your
+freedom to share and change it. By contrast, the GNU General Public
+License is intended to guarantee your freedom to share and change free
+software--to make sure the software is free for all its users. This
+General Public License applies to most of the Free Software
+Foundation's software and to any other program whose authors commit to
+using it. (Some other Free Software Foundation software is covered by
+the GNU Library General Public License instead.) You can apply it to
+your programs, too.
+
+ When we speak of free software, we are referring to freedom, not
+price. Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+this service if you wish), that you receive source code or can get it
+if you want it, that you can change the software or use pieces of it
+in new free programs; and that you know you can do these things.
+
+ To protect your rights, we need to make restrictions that forbid
+anyone to deny you these rights or to ask you to surrender the rights.
+These restrictions translate to certain responsibilities for you if you
+distribute copies of the software, or if you modify it.
+
+ For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must give the recipients all the rights that
+you have. You must make sure that they, too, receive or can get the
+source code. And you must show them these terms so they know their
+rights.
+
+ We protect your rights with two steps: (1) copyright the software, and
+(2) offer you this license which gives you legal permission to copy,
+distribute and/or modify the software.
+
+ Also, for each author's protection and ours, we want to make certain
+that everyone understands that there is no warranty for this free
+software. If the software is modified by someone else and passed on, we
+want its recipients to know that what they have is not the original, so
+that any problems introduced by others will not reflect on the original
+authors' reputations.
+
+ Finally, any free program is threatened constantly by software
+patents. We wish to avoid the danger that redistributors of a free
+program will individually obtain patent licenses, in effect making the
+program proprietary. To prevent this, we have made it clear that any
+patent must be licensed for everyone's free use or not licensed at all.
+
+ The precise terms and conditions for copying, distribution and
+modification follow.
+
+ GNU GENERAL PUBLIC LICENSE
+ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+ 0. This License applies to any program or other work which contains
+a notice placed by the copyright holder saying it may be distributed
+under the terms of this General Public License. The "Program", below,
+refers to any such program or work, and a "work based on the Program"
+means either the Program or any derivative work under copyright law:
+that is to say, a work containing the Program or a portion of it,
+either verbatim or with modifications and/or translated into another
+language. (Hereinafter, translation is included without limitation in
+the term "modification".) Each licensee is addressed as "you".
+
+Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope. The act of
+running the Program is not restricted, and the output from the Program
+is covered only if its contents constitute a work based on the
+Program (independent of having been made by running the Program).
+Whether that is true depends on what the Program does.
+
+ 1. You may copy and distribute verbatim copies of the Program's
+source code as you receive it, in any medium, provided that you
+conspicuously and appropriately publish on each copy an appropriate
+copyright notice and disclaimer of warranty; keep intact all the
+notices that refer to this License and to the absence of any warranty;
+and give any other recipients of the Program a copy of this License
+along with the Program.
+
+You may charge a fee for the physical act of transferring a copy, and
+you may at your option offer warranty protection in exchange for a fee.
+
+ 2. You may modify your copy or copies of the Program or any portion
+of it, thus forming a work based on the Program, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+ a) You must cause the modified files to carry prominent notices
+ stating that you changed the files and the date of any change.
+
+ b) You must cause any work that you distribute or publish, that in
+ whole or in part contains or is derived from the Program or any
+ part thereof, to be licensed as a whole at no charge to all third
+ parties under the terms of this License.
+
+ c) If the modified program normally reads commands interactively
+ when run, you must cause it, when started running for such
+ interactive use in the most ordinary way, to print or display an
+ announcement including an appropriate copyright notice and a
+ notice that there is no warranty (or else, saying that you provide
+ a warranty) and that users may redistribute the program under
+ these conditions, and telling the user how to view a copy of this
+ License. (Exception: if the Program itself is interactive but
+ does not normally print such an announcement, your work based on
+ the Program is not required to print an announcement.)
+
+These requirements apply to the modified work as a whole. If
+identifiable sections of that work are not derived from the Program,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works. But when you
+distribute the same sections as part of a whole which is a work based
+on the Program, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Program.
+
+In addition, mere aggregation of another work not based on the Program
+with the Program (or with a work based on the Program) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+ 3. You may copy and distribute the Program (or a work based on it,
+under Section 2) in object code or executable form under the terms of
+Sections 1 and 2 above provided that you also do one of the following:
+
+ a) Accompany it with the complete corresponding machine-readable
+ source code, which must be distributed under the terms of Sections
+ 1 and 2 above on a medium customarily used for software interchange; or,
+
+ b) Accompany it with a written offer, valid for at least three
+ years, to give any third party, for a charge no more than your
+ cost of physically performing source distribution, a complete
+ machine-readable copy of the corresponding source code, to be
+ distributed under the terms of Sections 1 and 2 above on a medium
+ customarily used for software interchange; or,
+
+ c) Accompany it with the information you received as to the offer
+ to distribute corresponding source code. (This alternative is
+ allowed only for noncommercial distribution and only if you
+ received the program in object code or executable form with such
+ an offer, in accord with Subsection b above.)
+
+The source code for a work means the preferred form of the work for
+making modifications to it. For an executable work, complete source
+code means all the source code for all modules it contains, plus any
+associated interface definition files, plus the scripts used to
+control compilation and installation of the executable. However, as a
+special exception, the source code distributed need not include
+anything that is normally distributed (in either source or binary
+form) with the major components (compiler, kernel, and so on) of the
+operating system on which the executable runs, unless that component
+itself accompanies the executable.
+
+If distribution of executable or object code is made by offering
+access to copy from a designated place, then offering equivalent
+access to copy the source code from the same place counts as
+distribution of the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+ 4. You may not copy, modify, sublicense, or distribute the Program
+except as expressly provided under this License. Any attempt
+otherwise to copy, modify, sublicense or distribute the Program is
+void, and will automatically terminate your rights under this License.
+However, parties who have received copies, or rights, from you under
+this License will not have their licenses terminated so long as such
+parties remain in full compliance.
+
+ 5. You are not required to accept this License, since you have not
+signed it. However, nothing else grants you permission to modify or
+distribute the Program or its derivative works. These actions are
+prohibited by law if you do not accept this License. Therefore, by
+modifying or distributing the Program (or any work based on the
+Program), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Program or works based on it.
+
+ 6. Each time you redistribute the Program (or any work based on the
+Program), the recipient automatically receives a license from the
+original licensor to copy, distribute or modify the Program subject to
+these terms and conditions. You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties to
+this License.
+
+ 7. If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Program at all. For example, if a patent
+license would not permit royalty-free redistribution of the Program by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Program.
+
+If any portion of this section is held invalid or unenforceable under
+any particular circumstance, the balance of the section is intended to
+apply and the section as a whole is intended to apply in other
+circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system, which is
+implemented by public license practices. Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+ 8. If the distribution and/or use of the Program is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Program under this License
+may add an explicit geographical distribution limitation excluding
+those countries, so that distribution is permitted only in or among
+countries not thus excluded. In such case, this License incorporates
+the limitation as if written in the body of this License.
+
+ 9. The Free Software Foundation may publish revised and/or new versions
+of the General Public License from time to time. Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Program
+specifies a version number of this License which applies to it and "any
+later version", you have the option of following the terms and conditions
+either of that version or of any later version published by the Free
+Software Foundation. If the Program does not specify a version number of
+this License, you may choose any version ever published by the Free Software
+Foundation.
+
+ 10. If you wish to incorporate parts of the Program into other free
+programs whose distribution conditions are different, write to the author
+to ask for permission. For software which is copyrighted by the Free
+Software Foundation, write to the Free Software Foundation; we sometimes
+make exceptions for this. Our decision will be guided by the two goals
+of preserving the free status of all derivatives of our free software and
+of promoting the sharing and reuse of software generally.
+
+ NO WARRANTY
+
+ 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
+FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
+OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
+OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
+TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
+PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
+REPAIR OR CORRECTION.
+
+ 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
+REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
+INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
+OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
+TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
+YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
+PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGES.
+
+ END OF TERMS AND CONDITIONS
+
+ Appendix: How to Apply These Terms to Your New Programs
+
+ If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+ To do so, attach the following notices to the program. It is safest
+to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+ <one line to give the program's name and a brief idea of what it does.>
+ Copyright (C) 19yy <name of author>
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software
+ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+Also add information on how to contact you by electronic and paper mail.
+
+If the program is interactive, make it output a short notice like this
+when it starts in an interactive mode:
+
+ Gnomovision version 69, Copyright (C) 19yy name of author
+ Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+ This is free software, and you are welcome to redistribute it
+ under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License. Of course, the commands you use may
+be called something other than `show w' and `show c'; they could even be
+mouse-clicks or menu items--whatever suits your program.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a "copyright disclaimer" for the program, if
+necessary. Here is a sample; alter the names:
+
+ Yoyodyne, Inc., hereby disclaims all copyright interest in the program
+ `Gnomovision' (which makes passes at compilers) written by James Hacker.
+
+ <signature of Ty Coon>, 1 April 1989
+ Ty Coon, President of Vice
+
+This General Public License does not permit incorporating your program into
+proprietary programs. If your program is a subroutine library, you may
+consider it more useful to permit linking proprietary applications with the
+library. If this is what you want to do, use the GNU Library General
+Public License instead of this License.
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/MANIFEST b/Master/texmf-dist/doc/latex/cweb-latex/MANIFEST
new file mode 100644
index 00000000000..f6ae6a4dcc3
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/MANIFEST
@@ -0,0 +1,21 @@
+$Id: MANIFEST,v 1.5 1995/08/29 17:34:33 schrod Exp $ -*- Indented-Text -*-
+----------------------------------------------------------------------
+
+The following files are part of the cweb bundle:
+
+ README -- general intro, what's this package for
+ CATALOG -- catalog information about this bundle
+ INSTALL -- installation instructions
+
+ cweb.cls -- class macro file
+ keyvald.sty -- needed by cweb.cls
+ cwebarray.sty -- style option for ruled tables, needs array.sty
+ cweb-user.dvi -- typeset user manual
+ cweb-conf.dvi -- typeset description of internal interface
+
+ History -- revision history of this package
+ MANIFEST -- this file
+ License -- GPL
+ Makefile -- will help you for installation on Unix
+ src/ -- subdirectory with my development area
+ Please read the file README in that directory.
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/README b/Master/texmf-dist/doc/latex/cweb-latex/README
new file mode 100644
index 00000000000..348323fa78c
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/README
@@ -0,0 +1,92 @@
+$Id: README,v 1.6 1995/11/21 00:41:49 schrod Exp $
+------------------------------------------------------------
+
+
+This is the distribution of LaTeX CWEB, a bundle that allows you to
+use LaTeX as the documentation markup of your CWEB program. Your CWEB
+programs may be LaTeX documents now -- with sections, subsections,
+environments, figures, graphics inclusion, etc.
+
+You will want to use this bundle if you
+
+ -- are a CWEB programmer and use LaTeX elsewhere,
+ -- haven't used CWEB up to now, because you wanted structured markup
+ (and you didn't want to develop it yourself), or
+ -- want an hierarchic document structure for CWEB programs.
+
+Included in the distribution is the full source and DVI files for your
+convenience.
+ In addition, the distribution features a well known example: the
+``Counting Words'' program distributed with CWEB. I recoded it to
+LaTeX markup. You'll find both source and processed files in the
+examples/ directory.
+
+Note that the style file is rather large because the comments are not
+stripped yet. This will change in a future release. I'm still working
+on this class, the next release will support inclusion of CWEB program
+parts in other LaTeX documents. (Yes, I know that this was already
+promised for this release. Look in src/ for a preliminary version, in
+cwebparts.sty and cwebx.sty -- the end user documentation is missing
+and it has lots of rough edges, but it works!)
+ If you have more wishes or if you have any other gripes about the
+usage of this class, you should send them *now*, as you can still
+influence the development. When I've finished the realization of the
+inclusion, I'll freeze the code base and will concentrate on fixes for
+reported problems.
+
+Anyhow, feedback is always appreciated, I'd like to know if you that
+style. -- The more people tell me they use it, the more support I'll
+put in future versions.
+
+
+If you want to use this style file, you must use CWEB V3. (Current
+release, as of $Date: 1995/11/21 00:41:49 $: 3.4.) You must also use a
+more-or-less recent LaTeX2e, I cannot support LaTeX 2.09 any longer.
+(`More-or-less recent' means: if I need a feature that was introduced
+in the current LaTeX revision or in the previous one, I'll use it.)
+Most probably you also need a big TeX to use the cweb class, but they
+should be common by now anyhow.
+
+
+MANIFEST features a list of files in this distribution.
+For installation instructions, check the file INSTALL.
+src/TODO explicates my development plans.
+
+
+Reference anonymous ftp site:
+ ftp.th-darmstadt.de [130.83.47.112]
+ directory pub/programming/literate-programming/c.c++
+ file cweb-sty-<version>.tar.gz
+ where <version> is the current version number
+ (a gzipped tar file)
+
+You'll also find a version of this bundle on CTAN in
+macros/latex/contrib/supported/cweb/.
+
+
+LEGALESE
+
+This is freely distributable software; you can redistribute it
+and/or modify it under the terms of the GNU General Public License as
+published by the Free Software Foundation; either version 2 of the
+License, or (at your option) any later version.
+
+This software is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License in
+the file License along with this package; if not, write to the Free
+Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+
+Share and enjoy,
+
+ Joachim
+
+--
+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
+Joachim Schrod Email: schrod@iti.informatik.th-darmstadt.de
+Computer Science Department
+Technical University of Darmstadt, Germany
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/Index b/Master/texmf-dist/doc/latex/cweb-latex/contrib/Index
new file mode 100644
index 00000000000..bf7a308a667
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/Index
@@ -0,0 +1,16 @@
+% $ITI: Index,v 1.2 1995/11/21 00:42:08 schrod Exp $ -*- Indented-Text -*-
+%----------------------------------------------------------------------
+
+The following files are contributed to LaTeX CWEB:
+
+ cwbl-french - by Andreas Scherer <scherer@physik.rwth-aachen.de>
+ Support for French CWEB documents.
+ cwbl-italian - by Andreas Scherer <scherer@physik.rwth-aachen.de>
+ Support for Italian CWEB documents.
+ cwbl-deutsch - by Andreas Scherer <scherer@physik.rwth-aachen.de>
+ Another support for German CWEB documents,
+ different texts than the one in src/.
+
+ cwebzw.sty - by Zden\v{e}k Wagner <wagner@csearn.bitnet>
+ Created for cweb.sty 1.8;
+ most probably doesn't run any more, moved to wagner/.
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-deutsch.sty b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-deutsch.sty
new file mode 100644
index 00000000000..9192ad883b2
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-deutsch.sty
@@ -0,0 +1,83 @@
+% $Id: dcweb.sty,v 1.3 1995/09/11 18:03:44 scherer Exp $
+%
+% THIS IS A TEST!
+%
+% I have no idea how to write a style-conformant package, so bear with me.
+% This `package' translates the `LaTeX CWEB' class to German.
+%
+% Andreas Scherer
+% Rolandstra{\ss}e 16
+% 52070 Aachen, Germany
+%
+% 31. August 1995
+%
+% $Log: dcweb.sty,v $
+% Revision 1.3 1995/09/11 18:03:44 scherer
+% Following Joachim Schrod's advice, `chunk' means `Baustein'.
+%
+% Revision 1.2 1995/09/11 18:01:24 scherer
+% Make this package usable with `german.sty'.
+%
+% Revision 1.1 1995/09/11 18:00:41 scherer
+% Initial revision
+%
+\renewcommand{\CwebCRAlso}{\CwebCrossRef{Siehe auch den Baustein}}
+ % See also chunk
+\renewcommand{\CwebCRsAlso}{\CwebCrossRef{Siehe auch die Bausteine}}
+ % See also chunks
+
+\renewcommand{\CwebCRCite}
+ {\CwebCrossRef{Dieser Programmteil wird zitiert im Baustein}}
+ % This code is cited in chunk
+\renewcommand{\CwebCRsCite}
+ {\CwebCrossRef{Dieser Programmteil wird zitiert in den Bausteinen}}
+ % This code is cited in chunks
+
+\renewcommand{\CwebCRUse}
+ {\CwebCrossRef{Dieser Programmteil wird verwendet im Baustein}}
+ % This code is used in chunk
+\renewcommand{\CwebCRsUse}
+ {\CwebCrossRef{Dieser Programmteil wird verwendet in den Bausteinen}}
+ % This code is used in chunks
+
+\renewcommand{\CwebCREt}{ und~}
+ % and~
+\renewcommand{\CwebCRsEt}{ und~}
+ % , and~
+
+\renewcommand{\CwebCRChanged}{%
+ \CwebCrossRef{Die folgenden Bausteine sind vom Change-File
+ ver\"andert worden:}%
+ }
+ % The following chunks were changed by the change file:
+\renewcommand{\CwebChangedListName}{Ver\"anderte Bausteine}
+ % Changed Chunks
+
+\renewcommand{\CwebRefListName}{Liste der Verfeinerungen}
+ % List of Refinements
+
+\renewcommand{\CwebXRCite}{\CwebCrossRef{Zitiert im Baustein}}
+ % Cited in chunk
+\renewcommand{\CwebXRCites}{\CwebCrossRef{Zitiert in den Bausteinen}}
+ % Cited in chunks
+\renewcommand{\CwebXRUse}{\CwebCrossRef{Verwendet im Baustein}}
+ % Used in chunk
+\renewcommand{\CwebXRUses}{\CwebCrossRef{Verwendet in den Bausteinen}}
+ % Used in chunks
+
+
+%% -js [07 Nov 95]:
+%% Discarded code below, this problem has been repaired in the
+%% cwebbase module.
+%%
+%% \makeatletter
+%% \catcode`\_=\CatLetter
+%% \def\CwebString#1{%
+%% \hbox{\tt
+%% \cweb@string_setup
+%% \ifx\originalTeX\undefined\else\originalTeX\fi % for `german.sty'
+%% #1\kern .05em
+%% }%
+%% }
+%% \catcode`\_=\CatUsCode
+%% \makeatother
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-french.sty b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-french.sty
new file mode 100644
index 00000000000..6dd3f6203ed
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-french.sty
@@ -0,0 +1,64 @@
+% $Id: fcweb.sty,v 1.2 1995/10/26 12:28:04 scherer Exp $
+%
+% THIS IS A TEST!
+%
+% I have no idea how to write a style-conformant package, so bear with me.
+% This `package' translates the `LaTeX CWEB' class to French.
+%
+% Andreas Scherer
+% Rolandstra{\ss}e 16
+% 52070 Aachen, Germany
+%
+% 31. August 1995
+%
+% $Log: fcweb.sty,v $
+% Revision 1.2 1995/10/26 12:28:04 scherer
+% Some French corrections pointed out by <bronne@@montefiore.ulg.ac.be.
+% Merci beaucoup !
+%
+% Revision 1.1 1995/09/11 17:57:59 scherer
+% Initial revision
+%
+\renewcommand{\CwebCRAlso}{\CwebCrossRef{Voir aussi le fragment}}
+ % See also chunk
+\renewcommand{\CwebCRsAlso}{\CwebCrossRef{Voir aussi les fragments}}
+ % See also chunks
+
+\renewcommand{\CwebCRCite}
+ {\CwebCrossRef{Code cit\'e dans le fragment}}
+ % This code is cited in chunk
+\renewcommand{\CwebCRsCite}
+ {\CwebCrossRef{Code cit\'e dans les fragments}}
+ % This code is cited in chunks
+
+\renewcommand{\CwebCRUse}
+ {\CwebCrossRef{Code utilis\'e dans le fragment}}
+ % This code is used in chunk
+\renewcommand{\CwebCRsUse}
+ {\CwebCrossRef{Code utilis\'e dans les fragments}}
+ % This code is used in chunks
+
+\renewcommand{\CwebCREt}{ et~}
+ % and~
+\renewcommand{\CwebCRsEt}{ et~}
+ % , and~
+
+\renewcommand{\CwebCRChanged}{%
+ \CwebCrossRef{Les fragments suivants ont \'et\'e modifi\'es dans
+ le fichier de changements\,:}%
+ }
+ % The following chunks were changed by the change file:
+\renewcommand{\CwebChangedListName}{Fragments modifi\'es}
+ % Changed Chunks
+
+\renewcommand{\CwebRefListName}{Liste des fragments}
+ % List of refinements
+
+\renewcommand{\CwebXRCite}{\CwebCrossRef{Cit\'e dans le fragment}}
+ % Cited in chunk
+\renewcommand{\CwebXRCites}{\CwebCrossRef{Cit\'e dans les fragments}}
+ % Cited in chunks
+\renewcommand{\CwebXRUse}{\CwebCrossRef{Utilis\'e dans le fragment}}
+ % Used in chunk
+\renewcommand{\CwebXRUses}{\CwebCrossRef{Utilis\'e dans les fragments}}
+ % Used in chunks
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-italian.sty b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-italian.sty
new file mode 100644
index 00000000000..dc75642bff8
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cwbl-italian.sty
@@ -0,0 +1,63 @@
+% $Id: icweb.sty,v 1.2 1995/09/28 09:20:01 scherer Exp $
+%
+% THIS IS A TEST!
+%
+% I have no idea how to write a style-conformant package, so bear with me.
+% This `package' translates the `LaTeX CWEB' class to Italian.
+%
+% Andreas Scherer
+% Rolandstra{\ss}e 16
+% 52070 Aachen, Germany
+%
+% 31. August 1995
+%
+% $Log: icweb.sty,v $
+% Revision 1.2 1995/09/28 09:20:01 scherer
+% I must apologize for my poor Italian. Giuseppe Ghibò corrected this.
+%
+% Revision 1.1 1995/09/11 17:56:07 scherer
+% Initial revision
+%
+\renewcommand{\CwebCRAlso}{\CwebCrossRef{Vedi anche chunk}}
+ % See also chunk
+\renewcommand{\CwebCRsAlso}{\CwebCrossRef{Vedi anche chunk}}
+ % See also chunks
+
+\renewcommand{\CwebCRCite}
+ {\CwebCrossRef{Questo codice \`e citato nel chunk}}
+ % This code is cited in chunk
+\renewcommand{\CwebCRsCite}
+ {\CwebCrossRef{Questo codice \`e citato nei chunk}}
+ % This code is cited in chunks
+
+\renewcommand{\CwebCRUse}
+ {\CwebCrossRef{Questo codice \`e usato nel chunk}}
+ % This code is used in chunk
+\renewcommand{\CwebCRsUse}
+ {\CwebCrossRef{Questo codice \`e usato nei chunk}}
+ % This code is used in chunks
+
+\renewcommand{\CwebCREt}{ e~}
+ % and~
+\renewcommand{\CwebCRsEt}{ e~}
+ % , and~
+
+\renewcommand{\CwebCRChanged}{%
+ \CwebCrossRef{I seguenti chunk sono stati modificati tramite
+ il change-file:}%
+ }
+ % The following chunks were changed by the change file:
+\renewcommand{\CwebChangedListName}{Chunk modificati}
+ % Changed Chunks
+
+\renewcommand{\CwebRefListName}{Nomi dei chunk:}
+ % List of Refinements
+
+\renewcommand{\CwebXRCite}{\CwebCrossRef{Citato nel chunk}}
+ % Cited in chunk
+\renewcommand{\CwebXRCites}{\CwebCrossRef{Citato nei chunk}}
+ % Cited in chunks
+\renewcommand{\CwebXRUse}{\CwebCrossRef{Usato nel chunk}}
+ % Used in chunk
+\renewcommand{\CwebXRUses}{\CwebCrossRef{Usato nei chunk}}
+ % Used in chunks
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/README.txt b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/README.txt
new file mode 100644
index 00000000000..676de55038a
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/README.txt
@@ -0,0 +1,131 @@
+Dear User:
+
+I present to you an evolution of the excellent CWEB class, originally by
+Joachim Schrod.
+The purpose of the new files is twofold:
+
+1.- To allow a file written using this class to be included in other
+documents, as for instance as an appendix to some report. For this,
+the file nodoc.tex may be of help.
+
+2.- To insert hyperlinks from the places where a chunk of code is
+referenced to the chunk itself, including the indices that are produced at
+the end of cweb document. The result makes it a lot more easy to follow
+the code if it is typeset using PDFLaTeX, or is otherwise processed
+to obtain a PDF or an HTML file, and makes no harm in the paper version
+of the document.
+
+The installation is straightforward. Put the files cweb-hy.cls and
+cwbasehy.tex where TeX can find them. Since cweb.cls was installed
+under tex/latex/literate, may I suggest you put all these files under
+tex/latex/liter-hy, all paths relative to your TEXMF hierarchy.
+
+1 The Class as a Package.
+
+For using the class as a normal package (with \usepackage{cweb-hy}), you
+simply have to copy the file cweb-hy.cls with the name cweb-hy.sty, so
+that it is found by LaTeX.
+
+How this works I can explain as follows. Suppose you have a file called
+whatever.w such as
+
+--------------------------------------
+% Very simple CWEB
+\documentclass[]{cweb-hy}
+\usepackage[T1]{fontenc}
+\usepackage{ae}
+\begin{document}
+\title{Whatever}
+\author{Me}
+
+@* CWEB program.
+This is your program.
+@c
+@<Header files@>@/
+@<Main program@>
+
+@ If we want to use the system resources, we must insert the definition
+of the standard functions.
+
+@<Header...@>=
+#include <stdio.h>
+
+@ The code makes whatever.
+
+@<Main...@>=
+int
+main(void)
+{
+printf("Whatever\n");
+exit(0);
+}
+@ \end{document}
+--------------------------------------
+And suppose further that you want to make a report including your
+beautifully typeset program (with CWEB and TeX).
+Well, with the CWEB class as it stands, you can't. However, with the very
+simple changes I have implemented, you could say
+
+\documentclass{report}
+\title{All my life's programs}
+\author{Me}
+\begin{document}
+\part{Rationale}
+Blah, Blah...
+
+\part{Programs}
+\section{Makes Whatever}
+\input{whatever}
+\end{document}
+
+and the file whatever.tex will get just in with all your very skillful
+code and clear documentation, everything typeset by TeX. The package
+accepts almost all the options the class does, but I decided not to
+include some auxiliary files with the cross-references in the former
+case. For two reasons. I am unsure as to how to do it and it is a
+bit overwhelming if you are in a large document to have these sections.
+I guess it can be done, however, and I guess you may like it. Drop me a
+mail in this case.
+
+You may now say to yourself, how is this going to work, if in my
+file whatever.tex there is a \documentclass, etc. all forbidden in
+a document's body? You have two choices. One is to delete all the
+offending commands and \input the diminished file. Another is to use
+nodoc.tex which I also provide. This nodoc.tex inhibits all offending
+commands for a while ---long enough for LaTeX not noticing it---,
+and restores the definitions just in time for you to have your dose of
+error-messages if you dare putting a preamble macro in the wrong place.
+For using it, simply add an \input{nodoc} at the very beginning of your
+file, and you will then be allowed both to typeset your file with LaTeX
+directly and to include it in other files, at the expense of a very
+limited memory use.
+
+2. Making screen versions of your programs
+
+Programs are cumbersome to follow. With CWEB your programs get a
+better structure, are better typeset and are more easy to understand.
+They remain, however, hard to follow because you have to move paper to
+and fro, looking for the referenced code. Using this new package, and
+if the documentation file (the one obtained with cweave) is processed to
+provide a screen version that can use hyperlinks (such as with PDFLaTeX,
+xdvi, dvipdfm) with the help of the hyperref package, the screen version
+you get has all the program chunks linked and referenced, so that when
+one is used or cited, there is a link you can click on to get to the
+adequate place. If you do not use the hyperref package no harm is done
+and the changes take no effect.
+
+Enjoy yourself with this demonstration of my clumsy TeX programming,
+and let me know if you experience any problems. Please write to Enrique
+Meléndez <UEI0560130010@eurociber.es> in case of failure, being as
+precise as you can with the error, possibly with a short example
+that demonstrates the failure, and providing a workaround (in the best of
+the worlds :-))
+
+Regards,
+Enrique Meléndez
+
+By the way, no responsibility is held. I did this for fun (among other
+things) and it worked for me. it may not work for you or, if it works,
+it may not perform exactly as you intended. If you write to me I can
+try my best to solve your problem (and I probably will try), but I make no
+formal promises. See the GNU public license for details.
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cwbasehy.tex b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cwbasehy.tex
new file mode 100644
index 00000000000..1f7941323c5
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cwbasehy.tex
@@ -0,0 +1,2239 @@
+% $StyleId: cwebbase.doc,v 1.4 1995/11/07 17:55:19 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+
+%
+% LaTeX macros cwebbase
+% support for incorporation of CWEAVE markup in LaTeX documents
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+
+
+
+%%%%
+%%%%
+%%%% These TeX macros were documented with the documentation system
+%%%% MAKEPROG and automatically converted to the current form.
+%%%% If you have MAKEPROG available you may transform it back to
+%%%% the original input: Remove every occurence of three percents
+%%%% and one optional blank from the beginning of a line and remove
+%%%% every line which starts with four percents. The following lex
+%%%% program will do this:
+%%%%
+%%%% %%
+%%%%
+%%%% ^%%%\ ? ;
+%%%% ^%%%%.*\n ;
+%%%%
+%%%% If you just want to print the documentation you may fetch
+%%%% the archive print-makeprog.tar.Z from ftp.th-darmstadt.de (directory
+%%%% pub/tex/latex). It contains *all* used styles -- but beware, they
+%%%% may not be in a documented form...
+%%%%
+%%%%
+%%% \documentclass{progltx}
+
+%%% \usepackage{cweb-doc} % document-specific markup
+%%% \usepackage{fullpage}
+
+
+%%% \RCS $StyleRevision: 1.4 $
+%%% \RCS $StyleDate: 1995/11/07 17:55:19 $
+
+
+%%% \begin{document}
+
+
+%%% \title{Typesetting \cweave{} Output}
+%%% \author{% % LaTeX does not discard unnecessary glue...
+%%% Joachim Schrod%
+%%% \thanks{%
+%%% \protect\raggedright
+%%% TU~Darmstadt, Computer Science Department, WG Systems Programming,
+%%% Alexanderstr.~10, D-64283~Darmstadt, Germany.
+%%% Email: \texttt{schrod@iti.informatik.th-darmstadt.de}.%
+%%% }%
+%%% }
+%%% \date{%
+%%% Revision \RCSStyleRevision\\
+%%% (as of \RCSStyleDate)%
+%%% }
+
+%%% \maketitle
+
+
+%%% % doesn't work with progltx yet
+%%% %\tableofcontents
+
+
+
+
+%%% \chap Introduction.
+
+%%% This code shall interpret and render the tag set output by \cweave{}.
+%%% That's realized in this module as it's needed in multiple places --
+%%% when we want to use \LaTeX{} markup in \cweb{}, when we want to
+%%% incorporate \cweb{} pieces in \LaTeX{} documents (as examples, or in
+%%% textbooks), and when we want to incorporate whole \cweb{} documents in
+%%% a \LaTeX{} document. (That's by far the most difficult task.)
+
+%%% As \cweb{} operates in modes, for obvious reasons our implementation
+%%% is based on a finite state automaton (FSA). The whole tag set is
+%%% separated into subsets that are valid in certain states.
+
+%%% The complete set of tags does not obey any name conventions, some tags
+%%% are even in conflict with standard \LaTeX{} tags (e.g.\ |\\|). Most
+%%% tags are used to render code pieces. With the introduction of states
+%%% we can restrict the visibilities of these tags to those states where
+%%% we typeset program parts. The needed functionality will be available
+%%% in cseqs of our namespace, we'll map that to \cweave{}'s names in a
+%%% group.
+
+%%% Then we have to implement the rendering of chunks and sections, that's
+%%% (more or less) a straight-forward task. Actually, we only set up the
+%%% basic code pieces, the integration of section divisions in the actual
+%%% \LaTeX{} document is a task of higher-level code; that integration
+%%% depends heavily on the type of document. That also implies that we
+%%% won't implement front or back matter material (document markup, table
+%%% of contents, index, list of refinements, etc.)\ in this module.
+
+
+%%% \sect Let's identify this module.
+
+%%% \noindent The code below is explained in the implementation
+%%% documentation of the \pkg{rcs} package.
+
+%%% \beginprog
+\begingroup
+ \def\RCSFile#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesFile{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSFile{cwebbase.tex}{CWEAVE tags for LaTeX markup}
+ $StyleDate: 1995/11/07 17:55:19 $: 9999/00/00 \endRCS
+ $StyleRevision: 1.4 $: 0.0 \endRCS
+\endgroup
+%%% \endprog
+
+
+%%% \sect This module is part of the supported \textsl{cweb-sty} bundle.
+%%% Send bug reports, comments and repairs.
+
+%%% It assumes that the canonical setup of my modules has happened and can
+%%% be used herein. (In particular, the underscore is a letter.)
+
+
+%%% \sect This module reserves the namespace |cwbb|. It assumes that
+%%% |\cweb_warning| is bound either to |\PackageWarningNoLine| or
+%%% |\ClassWarningNoLine|, depending on the environment.
+
+%%% \beginprog
+\ifx \cwbb@loaded\undefined
+ \def\cwbb@loaded{$StyleRevision: 1.4 $}
+\else
+ \cweb_warning{cwebbase}%
+ {Some other package already uses namespace `cwbb'}
+\fi
+%%% \endprog
+
+
+%%% \sect This is freely distributable software; you can redistribute it
+%%% and/or modify it under the terms of the GNU General Public License as
+%%% published by the Free Software Foundation; either version~2 of the
+%%% License, or (at your option) any later version.
+
+%%% This software is distributed in the hope that it will be useful, but
+%%% \textbf{without any warranty}; without even the implied warranty of
+%%% \textbf{merchantability} or \textbf{fitness for a particular purpose}.
+%%% See the GNU General Public License for more details.
+
+%%% You should have received a copy of the GNU General Public License in
+%%% the file |License| along with this package; if not, write to the Free
+%%% Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+\ifx\hypertarget\undefined
+\def\hypertarget#1#2{#2}
+\def\hyperlink#1#2{#2}
+\fi
+
+
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The interface between CWEAVE and TeX
+%%% %
+
+%%% \input{cweave-spec}
+
+%%% %
+%%% % ------------------------------------------------------------
+
+
+
+
+%%% \chap Interface of \pkg{cwebbase} module.
+
+%%% \pkg{cwebbase} implements most of \cweave{} tags. It sets up tons of
+%%% redefinable cseqs, all of them start with |Cweb|.
+%%% %
+%%% \begin{fixme}
+%%% How do I document them? We need a configuration document anyhow. Do
+%%% I present here just a list of those cseqs realized in this module
+%%% and refer to the config docs? Need to wait until I have it, then
+%%% we'll see how the stuff here will be added.
+%%% \end{fixme}
+
+
+%%% \sect This module supports both hierarchic and flat section
+%%% structuring. You have to bind |\cweb@structure| to 0 (hierarchic) or 1
+%%% (flat) before importing it, to show which structure is wanted.
+
+
+%%% \sect This module supports suppression of different document parts.
+%%% You have to define flags |@cweb@suppress@|\<foo>|@| with
+%%% $\mbox{\<foo>} \in \{ |changehints|, |unchanged|, |format| \}$ to
+%%% denote if suppression shall happen.
+
+
+%%% \sect The counter |secnumdepth| must be~11 if chunks shall be
+%%% numbered. That counter doesn't need to be set when this module is
+%%% imported, it can be set locally during typesetting \cweb{} code.
+
+%%% The definitions for flat structure assume that |secnumdepth| is~11
+%%% during typesetting.
+
+
+%%% \sect |\part|, |\chapter| (if available), and |\section| are used for
+%%% \cweb{} section divisions (those with the lowest ranks). These
+%%% commands will change the respective counters globally and will
+%%% add toc lines and marks. You might want to prevent that by redefining
+%%% |\addcontentsline| and |\mark|.
+
+%%% Table of contents macros are not defined for the section divisions
+%%% introduced by this module. You must either prevent that they are
+%%% written to the |AUX| file at all; or you must define them.
+%%% %
+%%% \begin{fixme}
+%%% Add explanation of introduced section divisions.
+%%% \end{fixme}
+
+%%% For flat structure, all section divisions are of type `|chunk|', like
+%%% chunks. The section names start with `|\cwbbRank{|\<rank>|}|'. By
+%%% default, |\cwbbRank| takes one argument and expands to the empty list
+%%% -- this can be redefined for the table of contents.
+
+
+%%% \sect The document end, as explained in the \cweave{} interface specs
+%%% (chunk~\ref{spec:sec:docend}), is not implemented. That must be done
+%%% by the user of this module -- if the back matter is to be processed at
+%%% all.
+
+
+
+
+%%% \chap Processing states.
+
+%%% We have to typeset five different categories of material:
+%%% Documentation, program pieces---embedded within the documentation and
+%%% as large chunks, \TeX{} material within program pieces (i.e., comments
+%%% and refinement names), and cross reference information. Since we need
+%%% a complete other environment for the program pieces than for the rest
+%%% we design ``states'' where we switch to appropriately.
+%%% %
+%%% \begin{enumerate}
+
+%%% \item A chunk starts in the \textsl{documentation state}.
+
+%%% \item |\B| switches to \textsl{program state}. This can happen in
+%%% documentation and program state.
+
+%%% \item While we process the argument of |\PB| we're in \textsl{restricted
+%%% program state}; |\PB| may appear in documentation and in \TeX{}
+%%% state. Since \TeX{} state can be switched on within restricted
+%%% program state, |\PB| can appear within the argument of itself.
+
+%%% \item In the arguments of |\C|, |\SHC|, and |\X| we switch to
+%%% \textsl{\TeX{} state}. All these cseqs appear only in (restricted)
+%%% program state, their official names are actually different. I.e., only
+%%% in (restricted) program state these cseq are bound to the meaning
+%%% described here.
+
+%%% \item Cross reference information are attached to most chunks with
+%%% refinements. This information is processed in \textsl{CR state}.
+%%% After CR state material
+%%% comes always the next chunk or the document end, i.e., material in
+%%% documentation state.
+
+%%% \end{enumerate}
+%%% %
+%%% This FSA is illustrated by the diagram in figure~\ref{fig:fsa}.
+
+%%% \begin{figure}
+
+%%% \begin{center}
+%%% \chardef\\=`\\
+%%% \input{cweb-fsa.latex}
+%%% \end{center}
+
+%%% \caption{The processing state's FSA. The automaton starts and ends in
+%%% the ``documentation'' state.}
+%%% \label{fig:fsa}
+
+%%% \end{figure}
+
+%%% |\cwbb@documentation| will switch to documentation state,
+%%% |\cwbb@program| to program state, |\cwbb@Rprogram| to restricted
+%%% program state, |\cwbb@tex| to \TeX{} state, and |\cwbb@CR| to CR state.
+%%% If we're already in a state, the switch to this state shall be a
+%%% permissible null operation.
+
+
+%%% \sect The basic difference between these states can be named with
+%%% two parameters: (1)~The cseq bindings in effect and (2)~the layout
+%%% parameters used for paragraph makeup.
+
+%%% In (restricted) program state and in CR state the text is output
+%%% under the control of \cweave{}, and tagged by \cweave{}. The used
+%%% tags are from a global namespace and should only be in effect during
+%%% these states. We call this tag set the \textsl{\cweave{} bindings}. In the
+%%% other two states the tags are largely defined by the user, the tag
+%%% set is called the \textsl{user bindings}. The switch to another binding
+%%% is always done locally, i.e., if we switch from documentation to
+%%% restricted program state within a group we don't have to bother about
+%%% the restauration of the user binding; it will be done automatically
+%%% by \TeX{} at the end of the group. Nevertheless we must be able to
+%%% switch from the \cweave{} bindings back to the user bindings which
+%%% were in effect when we activated the \cweave{} bindings. This is
+%%% needed for the \TeX{} state which is always activated within
+%%% (restricted) program state.
+
+%%% The parameters for \textsl{program layout} are really special ones
+%%% since they need to support the indentation which shows the program
+%%% structure. These parameters are used in program and in \TeX{} state.
+%%% The \textsl{document layout} parameters established by the user are
+%%% used in the other three states.
+
+%%% The following table shall summarize this. $C$~denotes \cweave{}
+%%% bindings, $U$~user bindings, $P$~program layout, and $D$~document
+%%% layout. If an entry is empty, its value is not changed on entry in
+%%% this state.
+%%% %
+%%% \begin{center}
+%%% \DeleteShortVerb\|
+%%% \begin{tabular}{r|c|c}
+%%% \multicolumn1{c|}{\textsc{State}} & \textsc{Binding} & \textsc{Layout} \\[1pt]
+%%% \hline
+%%% documentation & $U$ & $D$ \\
+%%% program & $C$ & $P$ \\
+%%% restricted program & $C$ & \\
+%%% \TeX{} & $U$ & \\
+%%% CR & $C$ & $D$ \\
+%%% \end{tabular}
+%%% \MakeShortVerb\|
+%%% \end{center}
+
+
+%%% \sect Since the user bindings and the document layout is defined
+%%% initially, we don't have to do anything if it's requested. Only if we
+%%% change it, i.e., within |\cwbb@cweave_bindings| and
+%%% |\cwbb@program_layout|, we redefine |\cwbb@user_bindings| and
+%%% |\cwbb@doc_layout|. If they are eventually executed, they shall
+%%% rebind themselves back to |\relax|. This way we can switch to
+%%% documentation state as often as we want.
+
+%%% \beginprog
+\let\cwbb@user_bindings=\relax
+\let\cwbb@doc_layout=\relax
+
+\def\cwbb@documentation{%
+ \cwbb@user_bindings
+ \cwbb@doc_layout
+ }
+\def\cwbb@Rprogram{\cwbb@cweave_bindings}
+
+\def\cwbb@program{%
+ \cwbb@cweave_bindings
+ \cwbb@program_layout
+ }
+\def\cwbb@tex{\cwbb@user_bindings}
+
+\def\cwbb@CR{%
+ \cwbb@cweave_bindings
+ \cwbb@doc_layout
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Saving and restoring control sequences.
+
+%%% We have a lot of cseqs which are defined within the namespace of this
+%%% module and which will be used with other names. This usage is in a
+%%% controlled environment, namely neither in documentation nor in \TeX{}
+%%% state. (I.e., the text processed is tagged by \cweave{}, not by humans;
+%%% therefore we have a precise specification of the cseqs we have to
+%%% accept.) We cannot work with groups where a cseq is just redefined
+%%% and \TeX{} takes care for establishing the old binding again; when we
+%%% switch from program state to \TeX{} state all the bindings which were
+%%% in effect before the program state get active, i.e., in the
+%%% documentation state, must be in effect again. We cannot simply
+%%% consider the \TeX{} state as something parallel to program state, it
+%%% must be a hierarchical relationship: In the program state values are
+%%% set up which must be available after switching back from \TeX{} to
+%%% program state.
+
+%%% We save the current binding of a cseq in another cseq, but only if
+%%% there exists a binding currently. This is done to save valuable
+%%% \TeX{} main memory. (Why don't all those people who talk about
+%%% extending \TeX{} not just implement dynamic memory management for the
+%%% existing \TeX{}? It would still be \TeX{} and would \emph{really} be
+%%% a worthy activity. But not enough fun, I assume\,\dots) Actually, one
+%%% can assume that nearly no cseq bindings must be saved at all---the
+%%% used names are strange enough. The bindings of |\foo|, i.e., of the
+%%% cseq with the name \fbox{\ttfamily foo}, is saved as the cseq with the
+%%% name \fbox{\ttfamily cweb\char `\\ foo}, i.e., as
+%%% |\csname cweb\string\foo \endcsname|.
+
+
+%%% \sect The save process is not done statically, but by the macro
+%%% |\cwbb@rebind| which interprets a list of tuples $({\it old\_name},
+%%% {\it new\_name})$, terminated by the tuple $(|\stop|, |\stop|)$.
+%%% Eventually it constructs two new lists, |\cwbb@to_restore| with the
+%%% cseqs which had a binding, and |\cwbb@undefined| with the names which
+%%% didn't have one.
+
+%%% The saving is actually done by |\cwbb@save_binding|, |\cwbb@rebind|
+%%% is responsible for the initialization, |\cwbb@do_rebind| for the
+%%% effective rebinding and the tail recursion on the list.
+
+%%% \begin{fixme}
+%%% We could pull the |\next| assignment in the |\else| branch out of
+%%% the loop to get a better performance. Should measure if this is of
+%%% interest.
+%%% \end{fixme}
+
+%%% \beginprog
+\newtoks\cwbb@undefined
+\newtoks\cwbb@to_restore
+
+\def\cwbb@rebind{%
+ \cwbb@undefined{}%
+ \cwbb@to_restore{}%
+ \cwbb@do_rebind
+ }
+\def\cwbb@do_rebind#1#2{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \cwbb@save_binding #2%
+ \let #2=#1%
+ \let\next\cwbb@do_rebind
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+%%% \sect If the cseq to be saved is undefined, it may just be added to
+%%% the ``undefined list.'' Otherwise its binding is saved and it's added
+%%% to the ``to-be-restored list.''
+
+%%% \TeX{}nical note: The cseq-name for the saved binding must be created
+%%% before the |\let| is executed.
+
+%%% \beginprog
+\def\cwbb@save_binding#1{%
+ \ifx #1\undefined
+ \cwbb@undefined \expandafter{\the\cwbb@undefined #1}%
+ \else
+ \expandafter\let \csname cweb\string#1\endcsname =#1%
+ \cwbb@to_restore \expandafter{\the\cwbb@to_restore #1}%
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect The restoration of rebound cseqs is a two-tied activity: All
+%%% previously undefined cseqs must be made undefined again, and all saved
+%%% cseqs must be restored. Actually, we don't need to reset the two token
+%%% lists, but we do it to save space.
+
+%%% Both |\cwbb@undefine| and |\cwbb@restore_binding| iterate over a list
+%%% of cseqs terminated by |\stop|.
+
+%%% \begin{fixme}
+%%% And here the |\next| assignment could be prepended to the loop as
+%%% well.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@restore_bindings{%
+ \expandafter\cwbb@undefine \the\cwbb@undefined \stop
+ \cwbb@undefined{}%
+ \expandafter\cwbb@restore_binding \the\cwbb@to_restore \stop
+ \cwbb@to_restore{}%
+ }
+
+\def\cwbb@undefine#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \let#1\undefined
+ \let\next\cwbb@undefine
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+%%% \sect \TeX{}nical note: As in |\cwbb@save_binding|, the cseq-name for
+%%% the saved binding must be created before the |\let|s are executed.
+
+%%% \begin{fixme}
+%%% Another |\next| assignment.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@restore_binding#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \expandafter\let \expandafter#1\csname cweb\string#1\endcsname
+ \expandafter\let \csname cweb\string#1\endcsname \undefined
+ \let\next\cwbb@restore_binding
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Chunks.
+
+%%% All chunks are numbered subsequently. The chunk number is supplied as
+%%% part of the markup. In addition, the `changed' tag |\*| may be
+%%% appended to the chunk number.
+
+%%% By default, chunks have approximately two picas vertical space in
+%%% front.
+
+%%% If we use an hierarchic structure, we output the chunk number at the
+%%% refinement. (It's only needed to identify the refinement for cross
+%%% references.) But chunks are important structural units and thus should
+%%% be marked by typographic means that's more visible than the space
+%%% mentioned above, we add a `chunk start marker'.
+
+%%% For a flat structure, we output the chunk number as the chunk start
+%%% marker, that is the traditional \WEB{} layout.
+
+%%% In any case, the documentation part shall be run in after the chunk
+%%% start marker.
+
+%%% \beginprog
+\newskip\CwebChunkPreSkip
+ \CwebChunkPreSkip=2pc plus 1pc minus 6pt
+%%% \endprog
+
+
+%%% \sect In \LaTeX{}, structural divisions are best realized by
+%%% |\@startsection|, we'll use it also for chunk starts. The chunk start
+%%% marker is the section heading. For hierarchic structure, we use a
+%%% traditional paragraph sign, used since medieval times to tag new text
+%%% units. Our first test showed that the paragraph sign has typically an
+%%% underlength, that doesn't look good at the start of a chunk. We raise
+%%% the sign by its depth.
+
+%%% For flat structure, we don't have text as the section heading,
+%%% but we have to cope for a deficiency of \LaTeX{}: It always leaves a
+%%% quad after the section number (by the way, this is not mentioned in
+%%% the documentation---you have to look in the macros to figure this
+%%% out). So we use a backskip of $\rm -1\,em$ as the chunk start marker
+%%% then.
+
+%%% Changed chunks are marked by a changeflag, by default a star lapping
+%%% to the right. We need the information if that chunk is changed to
+%%% decide if we shall output the changeflag. We demand from the chunk
+%%% macros that they record that info in the flag |@cwbb_changed_chunk@|.
+
+%%% Each structural division must be categorized, it belongs to a
+%%% \textsl{type}. This type determines also the name of the counter, the
+%%% mark handling for this category, and how this section tag is
+%%% represented in the table of contents. The type of chunks is `|chunk|'
+%%% (trivial, isn't it?).
+
+%%% Furthermore, a structural division in \LaTeX{} has an associated
+%%% \textsl{level}. That level is used to determine if that division get
+%%% numbered in text or table of contents. For sections, we'll compute the
+%%% level from the rank, it will not be larger than the rank.%
+%%% \footnote{Check the next section for a more precise definition of
+%%% `rank' and `level'.}
+%%% Since the highest section rank is~10, we may use 11 as chunk level.
+
+%%% For flat structure, we output the number for all section headings,
+%%% including chunks. For hierarchic structure we use the default maximum
+%%% level as supplied by our document class (e.g., 3~for \cls{article},
+%%% and 2~for \cls{report} or \cls{book}).
+
+%%% As a last point, we have to care already for sections in flat
+%%% structure. They will use the |chunk| category as well, we have to
+%%% distinguish them in the table of contents. For that, we'll record the
+%%% level in the text, it will get written to the table of contents.
+
+%%% \beginprog
+\newif\if@cwbb_changed_chunk@
+\let\CwebChangeFlag=*
+\if@cweb@suppress@changehints@
+ \let\cwbb@change_flag=\relax
+\else
+ \let\cwbb@change_flag=\CwebChangeFlag
+\fi
+\def\cwbb@lap_change_flag{\rlap{\cwbb@change_flag}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\CwebChunkHeadingStyle\relax
+ \newbox\cweb@chunk_marker
+ \setbox\cweb@chunk_marker=\hbox{\P} % tradition from medivial times
+ \setbox\cweb@chunk_marker=\hbox{%
+ \raise\dp\cweb@chunk_marker \box\cweb@chunk_marker
+ }
+ \def\CwebChunkHeadingMarker{\copy\cweb@chunk_marker}
+ \def\CwebChunkHeading#1{%
+ \CwebChunkHeadingMarker
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+ \or
+ %% flat
+ \def\cwbbRank#1{}
+ \def\CwebChunkHeadingStyle{\bfseries}
+ \def\CwebChunkHeading#1{%
+ \protect\cwbbRank{#1}%
+ \kern -1em
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect Chunks are tagged with |\M| and get passed the chunk number as
+%%% argument. Before we evaluate |\@startsection|, we'll start the chunk.
+%%% That's common code to be used by sections as well, as they also
+%%% implicitely start a chunk.
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es}) introduced some code to insert
+%%% hyperlinks.
+
+%%% \beginprog
+\def\M#1{%
+ \cwbb@start_chunk{#1}%
+ \@startsection
+ {chunk}% % category, == counter etc.
+ {11}% % level
+ {\z@}% % no indentation for heading
+ {\CwebChunkPreSkip}% % skip above heading
+ {\m@ne em}% % run-in heading, 1em distance to text
+ {\normalfont\CwebChunkHeadingStyle}% % layout
+ {\CwebChunkHeading{11}}% % text, level might go to toc file
+ \hypertarget{_hypersect_#1}{}%
+ }
+%%% \endprog
+
+
+%%% \sect But let's first declare how a chunk number is represented. The
+%%% principle rendering is controlled by |\thechunk|, namely, we get
+%%% arabic numbers. This rendering is used both for creation of the
+%%% section header (if in flat mode) and for cross references. I.e.,
+%%% |\ref| expands into it.
+
+%%% But for flat structure we need different expansions: In the section
+%%% header a period shall be appended, a cross reference shall appear
+%%% without such a period. We use two different macros: |\thechunk|
+%%% creates the number with the period, |\cwbb@refchunk| only the number.
+%%% The latter is used for a redefinition of |\@currentlabel|. Of course,
+%%% for hierarchic structure, both are the same.
+
+%%% \beginprog
+\newcounter{chunk}
+
+\def\cwbb@refchunk{\arabic{chunk}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\thechunk=\cwbb@refchunk
+ \or
+ %% flat
+ \def\thechunk{\arabic{chunk}.}
+\fi
+%%% \endprog
+
+
+%%% \sect Each chunk is started by |\cwbb@start_chunk|, which does the
+%%% common actions. First of all, we switch to documentation state.
+
+%%% Then we store the chunk number in the |chunk| counter. That implies
+%%% that the chunk start must not step this counter! We'll take care for
+%%% that later by redefining |\refstepcounter|. The actual number is
+%%% passed as the argument, including an optional |\*| used to denote
+%%% changed chunks. We'll redefine that cseq to update the flag
+%%% |@cwbb_changed_chunk@|.
+
+%%% Each chunk must be typeset with an open |\if| since it is finished by
+%%% |\fi|. If we shall suppress the output of unchanged chunks, we have
+%%% to use |\iffalse| if not |@cwbb_changed_chunk@|, otherwise |\iftrue|.
+%%% The flag |@cwbb_print_chunk@| is set up to provide an appropriate
+%%% test.
+
+%%% When we typeset a chunk, we evaluate |\cwbb@init_print_chunk|. For
+%%% hierarchic structure, that cseq will be used to initialize the
+%%% typesetting of the chunk number at program parts.
+
+%%% \beginprog
+\if@cweb@suppress@unchanged@
+ \def\if@cwbb_print_chunk@{\if@cwbb_changed_chunk@}
+ \@gobble\fi % close \if in case it's skipped
+\else
+ \let\if@cwbb_print_chunk@=\iftrue
+ \@gobble\fi % same here
+\fi
+
+\def\cwbb@start_chunk#1{%
+ \cwbb@documentation
+% \hypertarget{_hypersect_#1}{}%
+ %%\hypertarget{_hyperchunk_#1}{}%
+ \begingroup
+ \def\*{\global\@cwbb_changed_chunk@true}%
+ \global\@cwbb_changed_chunk@false
+ \global \c@chunk #1\relax % might expand \*
+ \endgroup
+ \if@cwbb_print_chunk@
+ \cwbb@init_print_chunk
+ }
+%%% \endprog
+
+
+%%% \sect |\@startsection| will step the section counter (i.e., |chunk|).
+%%% Since we just assigned it in |\cwbb@start_chunk|, that would leave us
+%%% with a counter that's one too high. And our output in cross references
+%%% may be wrong, as a period is appended in flat structure.
+
+%%% That damage all comes from one macro, |\refstepcounter|; let's
+%%% advice it to do the Right Thing for |chunk| counters.
+
+%%% \beginprog
+\let\cwbb@refstepcounter=\refstepcounter
+
+\def\cwbb@string@chunk{chunk}
+\def\refstepcounter#1{%
+ \def\@tempa{#1}%
+ \ifx \@tempa\cwbb@string@chunk
+ \protected@edef\@currentlabel{\cwbb@refchunk}%
+ \else
+ \cwbb@refstepcounter{#1}%
+ \fi
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Sections.
+
+%%% Important sections are started on a new page, normal sections have the
+%%% same space as chunks in front. The very first section does not
+%%% automatically start on a new page since a title may be in front of it.
+
+%%% Each section division has a \textsl{rank}, that denotes the section
+%%% depth as seen by \cweave{}.
+
+%%% For this depth we compute a \textsl{level}, used to denote the section
+%%% depth for \LaTeX{}. This section depth is used to decide where
+%%% numbering shall stop, or if that entry shall be presented in the table
+%%% of contents, etc.
+
+%%% These two numbers are not the same, due to the support of document
+%%% classes with chapters.
+
+%%% We consider a section as important if it's rank is smaller than~3,
+%%% i.e., then we'll start a new page. This limit is controlled by the
+%%% number |\CwebRankNoEject|. The rank is used as the section level
+%%% and is therefore available to configure depth of typeset numbering and
+%%% appearance in table of contents.
+
+%%% \beginprog
+\def\CwebRankNoEject{3}
+%%% \endprog
+
+
+%%% \sect Sections are started with the macro |\N|. It's first parameter
+%%% is the section's rank. The second parameter is the chunk number, the
+%%% third the title of the section. This last parameter must be terminated
+%%% by a dot.
+
+%%% Each section implicitly starts a chunk. As explained above, major
+%%% sections may eject the page, and we add our actual section heading
+%%% with |\CwebSection|.
+
+%%% We also tell the user that we have reached the next section by telling
+%%% him about the current chunk number. The message is output after the
+%%% section header is set, it shall be on the correct page. Since the
+%%% token list to be output is expanded at shipout time, we must take care
+%%% for the immediate expansion ourselves.
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es}) introduced some code to insert
+%%% hyperlinks.
+%%% \beginprog
+\def\N#1#2#3.{%
+ \cwbb@start_chunk{#2}%
+ \CwebEjectSection{#1}%
+ \hypertarget{_hypersect_#2}{}%
+ \CwebSection{#1}{#3}%
+ \begingroup
+ \def\*{}%
+ \edef\tmp{*#2}%
+ \message \expandafter{\tmp}%
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect At the start of a section we usually eject the page if our rank
+%%% is not too low. But if the user had a title above the first section,
+%%% he won't want this page break. So we try to guess it. If we guess
+%%% wrong, the user may either add |\newpage| himself or he may redefine
+%%% |\CwebEjectSection| in a package.
+
+%%% \beginprog
+\def\CwebEjectSection#1{% % #1 == rank
+ \gdef\CwebEjectSection##1{%
+ \relax
+ \ifnum ##1<\CwebRankNoEject\relax
+ \newpage
+ \fi
+ }%
+ }
+
+\if@titlepage
+ \CwebEjectSection{}
+\fi
+%%% \endprog
+
+
+%%% \sect Now comes a heary piece. We have 11~section ranks, and need 11
+%%% different section categories, section counters, etc. For flat
+%%% structure, they shall all do the same; for hierarchic structure, they
+%%% shall be really different.
+
+%%% \noindent Let's have a look at the hierarchic structure first, it's
+%%% more complex.
+%%% %
+%%% \begin{itemize}
+
+%%% \item We associate section rank~0 (that's `|@**|') with document
+%%% parts, i.e., map it to |\part|. Many smaller programs won't use it.
+
+%%% \item If we are used in a class that has a bound cseq |\chapter|, we
+%%% associate rank~1 (`|@*0|') with it and rank~2 (`|@*1|') with
+%%% |\section|. Otherwise we associate rank~1 with |\section|. These
+%%% tags (|\chapter| and/or |\section|) must be defined because many
+%%% macros that want to create section divisions will use them.
+
+%%% \item All other section ranks are placed below |\section|.
+%%% |\subsection| and other \LaTeX{} style markup is not supported.
+
+%%% For hierarchic structure, we identify these sections with category
+%%% names `|section|$l$', where $l$ is the respective level, computed
+%%% from the rank. Those section names do not exist yet, we have to
+%%% create them.
+
+%%% \end{itemize}
+
+%%% Now we come to the point why we can't use the rank as the section
+%%% level passed to \LaTeX{}: When we assume the level usage of \LaTeX{}'s
+%%% standard classes, |\section| is \emph{always} on level~1, independent
+%%% of the existence of |\chapter|. The latter is placed on level~0, its
+%%% existence only influences the section level of |\part|, which is
+%%% either $-1$ or~0.
+
+%%% This implies that |\subsection| in a standard class is always on
+%%% level~2, but our equivalent section division is either `|@*0|'
+%%% (rank~1) or `|@*1|' (rank~2). I.e., when we have chapters, we use a
+%%% level that's one lower than the rank, otherwise we use the rank.
+
+%%% For flat structure, we use type |chunk|. This way we'll use the chunk
+%%% numbers, they are already set up.
+
+%%% For flat structure, we will map all ranks to the same section markup.
+%%% This markup will also be accessible by |\section| or |\chapter|. They
+%%% will be declared to be on a level that is the same as the rank.
+
+
+%%% \sect The printed representation of a section divison number is simply
+%%% the presentation from the next upper level, a full stop, and an arabic
+%%% section number. When we compute the name of the previous level we must
+%%% pay attention to our first section division created this way. This
+%%% division must be placed below |section|, without any following number.
+
+%%% Since registers are a valuable resource in \TeX{}, we have to take
+%%% care not to allocate a superfluous one.
+
+%%% \beginprog
+\def\toclevel@chunk{12}
+\def\theHchunk{chunk.\arabic{chunk}}
+\ifcase \cweb@structure
+ %% hierarchic
+ \newcounter{section2}[section]
+ \@namedef{thesection2}{\@nameuse{thesection}.\arabic{section2}}
+ \newcounter{section3}[section2]
+ \@namedef{thesection3}{\@nameuse{thesection2}.\arabic{section3}}
+ \newcounter{section4}[section3]
+ \@namedef{thesection4}{\@nameuse{thesection3}.\arabic{section4}}
+ \newcounter{section5}[section4]
+ \@namedef{thesection5}{\@nameuse{thesection4}.\arabic{section5}}
+ \newcounter{section6}[section5]
+ \@namedef{thesection6}{\@nameuse{thesection5}.\arabic{section6}}
+ \newcounter{section7}[section6]
+ \@namedef{thesection7}{\@nameuse{thesection6}.\arabic{section7}}
+ \newcounter{section8}[section7]
+ \@namedef{thesection8}{\@nameuse{thesection7}.\arabic{section8}}
+ \newcounter{section9}[section8]
+ \@namedef{thesection9}{\@nameuse{thesection8}.\arabic{section9}}
+%
+ \@namedef{theHsection2}{\@nameuse{thesection}.\arabic{section2}}
+ \@namedef{theHsection3}{\@nameuse{thesection2}.\arabic{section3}}
+ \@namedef{theHsection4}{\@nameuse{thesection3}.\arabic{section4}}
+ \@namedef{theHsection5}{\@nameuse{thesection4}.\arabic{section5}}
+ \@namedef{theHsection6}{\@nameuse{thesection5}.\arabic{section6}}
+ \@namedef{theHsection7}{\@nameuse{thesection6}.\arabic{section7}}
+ \@namedef{theHsection8}{\@nameuse{thesection7}.\arabic{section8}}
+ \@namedef{theHsection9}{\@nameuse{thesection8}.\arabic{section9}}
+ \@namedef{toclevel@section2}{2}
+ \@namedef{toclevel@section3}{3}
+ \@namedef{toclevel@section4}{4}
+ \@namedef{toclevel@section5}{5}
+ \@namedef{toclevel@section6}{6}
+ \@namedef{toclevel@section7}{7}
+ \@namedef{toclevel@section8}{8}
+ \@namedef{toclevel@section9}{9}
+ \ifx \chapter\undefined
+ \newcounter{section10}[section9]
+ \@namedef{thesection10}{\@nameuse{thesection9}.\arabic{section10}}
+ \@namedef{theHsection10}{\@nameuse{theHsection9}.\arabic{section10}}
+ \@namedef{toclevel@section10}{10}
+ \else
+ \fi
+\fi
+%%% \endprog
+
+
+%%% \sect Sections are typeset by |\CwebSection|; rank and title are the
+%%% arguments.
+
+%%% For hierarchic structure, we select the proper cseq according to rank:
+%%% |\part|, |\chapter| if this is a structure with chapters, |\section|,
+%%% and |\cwbb@section{|\<rank>|}| for other ranks. For the last cseq we
+%%% also have to supply the section category, it's `|section|$l$' where
+%%% $l$ is the rank~$r$ if we don't have chapters, $r-1$ otherwise.
+
+%%% That category cannot be computed by |\cwbb@section| itself since we
+%%% want to use that macro also for flat structure. But there all sections
+%%% are basically chunks, we have to use the |chunk| category there. Since
+%%% we want to differentiate them in the table of contents nevertheless,
+%%% we record the current rank in the title.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \ifcase\cweb@cls
+ \ifx \chapter\undefined
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\section
+ \else %% level = rank
+ \def\cwbb@sect_name{section#1}%
+ \def\next{\cwbb@section{#1}}%
+ \fi
+ \next{#2}%
+ }
+ \else
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\chapter
+ \or %% 2
+ \let\next\section
+ \else %% level = rank - 1
+ \count@=#1 \advance \count@\m@ne
+ \edef\cwbb@sect_name{section\number\count@}%
+ \edef\next{\noexpand\cwbb@section{\number\count@}}%
+ \fi
+ \next{#2}%
+ }
+ \fi
+ \or %% In case we are included from other classes
+ %%
+ \ifx \chapter\undefined
+ \def\CwebSection#1#2{%
+ \count@=#1 \advance \count@\@ne\advance \count@\@ne
+ \edef\cwbb@sect_name{section\number\count@}%
+ \edef\next{\noexpand\cwbb@section{\number\count@}}%
+ \next{#2}%
+ }
+ \else
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\chapter
+ \or %% 1
+ \let\next\section
+ \else %% level = rank
+ \def\cwbb@sect_name{section#1}%
+ \def\next{\cwbb@section{#1}}%
+ \fi
+ \next{#2}%
+ }
+ \fi
+ \fi
+ \or
+ %% flat
+ \def\CwebSection#1#2{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{#1}{\protect\cwbbRank{#1}#2}%
+ }
+ \def\section{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{1}%
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect We use the same layout for all minor sections. That layout can
+%%% be changed by ammended by defining |\CwebSect|$r$|Hook| to redefine
+%%% |\CwebSectPreSkip| and/or |\CwebSectLayout| locally. (Of couse, it may
+%%% be redefined once globally, too.) These are the values to be used for
+%%% arguments 4 and~5 of |\@startsection|, pay attention to the skip sign,
+%%% telling about indentation of the first paragraph. And don't forget
+%%% that the skip is only used for sections on ranks below
+%%% |\CwebRankNoEject|, the others start on a new page anyhow.
+
+%%% The default definition sets the section title in bold and leaves the
+%%% same space as before chunks. The first paragraph of the following
+%%% documentation has no indentation.
+
+%%% Other layout changes (indentation for heading, etc.)\ must be made by
+%%% redefining |\cwbb@section|. Or---email me, if you want to do it. Then
+%%% I might be enclined to introduce further possibilities.
+
+%%% \beginprog
+\newskip\CwebSectPreSkip
+ \CwebSectPreSkip = -\CwebChunkPreSkip
+\def\CwebSectLayout{\normalfont\bfseries}
+
+\def\cwbb@section#1{% % #2 (next token) is title
+ \csname CwebSect#1Hook\endcsname % \relax if undefined
+ \expandafter\@startsection
+ \expandafter{\cwbb@sect_name}% % category; for counter, etc.
+ {#1}% % level is rank
+ \z@ % no indentation for heading
+ \CwebSectPreSkip % skip above, maybe no next par indent
+ \medskipamount % skip below heading
+ \CwebSectLayout
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Marks.
+
+%%% \begin{fixme}
+%%% Since I don't know yet how to handle marks I ignore most of them for
+%%% the time being.
+
+%%% Have to change this later.
+%%% \end{fixme}
+
+%%% \beginprog
+\let\chunkmark\@gobble
+
+\count@=2
+\@whilenum \count@<11 \do {
+ \expandafter\let \csname section\number\count@ mark\endcsname \@gobble
+ \advance \count@ by 1
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Typesetting programs.
+
+%%% Program pieces come in two flavours: as argument of |\PB|, or as
+%%% material after |\B|. In the former case we can use a group for the
+%%% switch to the restricted program state, the group end will restore the
+%%% previous state again. In the latter case we use |\B| for the switch to
+%%% program state, the cross reference list or the next chunk will go to
+%%% another state. Since we have to do more for the material after |\B|,
+%%% we define this cseq later.
+
+%%% \beginprog
+\def\PB#1{%
+ \begingroup
+ \cwbb@Rprogram
+ \leavevmode
+ #1%
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect Note, that |\Y| cannot just be |\smallskip|, as in Plain
+%%% \cweb{}. We must assert that the current paragraph is ended before the
+%%% vertical glue is inserted, and that's not done by the \LaTeX{}
+%%% definition of |\smallskip|. In addition we add some negative penalty,
+%%% here is a good place for a page break. As the penalty value we use
+%%% half the chunk break penalty---of course a chunk start is an even
+%%% better place for a page break\,\dots
+
+%%% It might be that |\Y| is the very first token in a chunk. Then the
+%%% chunk start marker isn't set already since it is regarded as a run-in
+%%% section heading. In that case we don't set the vertical skip but
+%%% simply start the chunk with the program part.
+
+%%% \TeX{}nical note: The flag |@noskipsec| may be used to test if we're
+%%% immediately after a run-in section heading.
+
+%%% \beginprog
+\newcount\CwebProgPenalty
+ \CwebProgPenalty=\@secpenalty
+ \divide\CwebProgPenalty by 2
+\def\Y{%
+ \if@noskipsec
+ \else
+ \par
+ \penalty\CwebProgPenalty
+ \smallskip
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect Yep, let's unfold the ``official'' names of the cseqs used in
+%%% program state.
+
+%%% If we turn on the \cweave{} bindings, they might be in effect already,
+%%% we don't need to establish them again. We can test
+%%% |\cwbb@user_bindings| for this case, it will be redefined then.
+
+%%% \beginprog
+\def\cwbb@cweave_bindings{%
+ \ifx \cwbb@user_bindings\relax
+ \cwbb@rebind
+ % indentation and paragraph layout
+ \cwbb@incr_indent \1%
+ \cwbb@decr_indent \2%
+ \cwbb@expr_break \3%
+ \cwbb@backup \4%
+ \cwbb@opt_break \5%
+ \cwbb@break \6%
+ \cwbb@big_break \7%
+ \cwbb@noindent \8%
+ % C/C++ tokens
+ \CwebRel \?%
+ \CwebAddress \AND
+ \CwebComplement \CM
+ \CwebScope \DC
+ \CwebEquiv \E
+ \CwebGe \G
+ \CwebRightShift \GG
+ \CwebNe \I
+ \CwebAssign \K
+ \CwebLeftShift \LL
+ \CwebMod \MOD
+ \CwebNull \NULL
+ \CwebNot \R
+ \CwebBinOr \OR
+ \CwebMemberRef \PA
+ \CwebThis \this
+ \CwebOr \V
+ \CwebAnd \W
+ \CwebXor \XOR
+ \CwebLe \Z
+ \CwebPointer \MG
+ \CwebPointerMemberRef \MGA
+ \CwebDecr \MM
+ \CwebIncr \PP
+ % more tokens
+ \CwebId \\%
+ \CwebIdLetter \|%
+ \CwebRes \&%
+ \CwebString \.% %% ( ...Emacs...
+ \CwebStringBreak \)%
+ \CwebNumber \T
+ \CwebCombinedOp \MRL
+ % goes to TeX state
+ \CwebComment \C
+ \CwebCxxComment \SHC
+ \CwebRefName \X
+ % CWEB tokens
+ \CwebMacrosHere \ATH
+ \CwebDefine \D
+ \cwbb@format \F
+ \CwebIdCat \J
+ \CwebVerbString \vb
+ % cross reference tags
+ \cwbb@change_flag \*%
+ \CwebCRAlso \A
+ \CwebCRsAlso \As
+ \CwebCRCite \Q
+ \CwebCRsCite \Qs
+ \CwebCRUse \U
+ \CwebCRsUse \Us
+ \CwebCREt \ET
+ \CwebCRsEt \ETs
+ % finish the list
+ \stop\stop
+ \def\cwbb@user_bindings{%
+ \cwbb@restore_bindings
+ \let\cwbb@user_bindings\relax
+ }%
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect Since most of the \cweave{} bindings are simple and tedious
+%%% coding, we'll have a look at the program layout next. Between two
+%%% paragraphs there must not be any skip, the skip used in document
+%%% layout is saved in |\cwbb@save@parskip|. A few other layout
+%%% parameters from the document layout must be saved as well.
+
+%%% \beginprog
+\newskip\cwbb@save@parskip
+\newskip\cwbb@save@rightskip
+\newcount\cwbb@save@sem_sfcode
+\newcount\cwbb@save@pretolerance
+\newcount\cwbb@save@hyphenpenalty
+\newcount\cwbb@save@exhyphenpenalty
+
+\def\cwbb@save_doc_layout{%
+ \cwbb@save@parskip\parskip
+ \cwbb@save@rightskip\rightskip
+ \cwbb@save@sem_sfcode\sfcode`\;
+ \cwbb@save@pretolerance\pretolerance
+ \cwbb@save@hyphenpenalty\hyphenpenalty
+ \cwbb@save@exhyphenpenalty\exhyphenpenalty
+ }
+\def\cwbb@restore_doc_layout{%
+ \parskip\cwbb@save@parskip
+ \rightskip\cwbb@save@rightskip
+ \sfcode`\;=\cwbb@save@sem_sfcode
+ \pretolerance\cwbb@save@pretolerance
+ \hyphenpenalty\cwbb@save@hyphenpenalty
+ \exhyphenpenalty\cwbb@save@exhyphenpenalty
+ }
+%%% \endprog
+
+
+%%% \sect When |\cwbb@program_layout| is called for the very first time
+%%% we have to set up some variables. This cannot be done in advance
+%%% since they depend on values which might be changed by the user in the
+%%% preamble. This setup is done in |\cwbb@layout_init|.
+
+%%% If the program layout is already active, we must not switch it on
+%%% another time. This can be tested by the current binding of
+%%% |\cwbb@doc_layout|.
+
+%%% \beginprog
+\def\cwbb@program_layout{%
+ \ifx \cwbb@doc_layout\relax
+ \cwbb@layout_init
+ \cwbb@save_doc_layout
+ \def\cwbb@doc_layout{%
+ \cwbb@restore_doc_layout
+ \let\cwbb@doc_layout\relax
+ }%
+ % set new values
+ \parskip\z@skip
+ \rightskip\z@ plus 100\p@ minus 10\p@
+ \sfcode`\;=3000
+ \pretolerance\@M
+ \hyphenpenalty 1000 % strings can be broken this way
+ \exhyphenpenalty\@M
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect The unit of the basic indentation is stored in
+%%% |\CwebIndentUnit|. Continuation lines are indented two units further,
+%%% i.e., the initial hanging indentation is 3~units. The current hanging
+%%% indentation is kept in |\cwbb@indent|.
+
+%%% We need undiscardable items which can be used as backspaces, of one
+%%% and two units, respectively. This is done best by boxes. The
+%%% initialization of the boxes must be postponed until the user had the
+%%% chance to change |\CwebIndentUnit|, it's done in the program state
+%%% initialization.
+
+%%% \beginprog
+\newdimen\CwebIndentUnit
+ \CwebIndentUnit=1em
+\newdimen\cwbb@indent
+
+\newbox\cwbb@bak % backspace one unit
+\newbox\cwbb@bakk % backspace two units
+
+\def\cwbb@layout_init{%
+ \global\setbox\cwbb@bak \hbox to -1\CwebIndentUnit{}%
+ \global\setbox\cwbb@bakk \hbox to -2\CwebIndentUnit{}%
+ \global\let\cwbb@layout_init\relax
+ }
+%%% \endprog
+
+
+%%% \sect Now we can formulate how to start typesetting program pieces: We
+%%% switch to program state, set the initial indentation, and add the
+%%% basic indentation of the first code line. Then we might add a marker
+%%% to this program part; for hierarchic structure we want to typeset the
+%%% chunk number in the margin of the chunk's program part. That's not
+%%% really possible, 'though: definitions and format requests are also
+%%% introduced with |\B|, so the chunk number will happen to be at the
+%%% start of the definition part if one exists.
+
+%%% If the code line is the very first text to be typeset in this chunk,
+%%% we don't add the basic indentation---we're already indented from the
+%%% run-in chunk start marker. But we must not start our chunk with
+%%% |\noindent|, the marker would be typeset in the left margin then. (I
+%%% consider this as a \LaTeX{} error.) So we just emulate an empty
+%%% documentation part, the marker is now set correctly. |\B| can test for
+%%% this case: |@noskipsec| is still true then.
+
+%%% \beginprog
+\def\B{%
+ \cwbb@program
+ \cwbb@indent 3\CwebIndentUnit \hangindent\cwbb@indent
+ \ifvmode
+ \if@noskipsec
+ \indent % add an empty documentation part
+ \else
+ \noindent \kern\CwebIndentUnit
+ \fi
+ \fi
+ \CwebNumberProgramPart
+ }
+%%% \endprog
+
+
+%%% \sect The chunk number shall be added at the first program part. We
+%%% have a hook in the chunk start where we can set up
+%%% |\CwebNumberProgramPart|. It will redefine itself at first call, to
+%%% disable further output on more |\B| occurences in a chunk.
+
+%%% The chunk number should always be in the left margin. If it's in the
+%%% right margin, there would be too much horizontal space between the
+%%% code start and the actual number. (The first code line is typically
+%%% not very long.) \LaTeX{} does not allow to select the margin for
+%%% marginpars, therefore we have to typeset the marginal note `by hand'.
+%%% As we know that we're in horizontal mode when we're called, |\vadjust|
+%%% is used to create the note. The standard \LaTeX{} dimen register for
+%%% the distance between marginal notes and text matter is used, of
+%%% course.
+
+%%% Note that marginal notes may overwrite the chunk number. As it's very
+%%% unlikely that somebody issues marginal notes in program mode (that's
+%%% quite difficult and involves ingenious usage of the implementation of
+%%% this macros), I won't care for that case.
+
+%%% \begin{fixme}
+%%% The baseline of the marginal note shall be aligned to the baseline
+%%% of the preceding line. That does not happen by default. Currently, I
+%%% assume that the preceding line has a depth of appropriately the
+%%% depth of a~`g'. Experiments showed that this gives good results for
+%%% most \cweb{} documents. But nevertheless one should determine the
+%%% exact depth of the preceding line, to know how much |box0| should be
+%%% raised. If anybody knows how this can be done, please tell me. (I
+%%% know about the route over the output routine, implementing an own
+%%% marginal note category, but this seems too much effort.)
+%%% \end{fixme}
+
+%%% For flat structure, we don't add any program part marker and we
+%%% don't need to initialize it.
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es}) introduced some code to insert
+%%% hyperlinks.
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\cwbb@marginal_chunkno{%
+ \vadjust{%
+ \setbox\z@ \hbox to \z@{%
+ \hss \vphantom{g}%
+ \hypertarget{_hyperchunk_\thechunk}{%
+ \small \thechunk
+ }%
+ \hskip \marginparsep
+ }%
+ \vbox to \z@{%
+ \hsize\z@
+ \vss
+ \noindent \raise \dp\z@ \box\z@
+ }%
+ }%
+ }
+ \def\cwbb@init_print_chunk{%
+ \def\CwebNumberProgramPart{%
+ %\hypertarget{_hyperchunk_\thechunk}{%
+ \cwbb@marginal_chunkno%}
+ \let\CwebNumberProgramPart\relax
+ }%
+ }
+ \or
+ %% flat
+ \let\cwbb@init_print_chunk\relax
+ \let\CwebNumberProgramPart\relax
+\fi
+%%% \endprog
+
+
+%%% \sect If a statement is finished, a new paragraph with the basic
+%%% indentation has to be started.
+
+%%% An optional statement break is implemented by a low penalty which
+%%% will be selected if the line has to be broken. We assume that the
+%%% hanging indentation of the new line is already set correctly, and
+%%% have to backup two units to get the basic indentation. It might be
+%%% that the line break is not chosen by \TeX{}, we compensate the
+%%% backspace for this case. This compensation is discarded at the start
+%%% of a new line.
+
+%%% \beginprog
+\def\cwbb@break{% % forced break, between statements
+ \ifmmode\else
+ \endgraf % in LaTeX it isn't sure what \par is really...
+ \noindent
+ \hangindent\cwbb@indent \kern\cwbb@indent
+ \copy\cwbb@bakk % go back to basic indentation
+ \ignorespaces
+ \fi
+ }
+\def\cwbb@opt_break{% % optional break between statements
+ \hfil \penalty\m@ne
+ \hfilneg \kern .5em \kern 2\CwebIndentUnit % discarded on line break
+ \copy\cwbb@bakk
+ \ignorespaces
+ }
+\def\cwbb@big_break{% % forced break and a little extra space
+ \Y
+ \cwbb@break
+ }
+\def\cwbb@expr_break#1{% % break with penalty #1 * 10
+ \hfil \penalty#10
+ \hfilneg % discarded on line break
+ }
+%%% \endprog
+
+
+%%% \sect When we increment the indentation, we must not forget to set
+%%% the hanging indentation immediately since |\cwbb@opt_break| relies on
+%%% the new value. If a continuation line is needed it will also be
+%%% indented one unit more, which is ok since it should be distinguishable
+%%% from the next line.
+%%% %
+%%% \begin{fixme}
+%%% Shouldn't the hanging indentation be set anew at the end of
+%%% |\cwbb@decr_indent| as well?
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@incr_indent{%
+ \global\advance\cwbb@indent\CwebIndentUnit
+ \hangindent\cwbb@indent
+ }
+\def\cwbb@decr_indent{\global\advance\cwbb@indent -\CwebIndentUnit}
+\def\cwbb@backup{\copy\cwbb@bak}
+\def\cwbb@noindent{% % no indentation
+ \hskip -\cwbb@indent \hskip 2\CwebIndentUnit
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Program (C or \C++) tokens.
+
+%%% Since the user might want to change the (somewhat unusual) way some
+%%% of the operators are typeset, we supply names for them.
+
+%%% \beginprog
+\let\CwebAnd=\land % logical and, &&
+\let\CwebEquiv=\equiv % equiv sign, for ==
+\let\CwebGe=\ge % greater or equal
+\let\CwebLe=\le % less or equal
+\let\CwebNe=\ne % unequal, !=
+\let\CwebNull=\Lambda % NULL pointer
+\let\CwebNot=\lnot % logical not, !
+\let\CwebOr=\lor % logical or, ||
+\let\CwebXor=\oplus % bitwise exclusive or, ^
+%%% \endprog
+
+
+%%% \sect Some symbols have to be shifted around, to save computation time
+%%% we put them in boxes.
+
+%%% \begin{fixme}
+%%% It's not sure if the following symbols still look good in other font
+%%% families than Computer Modern. The movements might need to get
+%%% adapted when more experience is available. They were chosen from
+%%% Plain \cweb{}, where these were explicit movements measured in
+%%% points. I changed the dimension to quads, to achieve at least the
+%%% support for different sizes in the Computer Modern family.
+%%% \end{fixme}
+
+%%% \TeX{}nical note: The pointer symbol must not be declared as a math
+%%% symbol (e.g., with |\DeclareMathSymbol|). Then \TeX{} would insert an
+%%% italic correction behind it, according to rule~17 from appendix~G.
+%%% (Please send email if you know of an easy way to prevent that
+%%% addition.) So I use the symbol as an accent, but then I have to supply
+%%% a nucleus for this math list. That nucleus also determines the total
+%%% width of this construct. As the vector accent has a width of
+%%% approximately 5\,pt in |cmmi10|, I'm going to use 0.5\,em. (That
+%%% heuristic is as good as any other I can think of.)
+
+%%% \beginprog
+%% pointer to struct component (`->'), use symbol of \vec accent
+\newbox\cwbb@pointer
+\setbox\cwbb@pointer=\hbox{%
+ \kern -.2em
+ \lower .3em \hbox{$\vec{\kern .5em}$}%
+ \kern .1em
+ }
+
+\newbox\cwbb@decr % decrement, --
+\setbox\cwbb@decr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {-}\kern -.1em{-}$}%
+ \kern .05em
+ }
+\newbox\cwbb@incr % increment, ++
+\setbox\cwbb@incr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {+}\kern -.1em{+}$}%
+ \kern .05em
+ }
+%%% \endprog
+
+
+%%% \sect Although the following symbols are typeset like an ``ordinary
+%%% \C++ programmer'' would expect them, we provide own module names
+%%% nevertheless. They can now be changed as well, i.e., orthogonality is
+%%% enhanced. (And we can use |\cwbb@rebind| for assigning them to their
+%%% names while we switch to program state\,\dots).
+
+%%% Note: The comment at |\CwebMod| says ``modulo/remainder'', as this is
+%%% not defined by the C standard, i.e., implementation-dependent.
+
+%%% \beginprog
+\mathchardef\CwebAddress="2026 % `&', as binary op
+\let\CwebAssign== % assignment
+\let\CwebBinOr=\mid % binary or
+\def\CwebComplement{{\sim}} % `~', as ordinary symbol
+\def\CwebDecr{\copy\cwbb@decr} % decrement
+\def\CwebIncr{\copy\cwbb@incr} % increment
+\let\CwebLeftShift=\ll % left shift, <<
+\def\CwebMod{\mathbin{\hbox{\footnotesize\rm\%}}} % modulo/remainder, %
+\def\CwebMemberRef{\mathbin{.*}} % ptr to member (on object)
+\def\CwebPointer{\copy\cwbb@pointer}
+\def\CwebPointerMemberRef{\mathbin{\CwebPointer*}} % ptr to member (on ptr)
+\def\CwebRel{\mathrel?} % relation operator
+\let\CwebRightShift=\gg % right shift, >>
+\def\CwebScope{\kern.1em{::}\kern.1em} % scope resolution
+
+\def\CwebThis{\CwebRes{this}} % reserved identifier `this'
+%%% \endprog
+
+
+%%% \sect The special identifier |\TeX| remains. In math mode it shall be
+%%% the identifier, in other modes the usual logo. We use the definition
+%%% supplied by DEK (or \textsc{Silvio Levy}?)\ from Plain \cweb{}.
+
+%%% \beginprog
+\def\TeX{%
+ {%
+ \ifmmode\it\fi
+ \mbox{T\kern-.1667em\lower.424ex\hbox{E}\hskip-.125emX}%
+ }}
+%%% \endprog
+
+
+%%% \sect Some tokens don't have constant names, the name is supplied as
+%%% the argument.
+
+%%% Identifiers are typeset in italics, reserved words and type names in
+%%% boldface, and strings in typewriter. For an underscore in the bold
+%%% words we use a line that's 50\,\% thicker as the usual line thickness.
+%%% -- But what is the usual line thickness. As Knuth uses 0.4\,pt for a
+%%% 10\,pt Computer Modern font, we'll use 0.06\,em for a CM font in an
+%%% arbitrary size. If that line thickness is still OK for other fonts is
+%%% doubtful, as it should fit to the stem width. (CM is a rather light
+%%% font, after all.) But since there is no font dimension that gives a
+%%% hint to an appropriate line thickness, it's our best try.
+
+%%% \beginprog
+\def\CwebCombinedOp#1{\mathrel{\let\K==#1}} % e.g., += operator
+\def\CwebId#1{\hbox{\it#1\/\kern.05em}} % identifier, more than one char
+\def\CwebIdLetter#1{\hbox{$#1$}} % identifier, one letter
+\def\CwebRes#1{% % reserverd words and type names
+ \hbox{\bf
+ \def\_{\kern.04em\vbox{\hrule width.3em height.06em}\kern.08em}%
+ #1\/\kern .05em
+ }%
+ }
+%%% \endprog
+
+
+%%% \sect In Plain \cweb{} strings are typeset in the font |cmtex10|,
+%%% which is a typewriter font with extended ASCII characters. This font
+%%% does not have the usual accents and can therefore not used for
+%%% typesetting national characters. (Very often they are input in some
+%%% 8-bit encoding, the respective character code is made active and is
+%%% substituted by a cseq which expands to the correct glyph. Of course,
+%%% there are better ways, but that's the reality we have to cope with.)
+%%% Instead of this special font we use the standard typewriter font.
+%%% This has the further advantage that |cmtex10| may not be introduced
+%%% to NFSS\@.
+
+%%% \beginprog
+\def\CwebString{%
+ \hbox \bgroup
+ \tt
+ \cwbb@string_setup
+ \cwbb@string
+ }
+\def\cwbb@string#1{%
+ #1\kern .05em
+ \egroup
+ }
+%%% \endprog
+
+
+%%% \sect Within strings certain cseqs have a special meaning; this is
+%%% introduced by |\cwbb@string_setup|. The cseqs within strings are
+%%% mostly special \TeX{} characters that are to be typeset by cseqs with
+%%% the same name.
+
+%%% But a problem remains: active characters. They will still expand, and
+%%% will mess up the string visualization. This is a particular problem
+%%% with |german.sty| where the double quote is active. (Of course, that's
+%%% a problem -- I'm German. :--) When one typesets strings, a double
+%%% quote shall be typeset as double quotes, and must not expand to the
+%%% special meaning. Remember that double quotes happen to appear quite
+%%% often in strings. As the string argument is already tokenized, one has
+%%% to redefine the binding of $(\textit{\textup{symbol}}\ \textbf{.}\
+%%% |#\"|)$.
+
+%%% \begin{fixme}
+%%% Perhaps one might use a loop over all characters except the braces,
+%%% and set their catcode to other. That would give us enough security.
+
+%%% For now, we handle only the double quote.
+%%% \end{fixme}
+
+%%% Every once in a while we have discretionary breaks in a string,
+%%% denoted by |\CwebStringBreak|. This break is shown with the C
+%%% convention of escaped newlines.
+
+%%% \beginprog
+\def\cwbb@string_setup{%
+ \chardef\ =`\ % % <-- two spaces !
+ \chardef\&=`\&
+ \chardef\\=`\\
+ \chardef\^=`\^
+ \chardef\_=`\_
+ \chardef\{=`\{
+ \chardef\}=`\}
+ \chardef\~=`\~
+ \catcode`\"=\CatOther
+ }
+
+\def\CwebStringBreak{\discretionary{\hbox{\tt\char`\\}}{}{}}
+%%% \endprog
+
+
+%%% \sect Numbers are typeset in different ways. We use the definition of
+%%% Plain \cweb{}.
+%%% %
+%%% \begin{fixme}
+%%% Should add a specification of the possible input and an explanation
+%%% of the macros below. In particular, that the closing brace after
+%%% |\aftergroup| is used much later is probably not grokked by most
+%%% \TeX{} programmers.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@oct{\hbox{$^\circ$\kern-.1em\it\aftergroup\?\aftergroup}}
+\def\cwbb@hex{\hbox{$^{\scriptscriptstyle\#}$\tt\aftergroup}}
+
+\def\CwebNumber#1{% % octal, hex, or decimal constant
+ \hbox{%
+ $%
+ \def\?{\kern.2em}%
+ \def\$##1{\egroup\sb{\,\rm##1}\bgroup}% suffix to constant
+ \def\_{\cdot 10^{\aftergroup}}% power of ten (via dirty trick)
+ \let\~\cwbb@oct \let\^\cwbb@hex
+ {#1}%
+ $}%
+ }
+%%% \endprog
+
+
+%%% \sect Comments are typeset in \TeX{} state. We add a hook, the user
+%%% shall be able to change the layout (e.g., he might want another font).
+%%% %
+%%% \begin{fixme}
+%%% Currently \C++ comments are typeset like C comments. This is
+%%% horrible in usual circumstances, i.e., when complete blocks of text
+%%% are prefixed with |//|. We should simply catenate all these text and
+%%% typeset it as one paragraph, each line prefixed by |//|. But then we
+%%% have to implement an |\everyline| first, and since that's not so
+%%% easy we postpone it\,\dots
+%%% \end{fixme}
+%%% %
+%%% In front of a C comment there is an optional stmt break, with 2\,quad
+%%% in front if the line is not broken and 1.5\,quad if the line is broken.
+
+%%% \beginprog
+\def\CwebComment#1{%
+ \5% % 0.5em will be discarded on line break
+ \hskip 1.5em
+ $/\ast\,$%
+ {\cwbb@tex
+ \CwebCommentHook
+ #1%
+ }%
+ $\,\ast/$%
+ }
+\let\CwebCommentHook\relax
+\let\CwebCxxComment\CwebComment
+%%% \endprog
+
+
+
+
+
+%%% \chap \cweb{} tokens.
+
+%%% We distinguish three categories of \cweb{} tokens: (1) Those which
+%%% output constant text, (2) those which have attributes to be displayed
+%%% in a special way, and (3) those which start a new structure element,
+%%% namely |@d| and~|@f|. Let's consider them in this order.
+
+
+%%% \sect \cweb{} tokens which expand in a constant string are the
+%%% identifier catenation operator~(`|@&|') and the macro placement
+%%% directive~(`|@h|').
+
+%%% \beginprog
+\def\CwebIdCat{\CwebString{@\&}}
+\def\CwebMacrosHere{%
+ \begingroup
+ \def\CwebRefNumber##1{}%
+ \CwebRefName :Preprocessor definitions\X
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect Verbatim program strings, i.e., strings passed verbatim by
+%%% \ctangle{}~(`|@=|') are typeset like normal strings, but within a box.
+%%% We use 2\,pt as the separating distance, this is set locally.
+
+%%% \beginprog
+\def\CwebVerbString#1{{\fboxsep\tw@\p@ \fbox{\CwebString{#1}}}}
+%%% \endprog
+
+
+%%% \sect The refinement names are typeset in angles, this has a long
+%%% tradition. We burry the typesetting of the chunk number (which is
+%%% the first parameter) in a macro call; the user may change this to
+%%% achieve special effects.
+%%% %
+%%% \begin{fixme}
+%%% One could think about handling special values of the chunk numbers
+%%% differently. E.g., an empty argument is not typeset at all, a~0
+%%% triggers a marginal note about a missing definition, etc.
+%%% \end{fixme}
+
+%%% A refinement name may be typeset both in math and horizontal mode.
+%%% The name itself is typeset in horizontal mode, of course; for the
+%%% angles we need math mode. Therefore we assert at the start of the
+%%% macro that we're not in math mode any more. At the end we switch back
+%%% to math mode if we've started in it. This conditional switch from and
+%%% back to math mode is done by |\cwbb@toggle_text|.
+
+%%% \TeX{}nical note: If we're in math mode |\cwbb@toggle_text| must be
+%%% defined globally, as it will turn off math mode and the definition
+%%% would be un-made then. The second invocation of |\cwbb@toggle_text|
+%%% would be undefined then.
+
+%%% A refinement may also consist of the file name the expansion of this
+%%% refinement shall be written to.
+%%% %
+%%% \label{sec:refname-dot}
+%%% %
+%%% This file name is tagged with |\.|, it shall be typeset like a string.
+%%% But the user shall be able to use the dot accent in the refinement
+%%% name as well. We check if the text consists solely of the tag and its
+%%% argument; in this case we substitute |\.| with |\CwebString|.
+%%% Otherwise we leave it as it is. Then a refinement name may not
+%%% consist of a single dot-accented expression---well, that's highly
+%%% unlikely. (Nevertheless I'll document it in the user's manual\,\dots)
+
+%%% \begin{fixme}
+%%% One should be able to use |\EnsureMath| here. But the refinement
+%%% name must not be set in math mode and must not be set in an hbox;
+%%% the latter would prevent line breaks.
+%%% \end{fixme}
+%%% Enrique Mel\'endez (\texttt{ema@csn.es}) added the macro |\cweb@num@lnk|
+%%% include hyperlinks into the cross references.
+
+%%% \beginprog
+\def\CwebRefName#1:#2\X{%
+ \ifmmode \gdef\cwbb@toggle_text{\null$\null}%
+ \else \let\cwbb@toggle_text\relax
+ \fi
+ \cwbb@toggle_text
+ $\langle\,${\cwbb@tex \cwbb@check_dot{#2}\CwebRefNumber{#1}}$\,\rangle$%
+ %}
+ \cwbb@toggle_text
+ }
+\def\CwebRefNumber#1{%
+ {\reset@font \footnotesize \kern .5em\cweb@num@lnk#1, |}%
+ }
+%%% \endprog
+
+%%% \sect This is the implementation of |\cweb@num@lnk|. It essentially
+%%% processes the incoming list, takink care of whether one element of the
+%%% list is processed with the macro |\[|. It borrows the phylosophy from
+%%% |\check_dot| below in that we inhibit the behaviour of |\[| for a while,
+%%% so that we can test its presence. Then we proeceed differently when it is
+%%% present. I (\texttt{ema@csn.es}) have not a lot of experience in list
+%%% processing in \LaTeX. I have done my best, using only two macros, the
+%%% already named |\def\cweb@num@lnk| and a wrapper for it,
+%%% |\def\cweb@@num@lnk|. I need the strange termination code (\texttt{|}).
+
+%%% \beginprog
+\newif\if@lbracket@
+\def\hyper@i\[#1]{\hyperlink{_hyperchunk_#1}{\[#1]}}
+\def\cweb@num@lnk #1, #2|{%
+ \begingroup
+ \def\[##1]{}%
+ \expandafter\def\expandafter\next\expandafter{#1}%
+ \ifx\next\empty
+ \global\@lbracket@true
+ \else
+ \global\@lbracket@false
+ \fi
+ \endgroup
+ \if@lbracket@%
+ \hyper@i#1%
+ \else
+ \hyperlink{_hyperchunk_#1}{#1}%
+ \fi
+ \@lbracket@false
+ \edef\next{#2}%
+ \ifx\next\empty
+ \else
+ { } % Yep, a space by itself.
+ \cweb@@num@lnk #2|%
+ \fi
+}
+\def\cweb@@num@lnk #1, |{\cweb@num@lnk#1, |}
+%%% \endprog
+
+
+%%% \sect The next macro, |\cwbb@check_dot| is a hassle to implement. We
+%%% must check if it's argument consists solely of a `|\.|'~tag, together
+%%% with the argument to this tag. Then the `|\.|'-argument shall be
+%%% typeset like a string, |\.| itself must be ignored.
+
+%%% The check for |\.| must not evaluate the parameter. Especially |\PB|
+%%% must not be evaluated within, it would lead to havoc with the
+%%% redefinition of all bindings, etc. (At least that's the empirical
+%%% result---it happened; although I don't know where the problems are.
+%%% Anyhow, it wasn't in the specs of the rebinding process.) All other
+%%% cseqs which shouldn't be used in moving arguments may cause problems
+%%% here, too. Note that |\protect| is no solution here, the user often is
+%%% not aware of the presence of \cweave{}-generated tags. As an example,
+%%% consider the refinement \verb"@< local variables of |foo| @>|" which
+%%% will lead to a call with the argument `|local variables of \PB{\\{foo}}|'.
+%%% |\\| must not be evaluated here.
+
+%%% I use an approach where I hope that it will work---but to be honest,
+%%% I'm not sure. I want to define a macro |\next| which shall have an
+%%% empty expansion iff |#1| was a `|\.|'~tag with its argument. Empty
+%%% refinement names are not allowed so we can assume that |#1| consists
+%%% of at least one token. I evaluate the first token of |#1| before the
+%%% |\def| is done. With an appropriate binding a |\.| will disappear,
+%%% perhaps leaving an empty token list. I also take care that |\PB| is
+%%% not evaluated then, by temporary rebinding it to |\relax|. Thereby I
+%%% have taken care for all cseqs \cweave{} might have introduced at the
+%%% very front of |#1|. Other cseqs are user-tags. If they cause
+%%% problems, the user might circumstance them by adding |\protect|.
+
+%%% \beginprog
+\def\cwbb@check_dot#1{%
+ \begingroup
+ \let\.\@gobble
+ \let\PB\relax
+ \expandafter\def \expandafter\next \expandafter{#1}%
+ \ifx \next\empty
+ \gdef\next##1{\CwebString}%
+ \else
+ \global\let\next\relax
+ \fi
+ \endgroup
+ \next #1%
+ }
+%%% \endprog
+
+
+%%% \sect \cweb{} macro definitions (done by~|@d|) and format directives are
+%%% basically program text. When they are started we are already in
+%%% program state, i.e., |\B| has appeared in front of it. But since we have
+%%% an introductionary identifier (either |#define| or |format|) to set at
+%%% the very front we increase the indentation by three units.
+
+%%% The output of format directives may be suppressed. Then we have to
+%%% gather both identifiers that will appear after |\F|. Both of them
+%%% consist of one group-token (a cseq and the identifier name), after
+%%% each identifier one other token separates them. I.e., we have to
+%%% gobble six group-tokens. As the very last token is |\par| if a format
+%%% directive follows directly on another one, the definition must be
+%%% long.
+
+%%% \TeX{}nical note: |\cwbb@format| must expand to |\CwebFormat|, not to
+%%% its expansion. Otherwise a redefinition of |\CwebFormat| in a package
+%%% would not be used.
+
+%%% \beginprog
+\def\CwebDefine{\cwbb@macro{\#define}}
+\def\CwebFormat{\cwbb@macro{format}}
+\if@cweb@suppress@format@
+ \long\def\cwbb@format#1#2#3#4#5#6{}
+\else
+ \def\cwbb@format{\CwebFormat}
+\fi
+
+\def\cwbb@macro#1{%
+ \global\advance\cwbb@indent \tw@\CwebIndentUnit
+ \cwbb@incr_indent
+ \CwebRes{#1 }% % <-- blank!
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Refinement cross references and changes.
+
+%%% At the end of chunks with refinements, \cweave{} may output cross
+%%% reference information: In which chunks additional definitions for
+%%% this refinement are found, where this refinement is used, and where it
+%%% is cited. This cross reference information is always introduced by a
+%%% tag followed by a non-empty list of chunk numbers. Different tags
+%%% are used for lists with only one element and for lists with more than
+%%% one element, this way the introductionary text may be adapted.
+
+%%% Within a list of $n$~chunk numbers the first $n-1$~numbers are
+%%% separated by commas with one following blank. The last two numbers
+%%% are separated by |\ET| if~$n=2$, and by |\ETs| if~$n>2$. If a chunk
+%%% is changed by the changefile, a changeflag~(`|\*|') is appended to
+%%% its number. The list is eventually terminated by a full stop.
+
+
+%%% \sect We separate the cross reference information by a smallskip from
+%%% the refinement or from a previous cross reference information. The
+%%% information itself is typeset in a smaller font, as it is auxilliary,
+%%% inserted stuff. The number list has a hanging indentation of
+%%% |\CwebNumberListHangindent|. But beware: This isn't a dimension
+%%% register, it's a macro. This way one can use ems as dimensions.
+
+%%% The actual size change is represented by a macro. We will redefine it
+%%% when we need an other size, e.g., in the lists at the end of the
+%%% document.
+
+%%% We must assert that we're in CR state while we set a cross reference.
+
+%%% The first parameter of a cross reference information unit is the
+%%% introductionary text, the second is the number list. The parameters
+%%% must be evaluated in a group---local parameter changes therein must
+%%% not influence the environment.
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es}) included some code to handle
+%%% putting in hyperlinks. In the case of only one item, a single
+%%% |\hyperlink| has been included. The processing of several
+%%% cross-references is somewhat more difficult. This amounts to inhibiting
+%%% the effect of |\ET| and |\ETs| so that the list can be processed
+%%% properly, and calling the macro |\cweb@cr@lnk| to process it. The
+%%% appropriate end-list conjunction will be added there.
+
+%%% \beginprog
+\def\CwebNumberListHangindent{2em}
+\let\CwebCRSize\footnotesize
+\def\CwebCrossRef#1#2.{%
+ \par\smallskip
+ \cwbb@CR
+ \begingroup
+ \reset@font \CwebCRSize
+ \noindent \hangindent\CwebNumberListHangindent\relax
+ #1~\hyperlink{_hypersect_#2}{#2}.\par
+ %#1~\hyperlink{_hyperchunk_#2}{#2}.\par
+ \endgroup
+ }
+\def\CwebCrossRefs#1#2.{%
+ \par\smallskip
+ \cwbb@CR
+ \begingroup
+ \reset@font \CwebCRSize
+ \noindent \hangindent\CwebNumberListHangindent\relax
+ \let\save@et\ET%
+ \let\save@ets\ETs
+ \def\ET{, }%
+ \let\ETs\ET%
+ \edef\@cr@lst{#2}%
+ \let\LastList@Conjunction\save@et
+ #1~\expandafter\cweb@cr@lnk\@cr@lst, ]\par
+ \endgroup
+ }
+%%% \endprog
+
+%%% \sect The cross references are added here for the case of a list of them.
+%%% We test whether we are at the last point of the list, and then add the
+%%% adequate conjunction, which is updated in the macro |\cweb@@cr@lnk|, which
+%%% takes care of the case of there being more than two elements. The
+%%% processing of the list goes along the same lines as |\cweb@num@lnk|.
+
+%%% \beginprog
+\def\cweb@cr@lnk #1, #2, #3]{%
+ \hyperlink{_hyperchunk_#1}{#1}%
+ \edef\next{#3}%
+ \ifx\next\empty
+ \LastList@Conjunction\hyperlink{_hyperchunk_#2}{#2}%
+ \else
+ , \hyperlink{_hyperchunk_#2}{#2}%
+ \cweb@@cr@lnk#3]%
+ \fi
+}
+\def\cweb@@cr@lnk #1, #2]{%
+ \let\LastList@Conjunction\save@ets
+ \edef\next{#2}%
+ \ifx\next\empty
+ \LastList@Conjunction\hyperlink{_hyperchunk_#1}{#1}%
+ \else
+ , \cweb@cr@lnk#1, #2]%
+ \fi
+}
+%%% \endprog
+
+
+%%% \sect Well, let's define all introducing tags, which in fact start the
+%%% cross reference information unit. The number list is gathered by
+%%% |\CwebCrossRef|.
+
+%%% \beginprog
+\def\CwebCRAlso{\CwebCrossRef{See also chunk}}
+\def\CwebCRsAlso{\CwebCrossRefs{See also chunks}}
+
+\def\CwebCRCite{\CwebCrossRef{This code is cited in chunk}}
+\def\CwebCRsCite{\CwebCrossRefs{This code is cited in chunks}}
+
+\def\CwebCRUse{\CwebCrossRef{This code is used in chunk}}
+\def\CwebCRsUse{\CwebCrossRefs{This code is used in chunks}}
+
+\def\CwebCREt{ and~}
+\def\CwebCRsEt{, and~}
+%%% \endprog
+
+
+
+
+%%% \chap Bells and whistles.
+
+%%% \begin{fixme}
+%%% Just copied from Plain \cweb{}. It doesn't work
+%%% anyhow---|\char"|\<xy> usually doesn't typeset the correct
+%%% character. That's because this character is most probably an active
+%%% character; at least that's typical for the way \TeX{} systems are
+%%% used in Europe. I have to think about the implementation.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\ATL{\par\noindent\bgroup\catcode`\_=12 \postATL} % print @l in limbo
+\def\postATL#1 #2 {\bf letter \\{\uppercase{\char"#1}}
+ tangles as \tentex "#2"\egroup\par}
+\def\noATL#1 #2 {}
+\def\noatl{\let\ATL=\noATL} % suppress output from @l
+%%% \endprog
+
+
+
+%%% \chap The end.
+
+%%% Well, after all we're finished with this module
+
+%%% \beginprog
+\endinput
+%%% \endprog
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+%%% \vskip \PltxPreSectSkip
+
+%%% \rcsLogIntro{After the actual revision log of this module, we give an
+%%% extract of log messages of the \cls{cweb} class. This module was
+%%% extracted from revision~2.8 of that class, we present to you the
+%%% log entries that concern the code of the new module.}
+
+%%% \begin{rcslog}
+%%% $StyleLog: cwebbase.doc,v $
+%%% \Revision 1.4 1995/11/07 17:55:19 schrod
+%%% \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+%%% versions; make it work with the previous-to-last one, version
+%%% \mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+%%% \mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+%%% |\newif|, and |\hb@xt@| must not be used.\\
+%%% Problem reported by Laurent Desnogues
+%%% \path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+%%% -- add name).
+
+%%% \Revision 1.3 1995/11/07 14:13:12 schrod
+%%% Reset section numbers when higher-level section begins.\\
+%%% Error reported by Bronne Louis \path|<bronne@montefiore.ulg.ac.be>|.
+
+%%% \Revision 1.2 1995/11/06 10:58:47 schrod
+%%% Support active double quotes in strings. That happens when
+%%% |german.sty| is used. Problems with other active characters may still
+%%% appear.
+
+%%% \Revision 1.1 1995/09/12 23:02:22 schrod
+%%% Moved all code that does the actual typesetting of \cweave{} tags and
+%%% is therefore also needed for a |cweb| environment. It's now an own
+%%% module named \pkg{cwebbase}.
+
+%%% Chunk markers and chunk marginal numbers don't hang below the
+%%% baseline any more.
+
+
+%%% \item[] \noindent\hrulefill\break
+%%% \textbf{Extract of the revision log for |cweb.doc|:}
+
+%%% \vskip 2ex
+
+%%% \Revision 2.5 1995/08/29 02:07:26 schrod
+%%% Discard dependencies on 10\,pt fonts.
+
+%%% Support suppression of format directives.
+
+%%% \Revision 2.4 1995/08/27 19:31:43 schrod
+%%% Make configuration of change flag easier.
+
+%%% Put section title of changed chunks list in marks, and tell the user
+%%% that it got typeset.
+
+%%% \Revision 2.3 1995/08/27 17:24:46 schrod
+%%% Make usage of baseclass with chapters work.
+
+%%% \Revision 2.2 1995/08/27 13:26:22 schrod
+%%% Add possibility to suppress change hints.
+
+%%% \Revision 2.1 1995/08/25 19:11:18 schrod
+%%% Hierarchic structures are supported now, in addition to the flat
+%%% structure of the beta-test version. One can choose with an option. For
+%%% that step, the terminology was cleaned up, too: Chunks are not named
+%%% sections any more. (That change involved reimplementation of almost
+%%% all the structure and toc stuff.)
+
+%%% The chunk number supplied by \cweave{} is used now, not some computed
+%%% number. Change flags are printed, too.
+
+%%% One can suppress output of unchanged sections.
+
+%%% \Revision 1.12 1993/08/10 14:15:43 schrod
+%%% New page on main section only if group level $<$ |\cwebSecNoEject|.
+%%% Default for the latter is 3.
+
+%%% \Revision 1.11 1993/08/10 11:21:07 schrod
+%%% Reference to section number does not render a period after the
+%%% number any more.
+
+%%% \Revision 1.10 1993/08/09 20:08:20 schrod
+%%% |\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+%%% effect already.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.9 1993/08/09 18:05:28 schrod
+%%% Left shift operator wasn't defined correctly.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.6 1993/06/15 08:49:23 schrod
+%%% |\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+%%% causes problems if a |\PB| is within. Now I try hard not to evaluate any
+%%% tokens outside of my control.
+
+%%% Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+%%% need an own macro.
+
+%%% \Revision 1.4 1993/06/14 15:54:18 schrod
+%%% Add FSA diagram about processing states. CR state is also switched
+%%% to from \TeX{} state (that happens with |\ch| at the document end).
+
+%%% \Revision 1.3 1993/05/13 17:51:21 schrod
+%%% Refinements may also be filenames (`|@(|'). Then the complete name
+%%% consists of a |\.| macro call, which is handled now.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Made the detection of `|@.|' index entries more robust.
+
+%%% \Revision 1.2 1993/05/12 18:28:59 schrod
+%%% Adapted to recent changes of \cweave{} (of April 93):
+
+%%% Main sections have a group level, represented in the table of
+%%% contents. This changed the complete implementation of section tags.
+
+%%% New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+%%% |\CwebCombinedOp|.
+
+%%% \Revision 1.1 1993/04/09 15:00:37 schrod
+%%% Initial revision
+
+%%% \end{rcslog}
+
+
+
+%%% \end{document}
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%% Local Variables:
+%%% mode: LaTeX
+%%% TeX-brace-indent-level: 4
+%%% indent-tabs-mode: t
+%%% TeX-auto-untabify: nil
+%%% TeX-auto-regexp-list: LaTeX-auto-regexp-list
+%%% compile-command: "make cwebbase.tex"
+%%% End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cweb-hy.cls b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cweb-hy.cls
new file mode 100644
index 00000000000..e1ca54752ae
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/cweb-hy.cls
@@ -0,0 +1,1538 @@
+% $StyleId: cweb.doc,v 3.6 1995/11/30 15:44:53 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+% Modified by Enrique Meléndez <ema@csn.es> to support hyperlinks and to allow
+% it to be included in other classes documnts.
+
+%
+% LaTeX class cweb
+% support for LaTeX markup in CWEB sources
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+% If you have received this style file without the user manual (in the
+% file cweb-user.dvi or the respective LaTeX file), it's incomplete and
+% near to useless. If it was given to you as something that you shall
+% use as an author -- complain bitterly to your provider. You need the
+% documentation and you have a right on it! (Below you can find info
+% where to get the reference version.)
+
+
+
+
+%%%%
+%%%%
+%%%% These TeX macros were documented with the documentation system
+%%%% MAKEPROG and automatically converted to the current form.
+%%%% If you have MAKEPROG available you may transform it back to
+%%%% the original input: Remove every occurence of three percents
+%%%% and one optional blank from the beginning of a line and remove
+%%%% every line which starts with four percents. The following lex
+%%%% program will do this:
+%%%%
+%%%% %%
+%%%%
+%%%% ^%%%\ ? ;
+%%%% ^%%%%.*\n ;
+%%%%
+%%%% If you just want to print the documentation you may fetch
+%%%% the archive print-makeprog.tar.Z from ftp.th-darmstadt.de (directory
+%%%% pub/tex/latex). It contains *all* used styles -- but beware, they
+%%%% may not be in a documented form...
+%%%%
+%%%%
+%%% \documentclass{progltx}
+
+%%% \usepackage{cweb-doc} % document-specific markup
+%%% \usepackage{fullpage}
+
+
+%%% \RCS $StyleRevision: 3.6 $
+%%% \RCS $StyleDate: 1995/11/30 15:44:53 $
+
+
+%%% \begin{document}
+
+
+%%% \title{The \texttt{cweb} Class\\
+%%% {\large (Implementation)}%
+%%% }
+%%% \author{% % LaTeX does not discard unnecessary glue...
+%%% Joachim Schrod%
+%%% \thanks{%
+%%% \protect\raggedright
+%%% TU~Darmstadt, Computer Science Department, WG Systems Programming,
+%%% Alexanderstr.~10, D-64283~Darmstadt, Germany.
+%%% Email: \texttt{jschrod@acm.org}%
+%%% }%
+%%% }
+%%% \date{%
+%%% Revision \RCSStyleRevision\\
+%%% (as of \RCSStyleDate)%
+%%% }
+
+%%% \maketitle
+
+
+%%% % doesn't work with progltx yet
+%%% %\tableofcontents
+
+
+
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The user interface of cweb.cls
+%%% %
+
+%%% \input{cweb-user}
+
+%%% %
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The internal interface
+%%% %
+
+%%% \input{cweb-conf}
+
+%%% %
+%%% % ------------------------------------------------------------
+
+
+
+
+%%% \chap Implementation.
+
+%%% This implementation must typeset complete files output by \cweave{}.
+%%% Large parts of this task are available from the module \pkg{cwebbase}.
+%%% It remains mainly to implement document-level markup and front and
+%%% back matter (table of contents, index, etc.) In particular, the back
+%%% matter material is not trivial; its tags are behind |\end{document}|,
+%%% we have to read carefully behind the end of our \cweb{} document to
+%%% detect if we have to typeset them at all.
+
+
+%%% \sect This module reserves the namespace |cweb|.
+
+%%% \beginprog
+\ifx \cweb@loaded\undefined
+ \def\cweb@loaded{$StyleRevision: 3.6 $}
+\else
+ \PackageWarningNoLine{cweb}%
+ {Some other package already uses namespace `cweb'}
+\fi
+%%% \endprog
+
+
+%%% \sect Let's identify this class.
+
+%%% \noindent The code below is explained in the implementation
+%%% documentation of the \pkg{rcs} package.
+
+%%% \beginprog
+\begingroup
+ \def\RCSClass#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesClass{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSClass{cweb-inc}{LaTeX markup for CWEB sources}
+ $StyleDate: 1995/11/30 15:44:53 $: 9999/00/00 \endRCS
+ $StyleRevision: 3.6 $: 0.0 \endRCS
+\endgroup
+%%% \endprog
+
+
+%%% \sect This module is from a supported bundle. Send bug reports,
+%%% comments and repairs.
+
+%%% The reference version may be retrieved via anonymous ftp from
+%%% |ftp.th-darmstadt.de| [130.83.47.112], directory
+%%% |pub/programming/literate-programming/c.c++/|. It's
+%%% placed there as a gzipped tar file. (The information on the IP~number
+%%% is dated August~29,~1995.
+%%% It might have changed, although this is very unlikely. Use
+%%% your friendly nameserver.)
+
+
+%%% \sect This is freely distributable software; you can redistribute it
+%%% and/or modify it under the terms of the GNU General Public License as
+%%% published by the Free Software Foundation; either version~2 of the
+%%% License, or (at your option) any later version.
+
+%%% This software is distributed in the hope that it will be useful, but
+%%% \textbf{without any warranty}; without even the implied warranty of
+%%% \textbf{merchantability} or \textbf{fitness for a particular purpose}.
+%%% See the GNU General Public License for more details.
+
+%%% You should have received a copy of the GNU General Public License in
+%%% the file |License| along with this package; if not, write to the Free
+%%% Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+%%% \sect Before we start we declare some shorthands for category codes.
+%%% By declaring the underscore~`(|_|)' as letter we can use it in our
+%%% macros. (I agree with \textsc{D.~Knuth} that
+%%% |\identifier_several_words_long| is more readable than
+%%% |\IdentifierSeveralWordsLong| and in every case better than |\p@@@s|.)
+
+%%% As this is a \LaTeX{} style file the at sign is a letter anyhow; so
+%%% we can use the ``private'' \LaTeX{} macros; and with the underscore we
+%%% can make our own macros more readable. But as we have to restore this
+%%% category code at the end of this macro file we store its former value
+%%% in the control sequence |\CatUsCode|. This method is better than to
+%%% use a group because not all macros have to be defined global this way.
+
+%%% Since somebody might use more styles from me, this cseqs might be
+%%% defined already.
+
+%%% \beginprog
+\ifx \CatEscape\undefined
+ \chardef\CatEscape=0
+ \chardef\CatOpen=1
+ \chardef\CatClose=2
+ \chardef\CatIgnore=9
+ \chardef\CatLetter=11
+ \chardef\CatOther=12
+ \chardef\CatActive=13 % is defined in Plain already
+
+ \chardef\CatUsCode=\catcode`\_
+\fi
+
+\catcode`\_=\CatLetter % top level macro file
+%%% \endprog
+
+
+
+%%% See if we are a class or a package. If |cweb| is a class, |section| is
+%%% still undefined, as it will be defined when the base package is loaded
+%%% later on. This allows us to define |cweb@cls| to be true and act
+%%% under this condition.[ema@csn.es]
+%%% \beginprog
+
+\ifx\section\undefined
+ \def\cweb@cls{0}
+\else
+ \def\cweb@cls{1}
+\fi
+%%% \endprog
+
+
+%%% \sect Problems over problems with different \LaTeX{} versions. I have
+%%% decided that I will take the same stand as the \LaTeX{} team: I will
+%%% only support the last two official \LaTeX{} versions, folks who run
+%%% older software might need to upgrade.
+
+%%% The implementation will use `star'-forms of \LaTeX{}'s command
+%%% definition macros. They were introduced with \LaTeX{}
+%%% \mbox{$\langle$1994/12/01$\rangle$}.
+
+%%% Currently (as of 07 Nov 95), 1995/06/01 is the most recent version.
+%%% I want to use a features introduced with that version: |\newif| is not
+%%% outer any more. So I check here for the version and supply the new
+%%% definition if necessary.
+
+%%% \beginprog
+\NeedsTeXFormat{LaTeX2e}[1994/12/01]
+\@ifl@t@r \fmtversion{1995/06/01}{}% else
+ {
+ \@namedef{newif}#1{%
+ \count@\escapechar \escapechar\m@ne
+ \let#1\iffalse
+ \@if#1\iftrue
+ \@if#1\iffalse
+ \escapechar\count@
+ }
+ \def\@if#1#2{%
+ \expandafter\def
+ \csname \expandafter\@gobbletwo\string#1%
+ \expandafter\@gobbletwo\string#2\endcsname {%
+ \let#1#2%
+ }%
+ }
+ }
+%%% \endprog
+
+
+%%% \chap Options.
+
+%%% Our options are of the ``keyword-value'' category. We realize them by
+%%% the \pkg{keyvald} package, an enhancement of the \pkg{keyval} package
+%%% from the Graphics bundle that supports defaulting for unknown keys.
+
+%%% \begin{fixme}
+%%% Most of the code below should be extracted and moved to a
+%%% \pkg{keyvalx} package (or a new version of \pkg{keyval}). In
+%%% particular, support for package/class option processing, enumeration
+%%% values, and set values for keyword options.
+%%% \end{fixme}
+
+%%% |\ProcessKeyValOptions| may be used instead of |\ProcessOptions|, it
+%%% takes the keyval set identifier as argument. At the end it evaluates
+%%% and resets a hook. That hook may be used at option definition to
+%%% define further dependencies. (E.g., enabling that option implies
+%%% enabling an other option, too.)
+%%% %
+%%% \begin{fixme}
+%%% Like |\ProcessOptions|, I define |\@curroptions|, since I don't know
+%%% if that's needed later. The only place where it could been needed is
+%%% in the check for unprocessed options.
+%%% \end{fixme}
+
+%%% \beginprog
+\RequirePackage{keyvald}
+
+\let\ProcessKeyValOptionsHook\empty
+\providecommand*\ProcessKeyValOptions[1]{%
+ \edef\@curroptions{\@ptionlist{\@currname.\@currext}}%
+ \edef\next{\noexpand\setkeys{#1}{\@curroptions}}%
+ \next
+ \ProcessKeyValOptionsHook
+ \let\ProcessKeyValOptionsHook\empty
+ \AtEndOfPackage{\let\@unprocessedoptions\relax}
+ }
+%%% \endprog
+
+
+%%% \sect The enumeration option |structure| selects hierarchic~(0) or
+%%% flat~(1) structure. The predefined value is `hierarchic', there is no
+%%% default. The option value is bound to |\cweb@structure|.
+%%%
+%%% I make this behaviour conditional on whether |cweb| is a class or a
+%%% package. [ema@csn.es]
+
+%%% \beginprog
+\ifcase\cweb@cls
+%%% \endprog
+
+%%% \beginprog
+%% \define@key@enum{cweb}{structure}{hierarchic,flat}[hierarchic]
+\@namedef{KV@cweb@structure@enum:hierarchic}{0 } % <-- space!
+\@namedef{KV@cweb@structure@enum:flat}{1 } % <-- space!
+\define@key{cweb}{structure}{%
+ \expandafter\let \expandafter\@tempa
+ \csname KV@cweb@structure@enum:#1\endcsname % \relax if undefined
+ \ifx \@tempa\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for option structure%
+ }{%
+Possible valid values are `hierarchic' and `flat'.\MessageBreak
+The predefined value is `hierarchic'.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \let\cweb@structure\@tempa
+ \fi
+ }
+\setkeys{cweb}{structure=hierarchic}
+%%% \endprog
+
+%%% Since usually this will be inserted in a `normal' LaTeX document, I
+%%% stick to the hierarchic structure.[ema@csn.es]
+%%% \beginprog
+\or
+\def\cweb@structure{0}
+\fi
+%%% \endprog
+
+
+%%% \sect With the set option |suppress| one can select suppression of
+%%% different document parts: |changehints| suppresses output of hints
+%%% that a changefile was involved, |unchanged| output of unchanged
+%%% chunks, |index| and |reflist| output of index and reference list, and
+%%% |format| output of `|@f|' directives.
+
+%%% The predefined value is no suppression. There is no default value.
+%%% %
+%%% \begin{fixme}
+%%% Shouldn't I supply a default value? `|{index,reflist}|'?
+%%% \end{fixme}
+
+%%% Set options are represented by flags named
+%%% |@cweb@|\<option>|@|\<value>|@|.
+
+%%% This options also receive a predetermined value for the `package' case
+%%% [ema@csn.es]
+
+
+%%% \beginprog
+%% \define@key@set{cweb}{suppress}{changehints,unchanged,index,reflist,format}
+\@for \@tempa :=changehints,unchanged,index,reflist,format\do {%
+ \expandafter\newif \csname if@cweb@suppress@\@tempa @\endcsname
+ \csname @cweb@suppress@\@tempa @false\endcsname
+ }%
+%%% \endprog
+%%% \beginprog
+\ifcase \cweb@cls
+%%% \endprog
+%%% \beginprog
+\define@key{cweb}{suppress}{%
+ \@for \@tempa :=#1\do {%
+ \expandafter\let \expandafter\@tempb
+ \csname @cweb@suppress@\@tempa @true\endcsname
+ %% \relax if undefined
+ \ifx \@tempb\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for set option suppress%
+ }{%
+Possible valid values are `changehints', `unchanged',\MessageBreak
+or a comma-separated list of these words enclosed in braces.\MessageBreak
+Specified values are turned on, you can't turn them off.\MessageBreak
+No value is on initially.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \@tempb
+ \fi
+ }%
+ }
+\g@addto@macro\ProcessKeyValOptionsHook{%
+ \if@cweb@suppress@unchanged@
+ \@cweb@suppress@changehints@true
+ \fi
+ }
+%%% \endprog
+
+%%% For the package case, I select to suppress the Index, the List of
+%%% Refinements and the change hints. [ema@csn.es]
+%%% \beginprog,
+ \or
+ \@for \@tempa :=changehints,index,reflist\do {%
+ \csname @cweb@suppress@\@tempa @true\endcsname
+ }%
+ \@for \@tempa :=unchanged,format\do {%
+ \csname @cweb@suppress@\@tempa @false\endcsname
+ }%
+ \fi
+%%% \endprog
+
+
+%%% \sect The base document class may be specified by the option
+%%% |baseclass|.
+
+%%% This class will not be loaded if we are a package.[ema@csn.es]
+%%% \beginprog
+\ifcase\cweb@cls
+%%% \endprog
+%%% \beginprog
+\def\CwebBaseClass{article}
+\define@key{cweb}{baseclass}[report]{\def\CwebBaseClass{#1}}
+%%% \endprog
+
+%%% Nothing to do for packages.[ema@csn.es]
+%%% \beginprog
+\fi
+%%% \endprog
+
+
+%%% \sect Language-specific adaptions are stored in files
+%%% \texttt{cwbl-{\it language}.sty}. They may be accessed as packages or
+%%% by the |language| option that takes the \textit{language} as value. We
+%%% have to require the respective package at the end of the class, then
+%%% all cseqs with default names are defined and may be overwritten.
+
+%%% For some languages, we provide Babel-compatible option names, to ease
+%%% the usage of global class options taken up by other packages (like
+%%% \pkg{varioref}), too.
+
+%%% \beginprog
+\def\cweb_set_language#1{%
+ \IfFileExists{cwbl-#1.sty}{%
+ %% true
+ \RequirePackage{cwbl-#1}%
+ }{%
+ %% false
+ \ClassError{cweb}{Language `#1' is unknown.}{%
+Language definitions are stored in files `cwbl-<language>.sty'.\MessageBreak
+I could not locate the file `cwbl-#1.sty'. If you didn't make a%
+\MessageBreak
+typo, please check if there is such a file in a directory searched\MessageBreak
+by LaTeX.%
+ }%
+ }%
+ }
+
+\define@key{cweb}{language}{\AtEndOfClass{\cweb_set_language{#1}}}
+
+\define@key{cweb}{german}[]{\setkeys{cweb}{language=german}}
+\define@key{cweb}{french}[]{\setkeys{cweb}{language=french}}
+\define@key{cweb}{italian}[]{\setkeys{cweb}{language=italian}}
+\define@key{cweb}{spanish}[]{\setkeys{cweb}{language=spanish}}
+%%% \endprog
+
+
+
+
+%%% \chap Configuration, inheritance, and aggregation.
+
+%%% This chapter sets up access to other modules, therefore we restore the
+%%% original catcodes for the time being.
+
+%%% \beginprog
+\catcode`\_=\CatUsCode
+%%% \endprog
+
+
+%%% \sect As usual, we provide the possibility to configure this class.
+
+%%% \beginprog
+\InputIfFileExists{cweb.cfg}{%
+ \typeout{******************************************^^J%
+ *^^J%
+ * Using local configuration file cweb.cfg^^J%
+ *^^J%
+ ******************************************%
+ }%
+ }{}
+%%% \endprog
+
+
+%%% \sect Inherit article class.
+
+%%% \beginprog
+\ifx \CwebBaseClass\undefined
+ \def\CwebBaseClass{article}
+\fi
+\define@key{cweb}*[]{\PassOptionsToClass{\KV@key}{\CwebBaseClass}}
+\ProcessKeyValOptions{cweb}
+%%% \endprog
+
+%%% The class will not be loaded if we are a package.[ema@csn.es]
+%%% \beginprog
+\ifcase\cweb@cls
+\LoadClass{\CwebBaseClass}
+%%% \endprog
+
+%%% Nothing to do for packages.[ema@csn.es]
+%%% \beginprog
+\fi
+%%% \endprog
+
+
+%%% \sect OK, now comes our own code; switch on again our lexical
+%%% conventions.
+
+%%% \beginprog
+\catcode`\_=\CatLetter
+%%% \endprog
+
+
+
+
+%%% \chap Typesetting \cweave{} output.
+
+%%% This task has been out-sourced to an other module, \pkg{cwebbase}. As
+%%% mentioned at the start of the implementation, there remain mainly some
+%%% bigger tasks, they will be implemented in further sections below. For
+%%% the most part, the necessary preconditions are actually our class
+%%% option definitions, we don't have to do anything. (Gosh, what a
+%%% coincidence. :--) We need to tell \pkg{cwebbase}, that it should use
+%%% class warnings on problems.
+
+%%% \beginprog
+\let\cweb_warning=\ClassWarningNoLine
+\input{cwbasehy}
+%%% \endprog
+
+
+
+%%% \sect But there remain also some small things to be done right now:
+%%% For flat structure, the counter |secnumdepth| must be set to~11, then
+%%% we want to number chunks in the whole document.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \or
+ %% flat
+ \setcounter{secnumdepth}{11}
+\fi
+%%% \endprog
+
+
+%%% \sect \pkg{cwebbase} uses |\part|, |\chapter| (if bound), and
+%%% |\section|. We have no use for any other section division that came
+%%% from the base class, let's discard the standard ones.
+
+%%% Unless we are a package, in which case we don't know if there will be
+%%% sections using them later.
+%%% \beginprog
+\ifcase\cweb@cls
+%%% \endprog
+%%% \beginprog
+\let\subsection\undefined
+\let\subsubsection\undefined
+\let\paragraph\undefined
+\let\subparagraph\undefined
+%%% \endprog
+
+%%% Nothing to do for packages.[ema@csn.es]
+%%% \beginprog
+\or\fi
+%%% \endprog
+
+
+
+
+%%% \chap Table of contents.
+
+%%% The table of contents features only sections, chunks don't have
+%%% titles, after all. Titles are indented proportionally to the rank,
+%%% with an basic indentation of |\CwebTocIndent|. An entry on major
+%%% ranks may be differentiated by layout instead of indentation, as it's
+%%% done in standard \LaTeX{} classes, too.
+
+%%% The counter |\CwebTocIndentMaxLevel| constitutes an upper limit for a
+%%% recognized indent (concerning indentation, that is).
+
+%%% \beginprog
+\newdimen\CwebTocIndent
+ \CwebTocIndent=1.5em
+\newcount\CwebTocIndentMaxLevel
+ \CwebTocIndentMaxLevel=6 % increase indentation if <= max level
+%%% \endprog
+
+
+%%% \sect For hierarchic structure, we can ignore chunk entries
+%%% completely, but have to supply a bunch of definitions for entries of
+%%% type `|section|$l$'. But for flat structure, sections are passed as
+%%% chunks, too. Then the current rank is stored in the title, as
+%%% |\cwbbRank{|$r$|}|. These differences are so large that we'll have
+%%% to make a very large |\ifcase|, over many chunks in this macro source
+%%% file. Let's start with the hierarchic stuff.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\l@chunk#1#2{}
+%%% \endprog
+
+
+%%% \sect Let's assume that we're subclassed from one of \LaTeX{}'s
+%%% standard classes; otherwise this stuff must be redefined, too. They
+%%% indent |\section| entries always by the same amount (1.5\,em), our
+%%% layout must fit to that.
+
+%%% If we have chapters, they are also indented by 1.5\,em and
+%%% distinguished by layout. The first minor section starts on rank~3 and
+%%% is placed on level~2, it should be indented larger. Let's simply add
+%%% 1.5\,em for each rank, that makes 3\,em for sections of type
+%%% |section2|. I.e., we use $f=l$ as a factor for our basic indentation.
+
+%%% If we don't have chapters, the first minor section starts on rank~2
+%%% and is placed also on level~2, it should be indented like sections
+%%% (1.5\,em), as it will be in another layout. I.e., we use $f=l-1$ as
+%%% the indentation factor.
+
+%%% The actual definitions are generated in a loop. `|\l@section|$l$'
+%%% expands to `|\cweb@toc_entry{|$l$|}|'. The indentation will be
+%%% computed within that macro.
+
+%%% \beginprog
+ \count@=2
+ \@whilenum \count@<11 \do {
+ \expandafter\edef \csname l@section\number\count@\endcsname{%
+ \noexpand\cweb@toc_entry{\number\count@}%
+ }
+ \advance \count@ by 1
+ }
+%%% \endprog
+
+
+%%% \sect A toc entry may be produced by |\@dottedtocline|. That cseq
+%%% needs the level (to decide if that entry is produced after all), the
+%%% entry indentation, the inner-entry indentation used for the number,
+%%% entry text, and page number.
+
+%%% As explained above, we use an indentation of $f \cdot
+%%% |\CwebTocIndent|$, if the level is not larger than
+%%% |\CwebTocIndentMaxLevel|. Otherwise we use the indentation of that
+%%% level.
+
+%%% Each entry is kind of an item with the number as the label. We need an
+%%% inner-entry indentation to be used for subsequent lines within this
+%%% entry. If a number is specified, it's tagged by |\numberline| in the
+%%% entry text. With a proper local definition and a pro-forma evaluation
+%%% of the text, we get at the width of the number (with a fitting space
+%%% for separation to text) and can use that as the inner-entry
+%%% indentation.
+
+%%% \beginprog
+ \def\cweb@toc_entry#1#2{% % #3 (page) will follow
+ %% \count@ == factor for basic indentation
+ %% min(#1,TocIndentMaxRank) - correct_level
+ \ifnum #1>\CwebTocIndentMaxLevel
+ \count@\CwebTocIndentMaxLevel
+ \else
+ \count@ #1 % <-- space!
+ \fi
+ \ifx \chapter\undefined
+ \advance \count@\m@ne
+ \fi
+ %% \dimen@ == inner-entry indentation
+ %% width of number & space, if existing
+ \begingroup
+ \global\dimen@ \z@ % assert value
+ \def\numberline##1{\settowidth{\global\dimen@}{##1\enskip}}%
+ \setbox\z@ \hbox{#2}% % evaluates \numberline if specified
+ \endgroup
+ \@dottedtocline
+ {#1}% % level
+ {\count@ \CwebTocIndent}% % basic indent
+ \dimen@ % num width
+ {#2}% % entry; page will follow
+ }
+%%% \endprog
+
+
+%%% \sect We're done with toc configuration for hierarchic structure,
+%%% since entry definitions for |chapter| and |section| types exist
+%%% already. Let's turn over to the flat structure.
+
+%%% \beginprog
+ \or
+ %% flat
+%%% \endprog
+
+
+%%% \sect The section rank determines how they are featured in the table.
+%%% Titles on rank~0 are typeset boldface, other titles are dotted
+%%% contents lines in roman.
+
+%%% The rank is specified within the title, i.e., in the
+%%% first argument to |\l@chunk|, as an argument for the |\cwbbRank| tag.
+%%% With an appropriate definition we might set |\count@| to the
+%%% respective value. If it is not set, an explicite invocation of
+%%% |\section| or |\addcontentsline| is responsible for this entry. Then
+%%% we assume that the rank is~0.
+
+%%% \beginprog
+ \def\l@chunk#1{% % page will be processed later
+ \count@\@ne % default value of group level
+ \begingroup
+ \let\numberline\@gobble % width not known yet
+ \def\cwbbRank##1{\global\count@ ##1\relax}%
+ \setbox\z@ \hbox{#1}%
+ \endgroup
+ \ifnum \count@=\z@
+ \let\next\cweb@bold_toc_line
+ \else
+ \let\next\cweb@normal_toc_line
+ \fi
+ \next{#1}%
+ }
+%%% \endprog
+
+
+%%% \sect A boldface line is typeset similar to the definition of
+%%% |\@dottedtocline|, only with fewer parameters. It takes two
+%%% parameters, the entry and the page number.
+
+%%% \beginprog
+ \def\cweb@bold_toc_line#1#2{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1em plus\p@}%
+ \begingroup
+ \noindent \bf
+ \hangindent\CwebTocIndent
+ \rightskip\@tocrmarg \parfillskip -\rightskip
+ \interlinepenalty\@M
+ \@tempdima\CwebTocIndent % for \numberline
+ #1\nobreak\hfill \hbox to\@pnumwidth{\hss #2}%
+ \par
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect A normal line uses |\@dottedtocline| like the hierarchic stuff.
+%%% Rank~1 shall still be on the same level as the bold toc line, so we
+%%% have to use $r-1$ as the proportional factor for |\CwebTocIndent|.
+
+%%% \beginprog
+ \def\cweb@normal_toc_line#1{% % page will be processed later
+ \edef\next{\noexpand\@dottedtocline{\the\count@}}% % toc entry level
+ \ifnum \count@>\CwebTocIndentMaxLevel
+ \count@ \CwebTocIndentMaxLevel
+ \fi
+ \advance\count@\m@ne % group level -= 1
+ \next{\count@\CwebTocIndent}% % basic indent
+ \CwebTocIndent % numwidth
+ {#1}% % entry
+ }
+%%% \endprog
+
+
+%%% \sect The first argument of |\@dottedtocline| is used to specify a
+%%% depth of the issued entry. All entries with a depth larger than
+%%% |tocdepth| are discarded. We typeset all entries, our |tocdepth|
+%%% value is the largest possible section level.
+
+%%% \beginprog
+ \setcounter{tocdepth}{10}
+%%% \endprog
+
+
+%%% \sect That's it. Close our distinction between hierarchic and flat
+%%% structure.
+
+%%% \beginprog
+\fi
+%%% \endprog
+
+
+
+
+%%% \chap The end of the document.
+
+%%% The end of a \cweave{} document is a rather complicated thing. It is
+%%% described in detail in the \cweave{} interface specification, we repeat the
+%%% most important points shortly.
+%%% %
+%%% \begin{itemize}
+
+%%% \item The user has issued an |\end{document}| tag in the documentation
+%%% part of the last chunk.
+
+%%% This tag is therefore part of a chunk, the chunk end (i.e., the
+%%% |\fi| cseq) does appear later in the document. Particularly, this
+%%% means that at the accurence of this tag there is still an open |\if|.
+
+%%% \item Between the |\end{document}| and the |\fi| a lot of text may
+%%% come which must be ignored. Within this text no |\fi| is allowed.
+
+%%% \item After the |\fi| the document may be finished by |\end|.
+%%% (Then the |-x| option of \cweave{} was used.)
+
+%%% Or additional information about the whole document is added: An
+%%% optional tag |\ch|, two mandatory tags |\inx\fin|, and another
+%%% optional tag |\con|.
+
+%%% \end{itemize}
+
+%%% Btw, the last chunk is marked as changed if any changefile entry has
+%%% been applied in the document. (In Plain \cweb{} that chunk is
+%%% colloquially used as the introduction to the index, and that shall be
+%%% printed always.) If we suppress unchanged chunks, the last
+%%% chunk will therefore always be processed, independent if it was really
+%%% changed or not. Yet another reason to use a whole chunk just for
+%%% |\end{document}|.
+
+%%% We'll use a the following approach for the implementation of the
+%%% iidentifier index and the refinement list (remember that we do not
+%%% produce a table of contents): We will check if there is any of these
+%%% two lists available. The available ones are typeset. Then we'll
+%%% finish the document. This functionality will be bound to |\fin|.
+%%% |\inx| will be simply a no-op. |\con| will never be reached, so we
+%%% don't have to cope with it.
+
+
+%%% \sect With the introduction of an end-document hook in \LaTeXe{}, one
+%%% might have hoped that we can utilize that hook for our task. But this
+%%% is not possible: We don't want to add some additional actions
+%%% \emph{into} the end-document handling, we want to look ahead and issue
+%%% lots of commands \emph{before} the end-document handling has happened.
+%%% (If you know \TeX{} by heart, you may skip the rest of the explanation
+%%% and go to the next chunk.)
+
+%%% As a macro language, \TeX{}'s cseqs take those tokens as arguments
+%%% that are behind them in the token stream. The token stream is
+%%% initially the document source, evaluation of a macro pushes its
+%%% expansion (a token list) on front of the stream. I.e., the evaluation
+%%% of |\end| will push the expansion on the token stream before
+%%% |\enddocument| is even looked at. And |\enddocument| will also push
+%%% its expansion before the hook is evaluated.
+
+%%% If we would allow |\enddocument| to push its expansion, all these
+%%% tokens have to be evaluated after we have typeset the backmatter --
+%%% after all, we need a proper document end, don't we? One would need to
+%%% skip over them, look at and eventually evaluate tokens that come
+%%% behind (from the initial document stream, the document source), and
+%%% then evaluate the skipped tokens. That's too hairy, we don't know
+%%% exactly how many tokens to skip, this is probable to change in future
+%%% \LaTeX{} revisions, etc.
+
+%%% The pushed tokens from |\end| don't pose so much problems. As
+%%% |\enddocument| will terminate the document processing, they are not
+%%% evaluated anyhow, we can just throw them away. And |\end| has changed
+%%% only once in the last few years, so it seems to be a stable macro --
+%%% necessary as we need to know its expansion for propper skipping.
+
+%%% A very first consequence of the demand to process tokens after
+%%% |\end{document}| is that we have to save the ``standard'' \LaTeX{}
+%%% document termination, since we will execute that later.
+
+%%% \beginprog
+\ifcase\cweb@cls
+\let\cweb@save@enddocument=\enddocument
+\or
+\let\cweb@save@enddocument=\relax
+\fi
+%%% \endprog
+
+
+%%% \sect When we encounter |\end{document}| we have to skip tokens
+%%% until we reach a |\fi|. Then we establish the correct
+%%% environment for the material behind.
+
+%%% The skip itself is actually a bit difficil, so we'll look at it more
+%%% closely. |\enddocument| is executed within |\end|. As mentioned above,
+%%% the original
+%%% definition of |\enddocument| cheats, it terminates \LaTeX{} before
+%%% the rest of |\end| is executed. |\end| first checks if we're in
+%%% the correct current environment, then it establishs the end
+%%% processing for paragraph building environments, closes the
+%%% group---each environment is in its own group, executes the end
+%%% processing established before, and it may ignore following white
+%%% space.
+
+%%% The |document| pseudo-environment is different in that it isn't in a
+%%% group. I.e., we have to open a group which can be closed by |\end|
+%%% afterwards. The environment check will be done in this group. With
+%%% |\aftergroup| we establish a macro which will skip the rest of the
+%%% |\end| processing. This processing is not needed anyhow: |document|
+%%% is no paragraph building environment and the following white space is
+%%% ignored by ourselves. Since |document| is no paragraph building
+%%% environment, the end processing described above is empty, i.e., not
+%%% existant. Then come a few tokens which ignore the white space. One of
+%%% these tokens is a |\fi|, so we have to skip them first separately.
+%%% They are discarded by |\cweb@skip_end|, we can start to look
+%%% for our |\fi| afterwards.
+
+%%% If |\end{document}| is the first (i.e., the only) text in this chunk
+%%% we discard the chunk start marker. This is possible since it isn't set
+%%% until now, it's in the |\everypar| token register. Of course, then we
+%%% have to do the house keeping work stored in |\everypar| ourselves.
+
+%%% No need of this for packages. [ema@csn.es]
+%%% \beginprog
+\ifcase\cweb@cls
+%%% \endprog
+
+%%% \beginprog
+\def\enddocument{%
+ \if@noskipsec
+ \global\everypar{}%
+ \global\@noskipsecfalse
+ \global\@nobreakfalse
+ \global\clubpenalty\@clubpenalty
+ \fi
+ \begingroup
+ \aftergroup\cweb@skip_end
+ }
+%%% \endprog
+
+%%% [ema@cwn.es]
+%%% \beginprog
+\fi
+%%% \endprog
+
+%%% \sect Let's make sure that the expansion of |\end| is the one that we
+%%% expect. Then we can define |\cweb@skip_end| appropriately, we have to
+%%% skip 5~tokens.
+
+%%% \beginprog
+\CheckCommand*\end[1]{%
+ \csname end#1\endcsname\@checkend{#1}%
+ \expandafter\endgroup\if@endpe\@doendpe\fi
+% 1 2 3 4 5
+ \if@ignore\@ignorefalse\ignorespaces\fi}
+\def\cweb@skip_end#1#2#3#4{\cweb@check_fi}
+%%% \endprog
+
+
+%%% \sect |\cweb@check_fi| gets the next token, and invokes
+%%% |\cweb@do_check_fi| who really checks it. If this token is |\fi| we
+%%% can prepare to finish our document. Otherwise we have to check the
+%%% next token.
+
+%%% Of course, first the token we have checked already must be discarded.
+%%% Therefore we insert the followup action by |\afterassignment|. It's
+%%% important that the assignment gets \emph{only} the next token, we
+%%% have to supply both the equals sign and the optional blank which are
+%%% allowed between the both arguments to |\let|.
+
+%%% \beginprog
+\def\cweb@check_fi{%
+ \futurelet\next \cweb@do_check_fi
+ }
+\def\cweb@do_check_fi{%
+ \ifx \next\fi
+ \afterassignment\cweb@finish
+ \else
+ \afterassignment\cweb@check_fi
+ \fi
+ \let\next= % <-- blank!!
+ }
+%%% \endprog
+
+
+%%% \sect When we found the |\fi|, we also discard it. But we're still
+%%% within a chunk, i.e., there is an open |\if|. So we must supply the
+%%% |\fi|.
+
+%%% \TeX{}nical note: |\next| is now equivalent to |\fi|. It must not be
+%%% used within a loop where it denotes the tail recursion's action. I.e.,
+%%% code like
+%%% %
+%%% \begin{verbatim}
+%%% \ifx \foo\bar
+%%% \let\next\relax
+%%% \else
+%%% \let\next\continue
+%%% \fi
+%%% \next
+%%% \end{verbatim}
+%%% %
+%%% will fail. This trap is closed by a redefinition of |\next|.
+
+%%% The output of all following lists may be suppressed by some option.
+%%% That is tested at the place where the respective list shall be
+%%% produced.
+
+%%% All following lists are typeset in CR state.
+
+%%% \beginprog
+\def\cweb@finish{%
+ \fi
+ \let\next\relax
+ \let\ch\cweb@changed_list
+ \let\inx\relax
+ \let\fin\cweb@end_document
+ \let\end\cweb@save@enddocument
+ \cwbb@CR
+ }
+%%% \endprog
+
+%%% But, if we are inside another document, we would not like to have all
+%%% this. As they are disabled from the arguments, I will skip
+%%% them.[ema@csn.es]
+
+%%% \beginprog
+\ifcase\cweb@cls\or
+\let\con\relax %<- Not sure!
+\let\inx\relax
+\let\fin\relax
+\fi
+%%% \endprog
+
+%%% \sect The following lists are started with unnumbered section
+%%% divisions. They are either chapters or sections.
+
+%%% \beginprog
+\ifx \chapter\undefined
+ \let\CwebListSectionDiv=\section
+\else
+ \let\CwebListSectionDiv=\chapter
+\fi
+\def\CwebListSection{\CwebListSectionDiv*}
+%%% \endprog
+
+
+%%% \sect The list of changed chunks is a cross reference list, (nearly)
+%%% like all others at the end of a chunk. The only difference is that
+%%% we do not show changeflags any more---each chunk number in this list
+%%% carries a change flag by definition.
+
+%%% \TeX{}nical note: The redefinition of |\*| is part of the second
+%%% argument of |\CwebCrossRef|. It is \emph{not} a global redefinition.
+
+%%% \beginprog
+\def\CwebCRChanged{%
+ \CwebCrossRef{The following chunks were changed by the change file:}%
+ }
+\def\CwebChangedListName{Changed Chunks}
+
+\if@cweb@suppress@changehints@
+ \def\cweb@changed_list#1.{}
+\else
+ \def\cweb@changed_list{%
+ \CwebListSection{\CwebChangedListName}%
+ \@mkboth{\uppercase{\CwebChangedListName}}%
+ {\uppercase{\CwebChangedListName}}%
+ \message{\CwebChangedListName}%
+ \let\CwebCRSize\relax
+ \CwebCRChanged \let\*\relax
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect As outlined above, we want to check if there are any entries in
+%%% the identifier index or the refinement list. The former is stored in
+%%% the file |\jobname.idx|, the latter in the file |\jobname.scn|.
+
+%%% We write a macro |\cweb@if_file_not_empty| that's like |\IfFileExists|.
+
+%%% \beginprog
+\def\cweb@end_document{%
+ \cweb@if_file_not_empty{\jobname.idx}{%
+ \if@cweb@suppress@index@ \else
+ \CwebIdIndex
+ \fi
+ }{}%
+ \cweb@if_file_not_empty{\jobname.scn}{%
+ \if@cweb@suppress@reflist@ \else
+ \CwebRefList
+ \fi
+ }{}%
+ \cweb@save@enddocument
+ }
+%%% \endprog
+
+
+%%% \sect The paragraph layout for both the index and the refinement list
+%%% is much the same, it is taken from Plain \cweb{}: Each entry is a
+%%% paragraph, nearly no skip between the paragraphs (just a bit to
+%%% prevent underfull vboxes), no paragraph indentation, ragged right.
+%%% Overfull hboxes in the lists don't make sense, so we prevent them. And
+%%% we don't allow hyphenation.
+
+%%% \beginprog
+\def\cweb@list_par_layout{%
+ \parskip \z@ plus .5\p@
+ \parindent\z@
+ \rightskip \z@ plus 2.5em
+ \tolerance\@M \hyphenpenalty\@M
+ }
+%%% \endprog
+
+
+%%% \sect We consider a file as ``not empty'' if (1)~the file exists,
+%%% (2)~is not empty, and (3)~has a non-empty line at the very front. (In
+%%% fact, an empty line at the front means that there should not be
+%%% anything behind it---but we can't test this portably.)
+
+%%% |\cweb@if_file_not_empty| tests the property, it's parametrized by the
+%%% file name and gets passed actions for the true and the false case.
+
+%%% We open the file first. Then we check if it doesn't exist or if it's
+%%% empty, both conditions deliver true on |\ifeof|; in this case we
+%%% pretend that there was an empty line. (The \LaTeX{} kernel already
+%%% once provided a macro for an empty line, |\@defpar|, we repeat that
+%%% definition here since it was discarded there.) Otherwise we read the
+%%% first line. At this state, the emptiness of the first line is
+%%% equivalent to the non-availability of entries, we can easily construct
+%%% an appropriate macro call to set the flag.
+
+%%% \beginprog
+\def\@defpar{\par}
+\def\cweb@if_file_not_empty#1#2#3{%
+ \openin\@inputcheck #1\relax
+ \ifeof \@inputcheck
+ \let\next\@defpar
+ \else
+ \read\@inputcheck to \next
+ \fi
+ \closein\@inputcheck
+ \ifx \next\@defpar
+ \def\reserved@a{#3}%
+ \else
+ \def\reserved@a{#2}%
+ \fi
+ \reserved@a
+ }
+%%% \endprog
+
+
+%%% \sect The identifier index is available in the file |\jobname.idx|.
+%%% The setup for the index is a mixture of the |theindex| environment of
+%%% the \cls{article} class and DEK's index macros. It's typeset in two
+%%% columns; the user may specify an introductionary text for the index
+%%% by |\cwebIndexIntro|. If there is any introductionary text we add a
+%%% medium skip below.
+
+%%% For the index, |\parfillskip| set in such a way that almost empty
+%%% lines are avoided.
+
+%%% Before we read in this file, we have to bind the special cseqs used
+%%% therein.
+
+%%% \begin{fixme}
+%%% \LaTeX{} feature or bug? The optional argument of |\twocolumn| is
+%%% typeset as a vbox, therefore the space at top of the section heading
+%%% is not discarded at the top of a page. But we know that |\twocolumn|
+%%% will start a page in any case, we must prevent the addition of this
+%%% space. Looking in the source, we see that setting the flag
+%%% |@nobreak| will prevent this addition, let's do that as a
+%%% workaround. File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \begin{fixme}
+%%% Surprise: If |@nobreak| is set \emph{and} if there are no index
+%%% entries, |\twocolumn| will drop its optional argument. (That
+%%% happened to be noticed when there was an error in
+%%% |\cweb@if_file_not_empty|.) File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \begin{fixme}
+%%% If |\CwebListSection| is a chapter tag of the standard classes, that
+%%% macro will check if it's in twocolumn mode and will try to set
+%%% itself at the top of the current page in full width, as it would
+%%% have been given in an optional argument of |\twocolumn|. As the
+%%% method that optional argument uses (a float) must not be nested, we
+%%% must prevent this to happen. We set the |@twocolumn| flag to false,
+%%% chapter macros will then think they're evaluated in a one-column
+%%% document. It's debatable if that is a kludge or even \emph{the}
+%%% solution -- after all, within the optional argument we aren't in
+%%% twocolumn mode any more. File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \beginprog
+\newtoks\cwebIndexIntro
+ \cwebIndexIntro={}
+\def\CwebIdIndex{%
+ %% two column format
+ \columnseprule\z@
+ \columnsep 35\p@ % value is from article.sty
+ %% introduction of index
+ \twocolumn[%
+ \cwbb@tex
+ \@nobreaktrue % don't add space in front of section
+ \@twocolumnfalse % \CwebListSection might be \chapter
+ \CwebListSection{\indexname}%
+ \edef\intro{\the\cwebIndexIntro}% % is a local def
+ \ifx \intro\empty
+ \else
+ \the\cwebIndexIntro\unskip
+ \par\medskip
+ \fi
+ ]%
+ \@mkboth{\uppercase{\indexname}}{\uppercase{\indexname}}%
+ \message{\indexname:}% % tell the user what we're doing
+ %% paragraph layout
+ \cweb@list_par_layout
+ \parfillskip \z@ plus .6\hsize
+ %% bindings
+ \let\I\CwebIndexEntry
+ \let\[\CwebIndexDeclared
+ \let\*\cwbb@change_flag
+ %% read the index
+ \CwebReadIndex
+ }
+\def\CwebReadIndex{\@input{\jobname.idx}}
+%%% \endprog
+
+
+%%% \sect An index entry is typeset with the same hanging indentation like
+%%% a cross reference list.
+
+%%% The entry is tagged with |\.|, if it was entered by the \cweb{}
+%%% operator~`|@.|'. Then it shall be typeset as a string. But an indexed
+%%% name may also want to use |\.| as an accent. This is the same
+%%% situation as we had at the refinement names (in module
+%%% \pkg{cwebbase}), where we introduced |\cwbb@check_dot| to handle this
+%%% case. The same minor restriction as there holds here, a refinement
+%%% name may not consist of a single dot-accented expression. (|@:|~helps
+%%% in this singular case.)
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es})introduced come code to
+%%% insert hyperlinks (with the macro |\cweb@num@lnk|).
+
+%%% \beginprog
+\def\CwebIndexEntry#1, #2.{%
+ \par
+ \hangindent\CwebNumberListHangindent
+ \leavevmode
+ \cwbb@check_dot{#1}:\quad\cweb@num@lnk#2, |.
+ }
+%%% \endprog
+
+
+%%% \sect The chunks where identifiers are declared are noted with
+%%% underlined numbers. We also must not forget the default declaration of
+%%% |\9|, the tag for the user definable index layout.
+
+%%% \beginprog
+\def\CwebIndexDeclared#1]{$\underline{#1}$}
+\def\9#1{}
+%%% \endprog
+
+
+%%% \sect The list of the refinement names is available in the file
+%%% |\jobname.scn|. The layout is taken from the plain version: ragged
+%%% right, each entry is a paragraph, the different cross reference
+%%% categories are separated by a quad.
+
+%%% We must initialize our whole paragraph layout, as the index might have
+%%% been suppressed and no initialization has happened there then. If
+%%% there was an index, it changed |\parfillskip| and |\*|, let's restore it.
+
+%%% Enrique Mel\'endez (\texttt{ema@csn.es})introduced come code to
+%%% insert hyperlinks (also in the macro |\CwebCrossRefs|).
+
+%%% \beginprog
+\def\CwebRefListName{List of Refinements}
+
+\def\CwebRefList{%
+ %% paragraph layout: like in index, but
+ \onecolumn
+ \cweb@list_par_layout
+ \parfillskip\@flushglue % must be reset before section heading
+ \CwebListSection{\CwebRefListName}%
+ \@mkboth{\uppercase{\CwebRefListName}}%
+ {\uppercase{\CwebRefListName}}%
+ \message{\CwebRefListName:}%
+ %% different bindings
+ \def\I{\par \hangindent\CwebNumberListHangindent}%
+ \def\CwebCrossRef##1##2.{\quad %
+ {\reset@font\footnotesize ##1~\hyperlink{_hyperchunk_##2}{##2}.}}%
+ \def\CwebCrossRefs##1##2.{\quad %
+ { \let\save@et\ET%
+ \let\save@ets\ETs
+ \def\ET{, }%
+ \let\ETs\ET%
+ \edef\@crlst{##2}%
+ \let\LastList@Conjunction\save@et
+ \reset@font\footnotesize ##1~%
+ \expandafter\cweb@cr@lnk\@crlst, ].}}%
+ \let\Q\CwebXRCite
+ \let\Qs\CwebXRCites
+ \let\U\CwebXRUse
+ \let\Us\CwebXRUses
+ \let\*\cwbb@change_flag
+ %% Read the list, allow configuration first.
+ \csname CwebRefListHook\endcsname
+ \CwebReadPPList
+ }
+\def\CwebReadPPList{\@input{\jobname.scn}}
+%%% \endprog
+
+%%% \sect Extra definitions for the sake of internationalization.
+
+%%% \beginprog
+\def\CwebXRCite{\CwebCrossRef{Cited in chunk}}
+\def\CwebXRCites{\CwebCrossRefs{Cited in chunks}}
+\def\CwebXRUse{\CwebCrossRef{Used in chunk}}
+\def\CwebXRUses{\CwebCrossRefs{Used in chunks}}
+%%% \endprog
+
+
+
+%%% \chap The end.
+
+%%% Well, after all we're finished with this class. We must not forget to
+%%% restore the underscore catcode.
+
+%%% \beginprog
+\catcode`\_=\CatUsCode
+
+\endinput
+%%% \endprog
+
+
+%%% \sect I would like to thank those who helped me to improve this module.
+%%% % In particular, XXX provided XXXsubstantial parts of the code.
+
+%%% \textsc{Michael M\"uller} and \textsc{Zden\v{e}k Wagner} did thorough
+%%% checks that helped me to improve the alpha test version.
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+%%% \vskip \PltxPreSectSkip
+
+%%% \begin{rcslog}
+%%% $StyleLog: cweb.doc,v $
+%%% \Revision 3.6 1995/11/30 15:44:53 schrod
+%%% Still didn't work with version \mbox{$\langle$1994/12/01$\rangle$},
+%%% the copy of |\newif| was not complete.\\
+%%% Problem reported by Michael Seel \path|<seel@mpi-sb.mpg.de>|.
+
+%%% \Revision 3.5 1995/11/20 22:34:18 schrod
+%%% Add language support.\\
+%%% Triggered by Christian Kumpf \path|<smurf@igd.fhg.de>| and Andreas
+%%% Scherer \path|<scherer@physik.rwth-aachen.de>|.
+
+%%% Use my `official' (ACM) email address.
+
+%%% \Revision 3.4 1995/11/07 17:55:18 schrod
+%%% \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+%%% versions; make it work with the previous-to-last one, version
+%%% \mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+%%% \mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+%%% |\newif|, and |\hb@xt@| must not be used.\\
+%%% Problem reported by Laurent Desnogues
+%%% \path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+%%% -- add name).
+
+%%% \Revision 3.3 1995/09/17 14:15:36 schrod
+%%% Initialize list paragraph layout in refinement list, too. With the
+%%% introduction of index suppression, the initialization -- located in
+%%% the index code -- might have been skipped.
+
+%%% \Revision 3.2 1995/09/15 10:22:59 schrod
+%%% The \cls{cweb} class needs at least \LaTeX{} version
+%%% \mbox{$\langle$1994/12/01$\rangle$}.\\
+%%% Problem reported by John S. Robinson \path|jsrobin@umiacs.umd.edu|.
+
+%%% \Revision 3.1 1995/09/12 23:00:48 schrod
+%%% Moved all code that does the actual typesetting of \cweave{} tags and
+%%% is therefore also needed for a |cweb| environment. It's now an own
+%%% module named \pkg{cwebbase}.
+
+%%% \Revision 2.8 1995/09/12 11:44:33 schrod
+%%% Standard base classes with chapters (i.e., report \& book) produced an
+%%% error at the start of the index: |\chapter| may not be called in the
+%%% optional argument of |\twocolumn|. Add a workaround to |\twocolumn|.\\
+%%% Problem reported by Christopher Higgins \path|<C.Higgins@cit.gu.edu.au>|.
+
+%%% \Revision 2.7 1995/08/29 17:32:36 schrod
+%%% Add hook for refinement list.
+
+%%% Allow internationalization of refinement list.\\
+%%% (Problem reported by Christian Kumpf \path|<kumpf@igd.fhg.de>|.)
+
+%%% Cweb bundle, version 1.0.
+
+%%% \Revision 2.6 1995/08/29 15:19:50 schrod
+%%% Added (hint to non-existence of) description of the internal interface.
+
+%%% \Revision 2.5 1995/08/29 02:07:26 schrod
+%%% Discard dependencies on 10\,pt fonts.
+
+%%% Support suppression of format directives.
+
+%%% \Revision 2.4 1995/08/27 19:31:43 schrod
+%%% Discard superfluous space at the top of index. Spacing in refinement
+%%% heading was wrong. Repair test on empty auxilliary files. Make
+%%% section heading for index and refinement list divisions (more)
+%%% configurable. Suppression of index and reference list is supported.
+
+%%% Make configuration of change flag easier.
+
+%%% Put section title of changed chunks list in marks, and tell the user
+%%% that it got typeset.
+
+%%% \Revision 2.3 1995/08/27 17:24:46 schrod
+%%% Make usage of baseclass with chapters work.
+
+%%% \Revision 2.2 1995/08/27 13:26:22 schrod
+%%% Add possibility to suppress change hints. Suppression of unchanged
+%%% chunks suppress change hints as well, they are meaningless as all
+%%% printed chunks are changed by definition.
+
+%%% Realize dependencies in options by |\ProcessKeyValOptionsHook|.
+%%% Illegal values for enumeration options don't alter the current value
+%%% any more.
+
+%%% List of changed chunks is an unnumbered section now.
+
+%%% \Revision 2.1 1995/08/25 19:11:18 schrod
+%%% Add keyword-value option style, with new \pkg{keyvald} package.
+
+%%% Hierarchic strucutures are supported now, in addition to the flat
+%%% structure of the beta-test version. One can choose with an option. For
+%%% that step, the terminology was cleaned up, too: Chunks are not named
+%%% sections any more. (That change involved reimplementation of almost
+%%% all the structure and toc stuff.)
+
+%%% The chunk number supplied by \cweave{} is used now, not some computed
+%%% number. Change flags are printed, too.
+
+%%% One can suppress output of unchanged sections.
+
+%%% One can select the baseclass with an option. That may be used to use
+%%% \cls{report} or \cls{book} to get chapter-style layout. Of course,
+%%% using an arbitrary baseclass is dangerous, it must conform to the
+%%% conventions of \LaTeX{} standard classes.
+
+%%% \Revision 1.13 1995/08/08 00:14:29 schrod
+%%% Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+%%% my standard templates for that, no changes in functionality.
+
+%%% \Revision 1.12 1993/08/10 14:15:43 schrod
+%%% New page on main section only if group level $<$ |\cwebSecNoEject|.
+%%% Default for the latter is 3.
+
+%%% Document that logos will not be defined in this style file.
+
+%%% Copy of plain macros for |\CwebNumber| does not work. Repaired the
+%%% most important one (subscript must be accessed via |\sb|).
+%%% Incompatibility to NFSS will be addressed later.\\
+%%% (Problems reported by Zden\v{e}k Wagner \path|<wagner@csearn.bitnet>|.)
+
+%%% \Revision 1.11 1993/08/10 11:21:07 schrod
+%%% Reference to section number does not render a period after the
+%%% number any more.
+
+%%% \Revision 1.10 1993/08/09 20:08:20 schrod
+%%% |\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+%%% effect already.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.9 1993/08/09 18:05:28 schrod
+%%% Left shift operator wasn't defined correctly.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.8 1993/06/15 15:22:30 schrod
+%%% \textbf{Version 0.2:} Moved from |pub/incoming| to official place, in
+%%% Literate Programming Archive.
+
+%%% \Revision 1.7 1993/06/15 13:25:48 schrod
+%%% First attempt to make it a style: |article| is imported. Warns the
+%%% user if he tries to use |cweb| as a style option.
+
+%%% \Revision 1.6 1993/06/15 08:49:23 schrod
+%%% |\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+%%% causes problems if a |\PB| is within. Now I try hard not to evaluate any
+%%% tokens outside of my control.
+
+%%% Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+%%% need an own macro.
+
+%%% \Revision 1.5 1993/06/14 17:50:30 schrod
+%%% Handle a missing |\con|: |\inx| and |\con| are now never looked at.
+%%% Instead |\fin| checks for the existence of the lists and typeset them
+%%% if there are entries available.
+
+%%% \Revision 1.4 1993/06/14 15:54:18 schrod
+%%% Made it work with older versions of \LaTeX{}, too. The |\enddocument|
+%%% implementation of cweb.sty depends on the |\end| implementation; check
+%%% if the current implementation is one we know of (currently: 91-01-14
+%%% and 92-03-25).\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Add FSA diagram about processing states. CR state is also switched
+%%% to from \TeX{} state (that happens with |\ch| at the document end).
+
+%%% \Revision 1.3 1993/05/13 17:51:21 schrod
+%%% Refinements may also be filenames (`|@(|'). Then the complete name
+%%% consists of a |\.| macro call, which is handled now.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Made the detection of `|@.|' index entries more robust.
+
+%%% \Revision 1.2 1993/05/12 18:28:59 schrod
+%%% Adapted to recent changes of \cweave{} (of April 93):
+
+%%% Main sections have a group level, represented in the table of
+%%% contents. This changed the complete implementation of section tags.
+
+%%% New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+%%% |\CwebCombinedOp|.
+
+%%% \Revision 1.1 1993/04/09 15:00:37 schrod
+%%% Initial revision
+
+%%% \end{rcslog}
+
+
+
+%%% \end{document}
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%% Local Variables:
+%%% mode: LaTeX
+%%% TeX-brace-indent-level: 4
+%%% indent-tabs-mode: t
+%%% TeX-auto-untabify: nil
+%%% TeX-auto-regexp-list: LaTeX-auto-regexp-list
+%%% compile-command: "make cweb.cls"
+%%% End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/nodoc.tex b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/nodoc.tex
new file mode 100644
index 00000000000..fe9e4b8658a
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/cweb-hy/nodoc.tex
@@ -0,0 +1,121 @@
+% File nodoc.tex
+% By Enrique Mel\'endez, <UEI0560130010@eurociber.es> April, 14, 1999.
+% Freely distributable under, f. i. GNU.
+% I do this for fun, but if you use it, I would like no know how it performs
+% Please, keep the author notice.
+% Put \include nodoc.tex as the very first line of your documents.
+% No responsibility. Usde this at your own risk. (That is, it may not work
+% as you intended, but I never designed it to.)
+
+
+\makeatletter
+\ifx\section\undefined
+ \makeatother
+ \expandafter\endinput
+\else
+
+% Things we suppress are `only preamble'
+% documentclass
+\let\n@documentclass\documentclass
+\def\documentclass{\@ifnextchar[\@documentclass@{\@documentclass@[]}}
+\def\@documentclass@[#1]#2{\relax}
+
+% usepackage
+\let\n@usepackage\usepackage
+\def\usepackage{\@ifnextchar[\@usepackage@{\@usepackage@[]}}
+\def\@usepackage@[#1]#2{\relax}
+% The following 4 should be allowed to change. I don't know how.
+% author
+\let\n@author\author
+\def\author#1{\gdef \@author{#1}%\relax
+}
+% date
+\let\n@date\date
+\def\date#1{\relax}
+% title
+\let\n@title\title
+\def\title#1{\gdef \@title{#1}%\relax
+}
+% maketitle
+\let\n@maketitle\maketitle
+\def\maketitle{%
+ \vskip 2em%
+ \begin{center}%
+ \let \footnote \thanks
+ {\LARGE \@title \par}%
+ \vskip 1.5em%
+ {\large
+ \lineskip .5em%
+ \begin{tabular}[t]{c}%
+ \@author
+ \end{tabular}\par}%
+ \vskip 1em%
+ {\large \@date}%
+ \end{center}%
+ \par
+ \vskip 1.5em
+ \global\let\@title\@empty
+ \global\let\@author\@empty
+ \global\let\title\relax
+ \global\let\author\relax
+%\relax
+}
+\def\pagestyle#1{\relax}
+\def\thispagestyle#1{\relax}
+\renewenvironment{titlepage}
+ {%
+ \if@twocolumn
+ \@restonecoltrue\onecolumn
+ \else
+ \@restonecolfalse\newpage
+ \fi
+ \thispagestyle{empty}%
+ %\setcounter{page}\@ne
+ }%
+ {\if@restonecol\twocolumn \else \newpage \fi
+ %\if@twoside\else
+ %\setcounter{page}\@ne
+ %\fi
+ }
+% tableofcontents
+\let\n@tableofcontents\tableofcontents
+\def\tableofcontents{\relax}
+% The begindocument, where we rebind some of the above
+\let\@document@=\document
+\def\document{\endgroup\relax%
+\let\document\@document@%
+\let\documentclass\n@documentclass%
+\let\usepackage\n@usepackage%
+}
+% The enddocument, where some more things are rebound
+\let\@enddoc@=\enddocument
+\def\enddocument{%
+\let\tableofcontents\n@tableofcontents%
+\let\maketitle\n@maketitle
+\let\title\n@title
+\let\author\n@author
+\let\date\n@date
+\let\enddocument\@enddoc@%
+\begingroup}
+\fi
+
+%% Borrowed from LaTeX general format.
+\def\@ifnextchar#1#2#3{%
+ \let\reserved@d=#1%
+ \def\reserved@a{#2}\def\reserved@b{#3}%
+ \futurelet\@let@token\@ifnch}
+\def\@ifnch{%
+ \ifx\@let@token\@sptoken
+ \let\reserved@c\@xifnch
+ \else
+ \ifx\@let@token\reserved@d
+ \let\reserved@c\reserved@a
+ \else
+ \let\reserved@c\reserved@b
+ \fi
+ \fi
+ \reserved@c}
+\def\:{\let\@sptoken= } \: % this makes \@sptoken a space token
+\def\:{\@xifnch} \expandafter\def\: {\futurelet\@let@token\@ifnch}
+
+\makeatother
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/contrib/wagner/cwebzw.sty b/Master/texmf-dist/doc/latex/cweb-latex/contrib/wagner/cwebzw.sty
new file mode 100644
index 00000000000..cff2d7c254e
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/contrib/wagner/cwebzw.sty
@@ -0,0 +1,323 @@
+%% $Id: cwebzw.sty,v 1.2 1993/08/10 14:14:33 schrod Exp $
+%%----------------------------------------------------------------------
+%% Changed by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>
+%% Original received by mail:
+%% Date: Thu, 05 Aug 93 12:34:21 MDT
+%% From: Zdenek Wagner <WAGNER@CSEARN.BITNET>
+
+%% This is file ``cwebzw.sty'' written by Z. Wagner as an extension and
+%% correction of ``cweb.sty'' Revision 1.8
+
+%% This file is self-documenting. If you process this file through LaTeX,
+%% you will get the documentation. You can use MakeIndex to create an
+%% index of macros used. In your cweb documents you can either use it as
+%% is or you can remove all comment lines by your editor or the docstrip
+%% utility (or a program of your own).
+
+%% See chapter ``Introduction'' for other remarks.
+
+\def\cwebzwrevision{1.8} % Do not remove this line!
+
+% You can remove the next two lines
+\ifcat a\noexpand @\let\next\relax\else\def\next{%
+ \documentstyle[doc]{article}\MakePercentIgnore}\fi\next
+
+% \EnableCrossrefs
+% \CodelineIndex
+% \catcode`\_=12
+% \DoNotIndex{\cweb,\orig,\p} ^^A doc.sty cannot handle underscore
+% \DoNotIndex{\#,\$,\%,\&,\@,\\,\{,\},\^,\_,\~,\ }
+% \DoNotIndex{\@ne,\tw@,\bgroup,\egroup,\endinput,\end,\kern,\hbox,\vbox,\vtop}
+% \DoNotIndex{\aftergroup,\arabic,\begin,\protect,\rm,\reset@font,\setbox}
+% \DoNotIndex{\show,\typeout,\errhelp,\unskip,\unvbox,\wd,\input,\hskip}
+% \DoNotIndex{\ifdim,\ifmmode,\ifnum,\long,\footnotesize,\newbox,\newpage}
+% \DoNotIndex{\@ifnextchar,\@ifundefined,\@namedef}
+% \DoNotIndex{\advance,\begingroup,\catcode,\closein,\closeout}
+% \DoNotIndex{\day,\def,\edef,\else,\empty,\endgroup,\errmessage}
+% \DoNotIndex{\expandafter,\fi,\futurelet,\gdef,\global,\if,\ifeof}
+% \DoNotIndex{\ifx,\immediate,\let,\loop,\m@ne,\message,\month,\newcount}
+% \DoNotIndex{\newif,\newlinechar,\newread,\newtoks,\newwrite}
+% \DoNotIndex{\noexpand,\openin,\openout,\par,\read,\relax,\repeat}
+% \DoNotIndex{\space,\the,\undefined,\write,\xdef,\year,\z@}
+% \advance\textwidth by 40pt
+% \def\pcws{{\tt cweb.sty}}
+% \def\cws{\protect\pcws}
+% \newbox\cwebincr
+% \setbox\cwebincr=\hbox{^^A
+% \kern .5pt
+% \raise .1em\hbox{$\scriptstyle{+}\kern -.1em{+}$}^^A
+% \kern .5pt}
+% \def\Cpp{C\copy\cwebincr}
+% \begin{document}
+% \MakeShortVerb\|
+% \title{Extension and correction of \cws\ revision~\cwebzwrevision}
+% \author{Zden\v{e}k Wagner\\
+% $\langle$wagner@csearn.bitnet$\rangle$ or
+% $\langle$wagner@earn.cvut.cz$\rangle$}\date{}
+% \maketitle\vspace{3ex}
+
+% \tableofcontents\thispagestyle{empty}\newpage
+
+% \section{Introduction}
+% The purpose of this file is to add new features to \cws\ and
+% correct some bugs. When doing this we intended not to make a faintest
+% change into \cws. Therefore some definitions may look strangely
+% cumbersome. However, this file should be treated as a test of new
+% features and bug fixes which should then be placed in the official
+% distribution of \cws\ and in the ideal state this file will one day
+% vanish at all.
+
+% {\bf Note by Joachim Schrod}:\quad I have incorporated most error
+% repairs and some of the enhancements into \cws{} (Revision~1.12). To
+% avoid problems, I deleted all the respective macros from this
+% option. The rest that you see here will not be incorporated (at
+% least not as-is) and is therefore provided as a contributed style
+% option for those who like it.
+
+
+% \section{Preliminaries}
+% First we make sure that this file is not input twice.
+% \begin{macrocode}
+\@ifundefined{\cweb@egroup}{}{\endinput}
+% \end{macrocode}
+
+% Throughout this file we will use tags defined in \cws. Therefore we
+% begin the file with identifying itself and displaying the revision of
+% \cws\ for which it is written. Afterwards we change catcode of
+% underscore in order to have the same conditions as \cws.
+% \begin{macrocode}
+\typeout{Extension of cweb style, revision \cwebzwrevision}
+\typeout{(if you have an older revision of cweb.sty, please upgrade)}
+\catcode`\_=\CatLetter
+% \end{macrocode}
+
+% \section{Extensions}
+% \begin{macro}{\C++}
+% \begin{macro}{\Cpp}
+% \begin{macro}{\CPLUSPLUS/}
+% The documentation of \cws\ says that |\C++| may now be used for \Cpp\
+% logo. However, |\C++| is not defined in \cws. The conservative users
+% of {\tt ccwebmac.tex} designed by Hans--Hermann Bode might be
+% accustomed to |\Cpp| while users of {\tt cwebmac.tex} from {\tt CWEB
+% 3.0} might like |\CPLUSPLUS/|. Therefore we supply all three tags and
+% moreover make them robust so that they can safely be used in moving
+% arguments.
+% \begin{macrocode}
+\def\p_cpp{C\CwebIncr}
+\def\C++{\protect\p_cpp}
+\def\Cpp{\protect\p_cpp}
+\def\CPLUSPLUS/{\protect\p_cpp}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+
+% \section{Improvements}
+% \begin{macro}{\CwebScope}
+% If your program contains description of member functions as
+% |TListBox::|\-|handleEvent|, the documentation part may easily contain
+% lots of overfull boxes. Therefore we try to allow the line break after
+% the scope operator. We make the change only if it is not typeset in
+% the math mode. The following solution is not perfect. The expression
+% ``|::close|'' may be incorrectly split after the scope resolution
+% operator. To make a robust macro we should probably fiddle with
+% |\lastskip| and |\lastpenalty|.
+% \begin{macrocode}
+\def\CwebScope{\ifmmode \kern.1em::\kern.1em\else
+ \kern.1em::\hskip.1em\fi}
+% \end{macrocode}
+% \end{macro}
+
+% \subsection{Index appearance}
+% When typesetting two or more columns it is more pleasant if the
+% columns are balanced. This may easily be achieved with Frank
+% Mittelbach's {\tt multicol.sty}. We try whether the style is
+% available.
+% \begin{macrocode}
+\newif\ifhave@multicol
+\openin\@ne multicol.sty
+\ifeof\@ne \else \have@multicoltrue \fi
+\closein\@ne \relax
+% \end{macrocode}
+
+% If we found {\tt multicol.sty}, we use it. It might me nice to delay
+% this until we know that the index is really requested.
+% \begin{macrocode}
+\ifhave@multicol \input{multicol.sty} \fi
+% \end{macrocode}
+
+% \begin{macro}{\CwebIdIndex}
+% \begin{macro}{\twocolumn}
+% \begin{macro}{\onecolumn}
+% We delay the redefinitions of the twocolumn format into the definition
+% of |\CwebIdIndex|. If we made the changes globally, we could
+% eventually break something in the user's code.
+
+% I dare say that I do not like the page break in between the index and
+% the list of refinements so I do not put it into the redefinition of
+% |\onecolumn|. Notice also that the new version of |\onecolumn|
+% restores everything for the rare case it is needed somewhere later.
+% \begin{macrocode}
+\let\orig_CwebIdIndex=\CwebIdIndex
+\let\orig_twocolumn=\twocolumn
+\let\orig_onecolumn=\onecolumn
+
+\ifhave@multicol
+ \def\CwebIdIndex{%
+ \def\twocolumn{\newpage
+ \@ifnextchar[{\zw@twocol}{\begin{multicols}{2}}}
+ \long\def\zw@twocol[##1]{\setbox\cweb_tempbox=\vbox{##1}%
+ \begin{multicols}{2}[\unvbox\cweb_tempbox]}
+ \def\onecolumn{%
+ \let\onecolumn=\orig_onecolumn
+ \let\twocolumn=\orig_twocolumn}
+ \orig_CwebIdIndex\cweb@restore_end\end{multicols}}
+\fi
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+
+% \begin{macro}{\cweb@restore_end}
+% The index is processed after the user supplied |\end{document}| where
+% the |\end| tag has unfortunately been redefined. We would always get a
+% \LaTeX\ error informing us about unmatched |\begin| and |\end|.
+% Therefore we prepare a macro which will be used inside a group to
+% restore temporarily the original meaning of |\end|.
+% \begin{macrocode}
+\let\orig_end=\end
+\def\cweb@restore_end{\let\end=\orig_end}
+% \end{macrocode}
+% \end{macro}
+
+% \section{Crazy inventions}
+% As we can see from the documentation of \cws, it can handle two
+% versions of \LaTeX. Older versions of \LaTeX\ are unlikely to be used
+% and \LaTeX3 will handle |\end|'s in a different way so the following
+% invention does not seem useful. I wrote it mainly as an exercise so
+% that I could proudly say with the White Knight from Through the
+% Looking Glass, ``It's my own invention.''
+
+% \begin{macro}{\cweb@end_tokens}
+% When skipping around the |\end{document}| we have to catch the tokens
+% which are placed after |\endgroup|. For this purpose we establish a
+% new token register.
+% \begin{macrocode}
+\newtoks\cweb@end_tokens
+% \end{macrocode}
+% \end{macro}
+
+% \begin{macro}{\cweb@report_error}
+% Before we start the job of catching tokens we prepare some macros. The
+% first of them will display our error message. Though we hope that the
+% catching algorithm will work with all \LaTeX's currently available, it
+% may according to Murphy's Laws happen that someone's obsolete (or
+% brand new) \LaTeX\ will break it. We do not want to bother the user
+% with strange error messages. We therefore supply ours. The text is
+% taken exactly from \cws. The macro will be issued from inside a group
+% therefore we take no care about restoring the original definitions.
+% \begin{macrocode}
+\def\cweb@report_error{\let\orig_errmsg=\errmessage
+ \def\errmessage##1{%
+ \errhelp={^^J%
+This style depends on an internal implementation detail of LaTeX.^^J%
+Sad to note, this detail has changed; you cannot use the cweb style.^^J%
+Please type "i\show\end" now. TeX will spit out a few lines of^^J%
+definitions and ask again for input. Type "x" then, this will finish TeX.^^J
+Then send the log file to the style's author, Joachim Schrod^^J%
+(Email: schrod@iti.informatik.th-darmstadt.de). He will adapt cweb.sty^^J%
+to the new LaTeX version.^^J}
+ \orig_errmsg{Sorry, you cannot use the cweb style with this LaTeX version}}
+}
+% \end{macrocode}
+% \end{macro}
+
+% \begin{macro}{\cweb@catch_tokens}
+% We also prepare the macro which will catch the tokens. This is an
+% incomplete assignment to the token register. Its function will be
+% explained later.
+% \begin{macrocode}
+\def\cweb@catch_tokens{\cweb@end_tokens=\bgroup}
+% \end{macrocode}
+% \end{macro}
+
+% \begin{macro}{\cweb_3_env}
+% \begin{macro}{\endcweb_3_env}
+% For catching the tokens we will invoke a very special environment with
+% the name |cweb_3_env|. \LaTeX\ would complain if the environment were
+% undefined. Therefore we place to |\cweb_3_env| everything which is
+% needed at the beginning of the catching job and similarly
+% |\endcweb_3_env| will contain the trap for the tokens. Notice that we
+% have to use |\@namedef| to define tags with such strange names.
+% \begin{macrocode}
+\@namedef{cweb_3_env}{\cweb@report_error}
+\@namedef{endcweb_3_env}{\aftergroup\cweb@catch_tokens}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+
+% \begin{macro}{\cweb@skip_end}
+% Now we can catch the tokens and define |\cweb@skip_end| accordingly.
+% First we write it and then explain how it works.
+% \begin{macrocode}
+\begin{cweb_3_env}
+{\egroup\end{cweb_3_env}}
+\expandafter\def\expandafter\cweb@skip_end\the\cweb@end_tokens{\cweb@check_fi}
+% \end{macrocode}
+% When we start the environment, \LaTeX\ will invoke our error reporting
+% macro. It is invoked inside a group as we have promised. The ending
+% command will tell \TeX\ that |\cweb@catch_tokens| should be carried
+% out after the end of group. When the group is closed (after some
+% checks), the tokens which stem from the expansion of |\end| are not
+% executed but stored in our token register. So we have to complete the
+% assignment. You might have temptation to just append |\egroup| after
+% the statement ending the environment. But beware! This will store
+% |\egroup| as the next token in |\cweb@end_tokens|. It will also eat up
+% the rest of the file. Fortunately we are redefining \cws\ so that the
+% official tags will do the job. We could start catching tokens with
+% |\bgroup| which is equivalent to \{ but we must put \} and nothing
+% else at the end of the assignment. We solve it by inserting an empty
+% group |{\egroup|. This is syntactically correct. In almost all
+% contexts \} and |\egroup| are interchangeable. So when \TeX\ scans the
+% text, it sees matched braces and when executing the macros it finds
+% balanced groups. The definition of |\cweb@skip_end| is now
+% straightforward. Isn't it a pretty nonsense code?
+% \end{macro}
+
+% \section{Bug corrections}
+% \begin{macro}{\egroup_}
+% A funny (and instructive) bug invaded into the definition of
+% |\CwebNumber|. This tag is defined in \cws\ where underscore is
+% recatcoded to letter. Therefore, what would work in a usual \TeX\
+% context, does not work here. We correct it by defining |\egroup_|
+% which will do the correct things. However, there is one more thing
+% which should be mentioned. When using NFSS, |\rm| has no effect in the
+% math mode. The easiest safe way for switching to roman alphabet from
+% inside math is to use |\hbox|. Placing |\,| as done in \cws\ seems
+% unpleasant to my eyes therefore I remove it. Please notice that the
+% catcode of underscore is restored to the original value within the
+% following definition.
+
+% {\bf Note from Joachim Schrod:}\quad This will not work any more
+% since the bug with the underscore has been repaired (one shouldn't
+% simply copy plain macros to \LaTeX{} code\,\dots). Nevertheless the
+% problem with NFSS remains. But I'm not (yet?)\ convinced that the
+% work-around presented here is the adequate solution. So the code is
+% here for reference, but it will not be used.
+
+% \begin{macrocode}
+\def\egroup_{\egroup\begingroup
+ \catcode`\_=\CatUsCode
+ \let\,=\relax
+ \def\rm{\reset@font\footnotesize}%
+ \cweb@egroup}
+
+\catcode`\_=\CatUsCode
+\def\cweb@egroup#1{_{\hbox{#1}}\endgroup}
+% \end{macrocode}
+% \end{macro}
+
+% \PrintIndex
+% \end{document}
+\endinput
+
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/cweb-conf.pdf b/Master/texmf-dist/doc/latex/cweb-latex/cweb-conf.pdf
new file mode 100644
index 00000000000..0978e160863
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/cweb-conf.pdf
Binary files differ
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/cweb-user.pdf b/Master/texmf-dist/doc/latex/cweb-latex/cweb-user.pdf
new file mode 100644
index 00000000000..4e655f848b7
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/cweb-user.pdf
Binary files differ
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/cwebbase.tex b/Master/texmf-dist/doc/latex/cweb-latex/cwebbase.tex
new file mode 100644
index 00000000000..dff1d3494b9
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/cwebbase.tex
@@ -0,0 +1,2080 @@
+% $StyleId: cwebbase.doc,v 1.4 1995/11/07 17:55:19 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+
+%
+% LaTeX macros cwebbase
+% support for incorporation of CWEAVE markup in LaTeX documents
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+
+
+
+%%%%
+%%%%
+%%%% These TeX macros were documented with the documentation system
+%%%% MAKEPROG and automatically converted to the current form.
+%%%% If you have MAKEPROG available you may transform it back to
+%%%% the original input: Remove every occurence of three percents
+%%%% and one optional blank from the beginning of a line and remove
+%%%% every line which starts with four percents. The following lex
+%%%% program will do this:
+%%%%
+%%%% %%
+%%%%
+%%%% ^%%%\ ? ;
+%%%% ^%%%%.*\n ;
+%%%%
+%%%% If you just want to print the documentation you may fetch
+%%%% the archive print-makeprog.tar.Z from ftp.th-darmstadt.de (directory
+%%%% pub/tex/latex). It contains *all* used styles -- but beware, they
+%%%% may not be in a documented form...
+%%%%
+%%%%
+%%% \documentclass{progltx}
+
+%%% \usepackage{cweb-doc} % document-specific markup
+%%% \usepackage{fullpage}
+
+
+%%% \RCS $StyleRevision: 1.4 $
+%%% \RCS $StyleDate: 1995/11/07 17:55:19 $
+
+
+%%% \begin{document}
+
+
+%%% \title{Typesetting \cweave{} Output}
+%%% \author{% % LaTeX does not discard unnecessary glue...
+%%% Joachim Schrod%
+%%% \thanks{%
+%%% \protect\raggedright
+%%% TU~Darmstadt, Computer Science Department, WG Systems Programming,
+%%% Alexanderstr.~10, D-64283~Darmstadt, Germany.
+%%% Email: \texttt{schrod@iti.informatik.th-darmstadt.de}.%
+%%% }%
+%%% }
+%%% \date{%
+%%% Revision \RCSStyleRevision\\
+%%% (as of \RCSStyleDate)%
+%%% }
+
+%%% \maketitle
+
+
+%%% % doesn't work with progltx yet
+%%% %\tableofcontents
+
+
+
+
+%%% \chap Introduction.
+
+%%% This code shall interpret and render the tag set output by \cweave{}.
+%%% That's realized in this module as it's needed in multiple places --
+%%% when we want to use \LaTeX{} markup in \cweb{}, when we want to
+%%% incorporate \cweb{} pieces in \LaTeX{} documents (as examples, or in
+%%% textbooks), and when we want to incorporate whole \cweb{} documents in
+%%% a \LaTeX{} document. (That's by far the most difficult task.)
+
+%%% As \cweb{} operates in modes, for obvious reasons our implementation
+%%% is based on a finite state automaton (FSA). The whole tag set is
+%%% separated into subsets that are valid in certain states.
+
+%%% The complete set of tags does not obey any name conventions, some tags
+%%% are even in conflict with standard \LaTeX{} tags (e.g.\ |\\|). Most
+%%% tags are used to render code pieces. With the introduction of states
+%%% we can restrict the visibilities of these tags to those states where
+%%% we typeset program parts. The needed functionality will be available
+%%% in cseqs of our namespace, we'll map that to \cweave{}'s names in a
+%%% group.
+
+%%% Then we have to implement the rendering of chunks and sections, that's
+%%% (more or less) a straight-forward task. Actually, we only set up the
+%%% basic code pieces, the integration of section divisions in the actual
+%%% \LaTeX{} document is a task of higher-level code; that integration
+%%% depends heavily on the type of document. That also implies that we
+%%% won't implement front or back matter material (document markup, table
+%%% of contents, index, list of refinements, etc.)\ in this module.
+
+
+%%% \sect Let's identify this module.
+
+%%% \noindent The code below is explained in the implementation
+%%% documentation of the \pkg{rcs} package.
+
+%%% \beginprog
+\begingroup
+ \def\RCSFile#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesFile{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSFile{cwebbase.tex}{CWEAVE tags for LaTeX markup}
+ $StyleDate: 1995/11/07 17:55:19 $: 9999/00/00 \endRCS
+ $StyleRevision: 1.4 $: 0.0 \endRCS
+\endgroup
+%%% \endprog
+
+
+%%% \sect This module is part of the supported \textsl{cweb-sty} bundle.
+%%% Send bug reports, comments and repairs.
+
+%%% It assumes that the canonical setup of my modules has happened and can
+%%% be used herein. (In particular, the underscore is a letter.)
+
+
+%%% \sect This module reserves the namespace |cwbb|. It assumes that
+%%% |\cweb_warning| is bound either to |\PackageWarningNoLine| or
+%%% |\ClassWarningNoLine|, depending on the environment.
+
+%%% \beginprog
+\ifx \cwbb@loaded\undefined
+ \def\cwbb@loaded{$StyleRevision: 1.4 $}
+\else
+ \cweb_warning{cwebbase}%
+ {Some other package already uses namespace `cwbb'}
+\fi
+%%% \endprog
+
+
+%%% \sect This is freely distributable software; you can redistribute it
+%%% and/or modify it under the terms of the GNU General Public License as
+%%% published by the Free Software Foundation; either version~2 of the
+%%% License, or (at your option) any later version.
+
+%%% This software is distributed in the hope that it will be useful, but
+%%% \textbf{without any warranty}; without even the implied warranty of
+%%% \textbf{merchantability} or \textbf{fitness for a particular purpose}.
+%%% See the GNU General Public License for more details.
+
+%%% You should have received a copy of the GNU General Public License in
+%%% the file |License| along with this package; if not, write to the Free
+%%% Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+
+
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The interface between CWEAVE and TeX
+%%% %
+
+%%% \input{cweave-spec}
+
+%%% %
+%%% % ------------------------------------------------------------
+
+
+
+
+%%% \chap Interface of \pkg{cwebbase} module.
+
+%%% \pkg{cwebbase} implements most of \cweave{} tags. It sets up tons of
+%%% redefinable cseqs, all of them start with |Cweb|.
+%%% %
+%%% \begin{fixme}
+%%% How do I document them? We need a configuration document anyhow. Do
+%%% I present here just a list of those cseqs realized in this module
+%%% and refer to the config docs? Need to wait until I have it, then
+%%% we'll see how the stuff here will be added.
+%%% \end{fixme}
+
+
+%%% \sect This module supports both hierarchic and flat section
+%%% structuring. You have to bind |\cweb@structure| to 0 (hierarchic) or 1
+%%% (flat) before importing it, to show which structure is wanted.
+
+
+%%% \sect This module supports suppression of different document parts.
+%%% You have to define flags |@cweb@suppress@|\<foo>|@| with
+%%% $\mbox{\<foo>} \in \{ |changehints|, |unchanged|, |format| \}$ to
+%%% denote if suppression shall happen.
+
+
+%%% \sect The counter |secnumdepth| must be~11 if chunks shall be
+%%% numbered. That counter doesn't need to be set when this module is
+%%% imported, it can be set locally during typesetting \cweb{} code.
+
+%%% The definitions for flat structure assume that |secnumdepth| is~11
+%%% during typesetting.
+
+
+%%% \sect |\part|, |\chapter| (if available), and |\section| are used for
+%%% \cweb{} section divisions (those with the lowest ranks). These
+%%% commands will change the respective counters globally and will
+%%% add toc lines and marks. You might want to prevent that by redefining
+%%% |\addcontentsline| and |\mark|.
+
+%%% Table of contents macros are not defined for the section divisions
+%%% introduced by this module. You must either prevent that they are
+%%% written to the |AUX| file at all; or you must define them.
+%%% %
+%%% \begin{fixme}
+%%% Add explanation of introduced section divisions.
+%%% \end{fixme}
+
+%%% For flat structure, all section divisions are of type `|chunk|', like
+%%% chunks. The section names start with `|\cwbbRank{|\<rank>|}|'. By
+%%% default, |\cwbbRank| takes one argument and expands to the empty list
+%%% -- this can be redefined for the table of contents.
+
+
+%%% \sect The document end, as explained in the \cweave{} interface specs
+%%% (chunk~\ref{spec:sec:docend}), is not implemented. That must be done
+%%% by the user of this module -- if the back matter is to be processed at
+%%% all.
+
+
+
+
+%%% \chap Processing states.
+
+%%% We have to typeset five different categories of material:
+%%% Documentation, program pieces---embedded within the documentation and
+%%% as large chunks, \TeX{} material within program pieces (i.e., comments
+%%% and refinement names), and cross reference information. Since we need
+%%% a complete other environment for the program pieces than for the rest
+%%% we design ``states'' where we switch to appropriately.
+%%% %
+%%% \begin{enumerate}
+
+%%% \item A chunk starts in the \textsl{documentation state}.
+
+%%% \item |\B| switches to \textsl{program state}. This can happen in
+%%% documentation and program state.
+
+%%% \item While we process the argument of |\PB| we're in \textsl{restricted
+%%% program state}; |\PB| may appear in documentation and in \TeX{}
+%%% state. Since \TeX{} state can be switched on within restricted
+%%% program state, |\PB| can appear within the argument of itself.
+
+%%% \item In the arguments of |\C|, |\SHC|, and |\X| we switch to
+%%% \textsl{\TeX{} state}. All these cseqs appear only in (restricted)
+%%% program state, their official names are actually different. I.e., only
+%%% in (restricted) program state these cseq are bound to the meaning
+%%% described here.
+
+%%% \item Cross reference information are attached to most chunks with
+%%% refinements. This information is processed in \textsl{CR state}.
+%%% After CR state material
+%%% comes always the next chunk or the document end, i.e., material in
+%%% documentation state.
+
+%%% \end{enumerate}
+%%% %
+%%% This FSA is illustrated by the diagram in figure~\ref{fig:fsa}.
+
+%%% \begin{figure}
+
+%%% \begin{center}
+%%% \chardef\\=`\\
+%%% \input{cweb-fsa.latex}
+%%% \end{center}
+
+%%% \caption{The processing state's FSA. The automaton starts and ends in
+%%% the ``documentation'' state.}
+%%% \label{fig:fsa}
+
+%%% \end{figure}
+
+%%% |\cwbb@documentation| will switch to documentation state,
+%%% |\cwbb@program| to program state, |\cwbb@Rprogram| to restricted
+%%% program state, |\cwbb@tex| to \TeX{} state, and |\cwbb@CR| to CR state.
+%%% If we're already in a state, the switch to this state shall be a
+%%% permissible null operation.
+
+
+%%% \sect The basic difference between these states can be named with
+%%% two parameters: (1)~The cseq bindings in effect and (2)~the layout
+%%% parameters used for paragraph makeup.
+
+%%% In (restricted) program state and in CR state the text is output
+%%% under the control of \cweave{}, and tagged by \cweave{}. The used
+%%% tags are from a global namespace and should only be in effect during
+%%% these states. We call this tag set the \textsl{\cweave{} bindings}. In the
+%%% other two states the tags are largely defined by the user, the tag
+%%% set is called the \textsl{user bindings}. The switch to another binding
+%%% is always done locally, i.e., if we switch from documentation to
+%%% restricted program state within a group we don't have to bother about
+%%% the restauration of the user binding; it will be done automatically
+%%% by \TeX{} at the end of the group. Nevertheless we must be able to
+%%% switch from the \cweave{} bindings back to the user bindings which
+%%% were in effect when we activated the \cweave{} bindings. This is
+%%% needed for the \TeX{} state which is always activated within
+%%% (restricted) program state.
+
+%%% The parameters for \textsl{program layout} are really special ones
+%%% since they need to support the indentation which shows the program
+%%% structure. These parameters are used in program and in \TeX{} state.
+%%% The \textsl{document layout} parameters established by the user are
+%%% used in the other three states.
+
+%%% The following table shall summarize this. $C$~denotes \cweave{}
+%%% bindings, $U$~user bindings, $P$~program layout, and $D$~document
+%%% layout. If an entry is empty, its value is not changed on entry in
+%%% this state.
+%%% %
+%%% \begin{center}
+%%% \DeleteShortVerb\|
+%%% \begin{tabular}{r|c|c}
+%%% \multicolumn1{c|}{\textsc{State}} & \textsc{Binding} & \textsc{Layout} \\[1pt]
+%%% \hline
+%%% documentation & $U$ & $D$ \\
+%%% program & $C$ & $P$ \\
+%%% restricted program & $C$ & \\
+%%% \TeX{} & $U$ & \\
+%%% CR & $C$ & $D$ \\
+%%% \end{tabular}
+%%% \MakeShortVerb\|
+%%% \end{center}
+
+
+%%% \sect Since the user bindings and the document layout is defined
+%%% initially, we don't have to do anything if it's requested. Only if we
+%%% change it, i.e., within |\cwbb@cweave_bindings| and
+%%% |\cwbb@program_layout|, we redefine |\cwbb@user_bindings| and
+%%% |\cwbb@doc_layout|. If they are eventually executed, they shall
+%%% rebind themselves back to |\relax|. This way we can switch to
+%%% documentation state as often as we want.
+
+%%% \beginprog
+\let\cwbb@user_bindings=\relax
+\let\cwbb@doc_layout=\relax
+
+\def\cwbb@documentation{%
+ \cwbb@user_bindings
+ \cwbb@doc_layout
+ }
+\def\cwbb@Rprogram{\cwbb@cweave_bindings}
+
+\def\cwbb@program{%
+ \cwbb@cweave_bindings
+ \cwbb@program_layout
+ }
+\def\cwbb@tex{\cwbb@user_bindings}
+
+\def\cwbb@CR{%
+ \cwbb@cweave_bindings
+ \cwbb@doc_layout
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Saving and restoring control sequences.
+
+%%% We have a lot of cseqs which are defined within the namespace of this
+%%% module and which will be used with other names. This usage is in a
+%%% controlled environment, namely neither in documentation nor in \TeX{}
+%%% state. (I.e., the text processed is tagged by \cweave{}, not by humans;
+%%% therefore we have a precise specification of the cseqs we have to
+%%% accept.) We cannot work with groups where a cseq is just redefined
+%%% and \TeX{} takes care for establishing the old binding again; when we
+%%% switch from program state to \TeX{} state all the bindings which were
+%%% in effect before the program state get active, i.e., in the
+%%% documentation state, must be in effect again. We cannot simply
+%%% consider the \TeX{} state as something parallel to program state, it
+%%% must be a hierarchical relationship: In the program state values are
+%%% set up which must be available after switching back from \TeX{} to
+%%% program state.
+
+%%% We save the current binding of a cseq in another cseq, but only if
+%%% there exists a binding currently. This is done to save valuable
+%%% \TeX{} main memory. (Why don't all those people who talk about
+%%% extending \TeX{} not just implement dynamic memory management for the
+%%% existing \TeX{}? It would still be \TeX{} and would \emph{really} be
+%%% a worthy activity. But not enough fun, I assume\,\dots) Actually, one
+%%% can assume that nearly no cseq bindings must be saved at all---the
+%%% used names are strange enough. The bindings of |\foo|, i.e., of the
+%%% cseq with the name \fbox{\ttfamily foo}, is saved as the cseq with the
+%%% name \fbox{\ttfamily cweb\char `\\ foo}, i.e., as
+%%% |\csname cweb\string\foo \endcsname|.
+
+
+%%% \sect The save process is not done statically, but by the macro
+%%% |\cwbb@rebind| which interprets a list of tuples $({\it old\_name},
+%%% {\it new\_name})$, terminated by the tuple $(|\stop|, |\stop|)$.
+%%% Eventually it constructs two new lists, |\cwbb@to_restore| with the
+%%% cseqs which had a binding, and |\cwbb@undefined| with the names which
+%%% didn't have one.
+
+%%% The saving is actually done by |\cwbb@save_binding|, |\cwbb@rebind|
+%%% is responsible for the initialization, |\cwbb@do_rebind| for the
+%%% effective rebinding and the tail recursion on the list.
+
+%%% \begin{fixme}
+%%% We could pull the |\next| assignment in the |\else| branch out of
+%%% the loop to get a better performance. Should measure if this is of
+%%% interest.
+%%% \end{fixme}
+
+%%% \beginprog
+\newtoks\cwbb@undefined
+\newtoks\cwbb@to_restore
+
+\def\cwbb@rebind{%
+ \cwbb@undefined{}%
+ \cwbb@to_restore{}%
+ \cwbb@do_rebind
+ }
+\def\cwbb@do_rebind#1#2{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \cwbb@save_binding #2%
+ \let #2=#1%
+ \let\next\cwbb@do_rebind
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+%%% \sect If the cseq to be saved is undefined, it may just be added to
+%%% the ``undefined list.'' Otherwise its binding is saved and it's added
+%%% to the ``to-be-restored list.''
+
+%%% \TeX{}nical note: The cseq-name for the saved binding must be created
+%%% before the |\let| is executed.
+
+%%% \beginprog
+\def\cwbb@save_binding#1{%
+ \ifx #1\undefined
+ \cwbb@undefined \expandafter{\the\cwbb@undefined #1}%
+ \else
+ \expandafter\let \csname cweb\string#1\endcsname =#1%
+ \cwbb@to_restore \expandafter{\the\cwbb@to_restore #1}%
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect The restoration of rebound cseqs is a two-tied activity: All
+%%% previously undefined cseqs must be made undefined again, and all saved
+%%% cseqs must be restored. Actually, we don't need to reset the two token
+%%% lists, but we do it to save space.
+
+%%% Both |\cwbb@undefine| and |\cwbb@restore_binding| iterate over a list
+%%% of cseqs terminated by |\stop|.
+
+%%% \begin{fixme}
+%%% And here the |\next| assignment could be prepended to the loop as
+%%% well.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@restore_bindings{%
+ \expandafter\cwbb@undefine \the\cwbb@undefined \stop
+ \cwbb@undefined{}%
+ \expandafter\cwbb@restore_binding \the\cwbb@to_restore \stop
+ \cwbb@to_restore{}%
+ }
+
+\def\cwbb@undefine#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \let#1\undefined
+ \let\next\cwbb@undefine
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+%%% \sect \TeX{}nical note: As in |\cwbb@save_binding|, the cseq-name for
+%%% the saved binding must be created before the |\let|s are executed.
+
+%%% \begin{fixme}
+%%% Another |\next| assignment.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@restore_binding#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \expandafter\let \expandafter#1\csname cweb\string#1\endcsname
+ \expandafter\let \csname cweb\string#1\endcsname \undefined
+ \let\next\cwbb@restore_binding
+ \fi
+ \next
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Chunks.
+
+%%% All chunks are numbered subsequently. The chunk number is supplied as
+%%% part of the markup. In addition, the `changed' tag |\*| may be
+%%% appended to the chunk number.
+
+%%% By default, chunks have approximately two picas vertical space in
+%%% front.
+
+%%% If we use an hierarchic structure, we output the chunk number at the
+%%% refinement. (It's only needed to identify the refinement for cross
+%%% references.) But chunks are important structural units and thus should
+%%% be marked by typographic means that's more visible than the space
+%%% mentioned above, we add a `chunk start marker'.
+
+%%% For a flat structure, we output the chunk number as the chunk start
+%%% marker, that is the traditional \WEB{} layout.
+
+%%% In any case, the documentation part shall be run in after the chunk
+%%% start marker.
+
+%%% \beginprog
+\newskip\CwebChunkPreSkip
+ \CwebChunkPreSkip=2pc plus 1pc minus 6pt
+%%% \endprog
+
+
+%%% \sect In \LaTeX{}, structural divisions are best realized by
+%%% |\@startsection|, we'll use it also for chunk starts. The chunk start
+%%% marker is the section heading. For hierarchic structure, we use a
+%%% traditional paragraph sign, used since medieval times to tag new text
+%%% units. Our first test showed that the paragraph sign has typically an
+%%% underlength, that doesn't look good at the start of a chunk. We raise
+%%% the sign by its depth.
+
+%%% For flat structure, we don't have text as the section heading,
+%%% but we have to cope for a deficiency of \LaTeX{}: It always leaves a
+%%% quad after the section number (by the way, this is not mentioned in
+%%% the documentation---you have to look in the macros to figure this
+%%% out). So we use a backskip of $\rm -1\,em$ as the chunk start marker
+%%% then.
+
+%%% Changed chunks are marked by a changeflag, by default a star lapping
+%%% to the right. We need the information if that chunk is changed to
+%%% decide if we shall output the changeflag. We demand from the chunk
+%%% macros that they record that info in the flag |@cwbb_changed_chunk@|.
+
+%%% Each structural division must be categorized, it belongs to a
+%%% \textsl{type}. This type determines also the name of the counter, the
+%%% mark handling for this category, and how this section tag is
+%%% represented in the table of contents. The type of chunks is `|chunk|'
+%%% (trivial, isn't it?).
+
+%%% Furthermore, a structural division in \LaTeX{} has an associated
+%%% \textsl{level}. That level is used to determine if that division get
+%%% numbered in text or table of contents. For sections, we'll compute the
+%%% level from the rank, it will not be larger than the rank.%
+%%% \footnote{Check the next section for a more precise definition of
+%%% `rank' and `level'.}
+%%% Since the highest section rank is~10, we may use 11 as chunk level.
+
+%%% For flat structure, we output the number for all section headings,
+%%% including chunks. For hierarchic structure we use the default maximum
+%%% level as supplied by our document class (e.g., 3~for \cls{article},
+%%% and 2~for \cls{report} or \cls{book}).
+
+%%% As a last point, we have to care already for sections in flat
+%%% structure. They will use the |chunk| category as well, we have to
+%%% distinguish them in the table of contents. For that, we'll record the
+%%% level in the text, it will get written to the table of contents.
+
+%%% \beginprog
+\newif\if@cwbb_changed_chunk@
+\let\CwebChangeFlag=*
+\if@cweb@suppress@changehints@
+ \let\cwbb@change_flag=\relax
+\else
+ \let\cwbb@change_flag=\CwebChangeFlag
+\fi
+\def\cwbb@lap_change_flag{\rlap{\cwbb@change_flag}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\CwebChunkHeadingStyle\relax
+ \newbox\cweb@chunk_marker
+ \setbox\cweb@chunk_marker=\hbox{\P} % tradition from medivial times
+ \setbox\cweb@chunk_marker=\hbox{%
+ \raise\dp\cweb@chunk_marker \box\cweb@chunk_marker
+ }
+ \def\CwebChunkHeadingMarker{\copy\cweb@chunk_marker}
+ \def\CwebChunkHeading#1{%
+ \CwebChunkHeadingMarker
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+ \or
+ %% flat
+ \def\cwbbRank#1{}
+ \def\CwebChunkHeadingStyle{\bfseries}
+ \def\CwebChunkHeading#1{%
+ \protect\cwbbRank{#1}%
+ \kern -1em
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect Chunks are tagged with |\M| and get passed the chunk number as
+%%% argument. Before we evaluate |\@startsection|, we'll start the chunk.
+%%% That's common code to be used by sections as well, as they also
+%%% implicitely start a chunk.
+
+%%% \beginprog
+\def\M#1{%
+ \cwbb@start_chunk{#1}%
+ \@startsection
+ {chunk}% % category, == counter etc.
+ {11}% % level
+ {\z@}% % no indentation for heading
+ {\CwebChunkPreSkip}% % skip above heading
+ {\m@ne em}% % run-in heading, 1em distance to text
+ {\normalfont\CwebChunkHeadingStyle}% % layout
+ {\CwebChunkHeading{11}}% % text, level might go to toc file
+ }
+%%% \endprog
+
+
+%%% \sect But let's first declare how a chunk number is represented. The
+%%% principle rendering is controlled by |\thechunk|, namely, we get
+%%% arabic numbers. This rendering is used both for creation of the
+%%% section header (if in flat mode) and for cross references. I.e.,
+%%% |\ref| expands into it.
+
+%%% But for flat structure we need different expansions: In the section
+%%% header a period shall be appended, a cross reference shall appear
+%%% without such a period. We use two different macros: |\thechunk|
+%%% creates the number with the period, |\cwbb@refchunk| only the number.
+%%% The latter is used for a redefinition of |\@currentlabel|. Of course,
+%%% for hierarchic structure, both are the same.
+
+%%% \beginprog
+\newcounter{chunk}
+
+\def\cwbb@refchunk{\arabic{chunk}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\thechunk=\cwbb@refchunk
+ \or
+ %% flat
+ \def\thechunk{\arabic{chunk}.}
+\fi
+%%% \endprog
+
+
+%%% \sect Each chunk is started by |\cwbb@start_chunk|, which does the
+%%% common actions. First of all, we switch to documentation state.
+
+%%% Then we store the chunk number in the |chunk| counter. That implies
+%%% that the chunk start must not step this counter! We'll take care for
+%%% that later by redefining |\refstepcounter|. The actual number is
+%%% passed as the argument, including an optional |\*| used to denote
+%%% changed chunks. We'll redefine that cseq to update the flag
+%%% |@cwbb_changed_chunk@|.
+
+%%% Each chunk must be typeset with an open |\if| since it is finished by
+%%% |\fi|. If we shall suppress the output of unchanged chunks, we have
+%%% to use |\iffalse| if not |@cwbb_changed_chunk@|, otherwise |\iftrue|.
+%%% The flag |@cwbb_print_chunk@| is set up to provide an appropriate
+%%% test.
+
+%%% When we typeset a chunk, we evaluate |\cwbb@init_print_chunk|. For
+%%% hierarchic structure, that cseq will be used to initialize the
+%%% typesetting of the chunk number at program parts.
+
+%%% \beginprog
+\if@cweb@suppress@unchanged@
+ \def\if@cwbb_print_chunk@{\if@cwbb_changed_chunk@}
+ \@gobble\fi % close \if in case it's skipped
+\else
+ \let\if@cwbb_print_chunk@=\iftrue
+ \@gobble\fi % same here
+\fi
+
+\def\cwbb@start_chunk#1{%
+ \cwbb@documentation
+ \begingroup
+ \def\*{\global\@cwbb_changed_chunk@true}%
+ \global\@cwbb_changed_chunk@false
+ \global \c@chunk #1\relax % might expand \*
+ \endgroup
+ \if@cwbb_print_chunk@
+ \cwbb@init_print_chunk
+ }
+%%% \endprog
+
+
+%%% \sect |\@startsection| will step the section counter (i.e., |chunk|).
+%%% Since we just assigned it in |\cwbb@start_chunk|, that would leave us
+%%% with a counter that's one too high. And our output in cross references
+%%% may be wrong, as a period is appended in flat structure.
+
+%%% That damage all comes from one macro, |\refstepcounter|; let's
+%%% advice it to do the Right Thing for |chunk| counters.
+
+%%% \beginprog
+\let\cwbb@refstepcounter=\refstepcounter
+
+\def\cwbb@string@chunk{chunk}
+\def\refstepcounter#1{%
+ \def\@tempa{#1}%
+ \ifx \@tempa\cwbb@string@chunk
+ \protected@edef\@currentlabel{\cwbb@refchunk}%
+ \else
+ \cwbb@refstepcounter{#1}%
+ \fi
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Sections.
+
+%%% Important sections are started on a new page, normal sections have the
+%%% same space as chunks in front. The very first section does not
+%%% automatically start on a new page since a title may be in front of it.
+
+%%% Each section division has a \textsl{rank}, that denotes the section
+%%% depth as seen by \cweave{}.
+
+%%% For this depth we compute a \textsl{level}, used to denote the section
+%%% depth for \LaTeX{}. This section depth is used to decide where
+%%% numbering shall stop, or if that entry shall be presented in the table
+%%% of contents, etc.
+
+%%% These two numbers are not the same, due to the support of document
+%%% classes with chapters.
+
+%%% We consider a section as important if it's rank is smaller than~3,
+%%% i.e., then we'll start a new page. This limit is controlled by the
+%%% number |\CwebRankNoEject|. The rank is used as the section level
+%%% and is therefore available to configure depth of typeset numbering and
+%%% appearance in table of contents.
+
+%%% \beginprog
+\def\CwebRankNoEject{3}
+%%% \endprog
+
+
+%%% \sect Sections are started with the macro |\N|. It's first parameter
+%%% is the section's rank. The second parameter is the chunk number, the
+%%% third the title of the section. This last parameter must be terminated
+%%% by a dot.
+
+%%% Each section implicitly starts a chunk. As explained above, major
+%%% sections may eject the page, and we add our actual section heading
+%%% with |\CwebSection|.
+
+%%% We also tell the user that we have reached the next section by telling
+%%% him about the current chunk number. The message is output after the
+%%% section header is set, it shall be on the correct page. Since the
+%%% token list to be output is expanded at shipout time, we must take care
+%%% for the immediate expansion ourselves.
+
+%%% \beginprog
+\def\N#1#2#3.{%
+ \cwbb@start_chunk{#2}%
+ \CwebEjectSection{#1}%
+ \CwebSection{#1}{#3}%
+ \begingroup
+ \def\*{}%
+ \edef\tmp{*#2}%
+ \message \expandafter{\tmp}%
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect At the start of a section we usually eject the page if our rank
+%%% is not too low. But if the user had a title above the first section,
+%%% he won't want this page break. So we try to guess it. If we guess
+%%% wrong, the user may either add |\newpage| himself or he may redefine
+%%% |\CwebEjectSection| in a package.
+
+%%% \beginprog
+\def\CwebEjectSection#1{% % #1 == rank
+ \gdef\CwebEjectSection##1{%
+ \relax
+ \ifnum ##1<\CwebRankNoEject\relax
+ \newpage
+ \fi
+ }%
+ }
+
+\if@titlepage
+ \CwebEjectSection{}
+\fi
+%%% \endprog
+
+
+%%% \sect Now comes a heary piece. We have 11~section ranks, and need 11
+%%% different section categories, section counters, etc. For flat
+%%% structure, they shall all do the same; for hierarchic structure, they
+%%% shall be really different.
+
+%%% \noindent Let's have a look at the hierarchic structure first, it's
+%%% more complex.
+%%% %
+%%% \begin{itemize}
+
+%%% \item We associate section rank~0 (that's `|@**|') with document
+%%% parts, i.e., map it to |\part|. Many smaller programs won't use it.
+
+%%% \item If we are used in a class that has a bound cseq |\chapter|, we
+%%% associate rank~1 (`|@*0|') with it and rank~2 (`|@*1|') with
+%%% |\section|. Otherwise we associate rank~1 with |\section|. These
+%%% tags (|\chapter| and/or |\section|) must be defined because many
+%%% macros that want to create section divisions will use them.
+
+%%% \item All other section ranks are placed below |\section|.
+%%% |\subsection| and other \LaTeX{} style markup is not supported.
+
+%%% For hierarchic structure, we identify these sections with category
+%%% names `|section|$l$', where $l$ is the respective level, computed
+%%% from the rank. Those section names do not exist yet, we have to
+%%% create them.
+
+%%% \end{itemize}
+
+%%% Now we come to the point why we can't use the rank as the section
+%%% level passed to \LaTeX{}: When we assume the level usage of \LaTeX{}'s
+%%% standard classes, |\section| is \emph{always} on level~1, independent
+%%% of the existence of |\chapter|. The latter is placed on level~0, its
+%%% existence only influences the section level of |\part|, which is
+%%% either $-1$ or~0.
+
+%%% This implies that |\subsection| in a standard class is always on
+%%% level~2, but our equivalent section division is either `|@*0|'
+%%% (rank~1) or `|@*1|' (rank~2). I.e., when we have chapters, we use a
+%%% level that's one lower than the rank, otherwise we use the rank.
+
+%%% For flat structure, we use type |chunk|. This way we'll use the chunk
+%%% numbers, they are already set up.
+
+%%% For flat structure, we will map all ranks to the same section markup.
+%%% This markup will also be accessible by |\section| or |\chapter|. They
+%%% will be declared to be on a level that is the same as the rank.
+
+
+%%% \sect The printed representation of a section divison number is simply
+%%% the presentation from the next upper level, a full stop, and an arabic
+%%% section number. When we compute the name of the previous level we must
+%%% pay attention to our first section division created this way. This
+%%% division must be placed below |section|, without any following number.
+
+%%% Since registers are a valuable resource in \TeX{}, we have to take
+%%% care not to allocate a superfluous one.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \newcounter{section2}[section]
+ \@namedef{thesection2}{\@nameuse{thesection}.\arabic{section2}}
+ \newcounter{section3}[section2]
+ \@namedef{thesection3}{\@nameuse{thesection2}.\arabic{section3}}
+ \newcounter{section4}[section3]
+ \@namedef{thesection4}{\@nameuse{thesection3}.\arabic{section4}}
+ \newcounter{section5}[section4]
+ \@namedef{thesection5}{\@nameuse{thesection4}.\arabic{section5}}
+ \newcounter{section6}[section5]
+ \@namedef{thesection6}{\@nameuse{thesection5}.\arabic{section6}}
+ \newcounter{section7}[section6]
+ \@namedef{thesection7}{\@nameuse{thesection6}.\arabic{section7}}
+ \newcounter{section8}[section7]
+ \@namedef{thesection8}{\@nameuse{thesection7}.\arabic{section8}}
+ \newcounter{section9}[section8]
+ \@namedef{thesection9}{\@nameuse{thesection8}.\arabic{section9}}
+ \ifx \chapter\undefined
+ \newcounter{section10}[section9]
+ \@namedef{thesection10}{\@nameuse{thesection9}.\arabic{section10}}
+ \fi
+\fi
+%%% \endprog
+
+
+%%% \sect Sections are typeset by |\CwebSection|; rank and title are the
+%%% arguments.
+
+%%% For hierarchic structure, we select the proper cseq according to rank:
+%%% |\part|, |\chapter| if this is a structure with chapters, |\section|,
+%%% and |\cwbb@section{|\<rank>|}| for other ranks. For the last cseq we
+%%% also have to supply the section category, it's `|section|$l$' where
+%%% $l$ is the rank~$r$ if we don't have chapters, $r-1$ otherwise.
+
+%%% That category cannot be computed by |\cwbb@section| itself since we
+%%% want to use that macro also for flat structure. But there all sections
+%%% are basically chunks, we have to use the |chunk| category there. Since
+%%% we want to differentiate them in the table of contents nevertheless,
+%%% we record the current rank in the title.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \ifx \chapter\undefined
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\section
+ \else %% level = rank
+ \def\cwbb@sect_name{section#1}%
+ \def\next{\cwbb@section{#1}}%
+ \fi
+ \next{#2}%
+ }
+ \else
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\chapter
+ \or %% 2
+ \let\next\section
+ \else %% level = rank - 1
+ \count@=#1 \advance \count@\m@ne
+ \edef\cwbb@sect_name{section\number\count@}%
+ \edef\next{\noexpand\cwbb@section{\number\count@}}%
+ \fi
+ \next{#2}%
+ }
+ \fi
+ \or
+ %% flat
+ \def\CwebSection#1#2{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{#1}{\protect\cwbbRank{#1}#2}%
+ }
+ \def\section{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{1}%
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect We use the same layout for all minor sections. That layout can
+%%% be changed by ammended by defining |\CwebSect|$r$|Hook| to redefine
+%%% |\CwebSectPreSkip| and/or |\CwebSectLayout| locally. (Of couse, it may
+%%% be redefined once globally, too.) These are the values to be used for
+%%% arguments 4 and~5 of |\@startsection|, pay attention to the skip sign,
+%%% telling about indentation of the first paragraph. And don't forget
+%%% that the skip is only used for sections on ranks below
+%%% |\CwebRankNoEject|, the others start on a new page anyhow.
+
+%%% The default definition sets the section title in bold and leaves the
+%%% same space as before chunks. The first paragraph of the following
+%%% documentation has no indentation.
+
+%%% Other layout changes (indentation for heading, etc.)\ must be made by
+%%% redefining |\cwbb@section|. Or---email me, if you want to do it. Then
+%%% I might be enclined to introduce further possibilities.
+
+%%% \beginprog
+\newskip\CwebSectPreSkip
+ \CwebSectPreSkip = -\CwebChunkPreSkip
+\def\CwebSectLayout{\normalfont\bfseries}
+
+\def\cwbb@section#1{% % #2 (next token) is title
+ \csname CwebSect#1Hook\endcsname % \relax if undefined
+ \expandafter\@startsection
+ \expandafter{\cwbb@sect_name}% % category; for counter, etc.
+ {#1}% % level is rank
+ \z@ % no indentation for heading
+ \CwebSectPreSkip % skip above, maybe no next par indent
+ \medskipamount % skip below heading
+ \CwebSectLayout
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Marks.
+
+%%% \begin{fixme}
+%%% Since I don't know yet how to handle marks I ignore most of them for
+%%% the time being.
+
+%%% Have to change this later.
+%%% \end{fixme}
+
+%%% \beginprog
+\let\chunkmark\@gobble
+
+\count@=2
+\@whilenum \count@<11 \do {
+ \expandafter\let \csname section\number\count@ mark\endcsname \@gobble
+ \advance \count@ by 1
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Typesetting programs.
+
+%%% Program pieces come in two flavours: as argument of |\PB|, or as
+%%% material after |\B|. In the former case we can use a group for the
+%%% switch to the restricted program state, the group end will restore the
+%%% previous state again. In the latter case we use |\B| for the switch to
+%%% program state, the cross reference list or the next chunk will go to
+%%% another state. Since we have to do more for the material after |\B|,
+%%% we define this cseq later.
+
+%%% \beginprog
+\def\PB#1{%
+ \begingroup
+ \cwbb@Rprogram
+ \leavevmode
+ #1%
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect Note, that |\Y| cannot just be |\smallskip|, as in Plain
+%%% \cweb{}. We must assert that the current paragraph is ended before the
+%%% vertical glue is inserted, and that's not done by the \LaTeX{}
+%%% definition of |\smallskip|. In addition we add some negative penalty,
+%%% here is a good place for a page break. As the penalty value we use
+%%% half the chunk break penalty---of course a chunk start is an even
+%%% better place for a page break\,\dots
+
+%%% It might be that |\Y| is the very first token in a chunk. Then the
+%%% chunk start marker isn't set already since it is regarded as a run-in
+%%% section heading. In that case we don't set the vertical skip but
+%%% simply start the chunk with the program part.
+
+%%% \TeX{}nical note: The flag |@noskipsec| may be used to test if we're
+%%% immediately after a run-in section heading.
+
+%%% \beginprog
+\newcount\CwebProgPenalty
+ \CwebProgPenalty=\@secpenalty
+ \divide\CwebProgPenalty by 2
+\def\Y{%
+ \if@noskipsec
+ \else
+ \par
+ \penalty\CwebProgPenalty
+ \smallskip
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect Yep, let's unfold the ``official'' names of the cseqs used in
+%%% program state.
+
+%%% If we turn on the \cweave{} bindings, they might be in effect already,
+%%% we don't need to establish them again. We can test
+%%% |\cwbb@user_bindings| for this case, it will be redefined then.
+
+%%% \beginprog
+\def\cwbb@cweave_bindings{%
+ \ifx \cwbb@user_bindings\relax
+ \cwbb@rebind
+ % indentation and paragraph layout
+ \cwbb@incr_indent \1%
+ \cwbb@decr_indent \2%
+ \cwbb@expr_break \3%
+ \cwbb@backup \4%
+ \cwbb@opt_break \5%
+ \cwbb@break \6%
+ \cwbb@big_break \7%
+ \cwbb@noindent \8%
+ % C/C++ tokens
+ \CwebRel \?%
+ \CwebAddress \AND
+ \CwebComplement \CM
+ \CwebScope \DC
+ \CwebEquiv \E
+ \CwebGe \G
+ \CwebRightShift \GG
+ \CwebNe \I
+ \CwebAssign \K
+ \CwebLeftShift \LL
+ \CwebMod \MOD
+ \CwebNull \NULL
+ \CwebNot \R
+ \CwebBinOr \OR
+ \CwebMemberRef \PA
+ \CwebThis \this
+ \CwebOr \V
+ \CwebAnd \W
+ \CwebXor \XOR
+ \CwebLe \Z
+ \CwebPointer \MG
+ \CwebPointerMemberRef \MGA
+ \CwebDecr \MM
+ \CwebIncr \PP
+ % more tokens
+ \CwebId \\%
+ \CwebIdLetter \|%
+ \CwebRes \&%
+ \CwebString \.% %% ( ...Emacs...
+ \CwebStringBreak \)%
+ \CwebNumber \T
+ \CwebCombinedOp \MRL
+ % goes to TeX state
+ \CwebComment \C
+ \CwebCxxComment \SHC
+ \CwebRefName \X
+ % CWEB tokens
+ \CwebMacrosHere \ATH
+ \CwebDefine \D
+ \cwbb@format \F
+ \CwebIdCat \J
+ \CwebVerbString \vb
+ % cross reference tags
+ \cwbb@change_flag \*%
+ \CwebCRAlso \A
+ \CwebCRsAlso \As
+ \CwebCRCite \Q
+ \CwebCRsCite \Qs
+ \CwebCRUse \U
+ \CwebCRsUse \Us
+ \CwebCREt \ET
+ \CwebCRsEt \ETs
+ % finish the list
+ \stop\stop
+ \def\cwbb@user_bindings{%
+ \cwbb@restore_bindings
+ \let\cwbb@user_bindings\relax
+ }%
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect Since most of the \cweave{} bindings are simple and tedious
+%%% coding, we'll have a look at the program layout next. Between two
+%%% paragraphs there must not be any skip, the skip used in document
+%%% layout is saved in |\cwbb@save@parskip|. A few other layout
+%%% parameters from the document layout must be saved as well.
+
+%%% \beginprog
+\newskip\cwbb@save@parskip
+\newskip\cwbb@save@rightskip
+\newcount\cwbb@save@sem_sfcode
+\newcount\cwbb@save@pretolerance
+\newcount\cwbb@save@hyphenpenalty
+\newcount\cwbb@save@exhyphenpenalty
+
+\def\cwbb@save_doc_layout{%
+ \cwbb@save@parskip\parskip
+ \cwbb@save@rightskip\rightskip
+ \cwbb@save@sem_sfcode\sfcode`\;
+ \cwbb@save@pretolerance\pretolerance
+ \cwbb@save@hyphenpenalty\hyphenpenalty
+ \cwbb@save@exhyphenpenalty\exhyphenpenalty
+ }
+\def\cwbb@restore_doc_layout{%
+ \parskip\cwbb@save@parskip
+ \rightskip\cwbb@save@rightskip
+ \sfcode`\;=\cwbb@save@sem_sfcode
+ \pretolerance\cwbb@save@pretolerance
+ \hyphenpenalty\cwbb@save@hyphenpenalty
+ \exhyphenpenalty\cwbb@save@exhyphenpenalty
+ }
+%%% \endprog
+
+
+%%% \sect When |\cwbb@program_layout| is called for the very first time
+%%% we have to set up some variables. This cannot be done in advance
+%%% since they depend on values which might be changed by the user in the
+%%% preamble. This setup is done in |\cwbb@layout_init|.
+
+%%% If the program layout is already active, we must not switch it on
+%%% another time. This can be tested by the current binding of
+%%% |\cwbb@doc_layout|.
+
+%%% \beginprog
+\def\cwbb@program_layout{%
+ \ifx \cwbb@doc_layout\relax
+ \cwbb@layout_init
+ \cwbb@save_doc_layout
+ \def\cwbb@doc_layout{%
+ \cwbb@restore_doc_layout
+ \let\cwbb@doc_layout\relax
+ }%
+ % set new values
+ \parskip\z@skip
+ \rightskip\z@ plus 100\p@ minus 10\p@
+ \sfcode`\;=3000
+ \pretolerance\@M
+ \hyphenpenalty 1000 % strings can be broken this way
+ \exhyphenpenalty\@M
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect The unit of the basic indentation is stored in
+%%% |\CwebIndentUnit|. Continuation lines are indented two units further,
+%%% i.e., the initial hanging indentation is 3~units. The current hanging
+%%% indentation is kept in |\cwbb@indent|.
+
+%%% We need undiscardable items which can be used as backspaces, of one
+%%% and two units, respectively. This is done best by boxes. The
+%%% initialization of the boxes must be postponed until the user had the
+%%% chance to change |\CwebIndentUnit|, it's done in the program state
+%%% initialization.
+
+%%% \beginprog
+\newdimen\CwebIndentUnit
+ \CwebIndentUnit=1em
+\newdimen\cwbb@indent
+
+\newbox\cwbb@bak % backspace one unit
+\newbox\cwbb@bakk % backspace two units
+
+\def\cwbb@layout_init{%
+ \global\setbox\cwbb@bak \hbox to -1\CwebIndentUnit{}%
+ \global\setbox\cwbb@bakk \hbox to -2\CwebIndentUnit{}%
+ \global\let\cwbb@layout_init\relax
+ }
+%%% \endprog
+
+
+%%% \sect Now we can formulate how to start typesetting program pieces: We
+%%% switch to program state, set the initial indentation, and add the
+%%% basic indentation of the first code line. Then we might add a marker
+%%% to this program part; for hierarchic structure we want to typeset the
+%%% chunk number in the margin of the chunk's program part. That's not
+%%% really possible, 'though: definitions and format requests are also
+%%% introduced with |\B|, so the chunk number will happen to be at the
+%%% start of the definition part if one exists.
+
+%%% If the code line is the very first text to be typeset in this chunk,
+%%% we don't add the basic indentation---we're already indented from the
+%%% run-in chunk start marker. But we must not start our chunk with
+%%% |\noindent|, the marker would be typeset in the left margin then. (I
+%%% consider this as a \LaTeX{} error.) So we just emulate an empty
+%%% documentation part, the marker is now set correctly. |\B| can test for
+%%% this case: |@noskipsec| is still true then.
+
+%%% \beginprog
+\def\B{%
+ \cwbb@program
+ \cwbb@indent 3\CwebIndentUnit \hangindent\cwbb@indent
+ \ifvmode
+ \if@noskipsec
+ \indent % add an empty documentation part
+ \else
+ \noindent \kern\CwebIndentUnit
+ \fi
+ \fi
+ \CwebNumberProgramPart
+ }
+%%% \endprog
+
+
+%%% \sect The chunk number shall be added at the first program part. We
+%%% have a hook in the chunk start where we can set up
+%%% |\CwebNumberProgramPart|. It will redefine itself at first call, to
+%%% disable further output on more |\B| occurences in a chunk.
+
+%%% The chunk number should always be in the left margin. If it's in the
+%%% right margin, there would be too much horizontal space between the
+%%% code start and the actual number. (The first code line is typically
+%%% not very long.) \LaTeX{} does not allow to select the margin for
+%%% marginpars, therefore we have to typeset the marginal note `by hand'.
+%%% As we know that we're in horizontal mode when we're called, |\vadjust|
+%%% is used to create the note. The standard \LaTeX{} dimen register for
+%%% the distance between marginal notes and text matter is used, of
+%%% course.
+
+%%% Note that marginal notes may overwrite the chunk number. As it's very
+%%% unlikely that somebody issues marginal notes in program mode (that's
+%%% quite difficult and involves ingenious usage of the implementation of
+%%% this macros), I won't care for that case.
+
+%%% \begin{fixme}
+%%% The baseline of the marginal note shall be aligned to the baseline
+%%% of the preceding line. That does not happen by default. Currently, I
+%%% assume that the preceding line has a depth of appropriately the
+%%% depth of a~`g'. Experiments showed that this gives good results for
+%%% most \cweb{} documents. But nevertheless one should determine the
+%%% exact depth of the preceding line, to know how much |box0| should be
+%%% raised. If anybody knows how this can be done, please tell me. (I
+%%% know about the route over the output routine, implementing an own
+%%% marginal note category, but this seems too much effort.)
+%%% \end{fixme}
+
+%%% For flat structure, we don't add any program part marker and we
+%%% don't need to initialize it.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\cwbb@marginal_chunkno{%
+ \vadjust{%
+ \setbox\z@ \hbox to \z@{%
+ \hss \vphantom{g}%
+ \small \thechunk
+ \hskip \marginparsep
+ }%
+ \vbox to \z@{%
+ \hsize\z@
+ \vss
+ \noindent \raise \dp\z@ \box\z@
+ }%
+ }%
+ }
+ \def\cwbb@init_print_chunk{%
+ \def\CwebNumberProgramPart{%
+ \cwbb@marginal_chunkno
+ \let\CwebNumberProgramPart\relax
+ }%
+ }
+ \or
+ %% flat
+ \let\cwbb@init_print_chunk\relax
+ \let\CwebNumberProgramPart\relax
+\fi
+%%% \endprog
+
+
+%%% \sect If a statement is finished, a new paragraph with the basic
+%%% indentation has to be started.
+
+%%% An optional statement break is implemented by a low penalty which
+%%% will be selected if the line has to be broken. We assume that the
+%%% hanging indentation of the new line is already set correctly, and
+%%% have to backup two units to get the basic indentation. It might be
+%%% that the line break is not chosen by \TeX{}, we compensate the
+%%% backspace for this case. This compensation is discarded at the start
+%%% of a new line.
+
+%%% \beginprog
+\def\cwbb@break{% % forced break, between statements
+ \ifmmode\else
+ \endgraf % in LaTeX it isn't sure what \par is really...
+ \noindent
+ \hangindent\cwbb@indent \kern\cwbb@indent
+ \copy\cwbb@bakk % go back to basic indentation
+ \ignorespaces
+ \fi
+ }
+\def\cwbb@opt_break{% % optional break between statements
+ \hfil \penalty\m@ne
+ \hfilneg \kern .5em \kern 2\CwebIndentUnit % discarded on line break
+ \copy\cwbb@bakk
+ \ignorespaces
+ }
+\def\cwbb@big_break{% % forced break and a little extra space
+ \Y
+ \cwbb@break
+ }
+\def\cwbb@expr_break#1{% % break with penalty #1 * 10
+ \hfil \penalty#10
+ \hfilneg % discarded on line break
+ }
+%%% \endprog
+
+
+%%% \sect When we increment the indentation, we must not forget to set
+%%% the hanging indentation immediately since |\cwbb@opt_break| relies on
+%%% the new value. If a continuation line is needed it will also be
+%%% indented one unit more, which is ok since it should be distinguishable
+%%% from the next line.
+%%% %
+%%% \begin{fixme}
+%%% Shouldn't the hanging indentation be set anew at the end of
+%%% |\cwbb@decr_indent| as well?
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@incr_indent{%
+ \global\advance\cwbb@indent\CwebIndentUnit
+ \hangindent\cwbb@indent
+ }
+\def\cwbb@decr_indent{\global\advance\cwbb@indent -\CwebIndentUnit}
+\def\cwbb@backup{\copy\cwbb@bak}
+\def\cwbb@noindent{% % no indentation
+ \hskip -\cwbb@indent \hskip 2\CwebIndentUnit
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Program (C or \C++) tokens.
+
+%%% Since the user might want to change the (somewhat unusual) way some
+%%% of the operators are typeset, we supply names for them.
+
+%%% \beginprog
+\let\CwebAnd=\land % logical and, &&
+\let\CwebEquiv=\equiv % equiv sign, for ==
+\let\CwebGe=\ge % greater or equal
+\let\CwebLe=\le % less or equal
+\let\CwebNe=\ne % unequal, !=
+\let\CwebNull=\Lambda % NULL pointer
+\let\CwebNot=\lnot % logical not, !
+\let\CwebOr=\lor % logical or, ||
+\let\CwebXor=\oplus % bitwise exclusive or, ^
+%%% \endprog
+
+
+%%% \sect Some symbols have to be shifted around, to save computation time
+%%% we put them in boxes.
+
+%%% \begin{fixme}
+%%% It's not sure if the following symbols still look good in other font
+%%% families than Computer Modern. The movements might need to get
+%%% adapted when more experience is available. They were chosen from
+%%% Plain \cweb{}, where these were explicit movements measured in
+%%% points. I changed the dimension to quads, to achieve at least the
+%%% support for different sizes in the Computer Modern family.
+%%% \end{fixme}
+
+%%% \TeX{}nical note: The pointer symbol must not be declared as a math
+%%% symbol (e.g., with |\DeclareMathSymbol|). Then \TeX{} would insert an
+%%% italic correction behind it, according to rule~17 from appendix~G.
+%%% (Please send email if you know of an easy way to prevent that
+%%% addition.) So I use the symbol as an accent, but then I have to supply
+%%% a nucleus for this math list. That nucleus also determines the total
+%%% width of this construct. As the vector accent has a width of
+%%% approximately 5\,pt in |cmmi10|, I'm going to use 0.5\,em. (That
+%%% heuristic is as good as any other I can think of.)
+
+%%% \beginprog
+%% pointer to struct component (`->'), use symbol of \vec accent
+\newbox\cwbb@pointer
+\setbox\cwbb@pointer=\hbox{%
+ \kern -.2em
+ \lower .3em \hbox{$\vec{\kern .5em}$}%
+ \kern .1em
+ }
+
+\newbox\cwbb@decr % decrement, --
+\setbox\cwbb@decr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {-}\kern -.1em{-}$}%
+ \kern .05em
+ }
+\newbox\cwbb@incr % increment, ++
+\setbox\cwbb@incr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {+}\kern -.1em{+}$}%
+ \kern .05em
+ }
+%%% \endprog
+
+
+%%% \sect Although the following symbols are typeset like an ``ordinary
+%%% \C++ programmer'' would expect them, we provide own module names
+%%% nevertheless. They can now be changed as well, i.e., orthogonality is
+%%% enhanced. (And we can use |\cwbb@rebind| for assigning them to their
+%%% names while we switch to program state\,\dots).
+
+%%% Note: The comment at |\CwebMod| says ``modulo/remainder'', as this is
+%%% not defined by the C standard, i.e., implementation-dependent.
+
+%%% \beginprog
+\mathchardef\CwebAddress="2026 % `&', as binary op
+\let\CwebAssign== % assignment
+\let\CwebBinOr=\mid % binary or
+\def\CwebComplement{{\sim}} % `~', as ordinary symbol
+\def\CwebDecr{\copy\cwbb@decr} % decrement
+\def\CwebIncr{\copy\cwbb@incr} % increment
+\let\CwebLeftShift=\ll % left shift, <<
+\def\CwebMod{\mathbin{\hbox{\footnotesize\rm\%}}} % modulo/remainder, %
+\def\CwebMemberRef{\mathbin{.*}} % ptr to member (on object)
+\def\CwebPointer{\copy\cwbb@pointer}
+\def\CwebPointerMemberRef{\mathbin{\CwebPointer*}} % ptr to member (on ptr)
+\def\CwebRel{\mathrel?} % relation operator
+\let\CwebRightShift=\gg % right shift, >>
+\def\CwebScope{\kern.1em{::}\kern.1em} % scope resolution
+
+\def\CwebThis{\CwebRes{this}} % reserved identifier `this'
+%%% \endprog
+
+
+%%% \sect The special identifier |\TeX| remains. In math mode it shall be
+%%% the identifier, in other modes the usual logo. We use the definition
+%%% supplied by DEK (or \textsc{Silvio Levy}?)\ from Plain \cweb{}.
+
+%%% \beginprog
+\def\TeX{%
+ {%
+ \ifmmode\it\fi
+ \mbox{T\kern-.1667em\lower.424ex\hbox{E}\hskip-.125emX}%
+ }}
+%%% \endprog
+
+
+%%% \sect Some tokens don't have constant names, the name is supplied as
+%%% the argument.
+
+%%% Identifiers are typeset in italics, reserved words and type names in
+%%% boldface, and strings in typewriter. For an underscore in the bold
+%%% words we use a line that's 50\,\% thicker as the usual line thickness.
+%%% -- But what is the usual line thickness. As Knuth uses 0.4\,pt for a
+%%% 10\,pt Computer Modern font, we'll use 0.06\,em for a CM font in an
+%%% arbitrary size. If that line thickness is still OK for other fonts is
+%%% doubtful, as it should fit to the stem width. (CM is a rather light
+%%% font, after all.) But since there is no font dimension that gives a
+%%% hint to an appropriate line thickness, it's our best try.
+
+%%% \beginprog
+\def\CwebCombinedOp#1{\mathrel{\let\K==#1}} % e.g., += operator
+\def\CwebId#1{\hbox{\it#1\/\kern.05em}} % identifier, more than one char
+\def\CwebIdLetter#1{\hbox{$#1$}} % identifier, one letter
+\def\CwebRes#1{% % reserverd words and type names
+ \hbox{\bf
+ \def\_{\kern.04em\vbox{\hrule width.3em height.06em}\kern.08em}%
+ #1\/\kern .05em
+ }%
+ }
+%%% \endprog
+
+
+%%% \sect In Plain \cweb{} strings are typeset in the font |cmtex10|,
+%%% which is a typewriter font with extended ASCII characters. This font
+%%% does not have the usual accents and can therefore not used for
+%%% typesetting national characters. (Very often they are input in some
+%%% 8-bit encoding, the respective character code is made active and is
+%%% substituted by a cseq which expands to the correct glyph. Of course,
+%%% there are better ways, but that's the reality we have to cope with.)
+%%% Instead of this special font we use the standard typewriter font.
+%%% This has the further advantage that |cmtex10| may not be introduced
+%%% to NFSS\@.
+
+%%% \beginprog
+\def\CwebString{%
+ \hbox \bgroup
+ \tt
+ \cwbb@string_setup
+ \cwbb@string
+ }
+\def\cwbb@string#1{%
+ #1\kern .05em
+ \egroup
+ }
+%%% \endprog
+
+
+%%% \sect Within strings certain cseqs have a special meaning; this is
+%%% introduced by |\cwbb@string_setup|. The cseqs within strings are
+%%% mostly special \TeX{} characters that are to be typeset by cseqs with
+%%% the same name.
+
+%%% But a problem remains: active characters. They will still expand, and
+%%% will mess up the string visualization. This is a particular problem
+%%% with |german.sty| where the double quote is active. (Of course, that's
+%%% a problem -- I'm German. :--) When one typesets strings, a double
+%%% quote shall be typeset as double quotes, and must not expand to the
+%%% special meaning. Remember that double quotes happen to appear quite
+%%% often in strings. As the string argument is already tokenized, one has
+%%% to redefine the binding of $(\textit{\textup{symbol}}\ \textbf{.}\
+%%% |#\"|)$.
+
+%%% \begin{fixme}
+%%% Perhaps one might use a loop over all characters except the braces,
+%%% and set their catcode to other. That would give us enough security.
+
+%%% For now, we handle only the double quote.
+%%% \end{fixme}
+
+%%% Every once in a while we have discretionary breaks in a string,
+%%% denoted by |\CwebStringBreak|. This break is shown with the C
+%%% convention of escaped newlines.
+
+%%% \beginprog
+\def\cwbb@string_setup{%
+ \chardef\ =`\ % % <-- two spaces !
+ \chardef\&=`\&
+ \chardef\\=`\\
+ \chardef\^=`\^
+ \chardef\_=`\_
+ \chardef\{=`\{
+ \chardef\}=`\}
+ \chardef\~=`\~
+ \catcode`\"=\CatOther
+ }
+
+\def\CwebStringBreak{\discretionary{\hbox{\tt\char`\\}}{}{}}
+%%% \endprog
+
+
+%%% \sect Numbers are typeset in different ways. We use the definition of
+%%% Plain \cweb{}.
+%%% %
+%%% \begin{fixme}
+%%% Should add a specification of the possible input and an explanation
+%%% of the macros below. In particular, that the closing brace after
+%%% |\aftergroup| is used much later is probably not grokked by most
+%%% \TeX{} programmers.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\cwbb@oct{\hbox{$^\circ$\kern-.1em\it\aftergroup\?\aftergroup}}
+\def\cwbb@hex{\hbox{$^{\scriptscriptstyle\#}$\tt\aftergroup}}
+
+\def\CwebNumber#1{% % octal, hex, or decimal constant
+ \hbox{%
+ $%
+ \def\?{\kern.2em}%
+ \def\$##1{\egroup\sb{\,\rm##1}\bgroup}% suffix to constant
+ \def\_{\cdot 10^{\aftergroup}}% power of ten (via dirty trick)
+ \let\~\cwbb@oct \let\^\cwbb@hex
+ {#1}%
+ $}%
+ }
+%%% \endprog
+
+
+%%% \sect Comments are typeset in \TeX{} state. We add a hook, the user
+%%% shall be able to change the layout (e.g., he might want another font).
+%%% %
+%%% \begin{fixme}
+%%% Currently \C++ comments are typeset like C comments. This is
+%%% horrible in usual circumstances, i.e., when complete blocks of text
+%%% are prefixed with |//|. We should simply catenate all these text and
+%%% typeset it as one paragraph, each line prefixed by |//|. But then we
+%%% have to implement an |\everyline| first, and since that's not so
+%%% easy we postpone it\,\dots
+%%% \end{fixme}
+%%% %
+%%% In front of a C comment there is an optional stmt break, with 2\,quad
+%%% in front if the line is not broken and 1.5\,quad if the line is broken.
+
+%%% \beginprog
+\def\CwebComment#1{%
+ \5% % 0.5em will be discarded on line break
+ \hskip 1.5em
+ $/\ast\,$%
+ {\cwbb@tex
+ \CwebCommentHook
+ #1%
+ }%
+ $\,\ast/$%
+ }
+\let\CwebCommentHook\relax
+\let\CwebCxxComment\CwebComment
+%%% \endprog
+
+
+
+
+
+%%% \chap \cweb{} tokens.
+
+%%% We distinguish three categories of \cweb{} tokens: (1) Those which
+%%% output constant text, (2) those which have attributes to be displayed
+%%% in a special way, and (3) those which start a new structure element,
+%%% namely |@d| and~|@f|. Let's consider them in this order.
+
+
+%%% \sect \cweb{} tokens which expand in a constant string are the
+%%% identifier catenation operator~(`|@&|') and the macro placement
+%%% directive~(`|@h|').
+
+%%% \beginprog
+\def\CwebIdCat{\CwebString{@\&}}
+\def\CwebMacrosHere{%
+ \begingroup
+ \def\CwebRefNumber##1{}%
+ \CwebRefName :Preprocessor definitions\X
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect Verbatim program strings, i.e., strings passed verbatim by
+%%% \ctangle{}~(`|@=|') are typeset like normal strings, but within a box.
+%%% We use 2\,pt as the separating distance, this is set locally.
+
+%%% \beginprog
+\def\CwebVerbString#1{{\fboxsep\tw@\p@ \fbox{\CwebString{#1}}}}
+%%% \endprog
+
+
+%%% \sect The refinement names are typeset in angles, this has a long
+%%% tradition. We burry the typesetting of the chunk number (which is
+%%% the first parameter) in a macro call; the user may change this to
+%%% achieve special effects.
+%%% %
+%%% \begin{fixme}
+%%% One could think about handling special values of the chunk numbers
+%%% differently. E.g., an empty argument is not typeset at all, a~0
+%%% triggers a marginal note about a missing definition, etc.
+%%% \end{fixme}
+
+%%% A refinement name may be typeset both in math and horizontal mode.
+%%% The name itself is typeset in horizontal mode, of course; for the
+%%% angles we need math mode. Therefore we assert at the start of the
+%%% macro that we're not in math mode any more. At the end we switch back
+%%% to math mode if we've started in it. This conditional switch from and
+%%% back to math mode is done by |\cwbb@toggle_text|.
+
+%%% \TeX{}nical note: If we're in math mode |\cwbb@toggle_text| must be
+%%% defined globally, as it will turn off math mode and the definition
+%%% would be un-made then. The second invocation of |\cwbb@toggle_text|
+%%% would be undefined then.
+
+%%% A refinement may also consist of the file name the expansion of this
+%%% refinement shall be written to.
+%%% %
+%%% \label{sec:refname-dot}
+%%% %
+%%% This file name is tagged with |\.|, it shall be typeset like a string.
+%%% But the user shall be able to use the dot accent in the refinement
+%%% name as well. We check if the text consists solely of the tag and its
+%%% argument; in this case we substitute |\.| with |\CwebString|.
+%%% Otherwise we leave it as it is. Then a refinement name may not
+%%% consist of a single dot-accented expression---well, that's highly
+%%% unlikely. (Nevertheless I'll document it in the user's manual\,\dots)
+
+%%% \begin{fixme}
+%%% One should be able to use |\EnsureMath| here. But the refinement
+%%% name must not be set in math mode and must not be set in an hbox;
+%%% the latter would prevent line breaks.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\CwebRefName#1:#2\X{%
+ \ifmmode \gdef\cwbb@toggle_text{\null$\null}%
+ \else \let\cwbb@toggle_text\relax
+ \fi
+ \cwbb@toggle_text
+ $\langle\,${\cwbb@tex \cwbb@check_dot{#2}\CwebRefNumber{#1}}$\,\rangle$%
+ \cwbb@toggle_text
+ }
+\def\CwebRefNumber#1{%
+ {\reset@font \footnotesize \kern .5em#1}%
+ }
+%%% \endprog
+
+
+%%% \sect The next macro, |\cwbb@check_dot| is a hassle to implement. We
+%%% must check if it's argument consists solely of a `|\.|'~tag, together
+%%% with the argument to this tag. Then the `|\.|'-argument shall be
+%%% typeset like a string, |\.| itself must be ignored.
+
+%%% The check for |\.| must not evaluate the parameter. Especially |\PB|
+%%% must not be evaluated within, it would lead to havoc with the
+%%% redefinition of all bindings, etc. (At least that's the empirical
+%%% result---it happened; although I don't know where the problems are.
+%%% Anyhow, it wasn't in the specs of the rebinding process.) All other
+%%% cseqs which shouldn't be used in moving arguments may cause problems
+%%% here, too. Note that |\protect| is no solution here, the user often is
+%%% not aware of the presence of \cweave{}-generated tags. As an example,
+%%% consider the refinement \verb"@< local variables of |foo| @>|" which
+%%% will lead to a call with the argument `|local variables of \PB{\\{foo}}|'.
+%%% |\\| must not be evaluated here.
+
+%%% I use an approach where I hope that it will work---but to be honest,
+%%% I'm not sure. I want to define a macro |\next| which shall have an
+%%% empty expansion iff |#1| was a `|\.|'~tag with its argument. Empty
+%%% refinement names are not allowed so we can assume that |#1| consists
+%%% of at least one token. I evaluate the first token of |#1| before the
+%%% |\def| is done. With an appropriate binding a |\.| will disappear,
+%%% perhaps leaving an empty token list. I also take care that |\PB| is
+%%% not evaluated then, by temporary rebinding it to |\relax|. Thereby I
+%%% have taken care for all cseqs \cweave{} might have introduced at the
+%%% very front of |#1|. Other cseqs are user-tags. If they cause
+%%% problems, the user might circumstance them by adding |\protect|.
+
+%%% \beginprog
+\def\cwbb@check_dot#1{%
+ \begingroup
+ \let\.\@gobble
+ \let\PB\relax
+ \expandafter\def \expandafter\next \expandafter{#1}%
+ \ifx \next\empty
+ \gdef\next##1{\CwebString}%
+ \else
+ \global\let\next\relax
+ \fi
+ \endgroup
+ \next #1%
+ }
+%%% \endprog
+
+
+%%% \sect \cweb{} macro definitions (done by~|@d|) and format directives are
+%%% basically program text. When they are started we are already in
+%%% program state, i.e., |\B| has appeared in front of it. But since we have
+%%% an introductionary identifier (either |#define| or |format|) to set at
+%%% the very front we increase the indentation by three units.
+
+%%% The output of format directives may be suppressed. Then we have to
+%%% gather both identifiers that will appear after |\F|. Both of them
+%%% consist of one group-token (a cseq and the identifier name), after
+%%% each identifier one other token separates them. I.e., we have to
+%%% gobble six group-tokens. As the very last token is |\par| if a format
+%%% directive follows directly on another one, the definition must be
+%%% long.
+
+%%% \TeX{}nical note: |\cwbb@format| must expand to |\CwebFormat|, not to
+%%% its expansion. Otherwise a redefinition of |\CwebFormat| in a package
+%%% would not be used.
+
+%%% \beginprog
+\def\CwebDefine{\cwbb@macro{\#define}}
+\def\CwebFormat{\cwbb@macro{format}}
+\if@cweb@suppress@format@
+ \long\def\cwbb@format#1#2#3#4#5#6{}
+\else
+ \def\cwbb@format{\CwebFormat}
+\fi
+
+\def\cwbb@macro#1{%
+ \global\advance\cwbb@indent \tw@\CwebIndentUnit
+ \cwbb@incr_indent
+ \CwebRes{#1 }% % <-- blank!
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Refinement cross references and changes.
+
+%%% At the end of chunks with refinements, \cweave{} may output cross
+%%% reference information: In which chunks additional definitions for
+%%% this refinement are found, where this refinement is used, and where it
+%%% is cited. This cross reference information is always introduced by a
+%%% tag followed by a non-empty list of chunk numbers. Different tags
+%%% are used for lists with only one element and for lists with more than
+%%% one element, this way the introductionary text may be adapted.
+
+%%% Within a list of $n$~chunk numbers the first $n-1$~numbers are
+%%% separated by commas with one following blank. The last two numbers
+%%% are separated by |\ET| if~$n=2$, and by |\ETs| if~$n>2$. If a chunk
+%%% is changed by the changefile, a changeflag~(`|\*|') is appended to
+%%% its number. The list is eventually terminated by a full stop.
+
+
+%%% \sect We separate the cross reference information by a smallskip from
+%%% the refinement or from a previous cross reference information. The
+%%% information itself is typeset in a smaller font, as it is auxilliary,
+%%% inserted stuff. The number list has a hanging indentation of
+%%% |\CwebNumberListHangindent|. But beware: This isn't a dimension
+%%% register, it's a macro. This way one can use ems as dimensions.
+
+%%% The actual size change is represented by a macro. We will redefine it
+%%% when we need an other size, e.g., in the lists at the end of the
+%%% document.
+
+%%% We must assert that we're in CR state while we set a cross reference.
+
+%%% The first parameter of a cross reference information unit is the
+%%% introductionary text, the second is the number list. The parameters
+%%% must be evaluated in a group---local parameter changes therein must
+%%% not influence the environment.
+
+%%% \beginprog
+\def\CwebNumberListHangindent{2em}
+\let\CwebCRSize\footnotesize
+\def\CwebCrossRef#1#2.{%
+ \par\smallskip
+ \cwbb@CR
+ \begingroup
+ \reset@font \CwebCRSize
+ \noindent \hangindent\CwebNumberListHangindent\relax
+ #1~#2.\par
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect Well, let's define all introducing tags, which in fact start the
+%%% cross reference information unit. The number list is gathered by
+%%% |\CwebCrossRef|.
+
+%%% \beginprog
+\def\CwebCRAlso{\CwebCrossRef{See also chunk}}
+\def\CwebCRsAlso{\CwebCrossRef{See also chunks}}
+
+\def\CwebCRCite{\CwebCrossRef{This code is cited in chunk}}
+\def\CwebCRsCite{\CwebCrossRef{This code is cited in chunks}}
+
+\def\CwebCRUse{\CwebCrossRef{This code is used in chunk}}
+\def\CwebCRsUse{\CwebCrossRef{This code is used in chunks}}
+
+\def\CwebCREt{ and~}
+\def\CwebCRsEt{, and~}
+%%% \endprog
+
+
+
+
+%%% \chap Bells and whistles.
+
+%%% \begin{fixme}
+%%% Just copied from Plain \cweb{}. It doesn't work
+%%% anyhow---|\char"|\<xy> usually doesn't typeset the correct
+%%% character. That's because this character is most probably an active
+%%% character; at least that's typical for the way \TeX{} systems are
+%%% used in Europe. I have to think about the implementation.
+%%% \end{fixme}
+
+%%% \beginprog
+\def\ATL{\par\noindent\bgroup\catcode`\_=12 \postATL} % print @l in limbo
+\def\postATL#1 #2 {\bf letter \\{\uppercase{\char"#1}}
+ tangles as \tentex "#2"\egroup\par}
+\def\noATL#1 #2 {}
+\def\noatl{\let\ATL=\noATL} % suppress output from @l
+%%% \endprog
+
+
+
+%%% \chap The end.
+
+%%% Well, after all we're finished with this module
+
+%%% \beginprog
+\endinput
+%%% \endprog
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+%%% \vskip \PltxPreSectSkip
+
+%%% \rcsLogIntro{After the actual revision log of this module, we give an
+%%% extract of log messages of the \cls{cweb} class. This module was
+%%% extracted from revision~2.8 of that class, we present to you the
+%%% log entries that concern the code of the new module.}
+
+%%% \begin{rcslog}
+%%% $StyleLog: cwebbase.doc,v $
+%%% \Revision 1.4 1995/11/07 17:55:19 schrod
+%%% \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+%%% versions; make it work with the previous-to-last one, version
+%%% \mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+%%% \mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+%%% |\newif|, and |\hb@xt@| must not be used.\\
+%%% Problem reported by Laurent Desnogues
+%%% \path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+%%% -- add name).
+
+%%% \Revision 1.3 1995/11/07 14:13:12 schrod
+%%% Reset section numbers when higher-level section begins.\\
+%%% Error reported by Bronne Louis \path|<bronne@montefiore.ulg.ac.be>|.
+
+%%% \Revision 1.2 1995/11/06 10:58:47 schrod
+%%% Support active double quotes in strings. That happens when
+%%% |german.sty| is used. Problems with other active characters may still
+%%% appear.
+
+%%% \Revision 1.1 1995/09/12 23:02:22 schrod
+%%% Moved all code that does the actual typesetting of \cweave{} tags and
+%%% is therefore also needed for a |cweb| environment. It's now an own
+%%% module named \pkg{cwebbase}.
+
+%%% Chunk markers and chunk marginal numbers don't hang below the
+%%% baseline any more.
+
+
+%%% \item[] \noindent\hrulefill\break
+%%% \textbf{Extract of the revision log for |cweb.doc|:}
+
+%%% \vskip 2ex
+
+%%% \Revision 2.5 1995/08/29 02:07:26 schrod
+%%% Discard dependencies on 10\,pt fonts.
+
+%%% Support suppression of format directives.
+
+%%% \Revision 2.4 1995/08/27 19:31:43 schrod
+%%% Make configuration of change flag easier.
+
+%%% Put section title of changed chunks list in marks, and tell the user
+%%% that it got typeset.
+
+%%% \Revision 2.3 1995/08/27 17:24:46 schrod
+%%% Make usage of baseclass with chapters work.
+
+%%% \Revision 2.2 1995/08/27 13:26:22 schrod
+%%% Add possibility to suppress change hints.
+
+%%% \Revision 2.1 1995/08/25 19:11:18 schrod
+%%% Hierarchic structures are supported now, in addition to the flat
+%%% structure of the beta-test version. One can choose with an option. For
+%%% that step, the terminology was cleaned up, too: Chunks are not named
+%%% sections any more. (That change involved reimplementation of almost
+%%% all the structure and toc stuff.)
+
+%%% The chunk number supplied by \cweave{} is used now, not some computed
+%%% number. Change flags are printed, too.
+
+%%% One can suppress output of unchanged sections.
+
+%%% \Revision 1.12 1993/08/10 14:15:43 schrod
+%%% New page on main section only if group level $<$ |\cwebSecNoEject|.
+%%% Default for the latter is 3.
+
+%%% \Revision 1.11 1993/08/10 11:21:07 schrod
+%%% Reference to section number does not render a period after the
+%%% number any more.
+
+%%% \Revision 1.10 1993/08/09 20:08:20 schrod
+%%% |\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+%%% effect already.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.9 1993/08/09 18:05:28 schrod
+%%% Left shift operator wasn't defined correctly.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.6 1993/06/15 08:49:23 schrod
+%%% |\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+%%% causes problems if a |\PB| is within. Now I try hard not to evaluate any
+%%% tokens outside of my control.
+
+%%% Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+%%% need an own macro.
+
+%%% \Revision 1.4 1993/06/14 15:54:18 schrod
+%%% Add FSA diagram about processing states. CR state is also switched
+%%% to from \TeX{} state (that happens with |\ch| at the document end).
+
+%%% \Revision 1.3 1993/05/13 17:51:21 schrod
+%%% Refinements may also be filenames (`|@(|'). Then the complete name
+%%% consists of a |\.| macro call, which is handled now.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Made the detection of `|@.|' index entries more robust.
+
+%%% \Revision 1.2 1993/05/12 18:28:59 schrod
+%%% Adapted to recent changes of \cweave{} (of April 93):
+
+%%% Main sections have a group level, represented in the table of
+%%% contents. This changed the complete implementation of section tags.
+
+%%% New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+%%% |\CwebCombinedOp|.
+
+%%% \Revision 1.1 1993/04/09 15:00:37 schrod
+%%% Initial revision
+
+%%% \end{rcslog}
+
+
+
+%%% \end{document}
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%% Local Variables:
+%%% mode: LaTeX
+%%% TeX-brace-indent-level: 4
+%%% indent-tabs-mode: t
+%%% TeX-auto-untabify: nil
+%%% TeX-auto-regexp-list: LaTeX-auto-regexp-list
+%%% compile-command: "make cwebbase.tex"
+%%% End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/Makefile b/Master/texmf-dist/doc/latex/cweb-latex/examples/Makefile
new file mode 100644
index 00000000000..e23e3056dbf
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/Makefile
@@ -0,0 +1,29 @@
+# $Id: Makefile,v 1.5 1995/11/21 00:42:38 schrod Exp $
+#----------------------------------------------------------------------
+
+
+all:: wcltx.dvi
+
+
+wcltx.dvi: wcltx.tex wcltx.bib ../cweb.cls
+ tex-it latex wcltx
+ rm -f wcltx.blg
+
+wcltx.tex: wcltx.w
+ cweave wcltx
+
+
+ AUX = *.aux *.bbl *.dvi *.idx *.log *.scn *.tex
+
+distrib::
+ decomment-tex /usr/local/lib/texmf/tex/latex/misc/rcs.sty \
+ 'from CTAN, directory macros/tex/latex/contrib/supported/rcs/.' \
+ >rcs.sty
+ cp /usr/local/bin/tex-it .
+ $(MAKE) all
+ mkdir compare
+ mv $(AUX) compare
+
+
+clean::
+ /bin/rm -f $(AUX)
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.aux b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.aux
new file mode 100644
index 00000000000..c602ac21793
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.aux
@@ -0,0 +1,26 @@
+\relax
+\citation{litprog:schrod:cweb}
+\@writefile{toc}{\contentsline {section}{\numberline {1}An example of {\tt CWEB\/}{}}{1}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{1}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{1}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{2}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{2}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{2}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{3}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{3}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{3}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{3}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{4}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{4}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{4}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{4}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{4}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{5}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{5}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{5}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{5}}
+\bibstyle{plain}
+\bibdata{wcltx}
+\bibcite{litprog:schrod:cweb}{1}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{6}}
+\@writefile{toc}{\contentsline {chunk}{\copy \cweb@chunk_marker }{6}}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.bbl b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.bbl
new file mode 100644
index 00000000000..1e95c84cced
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.bbl
@@ -0,0 +1,8 @@
+\begin{thebibliography}{1}
+
+\bibitem{litprog:schrod:cweb}
+Klaus Guntermann and Joachim Schrod.
+\newblock {\WEB} adapted to {C}.
+\newblock {\em TUGboat}, 7(3):134--137, 1986.
+
+\end{thebibliography}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.dvi b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.dvi
new file mode 100644
index 00000000000..468bc34d4bf
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.dvi
Binary files differ
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.idx b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.idx
new file mode 100644
index 00000000000..dc79615c804
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.idx
@@ -0,0 +1,40 @@
+\I\\{argc}, \[5], 7, 8.
+\I\\{argv}, \[5], 7, 10, 17.
+\I\\{buf\_end}, \[12], 13, 16.
+\I\\{buf\_size}, \[12], 16.
+\I\\{buffer}, \[12], 13, 16.
+\I\.{BUFSIZ}, 12.
+\I\|{c}, \[12].
+\I\.{cannot open file}, 10.
+\I\\{cannot\_open\_file}, \[4], 10.
+\I\\{char\_count}, \[12], 13, 16, 17, 18, \[20].
+\I\\{close}, 11.
+\I\\{exit}, 5.
+\I\\{fd}, \[9], 10, 11, 16.
+\I\\{file\_count}, \[6], 7, 10, 17, 19.
+\I\\{fprintf}, 10, 20.
+\I\\{in\_word}, \[12], 13, 15.
+\I{Joke}, 14.
+\I\\{line\_count}, \[12], 13, 15, 17, 18, \[20].
+\I\\{main}, \[5], 14.
+\I\.{OK}, \[4].
+\I\\{open}, 10.
+\I\\{print\_count}, \[20].
+\I\\{printf}, 17, 19, 20.
+\I\\{prog\_name}, \[4], 5, 10, 20.
+\I\\{ptr}, \[12], 13, 15, 16.
+\I\\{read}, 16.
+\I\.{READ\_ONLY}, \[10].
+\I\\{status}, \[4], 5, 10, 20.
+\I\\{stderr}, 3, 10, 20.
+\I\\{stdin}, 9, 17.
+\I\\{stdout}, 3.
+\I\\{tot\_char\_count}, \[14], 18, 19.
+\I\\{tot\_line\_count}, \[14], 18, 19.
+\I\\{tot\_word\_count}, \[14], 18, 19.
+\I\.{Usage: ...}, 20.
+\I\\{usage\_error}, \[4], 20.
+\I\\{wc\_print}, 17, 19, \[20].
+\I\\{which}, \[6], 7, 17, 19, \[20].
+\I\\{word\_count}, \[12], 13, 15, 17, 18, \[20].
+
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.log b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.log
new file mode 100644
index 00000000000..a3f0d6136fd
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.log
@@ -0,0 +1,114 @@
+This is TeX, Version 3.1415 (C version 6.1) (format=latex 95.7.29) 30 NOV 1995 16:52
+**wcltx
+(wcltx.tex
+LaTeX2e <1995/06/01> patch level 3
+Hyphenation patterns for english, german, loaded.
+(/usr/local/lib/texmf/tex/generic/misc/cwebmac.tex) (../cweb.cls
+Document Class: cweb 1995/11/30 v3.6 LaTeX markup for CWEB sources
+(../keyvald.sty
+Package: keyvald 1994/08/23 v1.08.1 key=value parser (DPC) w/ defaults (js)
+) (/usr/local/lib/texmf/tex/latex/base/article.cls
+Document Class: article 1995/06/26 v1.3g Standard LaTeX document class
+(/usr/local/lib/texmf/tex/latex/base/size10.clo
+File: size10.clo 1995/06/19 v1.3e Standard LaTeX file (size option)
+)
+\c@part=\count82
+\c@section=\count83
+\c@subsection=\count84
+\c@subsubsection=\count85
+\c@paragraph=\count86
+\c@subparagraph=\count87
+\c@figure=\count88
+\c@table=\count89
+\abovecaptionskip=\skip41
+\belowcaptionskip=\skip42
+\bibindent=\dimen102
+) (../cwebbase.tex
+File: cwebbase.tex 1995/11/07 v1.4 CWEAVE tags for LaTeX markup
+\cwbb@undefined=\toks15
+\cwbb@to_restore=\toks16
+\CwebChunkPreSkip=\skip43
+\cweb@chunk_marker=\box26
+LaTeX Font Info: Try loading font information for OMS+cmr on input line 598.
+
+
+(/usr/local/lib/texmf/tex/latex/base/OMScmr.fd
+File: OMScmr.fd 1995/04/22 v2.3f Standard LaTeX font definitions
+)
+LaTeX Font Info: Font shape `OMS/cmr/m/n' in size <10> not available
+(Font) Font shape `OMS/cmsy/m/n' tried instead on input line 598.
+\c@chunk=\count90
+\c@section2=\count91
+\c@section3=\count92
+\c@section4=\count93
+\c@section5=\count94
+\c@section6=\count95
+\c@section7=\count96
+\c@section8=\count97
+\c@section9=\count98
+\c@section10=\count99
+\CwebSectPreSkip=\skip44
+\CwebProgPenalty=\count100
+\cwbb@save@parskip=\skip45
+\cwbb@save@rightskip=\skip46
+\cwbb@save@sem_sfcode=\count101
+\cwbb@save@pretolerance=\count102
+\cwbb@save@hyphenpenalty=\count103
+\cwbb@save@exhyphenpenalty=\count104
+\CwebIndentUnit=\dimen103
+\cwbb@indent=\dimen104
+\cwbb@bak=\box27
+\cwbb@bakk=\box28
+\cwbb@pointer=\box29
+LaTeX Font Info: External font `cmex10' loaded for size
+(Font) <7> on input line 1454.
+LaTeX Font Info: External font `cmex10' loaded for size
+(Font) <5> on input line 1454.
+\cwbb@decr=\box30
+\cwbb@incr=\box31
+)
+\CwebTocIndent=\dimen105
+\CwebTocIndentMaxLevel=\count105
+\cwebIndexIntro=\toks17
+) (rcs.sty
+Package: rcs 1995/08/02 v2.9 typeset info from RCS fields
+\rcsLogIntro=\toks18
+) (wcltx.aux)
+LaTeX Font Info: Checking defaults for OML/cmm/m/it on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: Checking defaults for T1/cmr/m/n on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: Checking defaults for OT1/cmr/m/n on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: Checking defaults for OMS/cmsy/m/n on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: Checking defaults for OMX/cmex/m/n on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 39.
+LaTeX Font Info: ... okay on input line 39.
+LaTeX Font Info: External font `cmex10' loaded for size
+(Font) <8> on input line 47.
+LaTeX Font Info: External font `cmex10' loaded for size
+(Font) <6> on input line 47.
+LaTeX Font Info: External font `cmex10' loaded for size
+(Font) <12> on input line 51.
+ *1
+[1
+
+] [2] [3] [4] [5] (wcltx.bbl [6]) [7] Index: (wcltx.idx) [8
+
+
+]
+List of Refinements: (wcltx.scn) [9
+
+] (wcltx.aux) )
+Here is how much of TeX's memory you used:
+ 709 strings out of 10921
+ 7728 string characters out of 72294
+ 56323 words of memory out of 262141
+ 3583 multiletter control sequences out of 9500
+ 8738 words of font info for 32 fonts, out of 100000 for 255
+ 14 hyphenation exceptions out of 607
+ 29i,8n,22p,172b,316s stack positions out of 300i,40n,60p,3000b,4000s
+
+Output written on wcltx.dvi (9 pages, 24308 bytes).
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.scn b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.scn
new file mode 100644
index 00000000000..4b0fbe3d4a8
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.scn
@@ -0,0 +1,32 @@
+\I\X11:Close file\X
+\U8.
+\I\X16:Fill \PB{\\{buffer}} if it is empty; \PB{\&{break}} at end of file\X
+\U15.
+\I\X20:Functions\X
+\U2.
+\I\X4, 14:Global variables\X
+\U2.
+\I\X3:Header files to include\X
+\U2.
+\I\X10:If a file is given, try to open \PB{${*}(\PP\\{argv})$}; \PB{%
+\&{continue}} if unsuccessful\X
+\U8.
+\I\X13:Initialize pointers and counters\X
+\U8.
+\I\X19:Print the grand totals if there were multiple files\X
+\U5.
+\I\X8:Process all the files\X
+\U5.
+\I\X15:Scan file\X
+\U8.
+\I\X7:Set up option selection\X
+\U5.
+\I\X5:The main program\X
+\U2.
+\I\X18:Update grand totals\X
+\U8.
+\I\X6, 9, 12:Variables local to \PB{\\{main}}\X
+\U5.
+\I\X17:Write statistics for file\X
+\U8.
+
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.tex b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.tex
new file mode 100644
index 00000000000..58a91af0247
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/compare/wcltx.tex
@@ -0,0 +1,412 @@
+\input cwebmac
+% $Id: wcltx.w,v 1.4 1995/08/29 17:27:57 schrod Exp $
+%------------------------------------------------------------
+
+%
+% wc: An example of CWEB by Silvio Levy and Donald E. Knuth
+%
+% [LaTeX]
+% (history at end)
+
+
+%% -js:
+%% should rework it: ANSI C compliance, POSIX.2 compliant option
+%% parsing with getopt(), including *all* required header files (ie,
+%% POSIX.1 compliance), using POSIX return codes and types, etc.
+
+
+
+\documentclass{cweb}
+
+\usepackage{rcs}
+
+
+%
+% some tags for this document
+%
+\def\CEE/{C\spacefactor 1000 }
+\def\cweb{{\tt CWEB\/}}
+\def\SPARC{SPARC\-\kern.1em station}
+\def\UNIX/{{\small UNIX\/}}
+\def\wc{{\tt wc\/}}
+\def\WEB{{\tt WEB\/}}
+
+\hfuzz=0.5pt
+
+
+
+
+\begin{document}
+
+
+
+\title{Counting words}
+\author{Silvio Levy\and Donald E. Knuth}
+\begingroup
+\def\thefootnote{\hbox{$^*$}}
+\footnotetext[0]{\LaTeX{} markup by Joachim Schrod.}
+\endgroup
+\RCSdate $Date: 1995/08/29 17:27:57 $
+
+\maketitle
+
+% not very interesting, only one starred section
+% \tableofcontents
+
+
+
+
+\N{1}{1}An example of \cweb{}. This example, based on a program by
+Klaus Guntermann and Joachim Schrod~\cite{litprog:schrod:cweb}
+presents the ``word count'' program from \UNIX/, rewritten in
+\cweb{} to demonstrate literate programming in \CEE/.%
+\footnote{%
+Incidentally, Klaus Guntermann and Joachim Schrod are now
+using this \cweb{} and have stopped the support for their version.
+Joachim Schrod contributed the \LaTeX{} style used for the markup of
+this document.
+}
+The level of detail in this document is intentionally high, for
+didactic purposes; many of the things spelled out here don't need to
+be explained in other programs.
+
+The purpose of \wc{} is to count lines, words, and/or characters in a
+list of files. The number of lines in a file is the number of newline
+characters it contains. The number of characters is the file length in bytes.
+A ``word'' is a maximal sequence of consecutive characters other than
+newline, space, or tab, containing at least one visible ASCII code.
+(We assume that the standard ASCII code is in use.)
+
+\fi
+
+\M{2}Most \cweb{} programs share a common structure. It's probably a
+good idea to state the overall structure explicitly at the outset,
+even though the various parts could all be introduced in unnamed
+sections of the code if we wanted to add them piecemeal.
+
+Here, then, is an overview of the file \texttt{wcltx.c} that is
+defined by this \cweb{} program \texttt{wcltx.w}:
+
+\Y\B\X3:Header files to include\X\6
+\X4:Global variables\X\6
+\X20:Functions\X\6
+\X5:The main program\X\par
+\fi
+
+\M{3}We must include the standard I/O definitions, since we want to send
+formatted output to \PB{\\{stdout}} and \PB{\\{stderr}}.
+
+\Y\B\4\X3:Header files to include\X${}\E{}$\6
+\8\#\&{include} \.{<stdio.h>}\par
+\U2.\fi
+
+\M{4} The \PB{\\{status}} variable will tell the operating system if the run
+was
+successful or not, and \PB{\\{prog\_name}} is used in case there's an error
+message to
+be printed.
+
+\Y\B\4\D$\.{OK}$ \5
+\T{0}\C{ \PB{\\{status}} code for successful run }\par
+\B\4\D$\\{usage\_error}$ \5
+\T{1}\C{ \PB{\\{status}} code for improper syntax }\par
+\B\4\D$\\{cannot\_open\_file}$ \5
+\T{2}\C{ \PB{\\{status}} code for file access error }\par
+\Y\B\4\X4:Global variables\X${}\E{}$\6
+\&{int} \\{status}${}\K\.{OK}{}$;\C{ exit status of command, initially \PB{%
+\.{OK}} }\6
+\&{char} ${}{*}\\{prog\_name}{}$;\C{ who we are }\par
+\A14.
+\U2.\fi
+
+\M{5}Now we come to the general layout of the \PB{\\{main}} function.
+
+\Y\B\4\X5:The main program\X${}\E{}$\6
+$\\{main}(\\{argc},\39\\{argv}){}$\1\1\6
+\&{int} \\{argc};\C{ the number of arguments on the \UNIX/ command line }\6
+\&{char} ${}{*}{*}\\{argv}{}$;\C{ the arguments themselves, an array of strings
+}\2\2\6
+${}\{{}$\1\6
+\X6:Variables local to \PB{\\{main}}\X\6
+${}\\{prog\_name}\K\\{argv}[\T{0}];{}$\6
+\X7:Set up option selection\X;\6
+\X8:Process all the files\X;\6
+\X19:Print the grand totals if there were multiple files\X;\6
+\\{exit}(\\{status});\6
+\4${}\}{}$\2\par
+\U2.\fi
+
+\M{6}If the first argument begins with a `\texttt{-}', the user is choosing
+the desired counts and specifying the order in which they should be
+displayed. Each selection is given by the initial character
+(lines, words, or characters). For example, `\texttt{-cl}' would cause
+just the number of characters and the number of lines to be printed,
+in that order.
+
+We do not process this string now; we simply remember where it is.
+It will be used to control the formatting at output time.
+
+\Y\B\4\X6:Variables local to \PB{\\{main}}\X${}\E{}$\6
+\&{int} \\{file\_count};\C{ how many files there are }\6
+\&{char} ${}{*}\\{which}{}$;\C{ which counts to print }\par
+\As9\ET12.
+\U5.\fi
+
+\M{7}\B\X7:Set up option selection\X${}\E{}$\6
+$\\{which}\K\.{"lwc"}{}$;\C{ if no option is given, print all three values }\6
+\&{if} ${}(\\{argc}>\T{1}\W{*}\\{argv}[\T{1}]\E\.{'-'}){}$\5
+${}\{{}$\1\6
+${}\\{which}\K\\{argv}[\T{1}]+\T{1};{}$\6
+${}\\{argc}\MM;{}$\6
+${}\\{argv}\PP;{}$\6
+\4${}\}{}$\2\6
+${}\\{file\_count}\K\\{argc}-\T{1}{}$;\par
+\U5.\fi
+
+\M{8}Now we scan the remaining arguments and try to open a file, if
+possible. The file is processed and its statistics are given.
+We use a \PB{\&{do}}~\dots~\PB{\&{while}} loop because we should read from the
+standard input if no file name is given.
+
+\Y\B\4\X8:Process all the files\X${}\E{}$\6
+$\\{argc}\MM;{}$\6
+\&{do}\5
+${}\{{}$\1\6
+\X10:If a file is given, try to open \PB{${*}(\PP\\{argv})$}; \PB{\&{continue}}
+if unsuccessful\X;\6
+\X13:Initialize pointers and counters\X;\6
+\X15:Scan file\X;\6
+\X17:Write statistics for file\X;\6
+\X11:Close file\X;\6
+\X18:Update grand totals\X;\C{ even if there is only one file }\6
+\4${}\}{}$\5
+\2\5
+\&{while} ${}(\MM\\{argc}>\T{0}){}$;\par
+\U5.\fi
+
+\M{9}Here's the code to open the file. A special trick allows us to
+handle input from \PB{\\{stdin}} when no name is given.
+Recall that the file descriptor to \PB{\\{stdin}} is~0; that's what we
+use as the default initial value.
+
+\Y\B\4\X6:Variables local to \PB{\\{main}}\X${}\mathrel+\E{}$\6
+\&{int} \\{fd}${}\K\T{0}{}$;\C{ file descriptor, initialized to \PB{\\{stdin}}
+}\par
+\fi
+
+\M{10}\B\D$\.{READ\_ONLY}$ \5
+\T{0}\C{ read access code for system \PB{\\{open}} routine }\par
+\Y\B\4\X10:If a file is given, try to open \PB{${*}(\PP\\{argv})$}; \PB{%
+\&{continue}} if unsuccessful\X${}\E{}$\6
+\&{if} ${}(\\{file\_count}>\T{0}\W(\\{fd}\K\\{open}({*}(\PP\\{argv}),\39\.{READ%
+\_ONLY}))<\T{0}){}$\5
+${}\{{}$\1\6
+${}\\{fprintf}(\\{stderr},\39\.{"\%s:\ cannot\ open\ fil}\)\.{e\ \%s\\n"},\39%
+\\{prog\_name},\39{*}\\{argv});{}$\6
+${}\\{status}\MRL{{\OR}{\K}}\\{cannot\_open\_file};{}$\6
+${}\\{file\_count}\MM;{}$\6
+\&{continue};\6
+\4${}\}{}$\2\par
+\U8.\fi
+
+\M{11}\B\X11:Close file\X${}\E{}$\6
+\\{close}(\\{fd});\par
+\U8.\fi
+
+\M{12}We will do some homemade buffering in order to speed things up:
+Characters
+will be read into the \PB{\\{buffer}} array before we process them.
+To do this we set up appropriate pointers and counters.
+
+\Y\B\4\D$\\{buf\_size}$ \5
+\.{BUFSIZ}\C{ \texttt{stdio.h}'s \PB{\.{BUFSIZ}} is chosen for efficiency }\par
+\Y\B\4\X6:Variables local to \PB{\\{main}}\X${}\mathrel+\E{}$\6
+\&{char} \\{buffer}[\\{buf\_size}];\C{ we read the input into this array }\6
+\&{register} \&{char} ${}{*}\\{ptr}{}$;\C{ the first unprocessed character in %
+\PB{\\{buffer}} }\6
+\&{register} \&{char} ${}{*}\\{buf\_end}{}$;\C{ the first unused position in %
+\PB{\\{buffer}} }\6
+\&{register} \&{int} \|c;\C{ current character, or number of characters just
+read }\6
+\&{int} \\{in\_word};\C{ are we within a word? }\6
+\&{long} \\{word\_count}${},{}$ \\{line\_count}${},{}$ \\{char\_count};\C{
+number of words, lines, and characters found in the file so far }\par
+\fi
+
+\M{13}\B\X13:Initialize pointers and counters\X${}\E{}$\6
+$\\{ptr}\K\\{buf\_end}\K\\{buffer};{}$\6
+${}\\{line\_count}\K\\{word\_count}\K\\{char\_count}\K\T{0};{}$\6
+${}\\{in\_word}\K\T{0}{}$;\par
+\U8.\fi
+
+\M{14}The grand totals must be initialized to zero at the beginning of the
+program. If we made these variables local to \PB{\\{main}}, we would have to
+do this initialization explicitly; however, \CEE/'s globals are automatically
+zeroed. (Or rather, ``statically zeroed.'') (Get it?)
+%% -js: The joke would be better if the vars would be static, which
+%% they aren't in the C sense...
+
+\Y\B\4\X4:Global variables\X${}\mathrel+\E{}$\6
+\&{long} \\{tot\_word\_count}${},{}$ \\{tot\_line\_count}${},{}$ \\{tot\_char%
+\_count};\C{ total number of words, lines, and chars }\par
+\fi
+
+\M{15}The present section, which does the counting that is \wc{}'s \emph{raison
+d'\^etre}, was actually one of the simplest to write. We look at each
+character and change state if it begins or ends a word.
+
+\Y\B\4\X15:Scan file\X${}\E{}$\6
+\&{while} (\T{1})\5
+${}\{{}$\1\6
+\X16:Fill \PB{\\{buffer}} if it is empty; \PB{\&{break}} at end of file\X;\6
+${}\|c\K{*}\\{ptr}\PP;{}$\6
+\&{if} ${}(\|c>\.{'\ '}\W\|c<\T{\~177}){}$\5
+${}\{{}$\C{ visible ASCII codes }\1\6
+\&{if} ${}(\R\\{in\_word}){}$\5
+${}\{{}$\1\6
+${}\\{word\_count}\PP;{}$\6
+${}\\{in\_word}\K\T{1};{}$\6
+\4${}\}{}$\2\6
+\&{continue};\6
+\4${}\}{}$\2\6
+\&{if} ${}(\|c\E\.{'\\n'}){}$\1\5
+${}\\{line\_count}\PP;{}$\2\6
+\&{else} \&{if} ${}(\|c\I\.{'\ '}\W\|c\I\.{'\\t'}){}$\1\5
+\&{continue};\2\6
+${}\\{in\_word}\K\T{0}{}$;\C{ \PB{\|c} is newline, space, or tab }\6
+\4${}\}{}$\2\par
+\U8.\fi
+
+\M{16}Buffered I/O allows us to count the number of characters almost for free.
+
+\Y\B\4\X16:Fill \PB{\\{buffer}} if it is empty; \PB{\&{break}} at end of file%
+\X${}\E{}$\6
+\&{if} ${}(\\{ptr}\G\\{buf\_end}){}$\5
+${}\{{}$\1\6
+${}\\{ptr}\K\\{buffer};{}$\6
+${}\|c\K\\{read}(\\{fd},\39\\{ptr},\39\\{buf\_size});{}$\6
+\&{if} ${}(\|c\Z\T{0}){}$\1\5
+\&{break};\2\6
+${}\\{char\_count}\MRL{+{\K}}\|c;{}$\6
+${}\\{buf\_end}\K\\{buffer}+\|c;{}$\6
+\4${}\}{}$\2\par
+\U15.\fi
+
+\M{17}It's convenient to output the statistics by defining a new function
+\PB{\\{wc\_print}}; then the same function can be used for the totals.
+Additionally we must decide here if we know the name of the file
+we have processed or if it was just \PB{\\{stdin}}.
+
+\Y\B\4\X17:Write statistics for file\X${}\E{}$\6
+$\\{wc\_print}(\\{which},\39\\{char\_count},\39\\{word\_count},\39\\{line%
+\_count});{}$\6
+\&{if} (\\{file\_count})\1\5
+${}\\{printf}(\.{"\ \%s\\n"},\39{*}\\{argv}){}$;\C{ not \PB{\\{stdin}} }\2\6
+\&{else}\1\5
+\\{printf}(\.{"\\n"});\C{ \PB{\\{stdin}} }\2\par
+\U8.\fi
+
+\M{18}\B\X18:Update grand totals\X${}\E{}$\6
+$\\{tot\_line\_count}\MRL{+{\K}}\\{line\_count};{}$\6
+${}\\{tot\_word\_count}\MRL{+{\K}}\\{word\_count};{}$\6
+${}\\{tot\_char\_count}\MRL{+{\K}}\\{char\_count}{}$;\par
+\U8.\fi
+
+\M{19}We might as well improve a bit on \UNIX/'s \wc{} by displaying the
+number of files too.
+
+\Y\B\4\X19:Print the grand totals if there were multiple files\X${}\E{}$\6
+\&{if} ${}(\\{file\_count}>\T{1}){}$\5
+${}\{{}$\1\6
+${}\\{wc\_print}(\\{which},\39\\{tot\_char\_count},\39\\{tot\_word\_count},\39%
+\\{tot\_line\_count});{}$\6
+${}\\{printf}(\.{"\ total\ in\ \%d\ files\\}\)\.{n"},\39\\{file\_count});{}$\6
+\4${}\}{}$\2\par
+\U5.\fi
+
+\M{20}Here now is the function that prints the values according to the
+specified options. The calling routine is supposed to supply a
+newline. If an invalid option character is found we inform
+the user about proper usage of the command. Counts are printed in
+8-digit fields so that they will line up in columns.
+
+\Y\B\4\D$\\{print\_count}(\|n)$ \5
+$\\{printf}(\.{"\%8ld"},\39\|n{}$)\par
+\Y\B\4\X20:Functions\X${}\E{}$\6
+$\\{wc\_print}(\\{which},\39\\{char\_count},\39\\{word\_count},\39\\{line%
+\_count}){}$\1\1\6
+\&{char} ${}{*}\\{which}{}$;\C{ which counts to print }\6
+\&{long} \\{char\_count}${},{}$ \\{word\_count}${},{}$ \\{line\_count};\C{
+given totals }\2\2\6
+${}\{{}$\1\6
+\&{while} ${}({*}\\{which}){}$\1\6
+\&{switch} ${}({*}\\{which}\PP){}$\5
+${}\{{}$\1\6
+\4\&{case} \.{'l'}:\5
+\\{print\_count}(\\{line\_count});\6
+\&{break};\6
+\4\&{case} \.{'w'}:\5
+\\{print\_count}(\\{word\_count});\6
+\&{break};\6
+\4\&{case} \.{'c'}:\5
+\\{print\_count}(\\{char\_count});\6
+\&{break};\6
+\4\&{default}:\6
+\&{if} ${}((\\{status}\AND\\{usage\_error})\E\T{0}){}$\5
+${}\{{}$\1\6
+${}\\{fprintf}(\\{stderr},\39\.{"\\nUsage:\ \%s\ [-lwc]\ }\)\.{[filename\ ...]%
+\\n"},\39\\{prog\_name});{}$\6
+${}\\{status}\MRL{{\OR}{\K}}\\{usage\_error};{}$\6
+\4${}\}{}$\2\6
+\4${}\}{}$\2\2\6
+\4${}\}{}$\2\par
+\U2.\fi
+
+\M{21}Incidentally, a test of this program against the system \wc{}
+command on a \SPARC\ showed that the ``official'' \wc{} was slightly
+slower. Furthermore, although that \wc{} gave an appropriate error
+message for the options `\texttt{-abc}', it made no complaints about the
+options `\texttt{-labc}'! Dare we suggest that the system routine
+might have been
+better if its programmer had used a more literate approach?
+
+
+
+\bibliographystyle{plain}
+\bibliography{wcltx}
+
+
+\fi
+
+\M{22}
+
+\cwebIndexIntro{%
+Here is a list of the identifiers used, and where they appear.
+Underlined entries indicate the place of definition. Error messages
+are also shown.
+}
+
+\end{document}
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: wcltx.w,v $
+% Revision 1.4 1995/08/29 17:27:57 schrod
+% Recoded to LaTeX2e markup.
+%
+% Revision 1.3 1993/06/15 14:01:14 schrod
+% Renamed document to wcltx, must change \bibliography tag, too.
+%
+% Revision 1.2 1993/06/15 13:35:12 schrod
+% `cweb' is now a style, was an option formerly.
+%
+% Revision 1.1 1993/04/09 15:00:37 schrod
+% Initial revision
+%
+\fi
+
+
+\inx
+\fin
+\con
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/rcs.sty b/Master/texmf-dist/doc/latex/cweb-latex/examples/rcs.sty
new file mode 100644
index 00000000000..e3cfaec72ec
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/rcs.sty
@@ -0,0 +1,227 @@
+% This is a stripped version of rcs.sty,
+% all comment lines and empty lines were discarded to save space. YOU
+% ARE NOT ALLOWED to use this file for your own work. You may ONLY
+% redistribute it as part of the complete distribution you received it.
+% This file was created at Thu Nov 30 15:51:58 UTC 1995.
+
+% The original file may be retrieved
+% from CTAN, directory macros/tex/latex/contrib/supported/rcs/..
+
+% ======================================================================
+
+\ifx \rcs@loaded\undefined
+ \def\rcs@loaded{$StyleRevision: 2.9 $}
+\else
+ \PackageWarningNoLine{rcs}%
+ {Some other package already uses namespace `rcs'}
+\fi
+\begingroup
+ \def\RCSPackage#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesPackage{#1}[\date\space\id\space #2]%
+ }
+ \RCSPackage{rcs}{typeset info from RCS fields}
+ $StyleDate: 1995/08/02 12:09:07 $: 9999/00/00 \endRCS
+ $StyleRevision: 2.9 $: 0.0 \endRCS
+\endgroup
+\ifx \CatEscape\undefined
+ \chardef\CatEscape=0
+ \chardef\CatOpen=1
+ \chardef\CatClose=2
+ \chardef\CatIgnore=9
+ \chardef\CatLetter=11
+ \chardef\CatOther=12
+ \chardef\CatActive=13
+ \chardef\CatUsCode=\catcode`\_
+\fi
+\catcode`\_=\CatLetter % top level macro file
+\def\RcsEmptyValue{}
+\def\rcs_split_field #1: #2 :#3\end_value{%
+ \def\RCS_keyword{#1}%
+ \def\RCS_value{#2}%
+ \ifx \RCS_value\empty
+ \let\RCS_value\RcsEmptyValue
+ \fi
+ }
+\def\RCS $#1${%
+ \rcs_split_field #1: {} :\end_value
+ \expandafter\let \csname RCS\rcs_raw_if_handle\RCS_keyword\endcsname
+ \RCS_value
+ \csname RcsHandle\RCS_keyword\endcsname
+ }
+\def\rcs_raw_if_handle{%
+ \@ifundefined{RcsHandle\RCS_keyword}{}{Raw}%
+ }
+\def\RcsHandleDate{%
+ \ifx \today\undefined
+ \expandafter\let \csname RCS\RCS_keyword\endcsname \RCS_value
+ \else
+ \ifx \RCS_value\RcsEmptyValue
+ \@namedef{RCS\RCS_keyword}{\today}%
+ \let\RCSTime\empty
+ \else
+ \expandafter\rcs_set_date \RCS_value\end_date
+ \fi
+ \fi
+ }
+\def\rcs_set_date #1/#2/#3 #4\end_date{%
+ \begingroup
+ \day #3 % <-- space!
+ \month #2 % <-- space!
+ \year #1 % <-- space!
+ \expandafter\xdef \csname RCS\RCS_keyword\endcsname {\today}%
+ \endgroup
+ \def\RCSTime{#4}%
+ }
+\newtoks\rcsLogIntro
+ \rcsLogIntro={}
+\def\rcsAuthor#1#2{\@namedef{rcs_author:#1}{#2}}
+\def\RCS_get_author#1{%
+ \@ifundefined{rcs_author:#1}{#1}{\@nameuse{rcs_author:#1}}%
+ }
+\def\RcsLogStyle{%
+ \def\RcsLogHeading{\subsubsection*}%
+ \footnotesize
+ }
+\def\rcs_log_hdr#1,v\end_value{%
+ \@rcs_empty_log@false
+ \RcsLogHeading{\RcsLogHeadingName #1}%
+ \the\rcsLogIntro
+ }
+\def\rcs_empty_log_value{%
+ \global\@rcs_empty_log@true % that's most probably in a group!
+ \RcsUnknownFile % text for unknown file name
+ ,v% % assert calling env
+ }
+\newif\if@rcs_empty_log@
+\def\RcsLogHeadingName{Revision Log for \tt} % file name in monospace
+\def\RcsUnknownFile{\rm $\langle\,$Unknown file name$\,\rangle$}
+\def\RcsEmptyLog{{\itshape No log entries available.}}
+\def\RcsLogRevision#1#2#3#4{%
+ \item [Revision #1]%
+ (created at \RcsLogDate #2\endDate\space \RcsLogTime{#3} % <-- space!
+ by \RCS_get_author{#4})\\\relax
+ }
+\def\RcsLogDate #1/#2/#3\endDate{%
+ \begingroup
+ \day #3 % <-- space!
+ \month #2 % <-- space!
+ \year #1 % <-- space!
+ \today
+ \endgroup
+ }
+\def\RcsLogTime#1{\ignorespaces}
+\def\rcs_settime{%
+ \def\RcsLogTime##1{[##1]}%
+ }
+\def\rcslog{%
+ \@ifnextchar[% % ] (Emacs)
+ \rcslog_configure
+ {\rcslog_configure[]}%
+ }
+\def\rcslog_configure[#1]{%
+ \let\Revision\rcslog_revision
+ \let\settime\rcs_settime
+ \RcsLogStyle
+ #1%
+ \afterassignment\rcslog_skipcr
+ \let\next
+ }
+\def\rcslog_skipcr{%
+ \ifx \next $%
+ \def\next{\rcslog_field $}%
+ \else\ifx \next\par
+ \let\next\rcslog_field
+ \else
+ \PackageError{rcs}{Missing RCS Log field in environment}{%
+The RCS Log field must be the very first text in the rcslog\MessageBreak
+environment. I will discard the rest of your input line and try to\MessageBreak
+recover then. If you think this won't work, exit now.\MessageBreak
+Otherwise press <Return> and cross your fingers!}
+ \let\next\rcslog_discard_line
+ \fi\fi
+ \next
+ }
+\def\rcslog_discard_line{%
+ \begingroup
+ \obeylines
+ \rcslog_gobble_line
+ }
+\begingroup
+\obeylines
+\gdef\rcslog_gobble_line #1^^M{%
+ \endgroup%
+ \afterassignment\rcslog_skipcr%
+ \let\next%
+ }
+\endgroup
+\def\rcslog_field{%
+ \begingroup
+ \obeylines
+ \rcslog_get_field
+ }
+\begingroup
+\obeylines
+\gdef\rcslog_get_field #1^^M{%
+ \endgroup%
+ \rcslog_set_field{#1}%
+ }
+\endgroup
+\def\rcslog_set_field#1{%
+ \let\RcsEmptyValue\rcs_empty_log_value
+ \RCS #1%
+ \expandafter\rcs_log_hdr \RCS_value\end_value
+ \list{}\RcsLogListStyle % empty label, layout configurable
+ \if@rcs_empty_log@
+ \item []\RcsEmptyLog % supply missing item
+ \fi
+ }
+\def\RcsLogListStyle{%
+ \global\@listdepth\z@
+ \labelwidth\z@ % item label w/o fixed size
+ \itemindent -\leftmargin % move label before start of line
+ \advance\itemindent \labelsep % label was moved too much
+ \def\makelabel##1{\bf ##1}% % how to typeset labels
+ \parsep\z@skip % no space between pars in items
+ \listparindent 1em % <-- space! indent pars in items
+ }%
+\let\endrcslog\endtrivlist
+\def\rcslog_revision{%
+ \begingroup
+ \obeylines
+ \rcslog_get_revinfo
+ }
+\begingroup
+\obeylines
+\gdef\rcslog_get_revinfo #1 #2 #3 #4^^M{%
+ \endgroup%
+ \RcsLogRevision{#1}{#2}{#3}{#4}%
+ }
+\endgroup
+\def\RCSdate $#1${%
+ \RCS $#1$%
+ \expandafter\date \expandafter{\csname RCS\RCS_keyword\endcsname}%
+ }
+\def\RCSID{%
+ \@ifnextchar\bgroup
+ \rcsid_get_field
+ \rcs_id
+ }
+\def\rcsid_get_field#1{\rcs_id #1}
+\def\rcs_id $#1${%
+ \RCS $#1$%
+ \begingroup
+ \let\protect\noexpand
+ \xdef\@oddfoot{%
+ \reset@font
+ \protect\framebox[\textwidth]{\RCS_keyword: \RCS_value}%
+ }%
+ \endgroup
+ \global\let\@evenfoot\@oddfoot
+ }
+\def\RCSdef $#1${%
+ \typeout{RCS field: $#1$}%
+ \RCS $#1$%
+ }
+\catcode`\_=\CatUsCode
+\endinput
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/tex-it b/Master/texmf-dist/doc/latex/cweb-latex/examples/tex-it
new file mode 100644
index 00000000000..07a8e98ad31
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/tex-it
@@ -0,0 +1,203 @@
+#!/bin/sh
+# $ITI: tex-it,v 1.3 1995/06/14 10:49:12 schrod Exp $
+#----------------------------------------------------------------------
+
+#
+# tex-it --- transform a TeX document to a final DVI file
+#
+# (history at end)
+
+
+# SYNOPSIS
+#
+# tex-it [-bi] tex-cmd file
+#
+# tex-it runs 'tex-cmd' on 'file' as often as necessary to produce a
+# DVI file where all cross references are resolved. That convergence
+# is reached by analyzing auxilliary files:
+# -- If file.toc has changed, the table of contents is not up to date.
+# -- If file.idx has changed, the index is not up to date.
+# MakeIndex is assumed to be used for creating the index.
+# -- If file.aux has changed, cross references are not up to date.
+# -- If file.aux contains a "\bibdata" tag and if the set of
+# "\citation" tags have changed, then BibTeX is run.
+#
+# No other auxilliary files are handled. No subdocuments (via
+# "\include") are supported.
+#
+# 'tex-cmd' may be more than one word.
+#
+#
+# OPTIONS
+# `-b' skip test for BibTeX
+# `-i' skip test for MakeIndex
+
+# NOTE: On HP-UX this script is not functional, you have to substitute
+# /bin/sh by /bin/ksh in the first line. (/bin/sh doesn't have the
+# getopts builtin.)
+
+
+cmd=`basename $0`
+usage()
+{
+ echo "usage: $cmd [-bi] tex-cmd file" >&2
+ exit 1
+}
+
+
+# Normalize and check arguments.
+
+do_bibtex=true
+do_makeindex=true
+while getopts 'bi' option
+ do case "$option" in
+ b) do_bibtex=''
+ ;;
+ i) do_makeindex=''
+ ;;
+ *) usage
+ ;;
+ esac
+ done
+
+shift `expr $OPTIND - 1`
+test $# -gt 0 || usage
+
+
+# Get file name (last argument), discard extension if necessary.
+
+file=`eval "echo \\$$#"`
+file=`echo $file | sed 's/\.[^.]*//'`
+
+
+# Create temporary directory for auxilliary files to compare. Discard
+# that directory if the script is terminated. Note that trap handler
+# for 0 is executed at the end of the trap handler for the other
+# signals.
+
+tmp=/tmp/tex$$
+trap "/bin/rm -rf $tmp" 0
+trap "exit 4" 1 2 3 15
+mkdir $tmp
+
+
+# ------------------------------------------------------------
+#
+# Set up functions that save auxilliary information and check if they
+# have changed. Save functions are called `save_<type>', check
+# functions are called `check_<type>'. Check functions set $run_tex to
+# "true" if TeX must be run anew. Check functions may run programs to
+# create other auxilliary files.
+
+
+# aux: table of contents & cross references
+#
+# We don't check for *.toc files. LaTeX stores the information in the
+# AUX file anyhow, and plain-based macros typically produce the table
+# of contents at the end of the document, when it's guaranteed to be
+# OK.
+
+save_aux()
+{
+ if [ -f "$file.aux" ]
+ then cp "$file.aux" $tmp
+ else touch "$tmp/$file.aux"
+ fi
+}
+
+check_aux()
+{
+ test -f "$file.aux" || return 0
+ cmp -s "$file.aux" "$tmp/$file.aux" || run_tex=true
+}
+
+
+# bibtex: Citations
+
+save_bibtex()
+{
+ if [ -f "$file.aux" ]
+ then egrep '^\\(citation|bibdata)' "$file.aux" |
+ sort >"$tmp/$file.bib"
+ else touch "$tmp/$file.bib"
+ fi
+}
+
+
+check_bibtex()
+{
+ test "$do_bibtex" -a -f "$file.aux" || return 0
+ egrep '^\\(citation|bibdata)' "$file.aux" | sort >$tmp/bib.new
+ if grep '^\\bibdata' $tmp/bib.new >/dev/null
+ then if cmp -s "$tmp/$file.bib" $tmp/bib.new
+ then : do nothing
+ else bibtex "$file"
+ run_tex=true
+ fi
+ fi
+}
+
+
+# idx: raw index data
+
+save_idx()
+{
+ if [ -f "$file.idx" ]
+ then cp "$file.idx" $tmp
+ else touch "$tmp/$file.idx"
+ fi
+}
+
+check_idx()
+{
+ test "$do_makeindex" -a -f "$file.idx" || return 0
+ if cmp -s "$file.idx" "$tmp/$file.idx"
+ then : do nothing
+ else makeindex "$file"
+ run_tex=true
+ fi
+}
+
+
+# Two functions that collect the functions above.
+
+save_aux_info()
+{
+ save_aux
+ save_bibtex
+ save_idx
+}
+
+check_aux_info()
+{
+ check_aux
+ check_bibtex
+ check_idx
+}
+
+
+# Process the document by TeX. Determine if TeX must be run anew by
+# the functions realized above.
+
+run_tex=true
+while [ "$run_tex" ]
+ do save_aux_info
+ "$@"
+ run_tex=''
+ check_aux_info
+ done
+
+
+
+#======================================================================
+#
+# $ITIlog: tex-it,v $
+# Revision 1.3 1995/06/14 10:49:12 schrod
+# HP-UX /bin/sh doesn't grok getopts, use ksh instead.
+#
+# Revision 1.2 1995/04/19 14:31:06 schrod
+# Support MakeIndex call, too.
+#
+# Revision 1.1 1995/03/14 17:07:26 schrod
+# Added script to process a TeX document.
+#
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.bib b/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.bib
new file mode 100644
index 00000000000..007879df1f4
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.bib
@@ -0,0 +1,23 @@
+$Id: wcltx.bib,v 1.1 1993/04/09 15:36:52 schrod Exp $
+----------------------------------------------------------------------
+
+%
+% extracted from litprog.bib database
+%
+
+@string{ tub = "TUGboat"}
+
+@article{litprog:schrod:cweb,
+ author = {Klaus Guntermann and Joachim Schrod},
+ title = {{\WEB} adapted to {C}},
+ journal = tub,
+ volume = 7,
+ number = 3,
+ month = {},
+ year = 1986,
+ pages = {134-137},
+ annote = {\js{} My {\small CWEB}, an adaption of Knuth's Pascal
+\WEB{}, ie, written in \WEB{}. Created to write an IBM~/370
+assembler/emulater for undergrad CS courses. I was an graduate
+student, Klaus was my supervisor then.}
+}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.w b/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.w
new file mode 100644
index 00000000000..4b82bd04f65
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/examples/wcltx.w
@@ -0,0 +1,327 @@
+% $Id: wcltx.w,v 1.4 1995/08/29 17:27:57 schrod Exp $
+%------------------------------------------------------------
+
+%
+% wc: An example of CWEB by Silvio Levy and Donald E. Knuth
+%
+% [LaTeX]
+% (history at end)
+
+
+%% -js:
+%% should rework it: ANSI C compliance, POSIX.2 compliant option
+%% parsing with getopt(), including *all* required header files (ie,
+%% POSIX.1 compliance), using POSIX return codes and types, etc.
+
+
+
+\documentclass{cweb}
+
+\usepackage{rcs}
+
+
+%
+% some tags for this document
+%
+\def\CEE/{C\spacefactor 1000 }
+\def\cweb{{\tt CWEB\/}}
+\def\SPARC{SPARC\-\kern.1em station}
+\def\UNIX/{{\small UNIX\/}}
+\def\wc{{\tt wc\/}}
+\def\WEB{{\tt WEB\/}}
+
+\hfuzz=0.5pt
+
+
+
+
+\begin{document}
+
+
+
+\title{Counting words}
+\author{Silvio Levy\and Donald E. Knuth}
+\begingroup
+\def\thefootnote{\hbox{$^*$}}
+\footnotetext[0]{\LaTeX{} markup by Joachim Schrod.}
+\endgroup
+\RCSdate $Date: 1995/08/29 17:27:57 $
+
+\maketitle
+
+% not very interesting, only one starred section
+% \tableofcontents
+
+
+
+@* An example of \cweb{}. This example, based on a program by
+Klaus Guntermann and Joachim Schrod~\cite{litprog:schrod:cweb}
+presents the ``word count'' program from \UNIX/, rewritten in
+\cweb{} to demonstrate literate programming in \CEE/.%
+ \footnote{%
+ Incidentally, Klaus Guntermann and Joachim Schrod are now
+using this \cweb{} and have stopped the support for their version.
+Joachim Schrod contributed the \LaTeX{} style used for the markup of
+this document.
+ }
+The level of detail in this document is intentionally high, for
+didactic purposes; many of the things spelled out here don't need to
+be explained in other programs.
+
+The purpose of \wc{} is to count lines, words, and/or characters in a
+list of files. The number of lines in a file is the number of newline
+characters it contains. The number of characters is the file length in bytes.
+A ``word'' is a maximal sequence of consecutive characters other than
+newline, space, or tab, containing at least one visible ASCII code.
+(We assume that the standard ASCII code is in use.)
+
+@ Most \cweb{} programs share a common structure. It's probably a
+good idea to state the overall structure explicitly at the outset,
+even though the various parts could all be introduced in unnamed
+sections of the code if we wanted to add them piecemeal.
+
+Here, then, is an overview of the file \texttt{wcltx.c} that is
+defined by this \cweb{} program \texttt{wcltx.w}:
+
+@c
+@<Header files to include@>@/
+@<Global variables@>@/
+@<Functions@>@/
+@<The main program@>
+
+@ We must include the standard I/O definitions, since we want to send
+formatted output to |stdout| and |stderr|.
+
+@<Header files...@>=
+#include <stdio.h>
+
+@ The |status| variable will tell the operating system if the run was
+successful or not, and |prog_name| is used in case there's an error message to
+be printed.
+
+@d OK 0 /* |status| code for successful run */
+@d usage_error 1 /* |status| code for improper syntax */
+@d cannot_open_file 2 /* |status| code for file access error */
+
+@<Global variables@>=
+int status=OK; /* exit status of command, initially |OK| */
+char *prog_name; /* who we are */
+
+@ Now we come to the general layout of the |main| function.
+
+@<The main...@>=
+main (argc,argv)
+ int argc; /* the number of arguments on the \UNIX/ command line */
+ char **argv; /* the arguments themselves, an array of strings */
+{
+ @<Variables local to |main|@>@;
+ prog_name=argv[0];
+ @<Set up option selection@>;
+ @<Process all the files@>;
+ @<Print the grand totals if there were multiple files @>;
+ exit(status);
+}
+
+@ If the first argument begins with a `\texttt{-}', the user is choosing
+the desired counts and specifying the order in which they should be
+displayed. Each selection is given by the initial character
+(lines, words, or characters). For example, `\texttt{-cl}' would cause
+just the number of characters and the number of lines to be printed,
+in that order.
+
+We do not process this string now; we simply remember where it is.
+ It will be used to control the formatting at output time.
+
+@<Var...@>=
+int file_count; /* how many files there are */
+char *which; /* which counts to print */
+
+@ @<Set up o...@>=
+which="lwc"; /* if no option is given, print all three values */
+if (argc>1 && *argv[1] == '-') { which=argv[1]+1; argc--; argv++; }
+file_count=argc-1;
+
+@ Now we scan the remaining arguments and try to open a file, if
+possible. The file is processed and its statistics are given.
+We use a |do|~\dots~|while| loop because we should read from the
+standard input if no file name is given.
+
+@<Process...@>=
+argc--;
+do@+{
+ @<If a file is given, try to open |*(++argv)|; |continue| if unsuccessful@>;
+ @<Initialize pointers and counters@>;
+ @<Scan file@>;
+ @<Write statistics for file@>;
+ @<Close file@>;
+ @<Update grand totals@>; /* even if there is only one file */
+}@+while (--argc>0);
+
+@ Here's the code to open the file. A special trick allows us to
+handle input from |stdin| when no name is given.
+Recall that the file descriptor to |stdin| is~0; that's what we
+use as the default initial value.
+
+@<Variabl...@>=
+int fd=0; /* file descriptor, initialized to |stdin| */
+
+@ @d READ_ONLY 0 /* read access code for system |open| routine */
+
+@<If a file...@>=
+if (file_count>0 && (fd=open(*(++argv),READ_ONLY))<0) {
+ fprintf (stderr, "%s: cannot open file %s\n", prog_name, *argv);
+@.cannot open file@>
+ status|=cannot_open_file;
+ file_count--;
+ continue;
+}
+
+@ @<Close file@>=
+close(fd);
+
+@ We will do some homemade buffering in order to speed things up: Characters
+will be read into the |buffer| array before we process them.
+To do this we set up appropriate pointers and counters.
+
+@d buf_size BUFSIZ /* \texttt{stdio.h}'s |BUFSIZ| is chosen for efficiency */
+
+@<Var...@>=
+char buffer[buf_size]; /* we read the input into this array */
+register char *ptr; /* the first unprocessed character in |buffer| */
+register char *buf_end; /* the first unused position in |buffer| */
+register int c; /* current character, or number of characters just read */
+int in_word; /* are we within a word? */
+long word_count, line_count, char_count; /* number of words, lines,
+ and characters found in the file so far */
+
+@ @<Init...@>=
+ptr=buf_end=buffer; line_count=word_count=char_count=0; in_word=0;
+
+@ The grand totals must be initialized to zero at the beginning of the
+program. If we made these variables local to |main|, we would have to
+do this initialization explicitly; however, \CEE/'s globals are automatically
+zeroed. (Or rather, ``statically zeroed.'') (Get it?)
+@^Joke@>
+%% -js: The joke would be better if the vars would be static, which
+%% they aren't in the C sense...
+
+@<Global var...@>=
+long tot_word_count, tot_line_count, tot_char_count;
+ /* total number of words, lines, and chars */
+
+@ The present section, which does the counting that is \wc{}'s \emph{raison
+d'\^etre}, was actually one of the simplest to write. We look at each
+character and change state if it begins or ends a word.
+
+@<Scan...@>=
+while (1) {
+ @<Fill |buffer| if it is empty; |break| at end of file@>;
+ c=*ptr++;
+ if (c>' ' && c<0177) { /* visible ASCII codes */
+ if (!in_word) {word_count++; in_word=1;}
+ continue;
+ }
+ if (c=='\n') line_count++;
+ else if (c!=' ' && c!='\t') continue;
+ in_word=0; /* |c| is newline, space, or tab */
+}
+
+@ Buffered I/O allows us to count the number of characters almost for free.
+
+@<Fill |buff...@>=
+if (ptr>=buf_end) {
+ ptr=buffer; c=read(fd,ptr,buf_size);
+ if (c<=0) break;
+ char_count+=c; buf_end=buffer+c;
+}
+
+@ It's convenient to output the statistics by defining a new function
+|wc_print|; then the same function can be used for the totals.
+Additionally we must decide here if we know the name of the file
+we have processed or if it was just |stdin|.
+
+@<Write...@>=
+wc_print(which, char_count, word_count, line_count);
+if (file_count) printf (" %s\n", *argv); /* not |stdin| */
+else printf ("\n"); /* |stdin| */
+
+@ @<Upda...@>=
+tot_line_count+=line_count;
+tot_word_count+=word_count;
+tot_char_count+=char_count;
+
+@ We might as well improve a bit on \UNIX/'s \wc{} by displaying the
+number of files too.
+
+@<Print the...@>=
+if (file_count>1) {
+ wc_print(which, tot_char_count, tot_word_count, tot_line_count);
+ printf(" total in %d files\n",file_count);
+}
+
+@ Here now is the function that prints the values according to the
+specified options. The calling routine is supposed to supply a
+newline. If an invalid option character is found we inform
+the user about proper usage of the command. Counts are printed in
+8-digit fields so that they will line up in columns.
+
+@d print_count(n) printf("%8ld",n)
+
+@<Fun...@>=
+wc_print(which, char_count, word_count, line_count)
+char *which; /* which counts to print */
+long char_count, word_count, line_count; /* given totals */
+{
+ while (*which)
+ switch (*which++) {
+ case 'l': print_count(line_count); break;
+ case 'w': print_count(word_count); break;
+ case 'c': print_count(char_count); break;
+ default: if ((status & usage_error)==0) {
+ fprintf (stderr, "\nUsage: %s [-lwc] [filename ...]\n", prog_name);
+@.Usage: ...@>
+ status|=usage_error;
+ }
+ }
+}
+
+@ Incidentally, a test of this program against the system \wc{}
+command on a \SPARC\ showed that the ``official'' \wc{} was slightly
+slower. Furthermore, although that \wc{} gave an appropriate error
+message for the options `\texttt{-abc}', it made no complaints about the
+options `\texttt{-labc}'! Dare we suggest that the system routine
+might have been
+better if its programmer had used a more literate approach?
+
+
+
+\bibliographystyle{plain}
+\bibliography{wcltx}
+
+
+@
+
+\cwebIndexIntro{%
+ Here is a list of the identifiers used, and where they appear.
+Underlined entries indicate the place of definition. Error messages
+are also shown.
+ }
+
+\end{document}
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: wcltx.w,v $
+% Revision 1.4 1995/08/29 17:27:57 schrod
+% Recoded to LaTeX2e markup.
+%
+% Revision 1.3 1993/06/15 14:01:14 schrod
+% Renamed document to wcltx, must change \bibliography tag, too.
+%
+% Revision 1.2 1993/06/15 13:35:12 schrod
+% `cweb' is now a style, was an option formerly.
+%
+% Revision 1.1 1993/04/09 15:00:37 schrod
+% Initial revision
+%
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/Imakefile b/Master/texmf-dist/doc/latex/cweb-latex/src/Imakefile
new file mode 100644
index 00000000000..64bd313b8f5
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/Imakefile
@@ -0,0 +1,106 @@
+XCOMM $Id: Imakefile,v 1.7 1995/11/21 00:41:31 schrod Exp $
+/* ---------------------------------------------------------------------- */
+
+/*
+ * Imakefile for cweb bundle
+ *
+ * (history at end)
+ */
+
+
+/* needed modules */
+#include <TeX.imk>
+#include <makeprog.imk>
+
+#include <install/TeX.imk>
+
+XCOMM ------------------------------------------------------------
+XCOMM continue in Imakefile
+
+
+/* exported files */
+
+MpDoc2TeXTarget(cwebbase,tex)
+MpDoc2TeXTarget(cweb,cls)
+MpDoc2TeXTarget(cwebparts,sty)
+
+TeXTarget(latex,cweb-user)
+TeXTarget(latex,cweb-conf)
+
+
+/* for some time, until it's moved to an own bundle */
+
+AllTarget(keyvald.sty)
+
+keyvald.sty: keyvald.dtx
+ latex keyvald.ins
+
+clean::
+ $(RM) keyvald.sty
+
+
+/* installation */
+
+#ifdef ITI_Site
+ DESTDIR = /archives/tex/Install
+#endif
+
+InstallTeXLaTeX(cweb,cweb.cls cwebbase.tex cwebarray.sty keyvald.sty \
+ cwebparts.sty cwbl-german.sty)
+InstallTeXDoc(latex/styles/cweb,cweb-user.dvi)
+InstallTeXDoc(latex/styles/cweb,cweb-conf.dvi)
+
+
+/* internal documentation */
+
+TeXFileTarget(latex,cweb,doc)
+TeXFileTarget(latex,cwebbase,doc)
+TeXFileTarget(latex,cwebparts,doc)
+TeXTarget(latex,cweave-spec)
+
+
+/* distribution */
+
+distrib:: all cweb.dvi cwebbase.dvi cwebparts.dvi
+ mv cweb-*.dvi *.cls cwebbase.tex keyvald.sty ..
+ cp cwebarray.sty cwbl-german.sty ..
+ cp $(HOME)/Notes/README.latex.doc-src README
+ $(MAKE) realclean
+
+realclean:: clean
+ $(RM) Makefile
+
+
+
+/* ============================================================
+ *
+ * $Log: Imakefile,v $
+ * Revision 1.7 1995/11/21 00:41:31 schrod
+ * German language package.
+ *
+ * Revision 1.6 1995/09/17 14:21:25 schrod
+ * Add support for German CWEB documents.
+ *
+ * Revision 1.5 1995/09/16 17:00:18 schrod
+ * Extract cwebbase.tex from cweb.cls, add cwebparts.sty.
+ *
+ * Revision 1.4 1995/08/29 17:34:39 schrod
+ * LaTeX CWEB, version 1.0.
+ *
+ * Revision 1.3 1995/08/29 15:19:47 schrod
+ * Added (hint to non-existence of) description of the internal interface.
+ *
+ * Revision 1.2 1995/08/25 19:00:32 schrod
+ * Add enhanced version of keyval package, with ability to specify
+ * default action for unknown keyword options.
+ *
+ * Revision 1.1 1995/08/08 00:14:24 schrod
+ * Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+ * my standard templates for that, no changes in functionality.
+ *
+
+ * ------------------------------------------------------------
+ * Local Variables:
+ * mode: Indented-Text
+ * End:
+ */
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/README b/Master/texmf-dist/doc/latex/cweb-latex/src/README
new file mode 100644
index 00000000000..0e7a4836715
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/README
@@ -0,0 +1,31 @@
+latex-style:src/README 29 Aug 95
+----------------------------------------------------------------------
+
+Please note: This directory is part of the distribution primarily for your
+information. It's not prepared to be usable by you, sorry.
+
+In particular, probably you won't have tools and style files I take
+for granted during development. Please don't complain, either use the
+supplied installable files (as listed in ../MANIFEST) or search & get
+the missing files. I will neither stop adding this directory to the
+distribution, nor will I add all missing files.
+
+On the one hand I won't restrict my own development environment by not
+using available tools, on the other hand I'll want to make my source
+available for those who have a similar environment to mine.
+
+
+For your orientation:
+
+ *.doc -- master file for TeX macro files
+ *.dvi -- typeset version of *.doc
+ *-user.tex -- user manual
+ *-conf.tex -- (optional) internal interface description
+ *.tex -- (optional) other documentation parts
+ *.{fig,latex} -- (optional) figures prepared with xfig
+ *.sty -- needed to process the above documents
+
+ TODO -- (optional) personal notes for future revisions
+ Imakefile -- does not use MIT's Imake templates
+ style/ -- (optional) files for AUC-TeX
+ test/ -- test cases
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/TODO b/Master/texmf-dist/doc/latex/cweb-latex/src/TODO
new file mode 100644
index 00000000000..82fddd2ee5c
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/TODO
@@ -0,0 +1,168 @@
+% $Id: TODO,v 1.10 1995/11/30 15:44:51 schrod Exp $
+%----------------------------------------------------------------------
+
+
+----------------------------------------------------------------------
+
+functionality [problem]
+
+In plain CWEB chunks are not split over pages, LaTeX CWEB does so. Add
+an option that re-establishs the behaviour of plain CWEB.
+
+Problem reported by Denis B. R\"ogel <denis.roegel@loria.fr>
+
+----------------------------------------------------------------------
+
+functionality [problem]
+
+User index would use with standard LaTeX markup (\index) would use
+*.idx as raw index file. This file is used by CWEAVE as tagged index
+file.
+
+Shall I redefine \makeindex?
+
+----------------------------------------------------------------------
+
+functionality [proposal]
+
+Add support to create the illusion of more than one refinement in a
+chunk.
+
+\cwebNoChunkIntro; see cwebx.sty.
+Proposed by Christopher Higgins <C.Higgins@cit.gu.edu.au>.
+Ack him for that.
+
+----------------------------------------------------------------------
+
+functionality [proposal]
+
+Add support for CWEB code examples.
+
+For that code from cwebparts.doc must be moved to cwebbase.doc; see
+cwebx.sty. That code has also several improvements over cwebparts.doc!
+cwebparts.doc is needed later for typesetting sections, etc. Well, if
+that can really be realized in a document/project independent way.
+Don't know if I really need it. (Except as a user front end. But I
+can also use cwebbase for that.)
+
+----------------------------------------------------------------------
+
+functionality [proposal]
+
+Add support for HyperTeX.
+
+Proposed by Christopher Higgins <C.Higgins@cit.gu.edu.au>.
+
+----------------------------------------------------------------------
+
+list headers [error]
+
+Use \chaptermark & \sectionmark, in \CwebListSection.
+
+Maybe discard \@mkboth from \CwebIdIndex, \CwebRefList, etc. It's
+supposed to be changed by a pagestyle, but this doesn't happen with
+fancyheadings anyhow -- and face it, fancyheadings is the method of
+choice for head-/footline creation.
+
+The actual code was caused by stupid copying from LaTeX standard
+classes; with the change it's possible to have other list heading
+layout (add to toc, other marks, etc.)
+
+But check if I can use that in \CwebIdIndex: There \CwebListSection is
+used in an \twocolumn optional arg, that will eventually trigger the
+output routine and will be set as a float. Does marks correctly
+migrate out of floats?
+
+----------------------------------------------------------------------
+
+layout [error]
+
+In chunks without documentation, a refinement is too near to the chunk
+marker. In addition, the distance is not independent of
+\CwebIndentUnit, making it impossible to change the latter.
+
+NOTE: Might be a problem of \cwebNoChunk that should start a
+paragraph. *When* will CWEAVE add `\4' after `\B'? Have to make tests
+for that. How can I detect an empty documentation section?
+
+\B kind of repaired in cwebx.sty.
+
+----------------------------------------------------------------------
+
+layout [error]
+
+Chunk numbers are not exactly aligned in all circumstances, see fixme
+remark in cwebbase.doc.
+
+----------------------------------------------------------------------
+
+interface [idea]
+
+Name all \cwbb@... macros used in other modules \Cwbb... (or even
+\cwbb...).
+
+----------------------------------------------------------------------
+
+markup [proposal]
+
+Support delimiting section titles by braces.
+
+Maybe one needs an option to turn that on.
+
+Supported by Peter Wainwright <prw-b@ugrad.cs.york.ac.uk>.
+
+----------------------------------------------------------------------
+
+documentation [ack]
+
+Who sent the mail concerning \hb@xt@? Add to acks. Change log message
+for it, too (cweb.doc: 3.4; cwebbase.doc: 1.4).
+
+----------------------------------------------------------------------
+
+documentation [problem]
+
+cweb-conf.tex:
+
+Cseq name of chunk marker (paragraph symbol).
+Ack Christopher Higgins \path|<c.higgins@cit.gu.edu.au>|.
+
+\CwebRankNoEject. (Implies explication of rank numbers. Should be
+explicated in cwebbase.doc as well, btw.)
+Ack Nicholas Urbanik \path|<nicku@iohk.com>|
+
+Support for another language must be stored in cwbl-*.sty. Check
+cwbl-german.sty for the cseqs that must be redefined, list their
+original English text for clarification.
+
+----------------------------------------------------------------------
+
+distribution [problem]
+
+cweb.cls & cwebbase.tex is > 100KB now. Export it in uncommented form.
+
+----------------------------------------------------------------------
+
+distribution [problem]
+
+It's better to collect stuff from contrib, and have each file that
+shall be installed in top dir.
+
+This means a Makefile in contrib/; and changes in README, INSTALL, and
+maybe even MANIFEST.
+
+----------------------------------------------------------------------
+
+distribution [proposal]
+
+Add hint to TODO in README. Maybe point out that src/ sometimes holds
+prereleases of future styles.
+
+----------------------------------------------------------------------
+
+distribution [proposal]
+
+Install TODO in doc dir. The end user should see open problems and
+plans, too.
+
+----------------------------------------------------------------------
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cwbl-german.sty b/Master/texmf-dist/doc/latex/cweb-latex/src/cwbl-german.sty
new file mode 100644
index 00000000000..68a7ce6c054
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cwbl-german.sty
@@ -0,0 +1,109 @@
+% $Id: cwbl-german.sty,v 1.1 1995/09/17 14:21:25 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% LaTeX2e package cwbl-german
+% adapt CWEB documents to language `german'
+%
+% [LaTeX2e]
+% (history at end)
+
+
+% Identification. The code below is explained in the implementation
+% documentation of the rcs package.
+
+\begingroup
+ \def\RCSPackage#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesPackage{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSPackage{cwbl-german}{German CWEB documents}
+ $Date: 1995/09/17 14:21:25 $: 9999/00/00 \endRCS
+ $Revision: 1.1 $: 0.0 \endRCS
+\endgroup
+
+
+%
+% What's this package for?
+%
+% XXX
+%
+
+
+% This class is supported. Send bug reports, comments and repairs.
+%
+% The reference version may be retrieved via anonymous ftp from
+% |ftp.th-darmstadt.de| [130.83.47.112], directory |pub/tex/latex|. It's
+% placed there as a gzipped tar file. (The information on the
+% IP~number is dated 17 Sep 95. It might have changed, also
+% this is very unlikely. Use your friendly nameserver.)
+
+
+%%% ------------------------------------------------------------
+
+
+% at the end of chunks
+
+\def\CwebCRAlso{\CwebCrossRef{Siehe auch Baustein}}
+\def\CwebCRsAlso{\CwebCrossRef{Siehe auch Bausteine}}
+
+\def\CwebCRCite{\CwebCrossRef{Dies wird zitiert im Baustein}}
+\def\CwebCRsCite{\CwebCrossRef{Dies wird zitiert in den Bausteinen}}
+
+\def\CwebCRUse{\CwebCrossRef{Dies wird benutzt im Baustein}}
+\def\CwebCRsUse{\CwebCrossRef{Dies wird benutzt in den Bausteinen}}
+
+\def\CwebCREt{ und~}
+\def\CwebCRsEt{ und~}
+
+
+% back matter
+
+\def\CwebCRChanged{%
+ \CwebCrossRef{%
+ Die folgenden Bausteine wurden durch ein Changefile ge\"andert:%
+ }%
+ }
+\def\CwebChangedListName{Ge\"anderte Bausteine}
+
+
+\def\CwebRefListName{Verzeichnis der Verfeinerungen}
+
+\def\CwebXRCite{\CwebCrossRef{Zitiert im Baustein}}
+\def\CwebXRCites{\CwebCrossRef{Zitiert in den Bausteinen}}
+\def\CwebXRUse{\CwebCrossRef{Benutzt im Baustein}}
+\def\CwebXRUses{\CwebCrossRef{Benutzt in den Bausteinen}}
+
+
+\def\CwebMacrosHere{%
+ \begingroup
+ \def\CwebRefNumber##1{}%
+ \CwebRefName :Pr\"aprozessor Definitionen\X
+ \endgroup
+ }
+
+
+
+\endinput
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: cwbl-german.sty,v $
+% Revision 1.1 1995/09/17 14:21:25 schrod
+% Add support for German CWEB documents.
+%
+
+
+
+%%%------------------------------------------------------------
+
+% Local Variables:
+% mode: LaTeX
+% TeX-brace-indent-level: 4
+% TeX-parse-self: nil
+% TeX-auto-save: nil
+% TeX-master: t
+% page-delimiter: "^%%%"
+% End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweave-spec.tex b/Master/texmf-dist/doc/latex/cweb-latex/src/cweave-spec.tex
new file mode 100644
index 00000000000..c1f5b4f07ff
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweave-spec.tex
@@ -0,0 +1,403 @@
+% $DocId: cweave-spec.tex,v 1.8 1995/09/12 22:56:02 schrod Exp $
+%------------------------------------------------------------
+
+%
+% specification of CWEAVE/TeX interface
+%
+% [LaTeX]
+% (history at end)
+
+
+% This file is either a subdocument of the style doc or a document on
+% its own. In the former case it's a chapter, in the latter it's a
+% ``normal'' LaTeX progltx document.
+% If it's a subdocument, this file will be included after
+% \begin{document}. We can detect this: \document redefines
+% \documentclass to be \@twoclasseserror. Then we also have to define
+% how this document is ended: Either by \endinput or by an additional
+% revision log.
+% Of course, this test works only if LaTeX 2e is used for processing.
+
+\expandafter\ifx \csname @twoclasseserror\endcsname \documentclass
+
+ \let\endSubDocument=\endinput
+
+ \chap The interface between \cweave{} and \TeX{}.
+ \label{sec:cweave-spec}
+
+\else
+
+ \let\endSubDocument=\relax
+
+ \documentclass{progltx}
+
+ \usepackage{cweb-doc} % additional document-specific markup
+ \usepackage{a4-9} % Tschichold's A4 layout
+
+
+ \begin{document}
+
+ \title{The Interface between \cweave{} and \TeX{}}
+ \author{Joachim Schrod%
+ \thanks{Email: \texttt{schrod@iti.informatik.th-darmstadt.de}}%
+ }
+ \RCSdate $DocDate: 1995/09/12 22:56:02 $
+
+ \maketitle
+
+ \sect
+
+\fi
+
+
+
+
+Here we present all tags output by \cweave{}%
+ \footnote{This information refers to \cweb{}~3.4. Differences to
+ other 3.x versions of \cweave{} are minor and can be ignored in
+ our context. See the revision log of the complete interface specs
+ document for more information.}
+in an ordered fashion. First we look at those
+tags which are part of the `protected interface,' i.e., they are
+visible to a \cweb{} user, but he must not use them. Then we consider
+the private tags, some are used in the documentation part, others
+are needed to typeset program code, and there are a few tags for
+typesetting special characters in strings.
+
+
+\sect Some tags output by \cweave{} are part of the protected
+interface even though they are not prefixed by |cweb|. We'll present
+them in the order they'll arrive in the document instance.
+
+The following table specifies in the second column if this tag takes
+arguments. If the entry is non-empty, it's either a number listing
+just how many arguments are expected; then usual argument passing is
+used. Or it displays the context required.
+ %
+\begin{cseqtab}
+|\ATL| & \verb*|#1 #2 | & \cweb{} operator: |@l|
+ && (how to output non-ASCII chars in \ctangle{})
+ && \arg1 hex code of mapped character
+ && \arg2 string output by \ctangle{}
+|\M| & 1 & \cweb{} structure tag: start of a chunk
+ && \arg1 chunk number
+|\N| & |#1#2#3.| & \cweb{} structure tag: start of a section
+ && \arg1 section rank, $0 \le |#1| \le 10$
+ && \arg2 chunk number
+ && \arg3 section title
+|\PB| & 1 & restricted program mode material
+ && \arg1 program code
+|\Y| & & between major pieces of a program part
+|\B| & & start program material
+|\fi| & & \cweb{} structure tag: end of a chunk
+|\9| & 1 & index entry, user defined layout
+ && \arg1 text of index entry
+\end{cseqtab}
+
+|\ATL| does only appear in front of the very first chunk. The
+chunk number is an explicit \TeX{} number which might be followed
+by a `changed flag' (see chunk~\ref{sec:text-tag.priv}). Note the
+usage of |\fi|, ie, each chunk must open an according |\if|. |\PB|
+might appear within its own argument (created by restricted program
+mode material in a refinement name within restricted program mode).
+
+|\9| deserves a further explanation: It is expected, though not
+defined, that it expands to an empty token list. The parameter will
+be a sort key, actually; the real key to be typeset will appear
+afterwards. So a \cweb{} user might index \TeX{} as
+`|@:TeX}{\TeX@>|', the index will feature it in the ``T''~chunk,
+not in the ``|\|''~chunk.
+
+With the exception of |\PB| the tags above will not appear in math mode.
+
+
+\sect The following tags might appear to be public ones, but they
+are, in fact, never used. That's because they will be placed after
+the |\fi| which terminates the last chunk. Our problem is that we
+have to produce the information which were produced by these tags in
+the Plain macros. Namely the optional lists poses us the problems. We
+must read behind the document end to gather the information which
+lists to produce.
+
+%% This label is used in cweb.doc
+\label{spec:sec:docend}
+
+We can depend on the following rules.
+ %
+\begin{enumerate}
+
+\item If |\end| is found, there will be no other of these tags.
+
+\item |\ch| comes first, but is optional.
+
+\item |\inx\fin| come next, they are mandatory.
+
+\item |\con| is optional again; if it is found it's the very last
+token to look at in this document.
+
+\end{enumerate}
+ %
+\begin{fixme}
+
+Perhaps |\inx| should be defined as |\end{document}|, and both |\ch|
+and |\inx| should be added to the protected interface. Then the user
+may simply omit |\end{document}|. But we should take care of a
+possible occurence anyhow.
+
+\end{fixme}
+
+Here are the tags mentioned in this chunk, and their respective
+meaning in the Plain version:
+ %
+\begin{cseqtab}
+|\ch| & |#1.| & Note which chunks are changed
+ && \arg1 list of chunk numbers, like in |\U| and such.
+|\inx| & & Create index
+|\fin| & & Create the table of refinement names
+|\con| & & Create the table of contents
+|\end| & & output if index and all lists were suppressed
+\end{cseqtab}
+
+
+
+\sect \label{sec:text-tag.priv}
+ Some tags appear only in special circumstances and may therefore be
+considered as private tags. The largest part of them concern the tagging of
+program code, we'll have a look at them later. First we present the
+tags used in other areas.
+
+Lists of chunk numbers occurs on several places: At the chunk
+start (where the list has actually only one element), within
+refinement names, in the identifier index, and for cross reference
+purposes. Cross references can be made at the end of a chunk, and
+in the refinement name list at the very end. Everywhere where a
+chunk number can occur it can be followed by a tag which shows that
+this chunk was changed by the changefile.
+ %
+\begin{cseqtab}
+|\*| & & tag after chunk number: this chunk is changed.
+\end{cseqtab}
+
+Within the identifier index we have also special tags. The identifiers
+are tagged like in the program mode, ie, with |\\| and |\&|. Remember
+that |\9|, listed above, appears also in the index.
+ %
+\begin{cseqtab}
+|\I| & \verb*|#1, | & start of an index entry
+ && \arg1 index entry
+|\[| & |#1]| & underlined chunk number in index
+ && \arg1 chunk number
+|\.| & 1 & |@.| index entry
+ && \arg1 index entry
+\end{cseqtab}
+
+In the list of refinement names the entries are marked similar to
+the index entries. But note that |\I| has no arg here.
+ %
+\begin{cseqtab}
+|\I| & & start of a new refinement name
+\end{cseqtab}
+
+
+\sect \label{sec:prog-tag}
+ OK, now we can have a look at the large amount of tags used
+for tagging program code. First, we have those which represent
+directly C or \C++ tokens.
+ %
+\begin{cseqtab}
+|\?| & & C operator: conditional expression
+|\AND| & & C operator: logical and
+|\CM| & & C operator: binary complement
+|\DC| & & \C++ operator: scope resolution
+|\E| & & C operator: equivalence
+ && and equivalence sign after refinement name on it's definition
+|\G| & & C operator: greater or equal
+|\GG| & & C operator: shift right
+|\I| & & C operator: not equal
+|\K| & & C operator: assignment
+|\LL| & & C operator: shift left
+|\MG| & & C operator: pointer to struct component
+|\MGA| & & \C++ operator: pointer to pointer to member
+|\MM| & & C operator: decrement
+|\MOD| & & C operator: modulo (actually, remainder)
+|\NULL| & & `quoted' identifier
+|\OR| & & C operator: binary or
+|\PA| & & \C++ operator: pointer to member
+|\PP| & & C operator: increment
+|\R| & & C operator: logical negation
+|\this| & & `quoted' identifier
+|\TeX| & & `customized' identifier
+|\V| & & C operator: logical or
+|\W| & & C operator: logical and
+|\XOR| & & C operator: binary exclusive or
+|\Z| & & C operator: less or equal
+\end{cseqtab}
+
+Other tokens have variable parts, passed as arguments.
+ %
+\begin{cseqtab}
+|\.| & 1 & C string
+ && \arg1 string
+|\)| & & discretionary break between string parts
+|\&| & 1 & reserved identifier
+ && \arg1 identifier
+|\\| & 1 & ``normal'' identifier with more than one chars
+ && \arg1 identifier
+\verb"\|" & 1 & ``normal'' identifier with one char
+ && \arg1 character
+|\C| & 1 & C comment
+ && \arg1 comment text
+|\MRL| & 1 & C operator: combined binary operators
+ && \arg1 operators, |\K| must print as `$=$'
+|\SHC| & 1 & \C++ comment
+ && \arg1 comment text
+|\T| & 1 & numeric constants
+ && \arg1 constant
+|\X| & |#1: #2\X| & refinement name
+ && \arg1 chunk number
+ && \arg2 refinement name
+\end{cseqtab}
+ %
+The refinement name (second argument of |\X|) may be a file name
+tagged by `|\.|'. Then the name must be set like a C~string.
+
+\cweb{} itself introduces its own operators:
+ %
+\begin{cseqtab}
+|\ATH| & & |@h| (place the preprocessor definitions here)
+|\D| & & |@d| (define macro)
+|\F| & & |@f| (format identifier like another one)
+|\J| & & |@&| (join)
+|\vb| & 1 & |@=| (pass arg verbatim by \ctangle{})
+ && \arg1 string
+\end{cseqtab}
+
+And we have some tags used for cross referencing. Chunk
+cross-referencing (actually, for program parts) is started with some
+tag, then follows a list of chunk numbers.
+ %
+\begingroup
+ \hfuzz=22pt % 21.76602pt too wide (with a4-9.sty)
+\begin{cseqtab}
+|\A| & & one add-on definition (``See also chunk'')
+|\As| & & more than one add-on definition (``See also chunks'')
+|\ET| & & separator between the last two numbers if there are only two
+|\ETs| & & separator between the last two numbers if there are more than two
+|\Q| & & the chunk where it is cited (``This code is cited in chunk'')
+|\Qs| & & more than one chunk where it is cited (``This code is cited in chunk'')
+|\U| & & used in one chunk (``This code is used in chunk'')
+|\Us| & & used in more than one chunk (``This code is used in chunks'')
+\end{cseqtab}
+\endgroup
+
+Program code must be indented according to its structure. Each
+``larger'' statement is typeset as one paragraph, usually only one
+line long. The basic indentation of the next statement may be
+incremented and decremented in given units. If a statement has to be
+broken in more than one line nevertheless, a hanging indentation is
+added to the basic indentation for the subsequent lines. One can mark
+an `optional' statement start, ie, some kind of optional paragraph
+start. If a line break must be inserted it should be inserted here and
+no hanging indentation should be added to the basic indentation.
+ %
+\begin{cseqtab}
+|\1| & & future stmts indented one more unit
+|\2| & & future stmts indented one less unit
+|\3| & 1 & optional line break within a statement
+ && \arg1 digit, penalty for line break
+|\4| & & backspace one indentation unit
+|\5| & & optional line break or small space (between run-in statements)
+|\6| & & line break
+|\7| & & line break and additional vertical space
+|\8| & & line is a preprocessor directive (must be issued at start of line)
+\end{cseqtab}
+
+
+\sect Some args are designated as strings above. Within these args the
+following tags are used to represent special characters:
+ %
+\begin{cseqtab}
+\verb*|\ | & & space
+|\&| & & ampersand
+|\\| & & backslash
+|\^| & & hat
+|\_| & & underscore
+|\{| & & left brace
+|\}| & & right brace
+|\~| & & tilde
+\end{cseqtab}
+
+
+\sect Last, we have tags which are used in their ``standard''
+\LaTeX{} meaning. We don't have much work with them:
+ %
+\begin{cseqtab}
+|\#| & & hash mark in program or in string
+|\$| & & dollar
+|\%| & & percent
+|\{| & & block start (is set in math mode)
+|\}| & & block end (is set in math mode)
+|\,| & & thin space (like the \LaTeX{} default)
+|\langle| & & left delimiter in include directives, or in templates
+|\le| & & C operator: less or equal
+|\ldots| & & Function protoypes: Variable number of arguments
+|\rangle| & & right delimiter in include directives, or in templates
+\end{cseqtab}
+
+
+
+
+\endSubDocument
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\vskip \PltxPreSectSkip
+
+\begin{rcslog}
+$DocLog: cweave-spec.tex,v $
+\Revision 1.8 1995/09/12 22:56:02 schrod
+|cweave -x| doesn't output |\vfill| in front of |\end| any more.
+Change noticed in \cweb{} 3.4.
+
+\Revision 1.7 1995/08/29 01:40:24 schrod
+Checked the interface description for \cweb{}~3.4.
+
+\Revision 1.6 1995/08/25 19:09:57 schrod
+Cleaned up terminology: Chunks are not named sections any more.
+Explicate possible range of section ranks.
+
+\Revision 1.5 1995/08/08 00:14:26 schrod
+Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+my standard templates for that, no changes in functionality.
+
+\Revision 1.4 1993/06/17 15:09:28 schrod
+\cweb{} 3.0 was released officially on June 16, 1993. Mentioned in
+the documentation that this version is needed for the |cweb| style.
+
+\Revision 1.3 1993/06/14 17:50:14 schrod
+Clarify the occurence pattern of the lists after the last section:
+which are optional, the expected order, etc.
+
+Changed `cseq' to `tag'. On this level of abstraction (that's the
+specification!), we're still not considering \TeX{}nical details.
+
+\Revision 1.2 1993/06/14 15:40:02 schrod
+Adapted to April 93 changes to \cweave{}: |\N| has three parameters
+(added group level), |\M| param is no data-tag any more, |\Z| and
+|\MRL| are new.
+
+Added further explanation of |\9| and |\X|.
+
+Replaced `program part name' by `refinement name'.
+
+\Revision 1.1 1993/04/09 15:00:37 schrod
+Initial revision
+
+\end{rcslog}
+
+
+
+\end{document}
+
+% LocalWords:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-conf.tex b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-conf.tex
new file mode 100644
index 00000000000..86dd2ae04e4
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-conf.tex
@@ -0,0 +1,103 @@
+% $DocId: cweb-conf.tex,v 2.1 1995/08/29 15:19:45 schrod Exp $
+%------------------------------------------------------------
+
+%
+% internal interface of cweb class
+%
+% [LaTeX]
+% (history at end)
+
+
+% This file is either a subdocument of the style doc or a document on
+% its own. In the former case it's a chapter, in the latter it's a
+% ``normal'' LaTeX progltx document.
+% If it's a subdocument, this file will be included after
+% \begin{document}. We can detect this: \document redefines
+% \documentclass to be \@twoclasseserror. Then we also have to define
+% how this document is ended: Either by \endinput or by an additional
+% revision log.
+% Of course, this test works only if LaTeX 2e is used for processing.
+
+\expandafter\ifx \csname @twoclasseserror\endcsname \documentclass
+
+ \let\endSubDocument=\endinput
+
+ \chap The internal interface.
+
+\else
+
+ \let\endSubDocument=\relax
+
+ \documentclass{progltx}
+
+ \usepackage{cweb-doc} % additional document-specific markup
+ \usepackage{a4-9} % Tschichold's A4 layout
+
+ \nofiles % no crossreferences used
+
+ \begin{document}
+
+ \title{The \texttt{cweb} Class\\
+ {\large (Internal Interface)}%
+ }
+ \author{Joachim Schrod%
+ \thanks{Email: \texttt{schrod@iti.informatik.th-darmstadt.de}}%
+ }
+
+ \RCS $DocDate: 1995/08/29 15:19:45 $
+ \date{%
+ \RCSDocDate\\[3pt] % LaTeX Error: Paragraph terminated too early
+ (Revision \RCSStyleRevision{} of \texttt{cweb.cls})%
+ }
+
+ \maketitle
+
+ \sect
+
+\fi
+
+
+
+
+This
+%
+\ifx \endSubDocument\relax
+ document
+\else
+ section
+\fi
+%
+shall describe the internal interface for \TeX{} programmers who want
+to write packages or subclasses to enhance the \cls{cweb} class. In
+particular, it shall describe all control sequences a programmer may
+redefine to adapt this class to special needs. (There are an awful lot
+of them, all cseqs that start with `|Cweb|'.)
+
+Due to time constraints this description has not been written yet. It
+depends on the demand when it will be written. I.e., if \emph{you} are
+a package/class writer and if you want to enhance the \cls{cweb}
+class, send me email. Please mention the kind of enhancement you want
+to do. I will then add information to this description as needed.
+
+
+
+
+\endSubDocument
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\vskip \PltxPreSectSkip
+
+\begin{rcslog}
+$DocLog: cweb-conf.tex,v $
+\Revision 2.1 1995/08/29 15:19:45 schrod
+Added (hint to non-existence of) description of the internal interface.
+
+\end{rcslog}
+
+
+
+\end{document}
+
+% LocalWords:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-doc.sty b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-doc.sty
new file mode 100644
index 00000000000..2859584f6cb
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-doc.sty
@@ -0,0 +1,133 @@
+% $Id: cweb-doc.sty,v 1.3 1995/08/25 19:07:54 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% tags for the documentation of cweb class
+%
+
+
+\ProvidesPackage{cweb-doc}
+
+
+%
+% Used packages
+%
+
+\begingroup
+ \let\begingroup=\endinput
+ \input{cweb.cls} % get style revision
+ \global\let\cweb@loaded=\cweb@loaded
+\endgroup
+\providecommand*\cweb@loaded{$StyleRevision$} % bootstrap
+
+
+
+%
+% Configuration of RCS fields
+%
+
+% keywords in style are prefixed by `Style', in documentation by `Doc'.
+\let\RcsHandleStyleDate=\RcsHandleDate % transform StyleDate value
+\let\RcsHandleDocDate=\RcsHandleDate % transform DocDate field
+
+% the revision of cweb.cls is needed by the documentation
+\def\RcsEmptyValue{$\langle$unknown$\rangle$}
+\expandafter\RCS \cweb@loaded % creates \RCSStyleRevision
+\def\RcsEmptyValue{}
+
+% author name
+\rcsAuthor{schrod}{Joachim Schrod}
+
+
+
+%
+% new environments
+%
+
+\font\manfnt=manfnt % with the dangerous bend sign
+\chardef\bend="7F % I like the non-S curve ;-)
+\newenvironment{fixme}{%
+ \trivlist
+ \hangindent 2em
+ \hangafter -2
+ \item[\hbox to 0pt{\hss \manfnt \bend}]%
+ \ignorespaces
+ }{%
+ \endtrivlist
+ }
+
+
+\newenvironment{cseqtab}{%
+ \quote
+ \obeycr
+ \tabular{lrl}%
+ \@gobblecr % eat following newline
+ }{%
+ \endtabular
+ \endquote
+ }
+
+
+\newenvironment{options}{%
+ \list{}{%
+ \labelwidth\z@
+ \itemindent -.5\leftmargin
+ \labelsep 1em
+ \def\makelabel##1{%
+ \hspace\labelsep
+ \normalfont\ttfamily ##1%
+ }%
+ }%
+ }{%
+ \endlist
+ }
+
+
+
+%
+% new tags
+%
+
+% logos
+\def\cweb{\texttt{CWEB}}
+\def\ctangle{\texttt{CTANGLE}}
+\def\cweave{\texttt{CWEAVE}}
+
+% other
+\renewcommand*\arg[1]{\textsl{Arg.~#1}:} % I don't use the arg operator
+
+\def\<#1>{\textit{#1}} % syntactic category
+
+\newcommand\cls[1]{\textsf{#1}} % LaTeX classes
+\newcommand\pkg[1]{\textsf{#1}} % LaTeX packages
+
+
+%
+% change of LaTeX parameters, for layout etc.
+%
+
+% underfull hboxes are not such a problem in program documentation
+
+\hbadness=4000
+
+
+% figure
+
+\def\textfraction{0.1} % must be at least 10% text on page
+\def\topfraction{0.9} % rest may be top insertion
+\def\floatpagefraction{0.8} % float pages only if the float is larger than 80%
+
+
+
+
+\endinput
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+Local Variables:
+mode: LaTeX
+TeX-brace-indent-level: 4
+indent-tabs-mode: t
+TeX-auto-untabify: nil
+End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.fig b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.fig
new file mode 100644
index 00000000000..729c992d272
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.fig
@@ -0,0 +1,78 @@
+#FIG 2.1
+80 2
+6 35 82 134 122
+1 1 0 1 -1 0 0 0 0.00000 1 0.000 83 100 47 18 83 100 130 119
+4 1 0 10 0 -1 0 0.00000 0 7 63 83 104 Documentation
+-6
+6 35 152 134 192
+6 62 158 107 180
+4 1 0 10 0 -1 0 0.00000 0 7 43 83 167 Restricted
+4 1 0 10 0 -1 0 0.00000 0 9 36 83 180 Program
+-6
+1 1 0 1 -1 0 0 0 0.00000 1 0.000 83 170 47 18 83 170 130 189
+-6
+6 116 17 215 57
+6 140 23 188 48
+4 1 0 10 0 -1 0 0.00000 0 9 48 164 32 (CR) Cross
+4 1 0 10 0 -1 0 0.00000 0 7 40 164 48 Reference
+-6
+1 1 0 1 -1 0 0 0 0.00000 1 0.000 164 35 47 18 164 35 211 54
+-6
+6 203 152 302 192
+1 1 0 1 -1 0 0 0 0.00000 1 0.000 251 170 47 18 251 170 298 189
+4 1 0 10 0 -1 0 0.00000 2 7 21 251 174 \TeX
+-6
+6 203 82 302 122
+1 1 0 1 -1 0 0 0 0.00000 1 0.000 251 100 47 18 251 100 298 119
+4 1 0 10 0 -1 0 0.00000 0 9 36 251 104 Program
+-6
+6 257 122 299 143
+6 257 131 287 143
+4 0 5 10 0 -1 0 0.00000 2 10 30 257 143 \\SHC
+-6
+4 0 5 10 0 -1 0 0.00000 2 10 42 257 132 \\C \\X
+-6
+6 221 47 266 71
+4 0 5 10 0 -1 0 0.00000 2 10 42 223 57 \\A \\Q
+4 0 5 10 0 -1 0 0.00000 2 8 18 235 69 \\U
+-6
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 254 119 254 152 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 245 83 208 46 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 119 41 89 81 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 101 83 125 51 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 131 174 203 174 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 0 1
+ 0 0 1.000 4.000 8.000
+ 131 164 203 164 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 0 1
+ 0 0 1.000 4.000 8.000
+ 89 119 89 152 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 131 95 203 95 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 0 1
+ 0 0 1.000 4.000 8.000
+ 131 105 203 105 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 1 0
+ 0 0 1.000 4.000 8.000
+ 74 119 74 152 9999 9999
+2 1 0 1 -1 0 0 0 0.000 -1 0 1
+ 0 0 1.000 4.000 8.000
+ 239 119 239 152 9999 9999
+4 2 5 10 0 -1 0 0.00000 2 10 24 72 136 \\PB
+4 1 5 10 0 -1 0 0.00000 2 8 24 167 160 \\PB
+4 1 5 10 0 -1 0 0.00000 2 8 18 166 92 \\B
+4 1 5 10 0 -1 0 0.00000 2 8 42 166 117 \\M \\N
+4 2 5 10 0 -1 0 0.00000 2 8 42 101 60 \\M \\N
+4 0 5 10 0 -1 0 0.00000 2 8 24 113 76 \\ch
+4 1 5 10 0 -1 0 0.00000 2 8 18 167 188 \\X
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.latex b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.latex
new file mode 100644
index 00000000000..8f624ed30af
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.latex
@@ -0,0 +1,62 @@
+\setlength{\unitlength}{0.012500in}%
+%
+\ifx \SetFigFont\undefined
+ \begingroup\makeatletter
+ % extract first six characters in \fmtname
+ \def\x#1#2#3#4#5#6#7\relax{\def\x{#1#2#3#4#5#6}}%
+ \expandafter\x\fmtname xxxxxx\relax \def\y{splain}%
+ \ifx\x\y % LaTeX or SliTeX?
+ \gdef\SetFigFont#1#2#3{%
+ \ifnum #1<17\tiny\else \ifnum #1<20\small\else
+ \ifnum #1<24\normalsize\else \ifnum #1<29\large\else
+ \ifnum #1<34\Large\else \ifnum #1<41\LARGE\else
+ \huge\fi\fi\fi\fi\fi\fi
+ \csname #3\endcsname}%
+ \else
+ \gdef\SetFigFont#1#2#3{\begingroup
+ \count@#1\relax \ifnum 25<\count@\count@25\fi
+ \def\x{\endgroup\@setsize\SetFigFont{#2pt}}%
+ \expandafter\x
+ \csname \romannumeral\the\count@ pt\expandafter\endcsname
+ \csname @\romannumeral\the\count@ pt\endcsname
+ \csname #3\endcsname}%
+ \fi
+ \endgroup
+\fi
+\begin{picture}(262,171)(37,651)
+\put( 84,735){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}Documentation}}}
+\put( 84,672){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}Restricted}}}
+\put( 84,659){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}Program}}}
+\thinlines
+\put( 84,669){\oval(94,36)}
+\put(165,807){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}(CR) Cross}}}
+\put(165,791){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}Reference}}}
+\put(165,804){\oval(94,36)}
+\put(252,669){\oval(94,36)}
+\put(252,665){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}\TeX}}}
+\put(252,739){\oval(94,36)}
+\put(252,735){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{}Program}}}
+\put(258,696){\makebox(0,0)[lb]{\smash{\SetFigFont{10}{12.0}{tt}\\SHC}}}
+\put(258,707){\makebox(0,0)[lb]{\smash{\SetFigFont{10}{12.0}{tt}\\C \\X}}}
+\put(224,782){\makebox(0,0)[lb]{\smash{\SetFigFont{10}{12.0}{tt}\\A \\Q}}}
+\put(236,770){\makebox(0,0)[lb]{\smash{\SetFigFont{10}{12.0}{tt}\\U}}}
+\put(255,720){\vector( 0,-1){ 33}}
+\put( 84,739){\oval(94,36)}
+\put(246,756){\vector(-1, 1){ 37}}
+\put(168,651){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{tt}\\X}}}
+\put(120,798){\vector(-3,-4){ 30}}
+\put(102,756){\vector( 3, 4){ 24}}
+\put(132,665){\vector( 1, 0){ 72}}
+\put(204,675){\vector(-1, 0){ 72}}
+\put( 90,687){\vector( 0, 1){ 33}}
+\put(132,744){\vector( 1, 0){ 72}}
+\put(204,734){\vector(-1, 0){ 72}}
+\put( 75,720){\vector( 0,-1){ 33}}
+\put(240,687){\vector( 0, 1){ 33}}
+\put( 73,703){\makebox(0,0)[rb]{\smash{\SetFigFont{10}{12.0}{tt}\\PB}}}
+\put(168,679){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{tt}\\PB}}}
+\put(167,747){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{tt}\\B}}}
+\put(167,722){\makebox(0,0)[b]{\smash{\SetFigFont{10}{12.0}{tt}\\M \\N}}}
+\put(102,779){\makebox(0,0)[rb]{\smash{\SetFigFont{10}{12.0}{tt}\\M \\N}}}
+\put(114,763){\makebox(0,0)[lb]{\smash{\SetFigFont{10}{12.0}{tt}\\ch}}}
+\end{picture}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.ltx b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.ltx
new file mode 100644
index 00000000000..aa5cdce2f26
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-fsa.ltx
@@ -0,0 +1,29 @@
+% $Id: cweb-fsa.ltx,v 1.2 1995/08/08 00:14:27 schrod Exp $
+%------------------------------------------------------------
+
+%
+% test of xfig created pictures
+%
+% [LaTeX]
+
+
+
+
+\documentstyle{article}
+
+
+\begin{document}
+
+\noindent Test of figure \texttt{cweb-fsa.latex}: Finite State
+Automaton for different CWEB modi.
+
+\bigskip
+
+\begin{center}
+ \chardef\\=`\\
+ \input{cweb-fsa.latex}
+\end{center}
+
+
+
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-user.tex b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-user.tex
new file mode 100644
index 00000000000..2f21f53150c
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb-user.tex
@@ -0,0 +1,779 @@
+% $DocId: cweb-user.tex,v 2.10 1995/11/20 22:34:16 schrod Exp $
+%------------------------------------------------------------
+
+%
+% user manual/requirement definition for cweb class
+%
+% [LaTeX]
+% (history at end)
+
+
+% This file is either a subdocument of the style doc or a document on
+% its own. In the former case it's a chapter, in the latter it's a
+% ``normal'' LaTeX progltx document.
+% If it's a subdocument, this file will be included after
+% \begin{document}. We can detect this: \document redefines
+% \documentclass to be \@twoclasseserror. Then we also have to define
+% how this document is ended: Either by \endinput or by an additional
+% revision log.
+% Of course, this test works only if LaTeX 2e is used for processing.
+
+\expandafter\ifx \csname @twoclasseserror\endcsname \documentclass
+
+ \let\endSubDocument=\endinput
+
+ \chap What's this class for?.
+
+\else
+
+ \let\endSubDocument=\relax
+
+ \documentclass{progltx}
+
+ \usepackage{cweb-doc} % additional document-specific markup
+ \usepackage{a4-9} % Tschichold's A4 layout
+
+
+ \begin{document}
+
+ \title{The \texttt{cweb} Class}
+ \author{Joachim Schrod%
+ \thanks{Email: \texttt{jschrod@acm.org}}%
+ }
+
+ \RCS $DocDate: 1995/11/20 22:34:16 $
+ \date{%
+ \RCSDocDate\\[3pt] % LaTeX Error: Paragraph terminated too early
+ (Revision \RCSStyleRevision{} of \texttt{cweb.cls})%
+ }
+
+ \maketitle
+
+ \sect
+
+\fi
+
+\cweb{} is a Literate Programming tool that enables you to combine
+documentation and actual C or \C++ code in one document. The \cls{cweb}
+class allows you to use \LaTeX{} for the documentation of your \cweb{}
+programs. Now a \cweb{} document is like a usual \LaTeX{}
+document---as you have probably written dozens before. With all the
+features of \LaTeX{}, and with all the quirks of \LaTeX{}.
+
+Originally, \cweb{} was only usable with Plain \TeX{}, as \cweave{}
+outputs \TeX{} code that looks as if it were Plain \TeX{}. Well, it
+isn't---it's just not the normal \LaTeX{} markup. No change in
+\cweave{} is necessary to use \LaTeX{}.
+
+For the sake of those \cweb{} programmers who have used the original
+markup interface (i.e., the Plain \TeX{} macros) before, I'll
+sometimes contrast the approach presented here with the original one.
+Then I'll call the original interface ``Plain \cweb{}''. The new
+interface is called ``\LaTeX{} \cweb{}''.
+
+If you never used \cweb{} before, you won't find enough information in
+this manual. Please read the user manual of \cweb{} first, and then
+come back again. I assume a working knowledge, and will only present
+the new stuff introduced by this class.
+
+
+\sect This bundle does \emph{not} support inclusion of \cweb{} code in
+other (general) \LaTeX{} documents. This is not the task of the
+\cls{cweb} class -- its task is the usage of \LaTeX{} for creation of
+\cweb{} documents.
+
+Note also that you will not be able to \LaTeX{} your \cweb{} documents
+(i.e., the program files) directly. You must still process them first
+by \cweave{}, then you can use \LaTeX{} to process the resulting
+\TeX{} file.
+
+
+\sect You have to use \cweb{} version~3 and you need at least \LaTeX{}
+version \mbox{$\langle$1994/12/01$\rangle$} (or any later one), this
+class does not work with older versions.
+
+
+\sect A \cweb{} document has two concurrent structures, a
+\textsl{document structure} and a \textsl{program structure}.
+
+The basic building block of the document structure, the bricks that
+build up our document, are called \textsl{chunks}. We use chunks to
+present one point of our work to our reader, the presentation has both
+informal (explanation) and formal parts (code). This is the essence of
+Literate Programming, that we tell each point twice, once on an
+abstract and once on a technical level.
+
+A chunk consists of three parts: (1)~the \textsl{documentation part},
+(2)~the \textsl{definition part}, and (3)~the \textsl{program part}.
+Each of these parts can be empty. The documentation part is mostly
+text with \LaTeX{} tags. In this text material from \textsl{restricted
+program mode} can appear. The definition part consists of a series of
+either \textsl{macro} or \textsl{format definitions}. The program part
+is one piece of a refinement, identified by a name (see below).
+
+A chunk starts with `\verb*|@ |', and gets assigned a unique
+identification, a number; we'll use that number to refer to that
+chunk. In particular, the number is of interest when we want to refer
+to the code part of the chunk, otherwise it's often regarded as
+superfluous.
+
+\sect And we have higher-level structure elements as well. They combine a
+sequence of chunks and a sequence of structure elements on a lower
+level, both sequences may be empty. As usual in \LaTeX{}, we call
+these elements \textsl{section divisions} or short \textsl{sections}.
+We do not give them names
+like `part,' `chapter,' `subsection,' etc., to denote their level;
+most often we'll use level-numbers (sometimes called \textit{rank})
+instead.
+
+A section division starts with `|@*|' and is followed by a character, the rank
+indicator. The highest rank is~`|@**|', you can think of it as an
+equivalence to \LaTeX{}'s |\part|. (Often we don't use that rank in
+smaller programs, as we don't use |\part| in smaller articles.) A
+section on the next rank is started with~`\verb*|@* |' (synonym to
+`|@*0|'), roughly equivalent to chapters. Then we'll use numbers to
+mark the next ranks, from |@*1| to~|@*9| (sections to
+sub$^{9}$sections). That's outrageous---I'll hope you'll never use
+11~hierarchy levels in your documents, they would be unreadable.
+
+A section has a title, that title comes after the section start tag
+and ends with a full stop. If you need periods in your title, enclose
+them in braces. (That behavior can be changed if there's enough
+interest. I can also make it possible to use braces to delimit the
+section title, as in |\section|, etc. Send mail with your opinion.)
+
+As a matter of convenience, a section implicitly starts a chunk;
+i.e., one does not need to use `\verb*|@ |' after the section title.
+Of course, it doesn't hurt to use it either.
+
+\medskip
+
+The explanation might have been a bit too abstract, so let me make one
+thing plain clear: You should \emph{never ever} use |\part|,
+|\chapter|, |\section|, or any other \LaTeX{} sectioning command in
+your \cweb{} documents. Use the sectioning commands of \cweb{}, i.e.,
+`|@*|\textit{rank}'.
+
+
+
+\sect The description above does not mean that the output is fixed to
+the style you'll know from the \cls{book} or \cls{report} document
+classes. Quite in contrary, the default layout is oriented along the
+\cls{article} class.
+
+There is a very easy possibility to change that default layout: With
+the option |baseclass| (see chunk~\ref{chunk:baseclass}) you can
+specify the document class \cls{cweb} inherits from. If that document
+class has chapters (like \cls{report} or \cls{book}), you'll get the
+layout as you're used from these classes.
+
+
+\sect In Plain \cweb{}, the term \textit{section} was used for
+`chunk.' We abstain from that usage, as sections are used in \LaTeX{}
+(and in almost all other documentation systems, for that matter) to
+designate higher-level structure elements. We prefer to follow the
+traditional usage of terms here, even if that means that we'll have to
+confront Plain \cweb{} users with a slight shift in semantics.
+
+And you'll find the term \textit{starred section} in Plain \cweb{} and
+might think these are our sections. Well, almost. There's a subtle,
+but important difference between these starred sections and our
+terminology: A starred section is just a chunk with a special
+attribute (a title). That title will appear in a table of contents,
+but that does not mean that the starred section is an element to
+denote a structural \emph{hierarchy}.
+
+You won't find a section number like `1.5.2' in Plain \cweb{}, do you?
+--- In \LaTeX{} \cweb{}, you've got the choice. You have the new
+support of the hierarchical structure, or you may stay with Plain
+\cweb{}'s (flat) structure.
+
+
+\sect We still have one open point in our explanation of the
+document's structure. As mentioned at the start, we've got two
+concurrent structures, let's have a look at the program structure
+finally.
+
+A \cweb{} program consists of a tree of \textsl{refinements}. A
+refinement is a sequence of program parts with the same name, ordered
+in appearance in the document. The root of the tree is the refinement
+with the special name~|@c|.
+
+The complete program text is defined by the depth-first traversal
+(DFS, i.e.\ infix-order) of the tree, the tool \ctangle{} extracts the
+program text from a \cweb{} document.
+
+
+
+
+\chap Markup for your document structure.
+
+Like every \LaTeX{} document, a \cweb{} document starts with a
+document class specification; then comes the preamble, terminated by
+the document start. This way we brought some structure in the \cweb{}
+limbo, but this shouldn't be a problem for you. You must tag the end
+of the document, as you do in other \LaTeX{} files. But here you must
+take a bit more care:
+%
+\begin{quote}
+ \itshape
+
+ Assert that the\/ |\end{document}| is in the documentation part of
+ a\/ chunk, neither in the definition nor in the program part.
+
+\end{quote}
+%
+If it would be in the program part, \cweave{} would readily process it
+and \LaTeX{} would never see this tag. Nevertheless, if you make this
+mistake, you will detect it early: \ctangle{} will copy this tag to
+the C file as well, and your program will be erroneous\,\dots
+
+And there's another difference you'll have to take care of: Usually
+you can put arbitrary text after the end of your document, \LaTeX{}
+will not see it. Please note that both \ctangle{} and \cweave{} will
+still see it---don't put a program part or even complete chunks there.
+For technical reasons, there must not be any |\fi| token after the
+document end, too.
+
+
+\sect The resulting structure of your \cweb{} document is exemplified
+in figure~\ref{fig:structure}. Note that |\end{document}| is placed
+in a new, empty, chunk. This isn't necessary, you could write more
+documentation there---but it's considered good style to use a sole
+chunk for document finish.
+
+%% my little joke
+\begin{figure}
+
+\setbox0=\vbox{\parindent=0pt \hsize=.85\textwidth
+\begin{minipage}{\hsize}
+\begin{verbatim}
+\documentclass{cweb}
+
+\begin{document}
+
+\title{My Program, Doomed for the ACM Software Systems Award}
+\author{Joe L. User}
+
+\maketitle
+
+\tableofcontents % if you want
+
+
+@* A PSPACE solution for the Traveling Salesman.
+
+< insert your program here >
+
+
+
+@
+\end{document}
+\end{verbatim}
+\end{minipage}
+}
+
+\begin{center}
+\fboxsep=1em
+\fbox{\box0}
+\end{center}
+
+\caption{Exemplified \cweb{} document structure}
+\label{fig:structure}
+\end{figure}
+
+
+\sect If you don't call \cweave{} with the |-x| option, an identifier
+index and a list of all refinements is created at the end of the
+document. You can specify an introductionary text for the index with
+the tag |\cwebIndexIntro|, the introduction is the argument of this
+tag.
+
+
+
+
+\chap Configuration.
+
+You may configure the \cls{cweb} class in several ways, with class
+options, redefinitions in the preamble, supplying additional
+packages, and by subclassing it.
+
+At start of the processing, the file |cweb.cfg| is read if it exists.
+That file may contain configuration code that shall apply to all your
+\cweb{} documents. You can set the paper size there, etc. Please don't
+set the default structure there (see
+chunk~\ref{chunk:structure-option}), others would get different results
+when they process your document.
+
+
+\sect Options are specified with the new ``keyword$=$value''-scheme,
+first introduced with the \pkg{graphicx} package. I.e., the options have a
+set of possible values, you may choose one. For instance, the option
+|structure| has the possible values |flat| and |hierarchic|. You tell
+that you want a flat document structure by
+%
+\begin{quote}
+ |\documentclass[structure=flat]{cweb}|
+\end{quote}
+%
+As usual, options are separated by commas. From the set of possible
+option values, two ones are special: The \textsl{predefined value} and
+the \textsl{default value}. The predefined value is the one that's
+used when you don't specify the option at all. The default value is
+used of you don't specify the value in the option; e.g.,
+%
+\begin{quote}
+ |\documentclass[suppress]{cweb}|
+\end{quote}
+%
+selects the default value of the option |suppress|. (Not that there is
+one currently. :--)
+
+There is always a predefined value, a default value may not be there.
+If a default value exists, it is always different from the predefined
+one. The default value gives you an easy specification of an option
+value that is expected to be common. If that value would be the
+predefined one, you won't need to specify the option in the first
+hand.
+
+The following options are available and explained below:
+%
+\begin{options}
+\item[structure] Select the structure model used by the \cls{cweb}
+ class.
+\item[suppress] Suppress output of different document parts.
+\item[baseclass] Select the base document class.
+\item[language] Select a language for inserted texts.
+\end{options}
+
+
+\sect \label{chunk:structure-option}
+
+We have two possibilities to output chunks and sections, either
+as a flat or as a hierarchic structure. This configuration is selected
+by the option |structure|, either through the value |flat| or
+|hierarchic|.
+%
+\begin{options}
+
+\item[flat] The flat structure is the ``classic'' approach,
+ the way Plain \cweb{} renders its chunks. Then each chunk starts
+ with its number. The chunk numbers are output in boldface, followed
+ by a dot and a quad.
+
+ Sections show chunk numbers as well (remember, they implicitly
+ start a chunk). ``Important'' sections are added to the table of
+ contents.
+
+\item[hierarchic] The hierarchic structure is the
+ ``modernistic'' (aka \LaTeX{}) approach, then we mark the start of a
+ chunk by a start sign. (By default, that's a paragraph sign, `\P').
+ Chunk numbers are rendered in the margin of the program part. If
+ there's no program part, no chunk number will be rendered---after
+ all, they are only used for identification of program parts.
+
+ Section numbers are hierarchic, like in other \LaTeX{} document
+ classes.
+
+\end{options}
+%
+The predefined value of |structure| is |hierarchic|. There doesn't
+exist a default value.
+
+
+\sect The option |suppress| allows to select suppression of different
+document parts. You have different parts you can suppress. The value
+of the |suppress| option is a comma-separated list of identifiers,
+enclosed in braces. If you have only one thing you want to suppress,
+you don't need to surround that identifier by braces. Possible
+identifiers are:
+
+\begin{options}
+
+\item[changehints] suppress output of hints that a changefile was
+ used: The change flag (by default a star at the start of a chunk)
+ and the list of changed chunks at the end of the document.
+
+\item[unchanged] suppress output of chunks that are not changed by
+ a changefile. This value implies |changehints|---by definition all
+ printed chunks are changed, it doesn't make any sense to hint to
+ that fact in addition.
+
+\item[index] suppress the identifier index.
+
+\item[reflist] suppress the refinement list.
+
+\item[format] suppress output of `|@f|' directives.
+
+\end{options}
+
+The predefined value is the empty list, i.e., nothing is suppressed.
+There is no default value since I don't know which parts are
+suppressed most often. Send email with your demands, I'll add a
+default if I see that there is one major application of this option.
+
+
+\sect \label{chunk:baseclass} The option |baseclass| allows to select
+the base document class used by the \cls{cweb} class. In particular,
+if the base class has |\chapter| defined the layout of major section
+divisions (`|@**|' and `\verb*|@* |') will be changed to be in a
+chapter layout.
+
+The predefined value is |article|, the default value is |report|.
+
+
+\sect The option |language| allows to adapt inserted texts to
+different languages. E.g., with the option `|language=german|' German
+texts are inserted for cross references, headings, etc. That option
+will neither switch on language-specific hyphenation, nor will it
+trigger any other processing -- that's the task of Babel or any other
+\LaTeX{} language style.
+
+There is neither a predefined nor a default value. (Without the
+|language| option, inserted texts are in English.)
+
+That option actually triggers the usage of a package
+\pkg{cwbl-\textit{language}}. To see if a language is supported, you
+have to look for respective files in the \LaTeX{} \cweb{} macro
+directory.%
+ \footnote{That's |texmf/tex/latex/cweb|, on a TDS compliant
+ installation. Please beware that pseudo operating systems like DOS
+ cripple file names and shorten them to eight characters; if you
+ see a file |cwbl-ger.sty| you'll know that the option value
+ |german| is supported.}
+Of course, the easiest way to check the availability is to try it
+out -- you'll get a comprehensible error message if there is no
+respective \cweb{} language package. In November 1995, |german|,
+|french|, and |italian| were supported, more might have been added
+later.
+
+\smallskip
+
+For compatibility with Babel, some other options (e.g., |german| or
+|french|) are also supported as nicknames for
+|language=|\textit{option}. If you use packages that make also use of
+these options (e.g., \pkg{babel} or \pkg{varioref}), you'll probably
+prefer them over |language| since they are picked up by those packages
+as well.
+
+
+
+
+
+\chap Restrictions.
+
+The following restrictions will not be withdrawn as far as I can
+see---except if somebody will send me a patch with changes. I tried to
+sort the restrictions in order of severity.
+
+
+\sect \emph{(Restrictions due to the design of \cweave{}.)}
+
+Please be aware that the vertical bar (`\verb"|"') is used by \cweb{}
+to delimit small C code pieces in the documentation parts, and is
+therefore processed (and mangled) by \cweave{}. \emph{You cannot use
+it for \TeX{} any more.}
+
+In particular, you cannot specify rules for the |tabular| or the
+|array| environment. Since you most certainly want to do so: You have
+two choices left:
+%
+\begin{enumerate}
+
+\item Make sure you have the \pkg{array} package (by Frank Mittelbach and
+ David Carlisle, from the Tools bundle) installed. Then you may use
+ the package \pkg{cwebarray}, it defines `|I|' (that's an uppercase~`i')
+ as a specifier for rules. I.e., instead of
+ \verb"\begin{tabular}{l|l}" you have to write
+ |\begin{tabular}{lIl}|. Not the most elegant solution, but it
+ works\,\dots
+
+\item
+ Use `|^^7c|' instead of `\verb"|"'. I.e., instead of
+ \verb"\begin{tabular}{l|l}" you may write |\begin{tabular}{l^^7cl}|.
+
+\end{enumerate}
+%
+These two choices are compatible, you may use both in one document.
+Needless to say, I consider the first alternative the better one.
+
+
+\sect Neither a refinement name nor an index entry made by~|@^| may
+consist of a \emph{single} dot-accented term. I.e., you must not
+write `|@<\.O@>|', `|@^\.o@>|', or even `|@^\.{foolish}@>|'. Of
+course you may write `|@^\.o, accent@>|' or `|@< Handle accent \.o @>|'.
+
+
+\sect \emph{(Restriction due to the implementation of \LaTeX{}.)}
+
+One cannot use restricted C mode in moving arguments. Most notably,
+this is disturbing in the section titles and in |\caption|
+tags. Sorry, folks.
+
+(Basically, that's because the definition of |\addcontentsline|, etc.,
+in the \LaTeX{} kernel is brain damaged; it wants to expand its
+argument. And I don't want to maintain redefinitions of such basic
+macros.)
+
+
+\sect \emph{(Restrictions due to \TeX{}.)}
+
+\C++ comments (i.e., from |//| to the end of the line) are typeset as C
+comments. This is especially bad if they are used for a whole block of
+comment lines, as it is quite common. Please put such comment blocks
+in the documentation part. (An |\everyline| implementation would be
+needed to lift that restriction, and that's impossible in general.)
+
+
+
+
+\chap Details.
+
+The following tags are reserved and must neither be used nor redefined:
+ |\ATL|,
+ |\B|,
+ |\M|,
+ |\N|,
+ |\PB|,
+ and~|\Y|.
+
+|\9| is already explained in the \cweb{} user manual: It's a special
+control sequence used for the index entries tagged with~`|@:|'. Its
+default definition is setup in such a way that you can cheat
+\cweave{} concerning the sort order of this entry: If you enter
+`|@:sort}{print@>|' you will get an index entry ``print'' next to the
+place where the index entry ``sort'' would be. But you're allowed to
+change this default definition.%
+ \footnote{Remember: A package is a very good place to place such
+ redefinitions; your document should be concerned with contents, not
+ with appearance.}
+
+The names of all other control sequences defined by this
+class---besides the common \LaTeX{} control sequences---start with
+|cweb|, |Cweb|, or |cwbb|. Please don't define new control sequences starting
+with this prefix. (The control sequences starting with |Cweb| may be
+redefined in a package to change the appearance of the \cweb{}
+document or the behavior in a controlled manner, check the
+description of the internal interface for this possibility.)
+
+
+\sect Plain \cweb{} users
+should note that the macros from |cwebmac.tex| are not available any
+more. E.g., you cannot use |\.| to typeset typewriter material; use
+either |\texttt| or~|\verb|, as it fits the situation. On the other hand,
+now you're able to use |\.| for the dot accent, you can define |\3|
+for the `sharp~s', |\C++| for the \C++ logo,%
+ \footnote{I will not define any such logos in this class. A
+ package like \pkg{logos} with lots of name definitions is the
+ appropriate place for this.}
+etc.
+
+Another detail for ex-plainies: The table of contents is produced by
+the |\tableofcontents| tag (during the second \LaTeX{} run), not
+automatically. But this is the standard \LaTeX{} behavior.
+
+
+
+
+\chap Known Problems.
+\label{sec:problems}
+
+There is a bunch of known bugs, problems, and ommisions.
+
+\bigbreak
+
+\noindent \textit{Bugs:}
+ %
+\begin{itemize}
+
+\item The presentation of |@l| redefinitions is not proper. But it
+wasn't in Plain \cweb{}, either.
+
+\end{itemize}
+
+\noindent \textit{Problems:}
+ %
+\begin{itemize}
+
+\item You can't use |\index| and |\makeindex| as-is to create an
+ index. Add your entries instead to \cweave{}'s index using `|@^|',
+ `|@.|', or `|@:|'. That problem is caused by a clash in the usage of
+ the filename extension |idx| by \cweave{} and \LaTeX{}. The
+ \pkg{index} package by \textsc{David Jones} will help you to define
+ additional index categories (that use an other extension) if you
+ don't want to mix identifiers and your other index entries.
+
+\end{itemize}
+
+\noindent \textit{Ommisions} (``wouldn't it be nice, if''s):
+ %
+\begin{itemize}
+
+\item The current chunk number and most section titles (below rank 1
+ or~2, depending on the base class) are not available as a mark for
+ inclusion in a running head.
+
+\item Better integration of \pkg{rcs} package. That package is used to
+ support including information from the revision management system
+ RCS into the document. E.g., the revision log at the end of that
+ document was produced by that package.
+
+\end{itemize}
+
+
+\sect I would like to thank those who helped me to improve this bundle.
+% In particular, XXX provided XXXsubstantial parts of the code.
+
+\textsc{Michael M\"uller} and \textsc{Zden\v{e}k Wagner} did thorough
+checks that helped me to improve the alpha test version.
+
+\textsc{Christian Kumpf} triggered the addition of language support
+and pointed out where configuration could be made easier by supplying
+more macros in the protected interface. \textsc{Andreas Scherer}
+provided packages for French and Italian \cweb{} documents.
+
+\textsc{Laurent Desnogues},
+\textsc{Felix G\"artner},
+and
+\textsc{John S. Robinson}
+provided error and problem reports.
+\textsc{Bronne Louis} pointed out the embarrassing error that section
+numbers were not reset when a higher-level section began.
+
+
+
+\endSubDocument
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\newpage
+
+\begin{rcslog}
+$DocLog: cweb-user.tex,v $
+\Revision 2.10 1995/11/20 22:34:16 schrod
+Add language support.\\
+Triggered by Christian Kumpf \path|<smurf@igd.fhg.de>| and Andreas
+Scherer \path|<scherer@physik.rwth-aachen.de>|.
+
+Use my `official' (ACM) email address.
+
+\Revision 2.9 1995/11/20 10:39:34 schrod
+ Explicate that one must not use |\section| etc., that one cannot
+\LaTeX{} the \cweb{} program directly but must run \cweave{} first,
+and that the \cls{cweb} class does not provide inclusion of \cweb{}
+programs in other \LaTeX{} documents.\\
+Clarification triggered by Laurent Desnogues
+\path|<laurent.desnogues@aiguemarine.unice.fr>|.
+
+\Revision 2.8 1995/09/15 10:22:58 schrod
+The \cls{cweb} class needs at least \LaTeX{} version
+\mbox{$\langle$1994/12/01$\rangle$}.\\
+Problem reported by John S. Robinson \path|jsrobin@umiacs.umd.edu|.
+
+\Revision 2.7 1995/09/12 22:58:38 schrod
+|cwbb| is also a reserved namespace.
+
+Added acknowledgements.
+
+\Revision 2.6 1995/08/29 17:34:14 schrod
+ \LaTeX{} \cweb{}, version 1.0.
+
+\Revision 2.5 1995/08/29 02:07:28 schrod
+Discard dependencies on 10\,pt fonts.
+
+Support suppression of format directives.
+
+\Revision 2.4 1995/08/27 19:31:11 schrod
+Suppression of index and reference list is supported.
+
+Discard documentation on |\cwebSecNoEject|, that's part of the
+protected interface.
+
+\Revision 2.3 1995/08/27 17:24:45 schrod
+Make usage of baseclass with chapters work.
+
+\Revision 2.2 1995/08/27 13:25:59 schrod
+Add possibility to suppress change hints. Suppression of unchanged
+chunks suppress change hints as well, they are meaningless as all
+printed chunks are changed by definition.
+
+Update problems section, they were partly resolved by the current
+changes. Added hint to \pkg{index} package that resolves the problem
+of extensions clashes with \MakeIndex{}.
+
+\Revision 2.1 1995/08/25 19:11:17 schrod
+Add keyword-value option style, with new \pkg{keyvald} package.
+
+Hierarchic strucutures are supported now, in addition to the flat
+structure of the beta-test version. One can choose with an option. For
+that step, the terminology was cleaned up, too: Chunks are not named
+sections any more. (That change involved reimplementation of almost
+all the structure and toc stuff.)
+
+The chunk number supplied by \cweave{} is used now, not some computed
+number. Change flags are printed, too.
+
+One can suppress output of unchanged sections.
+
+One can select the baseclass with an option. That may be used to use
+\cls{report} or \cls{book} to get chapter-style layout. Of course,
+using an arbitrary baseclass is dangerous, it must conform to the
+conventions of \LaTeX{} standard classes.
+
+\Revision 1.10 1995/08/08 00:14:28 schrod
+Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+my standard templates for that, no changes in functionality.
+
+\Revision 1.9 1993/08/10 14:15:40 schrod
+New page on main section only if group level $<$ |\cwebSecNoEject|.
+Default for the latter is 3.
+
+Document that logos will not be defined in this style file.
+
+Copy of plain macros for |\CwebNumber| does not work. Repaired the
+most important one (subscript must be accessed via |\sb|).
+Incompatibility to NFSS will be addressed later.\\
+(Problems reported by Zden\v{e}k Wagner \path|<wagner@csearn.bitnet>|.)
+
+\Revision 1.8 1993/08/10 11:21:03 schrod
+Reference to section number does not render a period after the
+number any more.
+
+\Revision 1.7 1993/08/09 18:05:54 schrod
+Mentioned that `\verb"|"' cannot be used for LaTeX purposes, in
+particular, not for ruled tables. Described workarounds, one of them
+is the new style option |cwebarray|.\\
+(Problem reported by Felix G\"artner
+\path|<theedge@rbg.informatik.th-darmstadt.de>|.)
+
+\Revision 1.6 1993/06/17 16:10:06 schrod
+`|cweb|' was still an option in the example.
+
+\Revision 1.5 1993/06/17 15:09:30 schrod
+\cweb{} 3.0 was released officially on June 16, 1993. Mentioned in
+the documentation that this version is needed for the |cweb| style.
+
+\Revision 1.4 1993/06/15 15:19:11 schrod
+It's a style now, was an option formerly.
+
+\Revision 1.3 1993/06/14 15:40:42 schrod
+Added section about restrictions (no restricted C mode material
+in moving arguments, no single dot-accented term in an index entry).
+
+\Revision 1.2 1993/05/12 20:23:32 schrod
+Adapted to recent changes of \cweave{} (of April 93): Main sections
+have a group level, represented in the table of contents.
+
+Boxed the example, it was not legible formerly.
+
+\Revision 1.1 1993/04/09 15:00:37 schrod
+Initial revision
+
+\end{rcslog}
+
+
+
+\end{document}
+
+% LocalWords: baseclass DFS PSPACE graphicx changehints changefile reflist idx
+% LocalWords: cwebarray cwebmac plainies ommisions if''s rcs cwbb uller Zden
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cweb.doc b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb.doc
new file mode 100644
index 00000000000..8fb7a0a5582
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cweb.doc
@@ -0,0 +1,1396 @@
+% $StyleId: cweb.doc,v 3.6 1995/11/30 15:44:53 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+
+%
+% LaTeX class cweb
+% support for LaTeX markup in CWEB sources
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+% If you have received this style file without the user manual (in the
+% file cweb-user.dvi or the respective LaTeX file), it's incomplete and
+% near to useless. If it was given to you as something that you shall
+% use as an author -- complain bitterly to your provider. You need the
+% documentation and you have a right on it! (Below you can find info
+% where to get the reference version.)
+
+
+
+
+\documentclass{progltx}
+
+\usepackage{cweb-doc} % document-specific markup
+\usepackage{fullpage}
+
+
+\RCS $StyleRevision: 3.6 $
+\RCS $StyleDate: 1995/11/30 15:44:53 $
+
+
+\begin{document}
+
+
+\title{The \texttt{cweb} Class\\
+ {\large (Implementation)}%
+ }
+\author{% % LaTeX does not discard unnecessary glue...
+ Joachim Schrod%
+ \thanks{%
+ \protect\raggedright
+ TU~Darmstadt, Computer Science Department, WG Systems Programming,
+ Alexanderstr.~10, D-64283~Darmstadt, Germany.
+ Email: \texttt{jschrod@acm.org}%
+ }%
+ }
+\date{%
+ Revision \RCSStyleRevision\\
+ (as of \RCSStyleDate)%
+ }
+
+\maketitle
+
+
+% doesn't work with progltx yet
+%\tableofcontents
+
+
+
+% ------------------------------------------------------------
+%
+% subdocument: The user interface of cweb.cls
+%
+
+\input{cweb-user}
+
+%
+% ------------------------------------------------------------
+%
+% subdocument: The internal interface
+%
+
+\input{cweb-conf}
+
+%
+% ------------------------------------------------------------
+
+
+
+
+\chap Implementation.
+
+This implementation must typeset complete files output by \cweave{}.
+Large parts of this task are available from the module \pkg{cwebbase}.
+It remains mainly to implement document-level markup and front and
+back matter (table of contents, index, etc.) In particular, the back
+matter material is not trivial; its tags are behind |\end{document}|,
+we have to read carefully behind the end of our \cweb{} document to
+detect if we have to typeset them at all.
+
+
+\sect This module reserves the namespace |cweb|.
+
+\beginprog
+\ifx \cweb@loaded\undefined
+ \def\cweb@loaded{$StyleRevision: 3.6 $}
+\else
+ \PackageWarningNoLine{cweb}%
+ {Some other package already uses namespace `cweb'}
+\fi
+\endprog
+
+
+\sect Let's identify this class.
+
+\noindent The code below is explained in the implementation
+documentation of the \pkg{rcs} package.
+
+\beginprog
+\begingroup
+ \def\RCSClass#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesClass{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSClass{cweb}{LaTeX markup for CWEB sources}
+ $StyleDate: 1995/11/30 15:44:53 $: 9999/00/00 \endRCS
+ $StyleRevision: 3.6 $: 0.0 \endRCS
+\endgroup
+\endprog
+
+
+\sect This module is from a supported bundle. Send bug reports,
+comments and repairs.
+
+The reference version may be retrieved via anonymous ftp from
+|ftp.th-darmstadt.de| [130.83.47.112], directory
+|pub/programming/literate-programming/c.c++/|. It's
+placed there as a gzipped tar file. (The information on the IP~number
+is dated August~29,~1995.
+It might have changed, although this is very unlikely. Use
+your friendly nameserver.)
+
+
+\sect This is freely distributable software; you can redistribute it
+and/or modify it under the terms of the GNU General Public License as
+published by the Free Software Foundation; either version~2 of the
+License, or (at your option) any later version.
+
+This software is distributed in the hope that it will be useful, but
+\textbf{without any warranty}; without even the implied warranty of
+\textbf{merchantability} or \textbf{fitness for a particular purpose}.
+See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License in
+the file |License| along with this package; if not, write to the Free
+Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+\sect Before we start we declare some shorthands for category codes.
+By declaring the underscore~`(|_|)' as letter we can use it in our
+macros. (I agree with \textsc{D.~Knuth} that
+|\identifier_several_words_long| is more readable than
+|\IdentifierSeveralWordsLong| and in every case better than |\p@@@s|.)
+
+As this is a \LaTeX{} style file the at sign is a letter anyhow; so
+we can use the ``private'' \LaTeX{} macros; and with the underscore we
+can make our own macros more readable. But as we have to restore this
+category code at the end of this macro file we store its former value
+in the control sequence |\CatUsCode|. This method is better than to
+use a group because not all macros have to be defined global this way.
+
+Since somebody might use more styles from me, this cseqs might be
+defined already.
+
+\beginprog
+\ifx \CatEscape\undefined
+ \chardef\CatEscape=0
+ \chardef\CatOpen=1
+ \chardef\CatClose=2
+ \chardef\CatIgnore=9
+ \chardef\CatLetter=11
+ \chardef\CatOther=12
+ \chardef\CatActive=13 % is defined in Plain already
+
+ \chardef\CatUsCode=\catcode`\_
+\fi
+
+\catcode`\_=\CatLetter % top level macro file
+\endprog
+
+
+\sect Problems over problems with different \LaTeX{} versions. I have
+decided that I will take the same stand as the \LaTeX{} team: I will
+only support the last two official \LaTeX{} versions, folks who run
+older software might need to upgrade.
+
+The implementation will use `star'-forms of \LaTeX{}'s command
+definition macros. They were introduced with \LaTeX{}
+\mbox{$\langle$1994/12/01$\rangle$}.
+
+Currently (as of 07 Nov 95), 1995/06/01 is the most recent version.
+I want to use a features introduced with that version: |\newif| is not
+outer any more. So I check here for the version and supply the new
+definition if necessary.
+
+\beginprog
+\NeedsTeXFormat{LaTeX2e}[1994/12/01]
+\@ifl@t@r \fmtversion{1995/06/01}{}% else
+ {
+ \@namedef{newif}#1{%
+ \count@\escapechar \escapechar\m@ne
+ \let#1\iffalse
+ \@if#1\iftrue
+ \@if#1\iffalse
+ \escapechar\count@
+ }
+ \def\@if#1#2{%
+ \expandafter\def
+ \csname \expandafter\@gobbletwo\string#1%
+ \expandafter\@gobbletwo\string#2\endcsname {%
+ \let#1#2%
+ }%
+ }
+ }
+\endprog
+
+
+
+
+\chap Options.
+
+Our options are of the ``keyword-value'' category. We realize them by
+the \pkg{keyvald} package, an enhancement of the \pkg{keyval} package
+from the Graphics bundle that supports defaulting for unknown keys.
+
+\begin{fixme}
+ Most of the code below should be extracted and moved to a
+ \pkg{keyvalx} package (or a new version of \pkg{keyval}). In
+ particular, support for package/class option processing, enumeration
+ values, and set values for keyword options.
+\end{fixme}
+
+|\ProcessKeyValOptions| may be used instead of |\ProcessOptions|, it
+takes the keyval set identifier as argument. At the end it evaluates
+and resets a hook. That hook may be used at option definition to
+define further dependencies. (E.g., enabling that option implies
+enabling an other option, too.)
+%
+\begin{fixme}
+ Like |\ProcessOptions|, I define |\@curroptions|, since I don't know
+ if that's needed later. The only place where it could been needed is
+ in the check for unprocessed options.
+\end{fixme}
+
+\beginprog
+\RequirePackage{keyvald}
+
+\let\ProcessKeyValOptionsHook\empty
+\providecommand*\ProcessKeyValOptions[1]{%
+ \edef\@curroptions{\@ptionlist{\@currname.\@currext}}%
+ \edef\next{\noexpand\setkeys{#1}{\@curroptions}}%
+ \next
+ \ProcessKeyValOptionsHook
+ \let\ProcessKeyValOptionsHook\empty
+ }
+\endprog
+
+
+\sect The enumeration option |structure| selects hierarchic~(0) or
+flat~(1) structure. The predefined value is `hierarchic', there is no
+default. The option value is bound to |\cweb@structure|.
+
+\beginprog
+%% \define@key@enum{cweb}{structure}{hierarchic,flat}[hierarchic]
+\@namedef{KV@cweb@structure@enum:hierarchic}{0 } % <-- space!
+\@namedef{KV@cweb@structure@enum:flat}{1 } % <-- space!
+\define@key{cweb}{structure}{%
+ \expandafter\let \expandafter\@tempa
+ \csname KV@cweb@structure@enum:#1\endcsname % \relax if undefined
+ \ifx \@tempa\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for option structure%
+ }{%
+Possible valid values are `hierarchic' and `flat'.\MessageBreak
+The predefined value is `hierarchic'.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \let\cweb@structure\@tempa
+ \fi
+ }
+\setkeys{cweb}{structure=hierarchic}
+\endprog
+
+
+\sect With the set option |suppress| one can select suppression of
+different document parts: |changehints| suppresses output of hints
+that a changefile was involved, |unchanged| output of unchanged
+chunks, |index| and |reflist| output of index and reference list, and
+|format| output of `|@f|' directives.
+
+The predefined value is no suppression. There is no default value.
+%
+\begin{fixme}
+ Shouldn't I supply a default value? `|{index,reflist}|'?
+\end{fixme}
+
+Set options are represented by flags named
+|@cweb@|\<option>|@|\<value>|@|.
+
+\beginprog
+%% \define@key@set{cweb}{suppress}{changehints,unchanged,index,reflist,format}
+\@for \@tempa :=changehints,unchanged,index,reflist,format\do {%
+ \expandafter\newif \csname if@cweb@suppress@\@tempa @\endcsname
+ \csname @cweb@suppress@\@tempa @false\endcsname
+ }%
+\define@key{cweb}{suppress}{%
+ \@for \@tempa :=#1\do {%
+ \expandafter\let \expandafter\@tempb
+ \csname @cweb@suppress@\@tempa @true\endcsname
+ %% \relax if undefined
+ \ifx \@tempb\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for set option suppress%
+ }{%
+Possible valid values are `changehints', `unchanged',\MessageBreak
+or a comma-separated list of these words enclosed in braces.\MessageBreak
+Specified values are turned on, you can't turn them off.\MessageBreak
+No value is on initially.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \@tempb
+ \fi
+ }%
+ }
+
+\g@addto@macro\ProcessKeyValOptionsHook{%
+ \if@cweb@suppress@unchanged@
+ \@cweb@suppress@changehints@true
+ \fi
+ }
+\endprog
+
+
+\sect The base document class may be specified by the option
+|baseclass|.
+
+\beginprog
+\def\CwebBaseClass{article}
+\define@key{cweb}{baseclass}[report]{\def\CwebBaseClass{#1}}
+\endprog
+
+
+\sect Language-specific adaptions are stored in files
+\texttt{cwbl-{\it language}.sty}. They may be accessed as packages or
+by the |language| option that takes the \textit{language} as value. We
+have to require the respective package at the end of the class, then
+all cseqs with default names are defined and may be overwritten.
+
+For some languages, we provide Babel-compatible option names, to ease
+the usage of global class options taken up by other packages (like
+\pkg{varioref}), too.
+
+\beginprog
+\def\cweb_set_language#1{%
+ \IfFileExists{cwbl-#1.sty}{%
+ %% true
+ \RequirePackage{cwbl-#1}%
+ }{%
+ %% false
+ \ClassError{cweb}{Language `#1' is unknown.}{%
+Language definitions are stored in files `cwbl-<language>.sty'.\MessageBreak
+I could not locate the file `cwbl-#1.sty'. If you didn't make a%
+\MessageBreak
+typo, please check if there is such a file in a directory searched\MessageBreak
+by LaTeX.%
+ }%
+ }%
+ }
+
+\define@key{cweb}{language}{\AtEndOfClass{\cweb_set_language{#1}}}
+
+\define@key{cweb}{german}[]{\setkeys{cweb}{language=german}}
+\define@key{cweb}{french}[]{\setkeys{cweb}{language=french}}
+\define@key{cweb}{italian}[]{\setkeys{cweb}{language=italian}}
+\endprog
+
+
+
+
+\chap Configuration, inheritance, and aggregation.
+
+This chapter sets up access to other modules, therefore we restore the
+original catcodes for the time being.
+
+\beginprog
+\catcode`\_=\CatUsCode
+\endprog
+
+
+\sect As usual, we provide the possibility to configure this class.
+
+\beginprog
+\InputIfFileExists{cweb.cfg}{%
+ \typeout{******************************************^^J%
+ *^^J%
+ * Using local configuration file cweb.cfg^^J%
+ *^^J%
+ ******************************************%
+ }%
+ }{}
+\endprog
+
+
+\sect Inherit article class.
+
+\beginprog
+\ifx \CwebBaseClass\undefined
+ \def\CwebBaseClass{article}
+\fi
+\define@key{cweb}*[]{\PassOptionsToClass{\KV@key}{\CwebBaseClass}}
+\ProcessKeyValOptions{cweb}
+\LoadClass{\CwebBaseClass}
+\endprog
+
+
+\sect OK, now comes our own code; switch on again our lexical
+conventions.
+
+\beginprog
+\catcode`\_=\CatLetter
+\endprog
+
+
+
+
+\chap Typesetting \cweave{} output.
+
+This task has been out-sourced to an other module, \pkg{cwebbase}. As
+mentioned at the start of the implementation, there remain mainly some
+bigger tasks, they will be implemented in further sections below. For
+the most part, the necessary preconditions are actually our class
+option definitions, we don't have to do anything. (Gosh, what a
+coincidence. :--) We need to tell \pkg{cwebbase}, that it should use
+class warnings on problems.
+
+\beginprog
+\let\cweb_warning=\ClassWarningNoLine
+\input{cwebbase}
+\endprog
+
+
+\sect But there remain also some small things to be done right now:
+For flat structure, the counter |secnumdepth| must be set to~11, then
+we want to number chunks in the whole document.
+
+\beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \or
+ %% flat
+ \setcounter{secnumdepth}{11}
+\fi
+\endprog
+
+
+\sect \pkg{cwebbase} uses |\part|, |\chapter| (if bound), and
+|\section|. We have no use for any other section division that came
+from the base class, let's discard the standard ones.
+
+\beginprog
+\let\subsection\undefined
+\let\subsubsection\undefined
+\let\paragraph\undefined
+\let\subparagraph\undefined
+\endprog
+
+
+
+
+\chap Table of contents.
+
+The table of contents features only sections, chunks don't have
+titles, after all. Titles are indented proportionally to the rank,
+with an basic indentation of |\CwebTocIndent|. An entry on major
+ranks may be differentiated by layout instead of indentation, as it's
+done in standard \LaTeX{} classes, too.
+
+The counter |\CwebTocIndentMaxLevel| constitutes an upper limit for a
+recognized indent (concerning indentation, that is).
+
+\beginprog
+\newdimen\CwebTocIndent
+ \CwebTocIndent=1.5em
+\newcount\CwebTocIndentMaxLevel
+ \CwebTocIndentMaxLevel=6 % increase indentation if <= max level
+\endprog
+
+
+\sect For hierarchic structure, we can ignore chunk entries
+completely, but have to supply a bunch of definitions for entries of
+type `|section|$l$'. But for flat structure, sections are passed as
+chunks, too. Then the current rank is stored in the title, as
+|\cwbbRank{|$r$|}|. These differences are so large that we'll have
+to make a very large |\ifcase|, over many chunks in this macro source
+file. Let's start with the hierarchic stuff.
+
+\beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\l@chunk#1#2{}
+\endprog
+
+
+\sect Let's assume that we're subclassed from one of \LaTeX{}'s
+standard classes; otherwise this stuff must be redefined, too. They
+indent |\section| entries always by the same amount (1.5\,em), our
+layout must fit to that.
+
+If we have chapters, they are also indented by 1.5\,em and
+distinguished by layout. The first minor section starts on rank~3 and
+is placed on level~2, it should be indented larger. Let's simply add
+1.5\,em for each rank, that makes 3\,em for sections of type
+|section2|. I.e., we use $f=l$ as a factor for our basic indentation.
+
+If we don't have chapters, the first minor section starts on rank~2
+and is placed also on level~2, it should be indented like sections
+(1.5\,em), as it will be in another layout. I.e., we use $f=l-1$ as
+the indentation factor.
+
+The actual definitions are generated in a loop. `|\l@section|$l$'
+expands to `|\cweb@toc_entry{|$l$|}|'. The indentation will be
+computed within that macro.
+
+\beginprog
+ \count@=2
+ \@whilenum \count@<11 \do {
+ \expandafter\edef \csname l@section\number\count@\endcsname{%
+ \noexpand\cweb@toc_entry{\number\count@}%
+ }
+ \advance \count@ by 1
+ }
+\endprog
+
+
+\sect A toc entry may be produced by |\@dottedtocline|. That cseq
+needs the level (to decide if that entry is produced after all), the
+entry indentation, the inner-entry indentation used for the number,
+entry text, and page number.
+
+As explained above, we use an indentation of $f \cdot
+|\CwebTocIndent|$, if the level is not larger than
+|\CwebTocIndentMaxLevel|. Otherwise we use the indentation of that
+level.
+
+Each entry is kind of an item with the number as the label. We need an
+inner-entry indentation to be used for subsequent lines within this
+entry. If a number is specified, it's tagged by |\numberline| in the
+entry text. With a proper local definition and a pro-forma evaluation
+of the text, we get at the width of the number (with a fitting space
+for separation to text) and can use that as the inner-entry
+indentation.
+
+\beginprog
+ \def\cweb@toc_entry#1#2{% % #3 (page) will follow
+ %% \count@ == factor for basic indentation
+ %% min(#1,TocIndentMaxRank) - correct_level
+ \ifnum #1>\CwebTocIndentMaxLevel
+ \count@\CwebTocIndentMaxLevel
+ \else
+ \count@ #1 % <-- space!
+ \fi
+ \ifx \chapter\undefined
+ \advance \count@\m@ne
+ \fi
+ %% \dimen@ == inner-entry indentation
+ %% width of number & space, if existing
+ \begingroup
+ \global\dimen@ \z@ % assert value
+ \def\numberline##1{\settowidth{\global\dimen@}{##1\enskip}}%
+ \setbox\z@ \hbox{#2}% % evaluates \numberline if specified
+ \endgroup
+ \@dottedtocline
+ {#1}% % level
+ {\count@ \CwebTocIndent}% % basic indent
+ \dimen@ % num width
+ {#2}% % entry; page will follow
+ }
+\endprog
+
+
+\sect We're done with toc configuration for hierarchic structure,
+since entry definitions for |chapter| and |section| types exist
+already. Let's turn over to the flat structure.
+
+\beginprog
+ \or
+ %% flat
+\endprog
+
+
+\sect The section rank determines how they are featured in the table.
+Titles on rank~0 are typeset boldface, other titles are dotted
+contents lines in roman.
+
+The rank is specified within the title, i.e., in the
+first argument to |\l@chunk|, as an argument for the |\cwbbRank| tag.
+With an appropriate definition we might set |\count@| to the
+respective value. If it is not set, an explicite invocation of
+|\section| or |\addcontentsline| is responsible for this entry. Then
+we assume that the rank is~0.
+
+\beginprog
+ \def\l@chunk#1{% % page will be processed later
+ \count@\@ne % default value of group level
+ \begingroup
+ \let\numberline\@gobble % width not known yet
+ \def\cwbbRank##1{\global\count@ ##1\relax}%
+ \setbox\z@ \hbox{#1}%
+ \endgroup
+ \ifnum \count@=\z@
+ \let\next\cweb@bold_toc_line
+ \else
+ \let\next\cweb@normal_toc_line
+ \fi
+ \next{#1}%
+ }
+\endprog
+
+
+\sect A boldface line is typeset similar to the definition of
+|\@dottedtocline|, only with fewer parameters. It takes two
+parameters, the entry and the page number.
+
+\beginprog
+ \def\cweb@bold_toc_line#1#2{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1em plus\p@}%
+ \begingroup
+ \noindent \bf
+ \hangindent\CwebTocIndent
+ \rightskip\@tocrmarg \parfillskip -\rightskip
+ \interlinepenalty\@M
+ \@tempdima\CwebTocIndent % for \numberline
+ #1\nobreak\hfill \hbox to\@pnumwidth{\hss #2}%
+ \par
+ \endgroup
+ }
+\endprog
+
+
+\sect A normal line uses |\@dottedtocline| like the hierarchic stuff.
+Rank~1 shall still be on the same level as the bold toc line, so we
+have to use $r-1$ as the proportional factor for |\CwebTocIndent|.
+
+\beginprog
+ \def\cweb@normal_toc_line#1{% % page will be processed later
+ \edef\next{\noexpand\@dottedtocline{\the\count@}}% % toc entry level
+ \ifnum \count@>\CwebTocIndentMaxLevel
+ \count@ \CwebTocIndentMaxLevel
+ \fi
+ \advance\count@\m@ne % group level -= 1
+ \next{\count@\CwebTocIndent}% % basic indent
+ \CwebTocIndent % numwidth
+ {#1}% % entry
+ }
+\endprog
+
+
+\sect The first argument of |\@dottedtocline| is used to specify a
+depth of the issued entry. All entries with a depth larger than
+|tocdepth| are discarded. We typeset all entries, our |tocdepth|
+value is the largest possible section level.
+
+\beginprog
+ \setcounter{tocdepth}{10}
+\endprog
+
+
+\sect That's it. Close our distinction between hierarchic and flat
+structure.
+
+\beginprog
+\fi
+\endprog
+
+
+
+
+\chap The end of the document.
+
+The end of a \cweave{} document is a rather complicated thing. It is
+described in detail in the \cweave{} interface specification, we repeat the
+most important points shortly.
+%
+\begin{itemize}
+
+\item The user has issued an |\end{document}| tag in the documentation
+part of the last chunk.
+
+This tag is therefore part of a chunk, the chunk end (i.e., the
+|\fi| cseq) does appear later in the document. Particularly, this
+means that at the accurence of this tag there is still an open |\if|.
+
+\item Between the |\end{document}| and the |\fi| a lot of text may
+come which must be ignored. Within this text no |\fi| is allowed.
+
+\item After the |\fi| the document may be finished by |\end|.
+(Then the |-x| option of \cweave{} was used.)
+
+Or additional information about the whole document is added: An
+optional tag |\ch|, two mandatory tags |\inx\fin|, and another
+optional tag |\con|.
+
+\end{itemize}
+
+Btw, the last chunk is marked as changed if any changefile entry has
+been applied in the document. (In Plain \cweb{} that chunk is
+colloquially used as the introduction to the index, and that shall be
+printed always.) If we suppress unchanged chunks, the last
+chunk will therefore always be processed, independent if it was really
+changed or not. Yet another reason to use a whole chunk just for
+|\end{document}|.
+
+We'll use a the following approach for the implementation of the
+iidentifier index and the refinement list (remember that we do not
+produce a table of contents): We will check if there is any of these
+two lists available. The available ones are typeset. Then we'll
+finish the document. This functionality will be bound to |\fin|.
+|\inx| will be simply a no-op. |\con| will never be reached, so we
+don't have to cope with it.
+
+
+\sect With the introduction of an end-document hook in \LaTeXe{}, one
+might have hoped that we can utilize that hook for our task. But this
+is not possible: We don't want to add some additional actions
+\emph{into} the end-document handling, we want to look ahead and issue
+lots of commands \emph{before} the end-document handling has happened.
+(If you know \TeX{} by heart, you may skip the rest of the explanation
+and go to the next chunk.)
+
+As a macro language, \TeX{}'s cseqs take those tokens as arguments
+that are behind them in the token stream. The token stream is
+initially the document source, evaluation of a macro pushes its
+expansion (a token list) on front of the stream. I.e., the evaluation
+of |\end| will push the expansion on the token stream before
+|\enddocument| is even looked at. And |\enddocument| will also push
+its expansion before the hook is evaluated.
+
+If we would allow |\enddocument| to push its expansion, all these
+tokens have to be evaluated after we have typeset the backmatter --
+after all, we need a proper document end, don't we? One would need to
+skip over them, look at and eventually evaluate tokens that come
+behind (from the initial document stream, the document source), and
+then evaluate the skipped tokens. That's too hairy, we don't know
+exactly how many tokens to skip, this is probable to change in future
+\LaTeX{} revisions, etc.
+
+The pushed tokens from |\end| don't pose so much problems. As
+|\enddocument| will terminate the document processing, they are not
+evaluated anyhow, we can just throw them away. And |\end| has changed
+only once in the last few years, so it seems to be a stable macro --
+necessary as we need to know its expansion for propper skipping.
+
+A very first consequence of the demand to process tokens after
+|\end{document}| is that we have to save the ``standard'' \LaTeX{}
+document termination, since we will execute that later.
+
+\beginprog
+\let\cweb@save@enddocument=\enddocument
+\endprog
+
+
+\sect When we encounter |\end{document}| we have to skip tokens
+until we reach a |\fi|. Then we establish the correct
+environment for the material behind.
+
+The skip itself is actually a bit difficil, so we'll look at it more
+closely. |\enddocument| is executed within |\end|. As mentioned above,
+the original
+definition of |\enddocument| cheats, it terminates \LaTeX{} before
+the rest of |\end| is executed. |\end| first checks if we're in
+the correct current environment, then it establishs the end
+processing for paragraph building environments, closes the
+group---each environment is in its own group, executes the end
+processing established before, and it may ignore following white
+space.
+
+The |document| pseudo-environment is different in that it isn't in a
+group. I.e., we have to open a group which can be closed by |\end|
+afterwards. The environment check will be done in this group. With
+|\aftergroup| we establish a macro which will skip the rest of the
+|\end| processing. This processing is not needed anyhow: |document|
+is no paragraph building environment and the following white space is
+ignored by ourselves. Since |document| is no paragraph building
+environment, the end processing described above is empty, i.e., not
+existant. Then come a few tokens which ignore the white space. One of
+these tokens is a |\fi|, so we have to skip them first separately.
+They are discarded by |\cweb@skip_end|, we can start to look
+for our |\fi| afterwards.
+
+If |\end{document}| is the first (i.e., the only) text in this chunk
+we discard the chunk start marker. This is possible since it isn't set
+until now, it's in the |\everypar| token register. Of course, then we
+have to do the house keeping work stored in |\everypar| ourselves.
+
+\beginprog
+\def\enddocument{%
+ \if@noskipsec
+ \global\everypar{}%
+ \global\@noskipsecfalse
+ \global\@nobreakfalse
+ \global\clubpenalty\@clubpenalty
+ \fi
+ \begingroup
+ \aftergroup\cweb@skip_end
+ }
+\endprog
+
+
+\sect Let's make sure that the expansion of |\end| is the one that we
+expect. Then we can define |\cweb@skip_end| appropriately, we have to
+skip 5~tokens.
+
+\beginprog
+\CheckCommand*\end[1]{%
+ \csname end#1\endcsname\@checkend{#1}%
+ \expandafter\endgroup\if@endpe\@doendpe\fi
+% 1 2 3 4 5
+ \if@ignore\global\@ignorefalse\ignorespaces\fi}
+
+\def\cweb@skip_end#1#2#3#4#5{\cweb@check_fi}
+\endprog
+
+
+\sect |\cweb@check_fi| gets the next token, and invokes
+|\cweb@do_check_fi| who really checks it. If this token is |\fi| we
+can prepare to finish our document. Otherwise we have to check the
+next token.
+
+Of course, first the token we have checked already must be discarded.
+Therefore we insert the followup action by |\afterassignment|. It's
+important that the assignment gets \emph{only} the next token, we
+have to supply both the equals sign and the optional blank which are
+allowed between the both arguments to |\let|.
+
+\beginprog
+\def\cweb@check_fi{%
+ \futurelet\next \cweb@do_check_fi
+ }
+\def\cweb@do_check_fi{%
+ \ifx \next\fi
+ \afterassignment\cweb@finish
+ \else
+ \afterassignment\cweb@check_fi
+ \fi
+ \let\next= % <-- blank!!
+ }
+\endprog
+
+
+\sect When we found the |\fi|, we also discard it. But we're still
+within a chunk, i.e., there is an open |\if|. So we must supply the
+|\fi|.
+
+\TeX{}nical note: |\next| is now equivalent to |\fi|. It must not be
+used within a loop where it denotes the tail recursion's action. I.e.,
+code like
+ %
+\begin{verbatim}
+ \ifx \foo\bar
+ \let\next\relax
+ \else
+ \let\next\continue
+ \fi
+ \next
+\end{verbatim}
+ %
+will fail. This trap is closed by a redefinition of |\next|.
+
+The output of all following lists may be suppressed by some option.
+That is tested at the place where the respective list shall be
+produced.
+
+All following lists are typeset in CR state.
+
+\beginprog
+\def\cweb@finish{%
+ \fi
+ \let\next\relax
+ \let\ch\cweb@changed_list
+ \let\inx\relax
+ \let\fin\cweb@end_document
+ \let\end\cweb@save@enddocument
+ \cwbb@CR
+ }
+\endprog
+
+
+\sect The following lists are started with unnumbered section
+divisions. They are either chapters or sections.
+
+\beginprog
+\ifx \chapter\undefined
+ \let\CwebListSectionDiv=\section
+\else
+ \let\CwebListSectionDiv=\chapter
+\fi
+\def\CwebListSection{\CwebListSectionDiv*}
+\endprog
+
+
+\sect The list of changed chunks is a cross reference list, (nearly)
+like all others at the end of a chunk. The only difference is that
+we do not show changeflags any more---each chunk number in this list
+carries a change flag by definition.
+
+\TeX{}nical note: The redefinition of |\*| is part of the second
+argument of |\CwebCrossRef|. It is \emph{not} a global redefinition.
+
+\beginprog
+\def\CwebCRChanged{%
+ \CwebCrossRef{The following chunks were changed by the change file:}%
+ }
+\def\CwebChangedListName{Changed Chunks}
+
+\if@cweb@suppress@changehints@
+ \def\cweb@changed_list#1.{}
+\else
+ \def\cweb@changed_list{%
+ \CwebListSection{\CwebChangedListName}%
+ \@mkboth{\uppercase{\CwebChangedListName}}%
+ {\uppercase{\CwebChangedListName}}%
+ \message{\CwebChangedListName}%
+ \let\CwebCRSize\relax
+ \CwebCRChanged \let\*\relax
+ }
+\fi
+\endprog
+
+
+\sect As outlined above, we want to check if there are any entries in
+the identifier index or the refinement list. The former is stored in
+the file |\jobname.idx|, the latter in the file |\jobname.scn|.
+
+We write a macro |\cweb@if_file_not_empty| that's like |\IfFileExists|.
+
+\beginprog
+\def\cweb@end_document{%
+ \cweb@if_file_not_empty{\jobname.idx}{%
+ \if@cweb@suppress@index@ \else
+ \CwebIdIndex
+ \fi
+ }{}%
+ \cweb@if_file_not_empty{\jobname.scn}{%
+ \if@cweb@suppress@reflist@ \else
+ \CwebRefList
+ \fi
+ }{}%
+ \cweb@save@enddocument
+ }
+\endprog
+
+
+\sect The paragraph layout for both the index and the refinement list
+is much the same, it is taken from Plain \cweb{}: Each entry is a
+paragraph, nearly no skip between the paragraphs (just a bit to
+prevent underfull vboxes), no paragraph indentation, ragged right.
+Overfull hboxes in the lists don't make sense, so we prevent them. And
+we don't allow hyphenation.
+
+\beginprog
+\def\cweb@list_par_layout{%
+ \parskip \z@ plus .5\p@
+ \parindent\z@
+ \rightskip \z@ plus 2.5em
+ \tolerance\@M \hyphenpenalty\@M
+ }
+\endprog
+
+
+\sect We consider a file as ``not empty'' if (1)~the file exists,
+(2)~is not empty, and (3)~has a non-empty line at the very front. (In
+fact, an empty line at the front means that there should not be
+anything behind it---but we can't test this portably.)
+
+|\cweb@if_file_not_empty| tests the property, it's parametrized by the
+file name and gets passed actions for the true and the false case.
+
+We open the file first. Then we check if it doesn't exist or if it's
+empty, both conditions deliver true on |\ifeof|; in this case we
+pretend that there was an empty line. (The \LaTeX{} kernel already
+once provided a macro for an empty line, |\@defpar|, we repeat that
+definition here since it was discarded there.) Otherwise we read the
+first line. At this state, the emptiness of the first line is
+equivalent to the non-availability of entries, we can easily construct
+an appropriate macro call to set the flag.
+
+\beginprog
+\def\@defpar{\par}
+\def\cweb@if_file_not_empty#1#2#3{%
+ \openin\@inputcheck #1\relax
+ \ifeof \@inputcheck
+ \let\next\@defpar
+ \else
+ \read\@inputcheck to \next
+ \fi
+ \closein\@inputcheck
+ \ifx \next\@defpar
+ \def\reserved@a{#3}%
+ \else
+ \def\reserved@a{#2}%
+ \fi
+ \reserved@a
+ }
+\endprog
+
+
+\sect The identifier index is available in the file |\jobname.idx|.
+The setup for the index is a mixture of the |theindex| environment of
+the \cls{article} class and DEK's index macros. It's typeset in two
+columns; the user may specify an introductionary text for the index
+by |\cwebIndexIntro|. If there is any introductionary text we add a
+medium skip below.
+
+For the index, |\parfillskip| set in such a way that almost empty
+lines are avoided.
+
+Before we read in this file, we have to bind the special cseqs used
+therein.
+
+\begin{fixme}
+ \LaTeX{} feature or bug? The optional argument of |\twocolumn| is
+ typeset as a vbox, therefore the space at top of the section heading
+ is not discarded at the top of a page. But we know that |\twocolumn|
+ will start a page in any case, we must prevent the addition of this
+ space. Looking in the source, we see that setting the flag
+ |@nobreak| will prevent this addition, let's do that as a
+ workaround. File a \LaTeX{} bug report?
+\end{fixme}
+
+\begin{fixme}
+ Surprise: If |@nobreak| is set \emph{and} if there are no index
+ entries, |\twocolumn| will drop its optional argument. (That
+ happened to be noticed when there was an error in
+ |\cweb@if_file_not_empty|.) File a \LaTeX{} bug report?
+\end{fixme}
+
+\begin{fixme}
+ If |\CwebListSection| is a chapter tag of the standard classes, that
+ macro will check if it's in twocolumn mode and will try to set
+ itself at the top of the current page in full width, as it would
+ have been given in an optional argument of |\twocolumn|. As the
+ method that optional argument uses (a float) must not be nested, we
+ must prevent this to happen. We set the |@twocolumn| flag to false,
+ chapter macros will then think they're evaluated in a one-column
+ document. It's debatable if that is a kludge or even \emph{the}
+ solution -- after all, within the optional argument we aren't in
+ twocolumn mode any more. File a \LaTeX{} bug report?
+\end{fixme}
+
+\beginprog
+\newtoks\cwebIndexIntro
+ \cwebIndexIntro={}
+\def\CwebIdIndex{%
+ %% two column format
+ \columnseprule\z@
+ \columnsep 35\p@ % value is from article.sty
+ %% introduction of index
+ \twocolumn[%
+ \cwbb@tex
+ \@nobreaktrue % don't add space in front of section
+ \@twocolumnfalse % \CwebListSection might be \chapter
+ \CwebListSection{\indexname}%
+ \edef\intro{\the\cwebIndexIntro}% % is a local def
+ \ifx \intro\empty
+ \else
+ \the\cwebIndexIntro\unskip
+ \par\medskip
+ \fi
+ ]%
+ \@mkboth{\uppercase{\indexname}}{\uppercase{\indexname}}%
+ \message{\indexname:}% % tell the user what we're doing
+ %% paragraph layout
+ \cweb@list_par_layout
+ \parfillskip \z@ plus .6\hsize
+ %% bindings
+ \let\I\CwebIndexEntry
+ \let\[\CwebIndexDeclared
+ \let\*\cwbb@change_flag
+ %% read the index
+ \CwebReadIndex
+ }
+\def\CwebReadIndex{\@input{\jobname.idx}}
+\endprog
+
+
+\sect An index entry is typeset with the same hanging indentation like
+a cross reference list.
+
+The entry is tagged with |\.|, if it was entered by the \cweb{}
+operator~`|@.|'. Then it shall be typeset as a string. But an indexed
+name may also want to use |\.| as an accent. This is the same
+situation as we had at the refinement names (in module
+\pkg{cwebbase}), where we introduced |\cwbb@check_dot| to handle this
+case. The same minor restriction as there holds here, a refinement
+name may not consist of a single dot-accented expression. (|@:|~helps
+in this singular case.)
+
+\beginprog
+\def\CwebIndexEntry#1, {%
+ \par
+ \hangindent\CwebNumberListHangindent
+ \leavevmode
+ \cwbb@check_dot{#1}:\quad
+ }
+\endprog
+
+
+\sect The chunks where identifiers are declared are noted with
+underlined numbers. We also must not forget the default declaration of
+|\9|, the tag for the user definable index layout.
+
+\beginprog
+\def\CwebIndexDeclared#1]{$\underline{#1}$}
+\def\9#1{}
+\endprog
+
+
+\sect The list of the refinement names is available in the file
+|\jobname.scn|. The layout is taken from the plain version: ragged
+right, each entry is a paragraph, the different cross reference
+categories are separated by a quad.
+
+We must initialize our whole paragraph layout, as the index might have
+been suppressed and no initialization has happened there then. If
+there was an index, it changed |\parfillskip| and |\*|, let's restore it.
+
+\beginprog
+\def\CwebRefListName{List of Refinements}
+
+\def\CwebRefList{%
+ %% paragraph layout: like in index, but
+ \onecolumn
+ \cweb@list_par_layout
+ \parfillskip\@flushglue % must be reset before section heading
+ \CwebListSection{\CwebRefListName}%
+ \@mkboth{\uppercase{\CwebRefListName}}%
+ {\uppercase{\CwebRefListName}}%
+ \message{\CwebRefListName:}%
+ %% different bindings
+ \def\I{\par \hangindent\CwebNumberListHangindent}%
+ \def\CwebCrossRef##1##2.{\quad {\reset@font\footnotesize ##1~##2.}}%
+ \let\Q\CwebXRCite
+ \let\Qs\CwebXRCites
+ \let\U\CwebXRUse
+ \let\Us\CwebXRUses
+ \let\*\cwbb@change_flag
+ %% Read the list, allow configuration first.
+ \csname CwebRefListHook\endcsname
+ \CwebReadPPList
+ }
+\def\CwebReadPPList{\@input{\jobname.scn}}
+\endprog
+
+
+\sect Extra definitions for the sake of internationalization.
+
+\beginprog
+\def\CwebXRCite{\CwebCrossRef{Cited in chunk}}
+\def\CwebXRCites{\CwebCrossRef{Cited in chunks}}
+\def\CwebXRUse{\CwebCrossRef{Used in chunk}}
+\def\CwebXRUses{\CwebCrossRef{Used in chunks}}
+\endprog
+
+
+
+\chap The end.
+
+Well, after all we're finished with this class. We must not forget to
+restore the underscore catcode.
+
+\beginprog
+\catcode`\_=\CatUsCode
+
+\endinput
+\endprog
+
+
+\sect I would like to thank those who helped me to improve this module.
+% In particular, XXX provided XXXsubstantial parts of the code.
+
+\textsc{Michael M\"uller} and \textsc{Zden\v{e}k Wagner} did thorough
+checks that helped me to improve the alpha test version.
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\vskip \PltxPreSectSkip
+
+\begin{rcslog}
+$StyleLog: cweb.doc,v $
+\Revision 3.6 1995/11/30 15:44:53 schrod
+Still didn't work with version \mbox{$\langle$1994/12/01$\rangle$},
+the copy of |\newif| was not complete.\\
+Problem reported by Michael Seel \path|<seel@mpi-sb.mpg.de>|.
+
+\Revision 3.5 1995/11/20 22:34:18 schrod
+Add language support.\\
+Triggered by Christian Kumpf \path|<smurf@igd.fhg.de>| and Andreas
+Scherer \path|<scherer@physik.rwth-aachen.de>|.
+
+Use my `official' (ACM) email address.
+
+\Revision 3.4 1995/11/07 17:55:18 schrod
+ \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+versions; make it work with the previous-to-last one, version
+\mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+\mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+|\newif|, and |\hb@xt@| must not be used.\\
+Problem reported by Laurent Desnogues
+\path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+-- add name).
+
+\Revision 3.3 1995/09/17 14:15:36 schrod
+Initialize list paragraph layout in refinement list, too. With the
+introduction of index suppression, the initialization -- located in
+the index code -- might have been skipped.
+
+\Revision 3.2 1995/09/15 10:22:59 schrod
+The \cls{cweb} class needs at least \LaTeX{} version
+\mbox{$\langle$1994/12/01$\rangle$}.\\
+Problem reported by John S. Robinson \path|jsrobin@umiacs.umd.edu|.
+
+\Revision 3.1 1995/09/12 23:00:48 schrod
+Moved all code that does the actual typesetting of \cweave{} tags and
+is therefore also needed for a |cweb| environment. It's now an own
+module named \pkg{cwebbase}.
+
+\Revision 2.8 1995/09/12 11:44:33 schrod
+Standard base classes with chapters (i.e., report \& book) produced an
+error at the start of the index: |\chapter| may not be called in the
+optional argument of |\twocolumn|. Add a workaround to |\twocolumn|.\\
+Problem reported by Christopher Higgins \path|<C.Higgins@cit.gu.edu.au>|.
+
+\Revision 2.7 1995/08/29 17:32:36 schrod
+Add hook for refinement list.
+
+Allow internationalization of refinement list.\\
+(Problem reported by Christian Kumpf \path|<kumpf@igd.fhg.de>|.)
+
+Cweb bundle, version 1.0.
+
+\Revision 2.6 1995/08/29 15:19:50 schrod
+Added (hint to non-existence of) description of the internal interface.
+
+\Revision 2.5 1995/08/29 02:07:26 schrod
+Discard dependencies on 10\,pt fonts.
+
+Support suppression of format directives.
+
+\Revision 2.4 1995/08/27 19:31:43 schrod
+Discard superfluous space at the top of index. Spacing in refinement
+heading was wrong. Repair test on empty auxilliary files. Make
+section heading for index and refinement list divisions (more)
+configurable. Suppression of index and reference list is supported.
+
+Make configuration of change flag easier.
+
+Put section title of changed chunks list in marks, and tell the user
+that it got typeset.
+
+\Revision 2.3 1995/08/27 17:24:46 schrod
+Make usage of baseclass with chapters work.
+
+\Revision 2.2 1995/08/27 13:26:22 schrod
+Add possibility to suppress change hints. Suppression of unchanged
+chunks suppress change hints as well, they are meaningless as all
+printed chunks are changed by definition.
+
+Realize dependencies in options by |\ProcessKeyValOptionsHook|.
+Illegal values for enumeration options don't alter the current value
+any more.
+
+List of changed chunks is an unnumbered section now.
+
+\Revision 2.1 1995/08/25 19:11:18 schrod
+Add keyword-value option style, with new \pkg{keyvald} package.
+
+Hierarchic strucutures are supported now, in addition to the flat
+structure of the beta-test version. One can choose with an option. For
+that step, the terminology was cleaned up, too: Chunks are not named
+sections any more. (That change involved reimplementation of almost
+all the structure and toc stuff.)
+
+The chunk number supplied by \cweave{} is used now, not some computed
+number. Change flags are printed, too.
+
+One can suppress output of unchanged sections.
+
+One can select the baseclass with an option. That may be used to use
+\cls{report} or \cls{book} to get chapter-style layout. Of course,
+using an arbitrary baseclass is dangerous, it must conform to the
+conventions of \LaTeX{} standard classes.
+
+\Revision 1.13 1995/08/08 00:14:29 schrod
+Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+my standard templates for that, no changes in functionality.
+
+\Revision 1.12 1993/08/10 14:15:43 schrod
+New page on main section only if group level $<$ |\cwebSecNoEject|.
+Default for the latter is 3.
+
+Document that logos will not be defined in this style file.
+
+Copy of plain macros for |\CwebNumber| does not work. Repaired the
+most important one (subscript must be accessed via |\sb|).
+Incompatibility to NFSS will be addressed later.\\
+(Problems reported by Zden\v{e}k Wagner \path|<wagner@csearn.bitnet>|.)
+
+\Revision 1.11 1993/08/10 11:21:07 schrod
+Reference to section number does not render a period after the
+number any more.
+
+\Revision 1.10 1993/08/09 20:08:20 schrod
+|\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+effect already.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+\Revision 1.9 1993/08/09 18:05:28 schrod
+Left shift operator wasn't defined correctly.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+\Revision 1.8 1993/06/15 15:22:30 schrod
+\textbf{Version 0.2:} Moved from |pub/incoming| to official place, in
+Literate Programming Archive.
+
+\Revision 1.7 1993/06/15 13:25:48 schrod
+First attempt to make it a style: |article| is imported. Warns the
+user if he tries to use |cweb| as a style option.
+
+\Revision 1.6 1993/06/15 08:49:23 schrod
+|\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+causes problems if a |\PB| is within. Now I try hard not to evaluate any
+tokens outside of my control.
+
+Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+need an own macro.
+
+\Revision 1.5 1993/06/14 17:50:30 schrod
+Handle a missing |\con|: |\inx| and |\con| are now never looked at.
+Instead |\fin| checks for the existence of the lists and typeset them
+if there are entries available.
+
+\Revision 1.4 1993/06/14 15:54:18 schrod
+Made it work with older versions of \LaTeX{}, too. The |\enddocument|
+implementation of cweb.sty depends on the |\end| implementation; check
+if the current implementation is one we know of (currently: 91-01-14
+and 92-03-25).\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+Add FSA diagram about processing states. CR state is also switched
+to from \TeX{} state (that happens with |\ch| at the document end).
+
+\Revision 1.3 1993/05/13 17:51:21 schrod
+Refinements may also be filenames (`|@(|'). Then the complete name
+consists of a |\.| macro call, which is handled now.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+Made the detection of `|@.|' index entries more robust.
+
+\Revision 1.2 1993/05/12 18:28:59 schrod
+Adapted to recent changes of \cweave{} (of April 93):
+
+Main sections have a group level, represented in the table of
+contents. This changed the complete implementation of section tags.
+
+New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+|\CwebCombinedOp|.
+
+\Revision 1.1 1993/04/09 15:00:37 schrod
+Initial revision
+
+\end{rcslog}
+
+
+
+\end{document}
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+Local Variables:
+mode: LaTeX
+TeX-brace-indent-level: 4
+indent-tabs-mode: t
+TeX-auto-untabify: nil
+TeX-auto-regexp-list: LaTeX-auto-regexp-list
+compile-command: "make cweb.cls"
+End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cwebarray.sty b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebarray.sty
new file mode 100644
index 00000000000..1eb73a0d48b
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebarray.sty
@@ -0,0 +1,50 @@
+% $Id: cwebarray.sty,v 1.2 1995/08/08 00:14:31 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% LaTeX style option cwebarray
+% depends on array style option
+%
+% [LaTeX]
+% (history at end)
+
+\ProvidesPackage{cwebarray}
+
+
+%
+% What's this style option for?
+%
+% If you use CWEB, you cannot use vertical bars (`|') as rule specifiers in
+% tables any more; CWEAVE processes them already. This style option
+% defines `I' (that's an uppercase i) as a replacement for the
+% vertical bar.
+%
+% I.e., write
+%
+% \begin{tabular}{lIl}
+%
+% instead of \begin{tabular}{l|l}.
+
+% This option won't work without the array style.
+
+\ifx \d@llarbegin\undefined
+ \RequirePackage{array}
+\fi
+
+\newcolumntype{I}{|}
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: cwebarray.sty,v $
+% Revision 1.2 1995/08/08 00:14:31 schrod
+% Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+% my standard templates for that, no changes in functionality.
+%
+% Revision 1.1 1993/08/09 18:05:55 schrod
+% Mentioned that `|' cannot be used for LaTeX purposes, in
+% particular, not for ruled tables. Described workarounds, one of them
+% is the new style option cwebarray.
+% [Reported by Felix G\"artner]
+%
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cwebbase.doc b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebbase.doc
new file mode 100644
index 00000000000..280e53f82fc
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebbase.doc
@@ -0,0 +1,2059 @@
+% $StyleId: cwebbase.doc,v 1.4 1995/11/07 17:55:19 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+
+%
+% LaTeX macros cwebbase
+% support for incorporation of CWEAVE markup in LaTeX documents
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+
+
+
+\documentclass{progltx}
+
+\usepackage{cweb-doc} % document-specific markup
+\usepackage{fullpage}
+
+
+\RCS $StyleRevision: 1.4 $
+\RCS $StyleDate: 1995/11/07 17:55:19 $
+
+
+\begin{document}
+
+
+\title{Typesetting \cweave{} Output}
+\author{% % LaTeX does not discard unnecessary glue...
+ Joachim Schrod%
+ \thanks{%
+ \protect\raggedright
+ TU~Darmstadt, Computer Science Department, WG Systems Programming,
+ Alexanderstr.~10, D-64283~Darmstadt, Germany.
+ Email: \texttt{schrod@iti.informatik.th-darmstadt.de}.%
+ }%
+ }
+\date{%
+ Revision \RCSStyleRevision\\
+ (as of \RCSStyleDate)%
+ }
+
+\maketitle
+
+
+% doesn't work with progltx yet
+%\tableofcontents
+
+
+
+
+\chap Introduction.
+
+This code shall interpret and render the tag set output by \cweave{}.
+That's realized in this module as it's needed in multiple places --
+when we want to use \LaTeX{} markup in \cweb{}, when we want to
+incorporate \cweb{} pieces in \LaTeX{} documents (as examples, or in
+textbooks), and when we want to incorporate whole \cweb{} documents in
+a \LaTeX{} document. (That's by far the most difficult task.)
+
+As \cweb{} operates in modes, for obvious reasons our implementation
+is based on a finite state automaton (FSA). The whole tag set is
+separated into subsets that are valid in certain states.
+
+The complete set of tags does not obey any name conventions, some tags
+are even in conflict with standard \LaTeX{} tags (e.g.\ |\\|). Most
+tags are used to render code pieces. With the introduction of states
+we can restrict the visibilities of these tags to those states where
+we typeset program parts. The needed functionality will be available
+in cseqs of our namespace, we'll map that to \cweave{}'s names in a
+group.
+
+Then we have to implement the rendering of chunks and sections, that's
+(more or less) a straight-forward task. Actually, we only set up the
+basic code pieces, the integration of section divisions in the actual
+\LaTeX{} document is a task of higher-level code; that integration
+depends heavily on the type of document. That also implies that we
+won't implement front or back matter material (document markup, table
+of contents, index, list of refinements, etc.)\ in this module.
+
+
+\sect Let's identify this module.
+
+\noindent The code below is explained in the implementation
+documentation of the \pkg{rcs} package.
+
+\beginprog
+\begingroup
+ \def\RCSFile#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesFile{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSFile{cwebbase.tex}{CWEAVE tags for LaTeX markup}
+ $StyleDate: 1995/11/07 17:55:19 $: 9999/00/00 \endRCS
+ $StyleRevision: 1.4 $: 0.0 \endRCS
+\endgroup
+\endprog
+
+
+\sect This module is part of the supported \textsl{cweb-sty} bundle.
+Send bug reports, comments and repairs.
+
+It assumes that the canonical setup of my modules has happened and can
+be used herein. (In particular, the underscore is a letter.)
+
+
+\sect This module reserves the namespace |cwbb|. It assumes that
+|\cweb_warning| is bound either to |\PackageWarningNoLine| or
+|\ClassWarningNoLine|, depending on the environment.
+
+\beginprog
+\ifx \cwbb@loaded\undefined
+ \def\cwbb@loaded{$StyleRevision: 1.4 $}
+\else
+ \cweb_warning{cwebbase}%
+ {Some other package already uses namespace `cwbb'}
+\fi
+\endprog
+
+
+\sect This is freely distributable software; you can redistribute it
+and/or modify it under the terms of the GNU General Public License as
+published by the Free Software Foundation; either version~2 of the
+License, or (at your option) any later version.
+
+This software is distributed in the hope that it will be useful, but
+\textbf{without any warranty}; without even the implied warranty of
+\textbf{merchantability} or \textbf{fitness for a particular purpose}.
+See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License in
+the file |License| along with this package; if not, write to the Free
+Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+
+
+% ------------------------------------------------------------
+%
+% subdocument: The interface between CWEAVE and TeX
+%
+
+\input{cweave-spec}
+
+%
+% ------------------------------------------------------------
+
+
+
+
+\chap Interface of \pkg{cwebbase} module.
+
+\pkg{cwebbase} implements most of \cweave{} tags. It sets up tons of
+redefinable cseqs, all of them start with |Cweb|.
+%
+\begin{fixme}
+ How do I document them? We need a configuration document anyhow. Do
+ I present here just a list of those cseqs realized in this module
+ and refer to the config docs? Need to wait until I have it, then
+ we'll see how the stuff here will be added.
+\end{fixme}
+
+
+\sect This module supports both hierarchic and flat section
+structuring. You have to bind |\cweb@structure| to 0 (hierarchic) or 1
+(flat) before importing it, to show which structure is wanted.
+
+
+\sect This module supports suppression of different document parts.
+You have to define flags |@cweb@suppress@|\<foo>|@| with
+$\mbox{\<foo>} \in \{ |changehints|, |unchanged|, |format| \}$ to
+denote if suppression shall happen.
+
+
+\sect The counter |secnumdepth| must be~11 if chunks shall be
+numbered. That counter doesn't need to be set when this module is
+imported, it can be set locally during typesetting \cweb{} code.
+
+The definitions for flat structure assume that |secnumdepth| is~11
+during typesetting.
+
+
+\sect |\part|, |\chapter| (if available), and |\section| are used for
+\cweb{} section divisions (those with the lowest ranks). These
+commands will change the respective counters globally and will
+add toc lines and marks. You might want to prevent that by redefining
+|\addcontentsline| and |\mark|.
+
+Table of contents macros are not defined for the section divisions
+introduced by this module. You must either prevent that they are
+written to the |AUX| file at all; or you must define them.
+%
+\begin{fixme}
+ Add explanation of introduced section divisions.
+\end{fixme}
+
+For flat structure, all section divisions are of type `|chunk|', like
+chunks. The section names start with `|\cwbbRank{|\<rank>|}|'. By
+default, |\cwbbRank| takes one argument and expands to the empty list
+-- this can be redefined for the table of contents.
+
+
+\sect The document end, as explained in the \cweave{} interface specs
+(chunk~\ref{spec:sec:docend}), is not implemented. That must be done
+by the user of this module -- if the back matter is to be processed at
+all.
+
+
+
+
+\chap Processing states.
+
+We have to typeset five different categories of material:
+Documentation, program pieces---embedded within the documentation and
+as large chunks, \TeX{} material within program pieces (i.e., comments
+and refinement names), and cross reference information. Since we need
+a complete other environment for the program pieces than for the rest
+we design ``states'' where we switch to appropriately.
+%
+\begin{enumerate}
+
+\item A chunk starts in the \textsl{documentation state}.
+
+\item |\B| switches to \textsl{program state}. This can happen in
+documentation and program state.
+
+\item While we process the argument of |\PB| we're in \textsl{restricted
+program state}; |\PB| may appear in documentation and in \TeX{}
+state. Since \TeX{} state can be switched on within restricted
+program state, |\PB| can appear within the argument of itself.
+
+\item In the arguments of |\C|, |\SHC|, and |\X| we switch to
+\textsl{\TeX{} state}. All these cseqs appear only in (restricted)
+program state, their official names are actually different. I.e., only
+in (restricted) program state these cseq are bound to the meaning
+described here.
+
+\item Cross reference information are attached to most chunks with
+refinements. This information is processed in \textsl{CR state}.
+After CR state material
+comes always the next chunk or the document end, i.e., material in
+documentation state.
+
+\end{enumerate}
+%
+This FSA is illustrated by the diagram in figure~\ref{fig:fsa}.
+
+\begin{figure}
+
+\begin{center}
+\chardef\\=`\\
+\input{cweb-fsa.latex}
+\end{center}
+
+\caption{The processing state's FSA. The automaton starts and ends in
+the ``documentation'' state.}
+\label{fig:fsa}
+
+\end{figure}
+
+|\cwbb@documentation| will switch to documentation state,
+|\cwbb@program| to program state, |\cwbb@Rprogram| to restricted
+program state, |\cwbb@tex| to \TeX{} state, and |\cwbb@CR| to CR state.
+If we're already in a state, the switch to this state shall be a
+permissible null operation.
+
+
+\sect The basic difference between these states can be named with
+two parameters: (1)~The cseq bindings in effect and (2)~the layout
+parameters used for paragraph makeup.
+
+In (restricted) program state and in CR state the text is output
+under the control of \cweave{}, and tagged by \cweave{}. The used
+tags are from a global namespace and should only be in effect during
+these states. We call this tag set the \textsl{\cweave{} bindings}. In the
+other two states the tags are largely defined by the user, the tag
+set is called the \textsl{user bindings}. The switch to another binding
+is always done locally, i.e., if we switch from documentation to
+restricted program state within a group we don't have to bother about
+the restauration of the user binding; it will be done automatically
+by \TeX{} at the end of the group. Nevertheless we must be able to
+switch from the \cweave{} bindings back to the user bindings which
+were in effect when we activated the \cweave{} bindings. This is
+needed for the \TeX{} state which is always activated within
+(restricted) program state.
+
+The parameters for \textsl{program layout} are really special ones
+since they need to support the indentation which shows the program
+structure. These parameters are used in program and in \TeX{} state.
+The \textsl{document layout} parameters established by the user are
+used in the other three states.
+
+The following table shall summarize this. $C$~denotes \cweave{}
+bindings, $U$~user bindings, $P$~program layout, and $D$~document
+layout. If an entry is empty, its value is not changed on entry in
+this state.
+%
+\begin{center}
+ \DeleteShortVerb\|
+\begin{tabular}{r|c|c}
+\multicolumn1{c|}{\textsc{State}} & \textsc{Binding} & \textsc{Layout} \\[1pt]
+\hline
+documentation & $U$ & $D$ \\
+program & $C$ & $P$ \\
+restricted program & $C$ & \\
+\TeX{} & $U$ & \\
+CR & $C$ & $D$ \\
+\end{tabular}
+ \MakeShortVerb\|
+\end{center}
+
+
+\sect Since the user bindings and the document layout is defined
+initially, we don't have to do anything if it's requested. Only if we
+change it, i.e., within |\cwbb@cweave_bindings| and
+|\cwbb@program_layout|, we redefine |\cwbb@user_bindings| and
+|\cwbb@doc_layout|. If they are eventually executed, they shall
+rebind themselves back to |\relax|. This way we can switch to
+documentation state as often as we want.
+
+\beginprog
+\let\cwbb@user_bindings=\relax
+\let\cwbb@doc_layout=\relax
+
+\def\cwbb@documentation{%
+ \cwbb@user_bindings
+ \cwbb@doc_layout
+ }
+\def\cwbb@Rprogram{\cwbb@cweave_bindings}
+
+\def\cwbb@program{%
+ \cwbb@cweave_bindings
+ \cwbb@program_layout
+ }
+\def\cwbb@tex{\cwbb@user_bindings}
+
+\def\cwbb@CR{%
+ \cwbb@cweave_bindings
+ \cwbb@doc_layout
+ }
+\endprog
+
+
+
+
+\chap Saving and restoring control sequences.
+
+We have a lot of cseqs which are defined within the namespace of this
+module and which will be used with other names. This usage is in a
+controlled environment, namely neither in documentation nor in \TeX{}
+state. (I.e., the text processed is tagged by \cweave{}, not by humans;
+therefore we have a precise specification of the cseqs we have to
+accept.) We cannot work with groups where a cseq is just redefined
+and \TeX{} takes care for establishing the old binding again; when we
+switch from program state to \TeX{} state all the bindings which were
+in effect before the program state get active, i.e., in the
+documentation state, must be in effect again. We cannot simply
+consider the \TeX{} state as something parallel to program state, it
+must be a hierarchical relationship: In the program state values are
+set up which must be available after switching back from \TeX{} to
+program state.
+
+We save the current binding of a cseq in another cseq, but only if
+there exists a binding currently. This is done to save valuable
+\TeX{} main memory. (Why don't all those people who talk about
+extending \TeX{} not just implement dynamic memory management for the
+existing \TeX{}? It would still be \TeX{} and would \emph{really} be
+a worthy activity. But not enough fun, I assume\,\dots) Actually, one
+can assume that nearly no cseq bindings must be saved at all---the
+used names are strange enough. The bindings of |\foo|, i.e., of the
+cseq with the name \fbox{\ttfamily foo}, is saved as the cseq with the
+name \fbox{\ttfamily cweb\char `\\ foo}, i.e., as
+|\csname cweb\string\foo \endcsname|.
+
+
+\sect The save process is not done statically, but by the macro
+|\cwbb@rebind| which interprets a list of tuples $({\it old\_name},
+{\it new\_name})$, terminated by the tuple $(|\stop|, |\stop|)$.
+Eventually it constructs two new lists, |\cwbb@to_restore| with the
+cseqs which had a binding, and |\cwbb@undefined| with the names which
+didn't have one.
+
+The saving is actually done by |\cwbb@save_binding|, |\cwbb@rebind|
+is responsible for the initialization, |\cwbb@do_rebind| for the
+effective rebinding and the tail recursion on the list.
+
+\begin{fixme}
+ We could pull the |\next| assignment in the |\else| branch out of
+ the loop to get a better performance. Should measure if this is of
+ interest.
+\end{fixme}
+
+\beginprog
+\newtoks\cwbb@undefined
+\newtoks\cwbb@to_restore
+
+\def\cwbb@rebind{%
+ \cwbb@undefined{}%
+ \cwbb@to_restore{}%
+ \cwbb@do_rebind
+ }
+\def\cwbb@do_rebind#1#2{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \cwbb@save_binding #2%
+ \let #2=#1%
+ \let\next\cwbb@do_rebind
+ \fi
+ \next
+ }
+\endprog
+
+
+\sect If the cseq to be saved is undefined, it may just be added to
+the ``undefined list.'' Otherwise its binding is saved and it's added
+to the ``to-be-restored list.''
+
+\TeX{}nical note: The cseq-name for the saved binding must be created
+before the |\let| is executed.
+
+\beginprog
+\def\cwbb@save_binding#1{%
+ \ifx #1\undefined
+ \cwbb@undefined \expandafter{\the\cwbb@undefined #1}%
+ \else
+ \expandafter\let \csname cweb\string#1\endcsname =#1%
+ \cwbb@to_restore \expandafter{\the\cwbb@to_restore #1}%
+ \fi
+ }
+\endprog
+
+
+\sect The restoration of rebound cseqs is a two-tied activity: All
+previously undefined cseqs must be made undefined again, and all saved
+cseqs must be restored. Actually, we don't need to reset the two token
+lists, but we do it to save space.
+
+Both |\cwbb@undefine| and |\cwbb@restore_binding| iterate over a list
+of cseqs terminated by |\stop|.
+
+\begin{fixme}
+ And here the |\next| assignment could be prepended to the loop as
+ well.
+\end{fixme}
+
+\beginprog
+\def\cwbb@restore_bindings{%
+ \expandafter\cwbb@undefine \the\cwbb@undefined \stop
+ \cwbb@undefined{}%
+ \expandafter\cwbb@restore_binding \the\cwbb@to_restore \stop
+ \cwbb@to_restore{}%
+ }
+
+\def\cwbb@undefine#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \let#1\undefined
+ \let\next\cwbb@undefine
+ \fi
+ \next
+ }
+\endprog
+
+
+\sect \TeX{}nical note: As in |\cwbb@save_binding|, the cseq-name for
+the saved binding must be created before the |\let|s are executed.
+
+\begin{fixme}
+ Another |\next| assignment.
+\end{fixme}
+
+\beginprog
+\def\cwbb@restore_binding#1{%
+ \ifx #1\stop
+ \let\next\relax
+ \else
+ \expandafter\let \expandafter#1\csname cweb\string#1\endcsname
+ \expandafter\let \csname cweb\string#1\endcsname \undefined
+ \let\next\cwbb@restore_binding
+ \fi
+ \next
+ }
+\endprog
+
+
+
+
+\chap Chunks.
+
+All chunks are numbered subsequently. The chunk number is supplied as
+part of the markup. In addition, the `changed' tag |\*| may be
+appended to the chunk number.
+
+By default, chunks have approximately two picas vertical space in
+front.
+
+If we use an hierarchic structure, we output the chunk number at the
+refinement. (It's only needed to identify the refinement for cross
+references.) But chunks are important structural units and thus should
+be marked by typographic means that's more visible than the space
+mentioned above, we add a `chunk start marker'.
+
+For a flat structure, we output the chunk number as the chunk start
+marker, that is the traditional \WEB{} layout.
+
+In any case, the documentation part shall be run in after the chunk
+start marker.
+
+\beginprog
+\newskip\CwebChunkPreSkip
+ \CwebChunkPreSkip=2pc plus 1pc minus 6pt
+\endprog
+
+
+\sect In \LaTeX{}, structural divisions are best realized by
+|\@startsection|, we'll use it also for chunk starts. The chunk start
+marker is the section heading. For hierarchic structure, we use a
+traditional paragraph sign, used since medieval times to tag new text
+units. Our first test showed that the paragraph sign has typically an
+underlength, that doesn't look good at the start of a chunk. We raise
+the sign by its depth.
+
+For flat structure, we don't have text as the section heading,
+but we have to cope for a deficiency of \LaTeX{}: It always leaves a
+quad after the section number (by the way, this is not mentioned in
+the documentation---you have to look in the macros to figure this
+out). So we use a backskip of $\rm -1\,em$ as the chunk start marker
+then.
+
+Changed chunks are marked by a changeflag, by default a star lapping
+to the right. We need the information if that chunk is changed to
+decide if we shall output the changeflag. We demand from the chunk
+macros that they record that info in the flag |@cwbb_changed_chunk@|.
+
+Each structural division must be categorized, it belongs to a
+\textsl{type}. This type determines also the name of the counter, the
+mark handling for this category, and how this section tag is
+represented in the table of contents. The type of chunks is `|chunk|'
+(trivial, isn't it?).
+
+Furthermore, a structural division in \LaTeX{} has an associated
+\textsl{level}. That level is used to determine if that division get
+numbered in text or table of contents. For sections, we'll compute the
+level from the rank, it will not be larger than the rank.%
+ \footnote{Check the next section for a more precise definition of
+ `rank' and `level'.}
+Since the highest section rank is~10, we may use 11 as chunk level.
+
+For flat structure, we output the number for all section headings,
+including chunks. For hierarchic structure we use the default maximum
+level as supplied by our document class (e.g., 3~for \cls{article},
+and 2~for \cls{report} or \cls{book}).
+
+As a last point, we have to care already for sections in flat
+structure. They will use the |chunk| category as well, we have to
+distinguish them in the table of contents. For that, we'll record the
+level in the text, it will get written to the table of contents.
+
+\beginprog
+\newif\if@cwbb_changed_chunk@
+\let\CwebChangeFlag=*
+\if@cweb@suppress@changehints@
+ \let\cwbb@change_flag=\relax
+\else
+ \let\cwbb@change_flag=\CwebChangeFlag
+\fi
+\def\cwbb@lap_change_flag{\rlap{\cwbb@change_flag}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\CwebChunkHeadingStyle\relax
+ \newbox\cweb@chunk_marker
+ \setbox\cweb@chunk_marker=\hbox{\P} % tradition from medivial times
+ \setbox\cweb@chunk_marker=\hbox{%
+ \raise\dp\cweb@chunk_marker \box\cweb@chunk_marker
+ }
+ \def\CwebChunkHeadingMarker{\copy\cweb@chunk_marker}
+ \def\CwebChunkHeading#1{%
+ \CwebChunkHeadingMarker
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+ \or
+ %% flat
+ \def\cwbbRank#1{}
+ \def\CwebChunkHeadingStyle{\bfseries}
+ \def\CwebChunkHeading#1{%
+ \protect\cwbbRank{#1}%
+ \kern -1em
+ \if@cwbb_changed_chunk@ \cwbb@lap_change_flag \fi
+ }
+\fi
+\endprog
+
+
+\sect Chunks are tagged with |\M| and get passed the chunk number as
+argument. Before we evaluate |\@startsection|, we'll start the chunk.
+That's common code to be used by sections as well, as they also
+implicitely start a chunk.
+
+\beginprog
+\def\M#1{%
+ \cwbb@start_chunk{#1}%
+ \@startsection
+ {chunk}% % category, == counter etc.
+ {11}% % level
+ {\z@}% % no indentation for heading
+ {\CwebChunkPreSkip}% % skip above heading
+ {\m@ne em}% % run-in heading, 1em distance to text
+ {\normalfont\CwebChunkHeadingStyle}% % layout
+ {\CwebChunkHeading{11}}% % text, level might go to toc file
+ }
+\endprog
+
+
+\sect But let's first declare how a chunk number is represented. The
+principle rendering is controlled by |\thechunk|, namely, we get
+arabic numbers. This rendering is used both for creation of the
+section header (if in flat mode) and for cross references. I.e.,
+|\ref| expands into it.
+
+But for flat structure we need different expansions: In the section
+header a period shall be appended, a cross reference shall appear
+without such a period. We use two different macros: |\thechunk|
+creates the number with the period, |\cwbb@refchunk| only the number.
+The latter is used for a redefinition of |\@currentlabel|. Of course,
+for hierarchic structure, both are the same.
+
+\beginprog
+\newcounter{chunk}
+
+\def\cwbb@refchunk{\arabic{chunk}}
+
+\ifcase \cweb@structure
+ %% hierarchic
+ \let\thechunk=\cwbb@refchunk
+ \or
+ %% flat
+ \def\thechunk{\arabic{chunk}.}
+\fi
+\endprog
+
+
+\sect Each chunk is started by |\cwbb@start_chunk|, which does the
+common actions. First of all, we switch to documentation state.
+
+Then we store the chunk number in the |chunk| counter. That implies
+that the chunk start must not step this counter! We'll take care for
+that later by redefining |\refstepcounter|. The actual number is
+passed as the argument, including an optional |\*| used to denote
+changed chunks. We'll redefine that cseq to update the flag
+|@cwbb_changed_chunk@|.
+
+Each chunk must be typeset with an open |\if| since it is finished by
+|\fi|. If we shall suppress the output of unchanged chunks, we have
+to use |\iffalse| if not |@cwbb_changed_chunk@|, otherwise |\iftrue|.
+The flag |@cwbb_print_chunk@| is set up to provide an appropriate
+test.
+
+When we typeset a chunk, we evaluate |\cwbb@init_print_chunk|. For
+hierarchic structure, that cseq will be used to initialize the
+typesetting of the chunk number at program parts.
+
+\beginprog
+\if@cweb@suppress@unchanged@
+ \def\if@cwbb_print_chunk@{\if@cwbb_changed_chunk@}
+ \@gobble\fi % close \if in case it's skipped
+\else
+ \let\if@cwbb_print_chunk@=\iftrue
+ \@gobble\fi % same here
+\fi
+
+\def\cwbb@start_chunk#1{%
+ \cwbb@documentation
+ \begingroup
+ \def\*{\global\@cwbb_changed_chunk@true}%
+ \global\@cwbb_changed_chunk@false
+ \global \c@chunk #1\relax % might expand \*
+ \endgroup
+ \if@cwbb_print_chunk@
+ \cwbb@init_print_chunk
+ }
+\endprog
+
+
+\sect |\@startsection| will step the section counter (i.e., |chunk|).
+Since we just assigned it in |\cwbb@start_chunk|, that would leave us
+with a counter that's one too high. And our output in cross references
+may be wrong, as a period is appended in flat structure.
+
+That damage all comes from one macro, |\refstepcounter|; let's
+advice it to do the Right Thing for |chunk| counters.
+
+\beginprog
+\let\cwbb@refstepcounter=\refstepcounter
+
+\def\cwbb@string@chunk{chunk}
+\def\refstepcounter#1{%
+ \def\@tempa{#1}%
+ \ifx \@tempa\cwbb@string@chunk
+ \protected@edef\@currentlabel{\cwbb@refchunk}%
+ \else
+ \cwbb@refstepcounter{#1}%
+ \fi
+ }
+\endprog
+
+
+
+
+\chap Sections.
+
+Important sections are started on a new page, normal sections have the
+same space as chunks in front. The very first section does not
+automatically start on a new page since a title may be in front of it.
+
+Each section division has a \textsl{rank}, that denotes the section
+depth as seen by \cweave{}.
+
+For this depth we compute a \textsl{level}, used to denote the section
+depth for \LaTeX{}. This section depth is used to decide where
+numbering shall stop, or if that entry shall be presented in the table
+of contents, etc.
+
+These two numbers are not the same, due to the support of document
+classes with chapters.
+
+We consider a section as important if it's rank is smaller than~3,
+i.e., then we'll start a new page. This limit is controlled by the
+number |\CwebRankNoEject|. The rank is used as the section level
+and is therefore available to configure depth of typeset numbering and
+appearance in table of contents.
+
+\beginprog
+\def\CwebRankNoEject{3}
+\endprog
+
+
+\sect Sections are started with the macro |\N|. It's first parameter
+is the section's rank. The second parameter is the chunk number, the
+third the title of the section. This last parameter must be terminated
+by a dot.
+
+Each section implicitly starts a chunk. As explained above, major
+sections may eject the page, and we add our actual section heading
+with |\CwebSection|.
+
+We also tell the user that we have reached the next section by telling
+him about the current chunk number. The message is output after the
+section header is set, it shall be on the correct page. Since the
+token list to be output is expanded at shipout time, we must take care
+for the immediate expansion ourselves.
+
+\beginprog
+\def\N#1#2#3.{%
+ \cwbb@start_chunk{#2}%
+ \CwebEjectSection{#1}%
+ \CwebSection{#1}{#3}%
+ \begingroup
+ \def\*{}%
+ \edef\tmp{*#2}%
+ \message \expandafter{\tmp}%
+ \endgroup
+ }
+\endprog
+
+
+\sect At the start of a section we usually eject the page if our rank
+is not too low. But if the user had a title above the first section,
+he won't want this page break. So we try to guess it. If we guess
+wrong, the user may either add |\newpage| himself or he may redefine
+|\CwebEjectSection| in a package.
+
+\beginprog
+\def\CwebEjectSection#1{% % #1 == rank
+ \gdef\CwebEjectSection##1{%
+ \relax
+ \ifnum ##1<\CwebRankNoEject\relax
+ \newpage
+ \fi
+ }%
+ }
+
+\if@titlepage
+ \CwebEjectSection{}
+\fi
+\endprog
+
+
+\sect Now comes a heary piece. We have 11~section ranks, and need 11
+different section categories, section counters, etc. For flat
+structure, they shall all do the same; for hierarchic structure, they
+shall be really different.
+
+\noindent Let's have a look at the hierarchic structure first, it's
+more complex.
+%
+\begin{itemize}
+
+\item We associate section rank~0 (that's `|@**|') with document
+ parts, i.e., map it to |\part|. Many smaller programs won't use it.
+
+\item If we are used in a class that has a bound cseq |\chapter|, we
+ associate rank~1 (`|@*0|') with it and rank~2 (`|@*1|') with
+ |\section|. Otherwise we associate rank~1 with |\section|. These
+ tags (|\chapter| and/or |\section|) must be defined because many
+ macros that want to create section divisions will use them.
+
+\item All other section ranks are placed below |\section|.
+ |\subsection| and other \LaTeX{} style markup is not supported.
+
+ For hierarchic structure, we identify these sections with category
+ names `|section|$l$', where $l$ is the respective level, computed
+ from the rank. Those section names do not exist yet, we have to
+ create them.
+
+\end{itemize}
+
+Now we come to the point why we can't use the rank as the section
+level passed to \LaTeX{}: When we assume the level usage of \LaTeX{}'s
+standard classes, |\section| is \emph{always} on level~1, independent
+of the existence of |\chapter|. The latter is placed on level~0, its
+existence only influences the section level of |\part|, which is
+either $-1$ or~0.
+
+This implies that |\subsection| in a standard class is always on
+level~2, but our equivalent section division is either `|@*0|'
+(rank~1) or `|@*1|' (rank~2). I.e., when we have chapters, we use a
+level that's one lower than the rank, otherwise we use the rank.
+
+For flat structure, we use type |chunk|. This way we'll use the chunk
+numbers, they are already set up.
+
+For flat structure, we will map all ranks to the same section markup.
+This markup will also be accessible by |\section| or |\chapter|. They
+will be declared to be on a level that is the same as the rank.
+
+
+\sect The printed representation of a section divison number is simply
+the presentation from the next upper level, a full stop, and an arabic
+section number. When we compute the name of the previous level we must
+pay attention to our first section division created this way. This
+division must be placed below |section|, without any following number.
+
+Since registers are a valuable resource in \TeX{}, we have to take
+care not to allocate a superfluous one.
+
+\beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \newcounter{section2}[section]
+ \@namedef{thesection2}{\@nameuse{thesection}.\arabic{section2}}
+ \newcounter{section3}[section2]
+ \@namedef{thesection3}{\@nameuse{thesection2}.\arabic{section3}}
+ \newcounter{section4}[section3]
+ \@namedef{thesection4}{\@nameuse{thesection3}.\arabic{section4}}
+ \newcounter{section5}[section4]
+ \@namedef{thesection5}{\@nameuse{thesection4}.\arabic{section5}}
+ \newcounter{section6}[section5]
+ \@namedef{thesection6}{\@nameuse{thesection5}.\arabic{section6}}
+ \newcounter{section7}[section6]
+ \@namedef{thesection7}{\@nameuse{thesection6}.\arabic{section7}}
+ \newcounter{section8}[section7]
+ \@namedef{thesection8}{\@nameuse{thesection7}.\arabic{section8}}
+ \newcounter{section9}[section8]
+ \@namedef{thesection9}{\@nameuse{thesection8}.\arabic{section9}}
+ \ifx \chapter\undefined
+ \newcounter{section10}[section9]
+ \@namedef{thesection10}{\@nameuse{thesection9}.\arabic{section10}}
+ \fi
+\fi
+\endprog
+
+
+\sect Sections are typeset by |\CwebSection|; rank and title are the
+arguments.
+
+For hierarchic structure, we select the proper cseq according to rank:
+|\part|, |\chapter| if this is a structure with chapters, |\section|,
+and |\cwbb@section{|\<rank>|}| for other ranks. For the last cseq we
+also have to supply the section category, it's `|section|$l$' where
+$l$ is the rank~$r$ if we don't have chapters, $r-1$ otherwise.
+
+That category cannot be computed by |\cwbb@section| itself since we
+want to use that macro also for flat structure. But there all sections
+are basically chunks, we have to use the |chunk| category there. Since
+we want to differentiate them in the table of contents nevertheless,
+we record the current rank in the title.
+
+\beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \ifx \chapter\undefined
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\section
+ \else %% level = rank
+ \def\cwbb@sect_name{section#1}%
+ \def\next{\cwbb@section{#1}}%
+ \fi
+ \next{#2}%
+ }
+ \else
+ \def\CwebSection#1#2{%
+ \ifcase #1
+ %% 0
+ \let\next\part
+ \or %% 1
+ \let\next\chapter
+ \or %% 2
+ \let\next\section
+ \else %% level = rank - 1
+ \count@=#1 \advance \count@\m@ne
+ \edef\cwbb@sect_name{section\number\count@}%
+ \edef\next{\noexpand\cwbb@section{\number\count@}}%
+ \fi
+ \next{#2}%
+ }
+ \fi
+ \or
+ %% flat
+ \def\CwebSection#1#2{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{#1}{\protect\cwbbRank{#1}#2}%
+ }
+ \def\section{%
+ \def\cwbb@sect_name{chunk}%
+ \cwbb@section{1}%
+ }
+\fi
+\endprog
+
+
+\sect We use the same layout for all minor sections. That layout can
+be changed by ammended by defining |\CwebSect|$r$|Hook| to redefine
+|\CwebSectPreSkip| and/or |\CwebSectLayout| locally. (Of couse, it may
+be redefined once globally, too.) These are the values to be used for
+arguments 4 and~5 of |\@startsection|, pay attention to the skip sign,
+telling about indentation of the first paragraph. And don't forget
+that the skip is only used for sections on ranks below
+|\CwebRankNoEject|, the others start on a new page anyhow.
+
+The default definition sets the section title in bold and leaves the
+same space as before chunks. The first paragraph of the following
+documentation has no indentation.
+
+Other layout changes (indentation for heading, etc.)\ must be made by
+redefining |\cwbb@section|. Or---email me, if you want to do it. Then
+I might be enclined to introduce further possibilities.
+
+\beginprog
+\newskip\CwebSectPreSkip
+ \CwebSectPreSkip = -\CwebChunkPreSkip
+\def\CwebSectLayout{\normalfont\bfseries}
+
+\def\cwbb@section#1{% % #2 (next token) is title
+ \csname CwebSect#1Hook\endcsname % \relax if undefined
+ \expandafter\@startsection
+ \expandafter{\cwbb@sect_name}% % category; for counter, etc.
+ {#1}% % level is rank
+ \z@ % no indentation for heading
+ \CwebSectPreSkip % skip above, maybe no next par indent
+ \medskipamount % skip below heading
+ \CwebSectLayout
+ }
+\endprog
+
+
+
+
+\chap Marks.
+
+\begin{fixme}
+ Since I don't know yet how to handle marks I ignore most of them for
+ the time being.
+
+ Have to change this later.
+\end{fixme}
+
+\beginprog
+\let\chunkmark\@gobble
+
+\count@=2
+\@whilenum \count@<11 \do {
+ \expandafter\let \csname section\number\count@ mark\endcsname \@gobble
+ \advance \count@ by 1
+ }
+\endprog
+
+
+
+
+\chap Typesetting programs.
+
+Program pieces come in two flavours: as argument of |\PB|, or as
+material after |\B|. In the former case we can use a group for the
+switch to the restricted program state, the group end will restore the
+previous state again. In the latter case we use |\B| for the switch to
+program state, the cross reference list or the next chunk will go to
+another state. Since we have to do more for the material after |\B|,
+we define this cseq later.
+
+\beginprog
+\def\PB#1{%
+ \begingroup
+ \cwbb@Rprogram
+ \leavevmode
+ #1%
+ \endgroup
+ }
+\endprog
+
+
+\sect Note, that |\Y| cannot just be |\smallskip|, as in Plain
+\cweb{}. We must assert that the current paragraph is ended before the
+vertical glue is inserted, and that's not done by the \LaTeX{}
+definition of |\smallskip|. In addition we add some negative penalty,
+here is a good place for a page break. As the penalty value we use
+half the chunk break penalty---of course a chunk start is an even
+better place for a page break\,\dots
+
+It might be that |\Y| is the very first token in a chunk. Then the
+chunk start marker isn't set already since it is regarded as a run-in
+section heading. In that case we don't set the vertical skip but
+simply start the chunk with the program part.
+
+\TeX{}nical note: The flag |@noskipsec| may be used to test if we're
+immediately after a run-in section heading.
+
+\beginprog
+\newcount\CwebProgPenalty
+ \CwebProgPenalty=\@secpenalty
+ \divide\CwebProgPenalty by 2
+\def\Y{%
+ \if@noskipsec
+ \else
+ \par
+ \penalty\CwebProgPenalty
+ \smallskip
+ \fi
+ }
+\endprog
+
+
+\sect Yep, let's unfold the ``official'' names of the cseqs used in
+program state.
+
+If we turn on the \cweave{} bindings, they might be in effect already,
+we don't need to establish them again. We can test
+|\cwbb@user_bindings| for this case, it will be redefined then.
+
+\beginprog
+\def\cwbb@cweave_bindings{%
+ \ifx \cwbb@user_bindings\relax
+ \cwbb@rebind
+ % indentation and paragraph layout
+ \cwbb@incr_indent \1%
+ \cwbb@decr_indent \2%
+ \cwbb@expr_break \3%
+ \cwbb@backup \4%
+ \cwbb@opt_break \5%
+ \cwbb@break \6%
+ \cwbb@big_break \7%
+ \cwbb@noindent \8%
+ % C/C++ tokens
+ \CwebRel \?%
+ \CwebAddress \AND
+ \CwebComplement \CM
+ \CwebScope \DC
+ \CwebEquiv \E
+ \CwebGe \G
+ \CwebRightShift \GG
+ \CwebNe \I
+ \CwebAssign \K
+ \CwebLeftShift \LL
+ \CwebMod \MOD
+ \CwebNull \NULL
+ \CwebNot \R
+ \CwebBinOr \OR
+ \CwebMemberRef \PA
+ \CwebThis \this
+ \CwebOr \V
+ \CwebAnd \W
+ \CwebXor \XOR
+ \CwebLe \Z
+ \CwebPointer \MG
+ \CwebPointerMemberRef \MGA
+ \CwebDecr \MM
+ \CwebIncr \PP
+ % more tokens
+ \CwebId \\%
+ \CwebIdLetter \|%
+ \CwebRes \&%
+ \CwebString \.% %% ( ...Emacs...
+ \CwebStringBreak \)%
+ \CwebNumber \T
+ \CwebCombinedOp \MRL
+ % goes to TeX state
+ \CwebComment \C
+ \CwebCxxComment \SHC
+ \CwebRefName \X
+ % CWEB tokens
+ \CwebMacrosHere \ATH
+ \CwebDefine \D
+ \cwbb@format \F
+ \CwebIdCat \J
+ \CwebVerbString \vb
+ % cross reference tags
+ \cwbb@change_flag \*%
+ \CwebCRAlso \A
+ \CwebCRsAlso \As
+ \CwebCRCite \Q
+ \CwebCRsCite \Qs
+ \CwebCRUse \U
+ \CwebCRsUse \Us
+ \CwebCREt \ET
+ \CwebCRsEt \ETs
+ % finish the list
+ \stop\stop
+ \def\cwbb@user_bindings{%
+ \cwbb@restore_bindings
+ \let\cwbb@user_bindings\relax
+ }%
+ \fi
+ }
+\endprog
+
+
+\sect Since most of the \cweave{} bindings are simple and tedious
+coding, we'll have a look at the program layout next. Between two
+paragraphs there must not be any skip, the skip used in document
+layout is saved in |\cwbb@save@parskip|. A few other layout
+parameters from the document layout must be saved as well.
+
+\beginprog
+\newskip\cwbb@save@parskip
+\newskip\cwbb@save@rightskip
+\newcount\cwbb@save@sem_sfcode
+\newcount\cwbb@save@pretolerance
+\newcount\cwbb@save@hyphenpenalty
+\newcount\cwbb@save@exhyphenpenalty
+
+\def\cwbb@save_doc_layout{%
+ \cwbb@save@parskip\parskip
+ \cwbb@save@rightskip\rightskip
+ \cwbb@save@sem_sfcode\sfcode`\;
+ \cwbb@save@pretolerance\pretolerance
+ \cwbb@save@hyphenpenalty\hyphenpenalty
+ \cwbb@save@exhyphenpenalty\exhyphenpenalty
+ }
+\def\cwbb@restore_doc_layout{%
+ \parskip\cwbb@save@parskip
+ \rightskip\cwbb@save@rightskip
+ \sfcode`\;=\cwbb@save@sem_sfcode
+ \pretolerance\cwbb@save@pretolerance
+ \hyphenpenalty\cwbb@save@hyphenpenalty
+ \exhyphenpenalty\cwbb@save@exhyphenpenalty
+ }
+\endprog
+
+
+\sect When |\cwbb@program_layout| is called for the very first time
+we have to set up some variables. This cannot be done in advance
+since they depend on values which might be changed by the user in the
+preamble. This setup is done in |\cwbb@layout_init|.
+
+If the program layout is already active, we must not switch it on
+another time. This can be tested by the current binding of
+|\cwbb@doc_layout|.
+
+\beginprog
+\def\cwbb@program_layout{%
+ \ifx \cwbb@doc_layout\relax
+ \cwbb@layout_init
+ \cwbb@save_doc_layout
+ \def\cwbb@doc_layout{%
+ \cwbb@restore_doc_layout
+ \let\cwbb@doc_layout\relax
+ }%
+ % set new values
+ \parskip\z@skip
+ \rightskip\z@ plus 100\p@ minus 10\p@
+ \sfcode`\;=3000
+ \pretolerance\@M
+ \hyphenpenalty 1000 % strings can be broken this way
+ \exhyphenpenalty\@M
+ \fi
+ }
+\endprog
+
+
+\sect The unit of the basic indentation is stored in
+|\CwebIndentUnit|. Continuation lines are indented two units further,
+i.e., the initial hanging indentation is 3~units. The current hanging
+indentation is kept in |\cwbb@indent|.
+
+We need undiscardable items which can be used as backspaces, of one
+and two units, respectively. This is done best by boxes. The
+initialization of the boxes must be postponed until the user had the
+chance to change |\CwebIndentUnit|, it's done in the program state
+initialization.
+
+\beginprog
+\newdimen\CwebIndentUnit
+ \CwebIndentUnit=1em
+\newdimen\cwbb@indent
+
+\newbox\cwbb@bak % backspace one unit
+\newbox\cwbb@bakk % backspace two units
+
+\def\cwbb@layout_init{%
+ \global\setbox\cwbb@bak \hbox to -1\CwebIndentUnit{}%
+ \global\setbox\cwbb@bakk \hbox to -2\CwebIndentUnit{}%
+ \global\let\cwbb@layout_init\relax
+ }
+\endprog
+
+
+\sect Now we can formulate how to start typesetting program pieces: We
+switch to program state, set the initial indentation, and add the
+basic indentation of the first code line. Then we might add a marker
+to this program part; for hierarchic structure we want to typeset the
+chunk number in the margin of the chunk's program part. That's not
+really possible, 'though: definitions and format requests are also
+introduced with |\B|, so the chunk number will happen to be at the
+start of the definition part if one exists.
+
+If the code line is the very first text to be typeset in this chunk,
+we don't add the basic indentation---we're already indented from the
+run-in chunk start marker. But we must not start our chunk with
+|\noindent|, the marker would be typeset in the left margin then. (I
+consider this as a \LaTeX{} error.) So we just emulate an empty
+documentation part, the marker is now set correctly. |\B| can test for
+this case: |@noskipsec| is still true then.
+
+\beginprog
+\def\B{%
+ \cwbb@program
+ \cwbb@indent 3\CwebIndentUnit \hangindent\cwbb@indent
+ \ifvmode
+ \if@noskipsec
+ \indent % add an empty documentation part
+ \else
+ \noindent \kern\CwebIndentUnit
+ \fi
+ \fi
+ \CwebNumberProgramPart
+ }
+\endprog
+
+
+\sect The chunk number shall be added at the first program part. We
+have a hook in the chunk start where we can set up
+|\CwebNumberProgramPart|. It will redefine itself at first call, to
+disable further output on more |\B| occurences in a chunk.
+
+The chunk number should always be in the left margin. If it's in the
+right margin, there would be too much horizontal space between the
+code start and the actual number. (The first code line is typically
+not very long.) \LaTeX{} does not allow to select the margin for
+marginpars, therefore we have to typeset the marginal note `by hand'.
+As we know that we're in horizontal mode when we're called, |\vadjust|
+is used to create the note. The standard \LaTeX{} dimen register for
+the distance between marginal notes and text matter is used, of
+course.
+
+Note that marginal notes may overwrite the chunk number. As it's very
+unlikely that somebody issues marginal notes in program mode (that's
+quite difficult and involves ingenious usage of the implementation of
+this macros), I won't care for that case.
+
+\begin{fixme}
+ The baseline of the marginal note shall be aligned to the baseline
+ of the preceding line. That does not happen by default. Currently, I
+ assume that the preceding line has a depth of appropriately the
+ depth of a~`g'. Experiments showed that this gives good results for
+ most \cweb{} documents. But nevertheless one should determine the
+ exact depth of the preceding line, to know how much |box0| should be
+ raised. If anybody knows how this can be done, please tell me. (I
+ know about the route over the output routine, implementing an own
+ marginal note category, but this seems too much effort.)
+\end{fixme}
+
+For flat structure, we don't add any program part marker and we
+don't need to initialize it.
+
+\beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\cwbb@marginal_chunkno{%
+ \vadjust{%
+ \setbox\z@ \hbox to \z@{%
+ \hss \vphantom{g}%
+ \small \thechunk
+ \hskip \marginparsep
+ }%
+ \vbox to \z@{%
+ \hsize\z@
+ \vss
+ \noindent \raise \dp\z@ \box\z@
+ }%
+ }%
+ }
+ \def\cwbb@init_print_chunk{%
+ \def\CwebNumberProgramPart{%
+ \cwbb@marginal_chunkno
+ \let\CwebNumberProgramPart\relax
+ }%
+ }
+ \or
+ %% flat
+ \let\cwbb@init_print_chunk\relax
+ \let\CwebNumberProgramPart\relax
+\fi
+\endprog
+
+
+\sect If a statement is finished, a new paragraph with the basic
+indentation has to be started.
+
+An optional statement break is implemented by a low penalty which
+will be selected if the line has to be broken. We assume that the
+hanging indentation of the new line is already set correctly, and
+have to backup two units to get the basic indentation. It might be
+that the line break is not chosen by \TeX{}, we compensate the
+backspace for this case. This compensation is discarded at the start
+of a new line.
+
+\beginprog
+\def\cwbb@break{% % forced break, between statements
+ \ifmmode\else
+ \endgraf % in LaTeX it isn't sure what \par is really...
+ \noindent
+ \hangindent\cwbb@indent \kern\cwbb@indent
+ \copy\cwbb@bakk % go back to basic indentation
+ \ignorespaces
+ \fi
+ }
+\def\cwbb@opt_break{% % optional break between statements
+ \hfil \penalty\m@ne
+ \hfilneg \kern .5em \kern 2\CwebIndentUnit % discarded on line break
+ \copy\cwbb@bakk
+ \ignorespaces
+ }
+\def\cwbb@big_break{% % forced break and a little extra space
+ \Y
+ \cwbb@break
+ }
+\def\cwbb@expr_break#1{% % break with penalty #1 * 10
+ \hfil \penalty#10
+ \hfilneg % discarded on line break
+ }
+\endprog
+
+
+\sect When we increment the indentation, we must not forget to set
+the hanging indentation immediately since |\cwbb@opt_break| relies on
+the new value. If a continuation line is needed it will also be
+indented one unit more, which is ok since it should be distinguishable
+from the next line.
+%
+\begin{fixme}
+ Shouldn't the hanging indentation be set anew at the end of
+ |\cwbb@decr_indent| as well?
+\end{fixme}
+
+\beginprog
+\def\cwbb@incr_indent{%
+ \global\advance\cwbb@indent\CwebIndentUnit
+ \hangindent\cwbb@indent
+ }
+\def\cwbb@decr_indent{\global\advance\cwbb@indent -\CwebIndentUnit}
+\def\cwbb@backup{\copy\cwbb@bak}
+\def\cwbb@noindent{% % no indentation
+ \hskip -\cwbb@indent \hskip 2\CwebIndentUnit
+ }
+\endprog
+
+
+
+
+\chap Program (C or \C++) tokens.
+
+Since the user might want to change the (somewhat unusual) way some
+of the operators are typeset, we supply names for them.
+
+\beginprog
+\let\CwebAnd=\land % logical and, &&
+\let\CwebEquiv=\equiv % equiv sign, for ==
+\let\CwebGe=\ge % greater or equal
+\let\CwebLe=\le % less or equal
+\let\CwebNe=\ne % unequal, !=
+\let\CwebNull=\Lambda % NULL pointer
+\let\CwebNot=\lnot % logical not, !
+\let\CwebOr=\lor % logical or, ||
+\let\CwebXor=\oplus % bitwise exclusive or, ^
+\endprog
+
+
+\sect Some symbols have to be shifted around, to save computation time
+we put them in boxes.
+
+\begin{fixme}
+ It's not sure if the following symbols still look good in other font
+ families than Computer Modern. The movements might need to get
+ adapted when more experience is available. They were chosen from
+ Plain \cweb{}, where these were explicit movements measured in
+ points. I changed the dimension to quads, to achieve at least the
+ support for different sizes in the Computer Modern family.
+\end{fixme}
+
+\TeX{}nical note: The pointer symbol must not be declared as a math
+symbol (e.g., with |\DeclareMathSymbol|). Then \TeX{} would insert an
+italic correction behind it, according to rule~17 from appendix~G.
+(Please send email if you know of an easy way to prevent that
+addition.) So I use the symbol as an accent, but then I have to supply
+a nucleus for this math list. That nucleus also determines the total
+width of this construct. As the vector accent has a width of
+approximately 5\,pt in |cmmi10|, I'm going to use 0.5\,em. (That
+heuristic is as good as any other I can think of.)
+
+\beginprog
+%% pointer to struct component (`->'), use symbol of \vec accent
+\newbox\cwbb@pointer
+\setbox\cwbb@pointer=\hbox{%
+ \kern -.2em
+ \lower .3em \hbox{$\vec{\kern .5em}$}%
+ \kern .1em
+ }
+
+\newbox\cwbb@decr % decrement, --
+\setbox\cwbb@decr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {-}\kern -.1em{-}$}%
+ \kern .05em
+ }
+\newbox\cwbb@incr % increment, ++
+\setbox\cwbb@incr=\hbox{%
+ \kern .05em
+ \raise .1em \hbox{$\scriptstyle {+}\kern -.1em{+}$}%
+ \kern .05em
+ }
+\endprog
+
+
+\sect Although the following symbols are typeset like an ``ordinary
+\C++ programmer'' would expect them, we provide own module names
+nevertheless. They can now be changed as well, i.e., orthogonality is
+enhanced. (And we can use |\cwbb@rebind| for assigning them to their
+names while we switch to program state\,\dots).
+
+Note: The comment at |\CwebMod| says ``modulo/remainder'', as this is
+not defined by the C standard, i.e., implementation-dependent.
+
+\beginprog
+\mathchardef\CwebAddress="2026 % `&', as binary op
+\let\CwebAssign== % assignment
+\let\CwebBinOr=\mid % binary or
+\def\CwebComplement{{\sim}} % `~', as ordinary symbol
+\def\CwebDecr{\copy\cwbb@decr} % decrement
+\def\CwebIncr{\copy\cwbb@incr} % increment
+\let\CwebLeftShift=\ll % left shift, <<
+\def\CwebMod{\mathbin{\hbox{\footnotesize\rm\%}}} % modulo/remainder, %
+\def\CwebMemberRef{\mathbin{.*}} % ptr to member (on object)
+\def\CwebPointer{\copy\cwbb@pointer}
+\def\CwebPointerMemberRef{\mathbin{\CwebPointer*}} % ptr to member (on ptr)
+\def\CwebRel{\mathrel?} % relation operator
+\let\CwebRightShift=\gg % right shift, >>
+\def\CwebScope{\kern.1em{::}\kern.1em} % scope resolution
+
+\def\CwebThis{\CwebRes{this}} % reserved identifier `this'
+\endprog
+
+
+\sect The special identifier |\TeX| remains. In math mode it shall be
+the identifier, in other modes the usual logo. We use the definition
+supplied by DEK (or \textsc{Silvio Levy}?)\ from Plain \cweb{}.
+
+\beginprog
+\def\TeX{%
+ {%
+ \ifmmode\it\fi
+ \mbox{T\kern-.1667em\lower.424ex\hbox{E}\hskip-.125emX}%
+ }}
+\endprog
+
+
+\sect Some tokens don't have constant names, the name is supplied as
+the argument.
+
+Identifiers are typeset in italics, reserved words and type names in
+boldface, and strings in typewriter. For an underscore in the bold
+words we use a line that's 50\,\% thicker as the usual line thickness.
+-- But what is the usual line thickness. As Knuth uses 0.4\,pt for a
+10\,pt Computer Modern font, we'll use 0.06\,em for a CM font in an
+arbitrary size. If that line thickness is still OK for other fonts is
+doubtful, as it should fit to the stem width. (CM is a rather light
+font, after all.) But since there is no font dimension that gives a
+hint to an appropriate line thickness, it's our best try.
+
+\beginprog
+\def\CwebCombinedOp#1{\mathrel{\let\K==#1}} % e.g., += operator
+\def\CwebId#1{\hbox{\it#1\/\kern.05em}} % identifier, more than one char
+\def\CwebIdLetter#1{\hbox{$#1$}} % identifier, one letter
+\def\CwebRes#1{% % reserverd words and type names
+ \hbox{\bf
+ \def\_{\kern.04em\vbox{\hrule width.3em height.06em}\kern.08em}%
+ #1\/\kern .05em
+ }%
+ }
+\endprog
+
+
+\sect In Plain \cweb{} strings are typeset in the font |cmtex10|,
+which is a typewriter font with extended ASCII characters. This font
+does not have the usual accents and can therefore not used for
+typesetting national characters. (Very often they are input in some
+8-bit encoding, the respective character code is made active and is
+substituted by a cseq which expands to the correct glyph. Of course,
+there are better ways, but that's the reality we have to cope with.)
+Instead of this special font we use the standard typewriter font.
+This has the further advantage that |cmtex10| may not be introduced
+to NFSS\@.
+
+\beginprog
+\def\CwebString{%
+ \hbox \bgroup
+ \tt
+ \cwbb@string_setup
+ \cwbb@string
+ }
+\def\cwbb@string#1{%
+ #1\kern .05em
+ \egroup
+ }
+\endprog
+
+
+\sect Within strings certain cseqs have a special meaning; this is
+introduced by |\cwbb@string_setup|. The cseqs within strings are
+mostly special \TeX{} characters that are to be typeset by cseqs with
+the same name.
+
+But a problem remains: active characters. They will still expand, and
+will mess up the string visualization. This is a particular problem
+with |german.sty| where the double quote is active. (Of course, that's
+a problem -- I'm German. :--) When one typesets strings, a double
+quote shall be typeset as double quotes, and must not expand to the
+special meaning. Remember that double quotes happen to appear quite
+often in strings. As the string argument is already tokenized, one has
+to redefine the binding of $(\textit{\textup{symbol}}\ \textbf{.}\
+|#\"|)$.
+
+\begin{fixme}
+ Perhaps one might use a loop over all characters except the braces,
+ and set their catcode to other. That would give us enough security.
+
+ For now, we handle only the double quote.
+\end{fixme}
+
+Every once in a while we have discretionary breaks in a string,
+denoted by |\CwebStringBreak|. This break is shown with the C
+convention of escaped newlines.
+
+\beginprog
+\def\cwbb@string_setup{%
+ \chardef\ =`\ % % <-- two spaces !
+ \chardef\&=`\&
+ \chardef\\=`\\
+ \chardef\^=`\^
+ \chardef\_=`\_
+ \chardef\{=`\{
+ \chardef\}=`\}
+ \chardef\~=`\~
+ \catcode`\"=\CatOther
+ }
+
+\def\CwebStringBreak{\discretionary{\hbox{\tt\char`\\}}{}{}}
+\endprog
+
+
+\sect Numbers are typeset in different ways. We use the definition of
+Plain \cweb{}.
+%
+\begin{fixme}
+ Should add a specification of the possible input and an explanation
+ of the macros below. In particular, that the closing brace after
+ |\aftergroup| is used much later is probably not grokked by most
+ \TeX{} programmers.
+\end{fixme}
+
+\beginprog
+\def\cwbb@oct{\hbox{$^\circ$\kern-.1em\it\aftergroup\?\aftergroup}}
+\def\cwbb@hex{\hbox{$^{\scriptscriptstyle\#}$\tt\aftergroup}}
+
+\def\CwebNumber#1{% % octal, hex, or decimal constant
+ \hbox{%
+ $%
+ \def\?{\kern.2em}%
+ \def\$##1{\egroup\sb{\,\rm##1}\bgroup}% suffix to constant
+ \def\_{\cdot 10^{\aftergroup}}% power of ten (via dirty trick)
+ \let\~\cwbb@oct \let\^\cwbb@hex
+ {#1}%
+ $}%
+ }
+\endprog
+
+
+\sect Comments are typeset in \TeX{} state. We add a hook, the user
+shall be able to change the layout (e.g., he might want another font).
+%
+\begin{fixme}
+ Currently \C++ comments are typeset like C comments. This is
+ horrible in usual circumstances, i.e., when complete blocks of text
+ are prefixed with |//|. We should simply catenate all these text and
+ typeset it as one paragraph, each line prefixed by |//|. But then we
+ have to implement an |\everyline| first, and since that's not so
+ easy we postpone it\,\dots
+\end{fixme}
+%
+In front of a C comment there is an optional stmt break, with 2\,quad
+in front if the line is not broken and 1.5\,quad if the line is broken.
+
+\beginprog
+\def\CwebComment#1{%
+ \5% % 0.5em will be discarded on line break
+ \hskip 1.5em
+ $/\ast\,$%
+ {\cwbb@tex
+ \CwebCommentHook
+ #1%
+ }%
+ $\,\ast/$%
+ }
+\let\CwebCommentHook\relax
+\let\CwebCxxComment\CwebComment
+\endprog
+
+
+
+
+
+\chap \cweb{} tokens.
+
+We distinguish three categories of \cweb{} tokens: (1) Those which
+output constant text, (2) those which have attributes to be displayed
+in a special way, and (3) those which start a new structure element,
+namely |@d| and~|@f|. Let's consider them in this order.
+
+
+\sect \cweb{} tokens which expand in a constant string are the
+identifier catenation operator~(`|@&|') and the macro placement
+directive~(`|@h|').
+
+\beginprog
+\def\CwebIdCat{\CwebString{@\&}}
+\def\CwebMacrosHere{%
+ \begingroup
+ \def\CwebRefNumber##1{}%
+ \CwebRefName :Preprocessor definitions\X
+ \endgroup
+ }
+\endprog
+
+
+\sect Verbatim program strings, i.e., strings passed verbatim by
+\ctangle{}~(`|@=|') are typeset like normal strings, but within a box.
+We use 2\,pt as the separating distance, this is set locally.
+
+\beginprog
+\def\CwebVerbString#1{{\fboxsep\tw@\p@ \fbox{\CwebString{#1}}}}
+\endprog
+
+
+\sect The refinement names are typeset in angles, this has a long
+tradition. We burry the typesetting of the chunk number (which is
+the first parameter) in a macro call; the user may change this to
+achieve special effects.
+%
+\begin{fixme}
+ One could think about handling special values of the chunk numbers
+ differently. E.g., an empty argument is not typeset at all, a~0
+ triggers a marginal note about a missing definition, etc.
+\end{fixme}
+
+A refinement name may be typeset both in math and horizontal mode.
+The name itself is typeset in horizontal mode, of course; for the
+angles we need math mode. Therefore we assert at the start of the
+macro that we're not in math mode any more. At the end we switch back
+to math mode if we've started in it. This conditional switch from and
+back to math mode is done by |\cwbb@toggle_text|.
+
+\TeX{}nical note: If we're in math mode |\cwbb@toggle_text| must be
+defined globally, as it will turn off math mode and the definition
+would be un-made then. The second invocation of |\cwbb@toggle_text|
+would be undefined then.
+
+A refinement may also consist of the file name the expansion of this
+refinement shall be written to.
+%
+\label{sec:refname-dot}
+%
+This file name is tagged with |\.|, it shall be typeset like a string.
+But the user shall be able to use the dot accent in the refinement
+name as well. We check if the text consists solely of the tag and its
+argument; in this case we substitute |\.| with |\CwebString|.
+Otherwise we leave it as it is. Then a refinement name may not
+consist of a single dot-accented expression---well, that's highly
+unlikely. (Nevertheless I'll document it in the user's manual\,\dots)
+
+\begin{fixme}
+ One should be able to use |\EnsureMath| here. But the refinement
+ name must not be set in math mode and must not be set in an hbox;
+ the latter would prevent line breaks.
+\end{fixme}
+
+\beginprog
+\def\CwebRefName#1:#2\X{%
+ \ifmmode \gdef\cwbb@toggle_text{\null$\null}%
+ \else \let\cwbb@toggle_text\relax
+ \fi
+ \cwbb@toggle_text
+ $\langle\,${\cwbb@tex \cwbb@check_dot{#2}\CwebRefNumber{#1}}$\,\rangle$%
+ \cwbb@toggle_text
+ }
+\def\CwebRefNumber#1{%
+ {\reset@font \footnotesize \kern .5em#1}%
+ }
+\endprog
+
+
+\sect The next macro, |\cwbb@check_dot| is a hassle to implement. We
+must check if it's argument consists solely of a `|\.|'~tag, together
+with the argument to this tag. Then the `|\.|'-argument shall be
+typeset like a string, |\.| itself must be ignored.
+
+The check for |\.| must not evaluate the parameter. Especially |\PB|
+must not be evaluated within, it would lead to havoc with the
+redefinition of all bindings, etc. (At least that's the empirical
+result---it happened; although I don't know where the problems are.
+Anyhow, it wasn't in the specs of the rebinding process.) All other
+cseqs which shouldn't be used in moving arguments may cause problems
+here, too. Note that |\protect| is no solution here, the user often is
+not aware of the presence of \cweave{}-generated tags. As an example,
+consider the refinement \verb"@< local variables of |foo| @>|" which
+will lead to a call with the argument `|local variables of \PB{\\{foo}}|'.
+|\\| must not be evaluated here.
+
+I use an approach where I hope that it will work---but to be honest,
+I'm not sure. I want to define a macro |\next| which shall have an
+empty expansion iff |#1| was a `|\.|'~tag with its argument. Empty
+refinement names are not allowed so we can assume that |#1| consists
+of at least one token. I evaluate the first token of |#1| before the
+|\def| is done. With an appropriate binding a |\.| will disappear,
+perhaps leaving an empty token list. I also take care that |\PB| is
+not evaluated then, by temporary rebinding it to |\relax|. Thereby I
+have taken care for all cseqs \cweave{} might have introduced at the
+very front of |#1|. Other cseqs are user-tags. If they cause
+problems, the user might circumstance them by adding |\protect|.
+
+\beginprog
+\def\cwbb@check_dot#1{%
+ \begingroup
+ \let\.\@gobble
+ \let\PB\relax
+ \expandafter\def \expandafter\next \expandafter{#1}%
+ \ifx \next\empty
+ \gdef\next##1{\CwebString}%
+ \else
+ \global\let\next\relax
+ \fi
+ \endgroup
+ \next #1%
+ }
+\endprog
+
+
+\sect \cweb{} macro definitions (done by~|@d|) and format directives are
+basically program text. When they are started we are already in
+program state, i.e., |\B| has appeared in front of it. But since we have
+an introductionary identifier (either |#define| or |format|) to set at
+the very front we increase the indentation by three units.
+
+The output of format directives may be suppressed. Then we have to
+gather both identifiers that will appear after |\F|. Both of them
+consist of one group-token (a cseq and the identifier name), after
+each identifier one other token separates them. I.e., we have to
+gobble six group-tokens. As the very last token is |\par| if a format
+directive follows directly on another one, the definition must be
+long.
+
+\TeX{}nical note: |\cwbb@format| must expand to |\CwebFormat|, not to
+its expansion. Otherwise a redefinition of |\CwebFormat| in a package
+would not be used.
+
+\beginprog
+\def\CwebDefine{\cwbb@macro{\#define}}
+\def\CwebFormat{\cwbb@macro{format}}
+\if@cweb@suppress@format@
+ \long\def\cwbb@format#1#2#3#4#5#6{}
+\else
+ \def\cwbb@format{\CwebFormat}
+\fi
+
+\def\cwbb@macro#1{%
+ \global\advance\cwbb@indent \tw@\CwebIndentUnit
+ \cwbb@incr_indent
+ \CwebRes{#1 }% % <-- blank!
+ }
+\endprog
+
+
+
+
+\chap Refinement cross references and changes.
+
+At the end of chunks with refinements, \cweave{} may output cross
+reference information: In which chunks additional definitions for
+this refinement are found, where this refinement is used, and where it
+is cited. This cross reference information is always introduced by a
+tag followed by a non-empty list of chunk numbers. Different tags
+are used for lists with only one element and for lists with more than
+one element, this way the introductionary text may be adapted.
+
+Within a list of $n$~chunk numbers the first $n-1$~numbers are
+separated by commas with one following blank. The last two numbers
+are separated by |\ET| if~$n=2$, and by |\ETs| if~$n>2$. If a chunk
+is changed by the changefile, a changeflag~(`|\*|') is appended to
+its number. The list is eventually terminated by a full stop.
+
+
+\sect We separate the cross reference information by a smallskip from
+the refinement or from a previous cross reference information. The
+information itself is typeset in a smaller font, as it is auxilliary,
+inserted stuff. The number list has a hanging indentation of
+|\CwebNumberListHangindent|. But beware: This isn't a dimension
+register, it's a macro. This way one can use ems as dimensions.
+
+The actual size change is represented by a macro. We will redefine it
+when we need an other size, e.g., in the lists at the end of the
+document.
+
+We must assert that we're in CR state while we set a cross reference.
+
+The first parameter of a cross reference information unit is the
+introductionary text, the second is the number list. The parameters
+must be evaluated in a group---local parameter changes therein must
+not influence the environment.
+
+\beginprog
+\def\CwebNumberListHangindent{2em}
+\let\CwebCRSize\footnotesize
+\def\CwebCrossRef#1#2.{%
+ \par\smallskip
+ \cwbb@CR
+ \begingroup
+ \reset@font \CwebCRSize
+ \noindent \hangindent\CwebNumberListHangindent\relax
+ #1~#2.\par
+ \endgroup
+ }
+\endprog
+
+
+\sect Well, let's define all introducing tags, which in fact start the
+cross reference information unit. The number list is gathered by
+|\CwebCrossRef|.
+
+\beginprog
+\def\CwebCRAlso{\CwebCrossRef{See also chunk}}
+\def\CwebCRsAlso{\CwebCrossRef{See also chunks}}
+
+\def\CwebCRCite{\CwebCrossRef{This code is cited in chunk}}
+\def\CwebCRsCite{\CwebCrossRef{This code is cited in chunks}}
+
+\def\CwebCRUse{\CwebCrossRef{This code is used in chunk}}
+\def\CwebCRsUse{\CwebCrossRef{This code is used in chunks}}
+
+\def\CwebCREt{ and~}
+\def\CwebCRsEt{, and~}
+\endprog
+
+
+
+
+\chap Bells and whistles.
+
+\begin{fixme}
+ Just copied from Plain \cweb{}. It doesn't work
+ anyhow---|\char"|\<xy> usually doesn't typeset the correct
+ character. That's because this character is most probably an active
+ character; at least that's typical for the way \TeX{} systems are
+ used in Europe. I have to think about the implementation.
+\end{fixme}
+
+\beginprog
+\def\ATL{\par\noindent\bgroup\catcode`\_=12 \postATL} % print @l in limbo
+\def\postATL#1 #2 {\bf letter \\{\uppercase{\char"#1}}
+ tangles as \tentex "#2"\egroup\par}
+\def\noATL#1 #2 {}
+\def\noatl{\let\ATL=\noATL} % suppress output from @l
+\endprog
+
+
+
+\chap The end.
+
+Well, after all we're finished with this module
+
+\beginprog
+\endinput
+\endprog
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\vskip \PltxPreSectSkip
+
+\rcsLogIntro{After the actual revision log of this module, we give an
+ extract of log messages of the \cls{cweb} class. This module was
+ extracted from revision~2.8 of that class, we present to you the
+ log entries that concern the code of the new module.}
+
+\begin{rcslog}
+$StyleLog: cwebbase.doc,v $
+\Revision 1.4 1995/11/07 17:55:19 schrod
+ \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+versions; make it work with the previous-to-last one, version
+\mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+\mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+|\newif|, and |\hb@xt@| must not be used.\\
+Problem reported by Laurent Desnogues
+\path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+-- add name).
+
+\Revision 1.3 1995/11/07 14:13:12 schrod
+Reset section numbers when higher-level section begins.\\
+Error reported by Bronne Louis \path|<bronne@montefiore.ulg.ac.be>|.
+
+\Revision 1.2 1995/11/06 10:58:47 schrod
+ Support active double quotes in strings. That happens when
+|german.sty| is used. Problems with other active characters may still
+appear.
+
+\Revision 1.1 1995/09/12 23:02:22 schrod
+Moved all code that does the actual typesetting of \cweave{} tags and
+is therefore also needed for a |cweb| environment. It's now an own
+module named \pkg{cwebbase}.
+
+Chunk markers and chunk marginal numbers don't hang below the
+baseline any more.
+
+
+\item[] \noindent\hrulefill\break
+ \textbf{Extract of the revision log for |cweb.doc|:}
+
+ \vskip 2ex
+
+\Revision 2.5 1995/08/29 02:07:26 schrod
+Discard dependencies on 10\,pt fonts.
+
+Support suppression of format directives.
+
+\Revision 2.4 1995/08/27 19:31:43 schrod
+Make configuration of change flag easier.
+
+Put section title of changed chunks list in marks, and tell the user
+that it got typeset.
+
+\Revision 2.3 1995/08/27 17:24:46 schrod
+Make usage of baseclass with chapters work.
+
+\Revision 2.2 1995/08/27 13:26:22 schrod
+Add possibility to suppress change hints.
+
+\Revision 2.1 1995/08/25 19:11:18 schrod
+Hierarchic structures are supported now, in addition to the flat
+structure of the beta-test version. One can choose with an option. For
+that step, the terminology was cleaned up, too: Chunks are not named
+sections any more. (That change involved reimplementation of almost
+all the structure and toc stuff.)
+
+The chunk number supplied by \cweave{} is used now, not some computed
+number. Change flags are printed, too.
+
+One can suppress output of unchanged sections.
+
+\Revision 1.12 1993/08/10 14:15:43 schrod
+New page on main section only if group level $<$ |\cwebSecNoEject|.
+Default for the latter is 3.
+
+\Revision 1.11 1993/08/10 11:21:07 schrod
+Reference to section number does not render a period after the
+number any more.
+
+\Revision 1.10 1993/08/09 20:08:20 schrod
+|\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+effect already.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+\Revision 1.9 1993/08/09 18:05:28 schrod
+Left shift operator wasn't defined correctly.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+\Revision 1.6 1993/06/15 08:49:23 schrod
+|\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+causes problems if a |\PB| is within. Now I try hard not to evaluate any
+tokens outside of my control.
+
+Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+need an own macro.
+
+\Revision 1.4 1993/06/14 15:54:18 schrod
+Add FSA diagram about processing states. CR state is also switched
+to from \TeX{} state (that happens with |\ch| at the document end).
+
+\Revision 1.3 1993/05/13 17:51:21 schrod
+Refinements may also be filenames (`|@(|'). Then the complete name
+consists of a |\.| macro call, which is handled now.\\
+(Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+Made the detection of `|@.|' index entries more robust.
+
+\Revision 1.2 1993/05/12 18:28:59 schrod
+Adapted to recent changes of \cweave{} (of April 93):
+
+Main sections have a group level, represented in the table of
+contents. This changed the complete implementation of section tags.
+
+New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+|\CwebCombinedOp|.
+
+\Revision 1.1 1993/04/09 15:00:37 schrod
+Initial revision
+
+\end{rcslog}
+
+
+
+\end{document}
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+Local Variables:
+mode: LaTeX
+TeX-brace-indent-level: 4
+indent-tabs-mode: t
+TeX-auto-untabify: nil
+TeX-auto-regexp-list: LaTeX-auto-regexp-list
+compile-command: "make cwebbase.tex"
+End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cwebparts.doc b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebparts.doc
new file mode 100644
index 00000000000..3eec51882b1
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebparts.doc
@@ -0,0 +1,452 @@
+% $StyleId: cwebparts.doc,v 1.1 1995/09/16 17:01:59 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below.
+
+%
+% LaTeX package cwebparts
+% include parts of CWEB programs in LaTeX documents
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+%% FIXME: The user manual is not in cwebparts-user.tex. Most probably
+%% I'll make one user manual for the whole bundle; due to the usage of
+%% cwebbase class and package have lots in common.
+%%
+%% % If you have received this style file without the user manual (in the
+%% % file cwebparts-user.tex or the respective DVI file), it's incomplete and
+%% % near to useless. If it was given to you as something that you shall
+%% % use as an author -- complain bitterly to your provider. You need the
+%% % documentation and you have a right on it! (Below you can find info
+%% % where to get the reference version.)
+
+
+
+
+\documentclass{progltx}
+
+\usepackage{cweb-doc} % document-specific markup
+\usepackage{fullpage}
+
+
+\RCS $StyleRevision: 1.1 $
+\RCS $StyleDate: 1995/09/16 17:01:59 $
+
+
+\begin{document}
+
+
+\title{The \texttt{cwebparts} Package\\
+ {\large (Implementation)}%
+ }
+\author{% % LaTeX does not discard unnecessary glue...
+ Joachim Schrod%
+ \thanks{%
+ \protect\raggedright
+ TU~Darmstadt, Computer Science Department, WG Systems Programming,
+ Alexanderstr.~10, D-64283~Darmstadt, Germany.
+ Email: \texttt{schrod@iti.informatik.th-darmstadt.de}.%
+ }%
+ }
+\date{%
+ Revision \RCSStyleRevision\\
+ (as of \RCSStyleDate)%
+ }
+
+\maketitle
+
+
+% doesn't work with progltx yet
+%\tableofcontents
+
+
+
+\chap What's this style for?.
+
+This style supports inclusion of \cweb{} program parts in \LaTeX{}
+documents. For that it provides the environment |cweb| where \cweave{}
+output may be stuffed in.
+
+A more versatile explanation is available in the user manual of the
+\cweb{} bundle; this package is part of that bundle.
+%
+\begin{fixme}
+ Read: ``should be available'' -- there is no bundle user manual
+ (yet). |cweb-user.tex| will be enhanced after first testing of this
+ code.
+\end{fixme}
+
+
+\sect The internal interface is also not described in this source,
+refer to the configuration documentation for this bundle.
+%
+\begin{fixme}
+ Even worse: I don't even have written the internal specs for the
+ \cls{cweb} class, not to speak of this package.
+\end{fixme}
+
+
+
+
+\chap Implementation.
+
+Most of the implementation is made available by the module
+\pkg{cwebbase}. We have to supply macros it demands from a user, and
+have to deliver the |cweb| environment.
+
+
+\sect This module reserves the namespace |cweb|. (This implies that it
+cannot be used with the \cls{cweb} class. But you don't want to do
+that anyhow. :--)
+
+\beginprog
+\ifx \cweb@loaded\undefined
+ \def\cweb@loaded{$StyleRevision: 1.1 $}
+\else
+ \PackageWarningNoLine{cwebparts}%
+ {Some other package already uses namespace `cweb'}
+\fi
+\endprog
+
+
+\sect Let's identify this package.
+
+\noindent The code below is explained in the implementation
+documentation of the |rcs| package.
+
+\beginprog
+\begingroup
+ \def\RCSPackage#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesPackage{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSPackage{cwebparts}{include CWEB program parts in LaTeX documents}
+ $StyleDate: 1995/09/16 17:01:59 $: 9999/00/00 \endRCS
+ $StyleRevision: 1.1 $: 0.0 \endRCS
+\endgroup
+\endprog
+
+
+\sect This module is supported. Send bug reports, comments and
+repairs.
+
+The reference version may be retrieved via anonymous ftp from
+|ftp.th-darmstadt.de| [130.83.47.112], directory |pub/tex/latex|. It's
+placed there as a gzipped tar file. (The information on the IP~number
+is dated September~13,~1995.
+It might have changed, although this is very unlikely. Use your
+friendly nameserver.)
+
+
+\sect This is freely distributable software; you can redistribute it
+and/or modify it under the terms of the GNU General Public License as
+published by the Free Software Foundation; either version~2 of the
+License, or (at your option) any later version.
+
+This software is distributed in the hope that it will be useful, but
+\textbf{without any warranty}; without even the implied warranty of
+\textbf{merchantability} or \textbf{fitness for a particular purpose}.
+See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License in
+the file |License| along with this package; if not, write to the Free
+Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+\sect Before we start we declare some shorthands for category codes.
+By declaring the underscore~`(|_|)' as letter we can use it in our
+macros. (I agree with \textsc{D.~Knuth} that
+|\identifier_several_words_long| is more readable than
+|\IdentifierSeveralWordsLong| and in every case better than |\p@@@s|.)
+
+As this is a \LaTeX{} style file the at sign is a letter anyhow; so
+we can use the ``private'' \LaTeX{} macros; and with the underscore we
+can make our own macros more readable. But as we have to restore this
+category code at the end of this macro file we store its former value
+in the control sequence |\CatUsCode|. This method is better than to
+use a group because not all macros have to be defined global this way.
+
+Since somebody might use more styles from me, this cseqs might be
+defined already.
+
+\beginprog
+\ifx \CatEscape\undefined
+ \chardef\CatEscape=0
+ \chardef\CatOpen=1
+ \chardef\CatClose=2
+ \chardef\CatIgnore=9
+ \chardef\CatLetter=11
+ \chardef\CatOther=12
+ \chardef\CatActive=13 % is defined in Plain already
+
+ \chardef\CatUsCode=\catcode`\_
+\fi
+
+\catcode`\_=\CatLetter % top level macro file
+\endprog
+
+
+
+
+\chap Typesetting \cweave{} output.
+
+The module \pkg{cwebbase} handles that. It demands some cseqs to be
+defined before it may be imported.
+%
+\begin{fixme}
+ These cseqs should actually be implemented as package options. Have
+ to extract the \pkg{keyvalx} code from the \cls{cweb} class first.
+ Now I fix the usage I need for our Literate Programming book.
+\end{fixme}
+
+\begin{fixme}
+ The options should not only be available as package options, but
+ also on a per environment base. We might to typeset one part with
+ flat and another one with hierarchic structure. Or we might to
+ suppress format directives only in some parts, while other parts
+ feature them. But this needs a change in \pkg{cwebbase}, there code
+ is \emph{defined} depending on the cseqs below. It must be evaluated
+ differently to support the stuff below. OTOH, this will cost
+ performance -- is it worth it?
+\end{fixme}
+
+\beginprog
+\def\cweb@structure{0 } % hierarchic structure
+\let\if@cweb@suppress@changehints@=\iffalse % no suppression whatsoever
+\let\if@cweb@suppress@unchanged@=\iffalse
+\let\if@cweb@suppress@format@=\iffalse
+
+\input{cwebbase}
+\endprog
+
+
+\sect Now we can typeset pieces of \cweave{} output -- in principle.
+If these pieces have sections in them, they'll write entries to the
+|TOC| file. These entries will either mess up the structure of the
+document's table of contents or lead to \TeX{} error messages. That's
+because either toc entries for section types |part|, |chapter|, or
+|section| are written (causing the conflict); or toc entries for
+section types |chunk| and `|section|$i$' are written (being
+undefined).
+
+The best way seems to stop the included source to add anything to the
+table of contents or to do any mark processing. For that, we provide a
+macro |\CwebNoLists| that will locally redefine the \LaTeX{} cseqs
+that cause such actions to happen. That macro will be used at the
+start of the environment.
+
+\beginprog
+\def\CwebNoLists{%
+ \let\addtocontents\@gobbletwo
+ \let\mark\@gobble
+%% \let\markboth\@gobbletwo
+%% \let\markright\@gobble
+ }
+\endprog
+
+
+\sect Furthermore, section divisions should not start a new page.
+%
+\begin{fixme}
+ Currently, sections in the included parts are typeset exactly the
+ same way as they're typeset in the main document. That should be
+ different -- but how do I change it in a general way?
+\end{fixme}
+
+\beginprog
+\let\CwebEjectSection\@gobble
+\endprog
+
+
+\sect OK, what needs to be done for the environment? We need to save
+and restore counters that are changed globally by section commands.
+
+\beginprog
+\ifx \chapter\undefined
+ \let\cweb@save_c@chapter\@gobble
+ \let\c@chapter\@gobble
+\else
+ \newcount\cweb@save_c@chapter
+\fi
+\newcount\cweb@save_c@part
+\newcount\cweb@save_c@section
+
+\def\cweb@save_for_env{%
+ %% save values changed globally by cwebbase macros
+ \cweb@save_c@chapter\c@chapter
+ \cweb@save_c@part\c@part
+ \cweb@save_c@section\c@section
+ \let\cweb@save_thechapter\thechapter
+ \let\cweb@save_thesection\thesection
+ %% set up typesetting of CWEB source
+ \@nobreaktrue % don't add space from first chunk
+ }
+\def\cweb@restore_env{%
+ \global\c@chapter\cweb@save_c@chapter
+ \global\c@part\cweb@save_c@part
+ \global\c@section\cweb@save_c@section
+ \global\let\thechapter\cweb@save_thechapter
+ \global\let\thesection\cweb@save_thesection
+ }
+\endprog
+
+
+\sect For the moment, we use list spacing around a |cweb| environment.
+\LaTeX{} supplies the |trivlist| environment for such purposes.
+
+But the environment content itself must not be typeset in a list
+context. Two problems exist: There must be at least one |\item|, it
+will actually add the vertical space and must therefore not be
+discarded. And list environments redefine |\par|, but the
+\pkg{cwebbase} macros need a more `normal' definition.
+
+An item label is stored by |\item| in the |\everypar| register. The
+label is only typeset if |@inlabel| is true. The flag |@newlist| tells
+most other \LaTeX{} macros, that there was no item yet; typically
+they'll add an empty paragraph if necessary. As the content will
+start with a \LaTeX{} section division, or even with document markup,
+we don't want that empty paragraph and should skip the empty item
+label. (Often enough, due to the initialisation of |\everypar| in
+|\@startsection|, it will be thrown away anyhow). So we establish its
+postconditions, the next macro won't notice that the item hasn't been
+set actually.
+
+\begin{fixme}
+ Experimentally, we reduce the skip in front of a chunk to 1\,pc,
+ otherwise the distance between two chunks is far too wide compared
+ to other vertical distances in the text around the included stuff.
+\end{fixme}
+
+\beginprog
+\def\CwebBeginEnvLayout{%
+ %% use spacing of list, but switch off list handling afterwards
+ \trivlist
+ \@inlabelfalse % don't go in horizontal mode
+ \item\relax
+ \global\@newlistfalse
+ \global\@inlabelfalse\the\everypar
+ \@setpar{\@@par}%
+ \CwebChunkPreSkip 1pc plus 6pt minus 3pt
+ }
+\let\CwebEndEnvLayout=\endtrivlist
+\endprog
+
+
+\sect The environment gets trivial now. We supply a hook for a whole
+document and also environment-specific ones (as an optional
+environment argument).
+
+\beginprog
+\newenvironment{cweb}[1][]%
+ {%
+ \CwebBeginEnvLayout
+ \cweb@save_for_env
+ \CwebNoLists
+ \csname CwebEnvHook\endcsname
+ #1
+ }{%
+ \CwebEndEnvLayout
+ }
+\endprog
+
+
+\sect An a convenience tag to include a complete \cweave{} file. Since
+we want to test the typesetting of our included \cweb{} program alone,
+we allow but ignore most kinds of document markup. Many of them have
+two arguments, one is optional.
+
+In the preambel, only |\documentclass| and |\usepackage| may appear.
+Titlepage material is handled, table of contents is discarded. The
+document environment does nothing, i.e., unlike the \cls{cweb} class,
+material behind |\end{document}| is not ignored.
+
+An |\end| tag is inserted by \cweave{} and must be redefined, we don't
+end an environment there. We'll open a group and will use that tag to
+close it again, as we know that we'll have reached the end of our
+\cweb{} file. Of course, |\end| must not be redefined during
+processing the actual program source -- environments are valid there.
+The redefinition is best be triggered by material inserted from
+\cweave{} at the very end, e.g., from |\inx|. But |\inx| may not be
+there (when \cweave{} is called with the `|-x|' option), we'll use
+|\enddocument| to add a |\inx| tag for safety reason. If the user
+didn't add an |\end{document}| to his file and used `|-x|', we're
+going to loose.
+
+\TeX{}nical note: |\document| must not have the same binding as
+|\relax|, as this would be interpreted as a non-existing environment
+by \LaTeX{}. (That's due to the |\csname| lossage.)
+
+\TeX{}nical note: Evaluating |\inx| in |\enddocument| will not have
+any effect, we're inside a group. That evaluation must be postponed
+after the group has been closed.
+
+\beginprog
+\newcommand\cweb@doc_markup_gobble[2][]{}
+\newcommand*\cwebInput[1]{%
+ \begin{cweb}
+ [%
+ \begingroup
+ \let\documentclass\cweb@doc_markup_gobble
+ \let\usepackage\cweb@doc_markup_gobble
+ \let\title\@gobble
+ \let\author\@gobble
+ \let\date\@gobble
+ \let\maketitle\relax
+ \let\tableofcontents\relax
+ \let\document\empty
+ \def\inx{\let\end\endgroup}%
+ \let\fin\relax
+ \let\con\relax
+ \def\enddocument{\aftergroup\inx}%
+ ]%
+ \input{#1}%
+ \end{cweb}%
+ }
+\endprog
+
+
+
+
+\chap The end.
+
+Well, after all we're finished with this style. We must not forget to
+restore the underscore catcode.
+
+\beginprog
+\catcode`\_=\CatUsCode
+
+\endinput
+\endprog
+
+
+%\sect I would like to thank those who helped me to improve this style.
+% In particular, XXX provided XXXsubstantial parts of the code.
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+\begin{rcslog}
+$StyleLog: cwebparts.doc,v $
+\Revision 1.1 1995/09/16 17:01:59 schrod
+Add support for inclusion of \cweb{} files in \LaTeX{} documents.
+
+\end{rcslog}
+
+
+
+\end{document}
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+Local Variables:
+mode: LaTeX
+TeX-brace-indent-level: 4
+indent-tabs-mode: t
+TeX-auto-untabify: nil
+TeX-auto-regexp-list: LaTeX-auto-regexp-list
+compile-command: "make cwebparts.sty"
+End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/cwebx.sty b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebx.sty
new file mode 100644
index 00000000000..d20db59b171
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/cwebx.sty
@@ -0,0 +1,406 @@
+% $Id: cwebx.sty,v 3.1 1995/11/21 00:42:49 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% LaTeX2e package cwebx
+% extra stuff for cweb bundle, to be added there sometimes
+%
+% [LaTeX2e]
+% (history at end)
+
+
+% Identification. The code below is explained in the implementation
+% documentation of the rcs package.
+
+\begingroup
+ \def\RCSPackage#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesPackage{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSPackage{cwebx}{extra stuff for cweb bundle}
+ $Date: 1995/11/21 00:42:49 $: 9999/00/00 \endRCS
+ $Revision: 3.1 $: 0.0 \endRCS
+\endgroup
+
+
+%
+% What's this package for?
+%
+% Additional macros for cweb.cls or cwebpart.sty, to be added there
+% sometimes.
+%
+
+
+% requires that cweb base is loaded
+\ifx \cwbb@loaded\undefined
+ \PackageError{cwebx}{cweb base is not loaded}
+ {This package requires the usage of either cweb.cls or
+ cwebparts.sty.}
+\fi
+
+
+% This style is supported. Send bug reports, comments and repairs.
+%
+% The reference version may be retrieved via anonymous ftp from
+% |ftp.th-darmstadt.de| [130.83.47.112], directory |pub/tex/latex|. It's
+% placed there as a gzipped tar file. (The information on the
+% IP~number is dated 27 Sep 95. It might have changed, also
+% this is very unlikely. Use your friendly nameserver.)
+
+
+
+\catcode`\_=\CatLetter
+
+
+
+%%% ======================================================================
+
+
+%%
+%% Errors in cwebbase.doc
+%%
+
+
+% As \4 follows \B always, the \kern\CwebIndentUnit to neutralize that
+% backspace must be made after run-in section, too. Between the chunk
+% marker and the program code start is 1em then (inserted by LaTeX
+% section start).
+
+% FIXME: \4 does _not_ follow \B all the time. Check when it is
+% issued. The code below was introduced to repair a problem that gets
+% visible if one uses \cwebNoChunk with an indentation unit of 2em,
+% obviously I have to look for a better repair possibility.
+
+\CheckCommand*\B{%
+ \cwbb@program
+ \cwbb@indent 3\CwebIndentUnit \hangindent\cwbb@indent
+ \ifvmode
+ \if@noskipsec
+ \indent % add an empty documentation part
+ \else
+ \noindent \kern\CwebIndentUnit
+ \fi
+ \fi
+ \CwebNumberProgramPart
+ }
+
+\renewcommand\B{%
+ \cwbb@program
+ \cwbb@indent 3\CwebIndentUnit \hangindent\cwbb@indent
+ \ifvmode
+ \if@noskipsec
+ \indent % add an empty documentation part
+ \else
+ \noindent
+ \fi
+ \fi
+ \kern\CwebIndentUnit
+ \CwebNumberProgramPart
+ }
+
+
+%%% ------------------------------------------------------------
+
+
+%%
+%% Macros from cwebparts.doc, to be moved to cwebbase.doc.
+%%
+
+% They are partly repaired and improved, where noted with `CHANGE'.
+
+
+% \sect Now we can typeset pieces of \cweave{} output -- in principle.
+% If these pieces have sections in them, they'll write entries to the
+% |TOC| file. These entries will either mess up the structure of the
+% document's table of contents or lead to \TeX{} error messages. That's
+% because either toc entries for section types |part|, |chapter|, or
+% |section| are written (causing the conflict); or toc entries for
+% section types |chunk| and `|section|$i$' are written (being
+% undefined).
+
+% The best way seems to stop the included source to add anything to the
+% table of contents or to do any mark processing. For that, we provide a
+% macro |\CwebNoLists| that will locally redefine the \LaTeX{} cseqs
+% that cause such actions to happen. That macro will be used at the
+% start of the environment.
+
+% CHANGE: Discard try code that redefines |\markboth| and |\markright|.
+
+% \beginprog
+\def\CwebNoLists{%
+ \let\addtocontents\@gobbletwo
+ \let\mark\@gobble
+ }
+% \endprog
+
+
+% \sect Furthermore, section divisions should not start a new page.
+% %
+% \begin{fixme}
+% Currently, sections in the included parts are typeset exactly the
+% same way as they're typeset in the main document. That should be
+% different -- but how do I change it in a general way?
+% \end{fixme}
+%
+% CHANGE: Moved to Cweb environment.
+
+
+% CHANGE: Improved question.
+%
+% \sect OK, what gets destroyed by the environment? We need to save
+% and restore counters that are changed globally by section commands.
+
+% \beginprog
+\ifx \chapter\undefined
+ \let\cweb@save_c@chapter\@gobble
+ \let\c@chapter\@gobble
+\else
+ \newcount\cweb@save_c@chapter
+\fi
+\newcount\cweb@save_c@part
+\newcount\cweb@save_c@section
+\newcount\cweb@save_c@chunk
+
+\def\cweb@save_for_env{%
+ %% save values changed globally by cwebbase macros
+ \cweb@save_c@chapter\c@chapter
+ \cweb@save_c@part\c@part
+ \cweb@save_c@section\c@section
+ \cweb@save_c@chunk\c@chunk
+ \let\cweb@save_thechapter\thechapter
+ \let\cweb@save_thesection\thesection
+ %% set up typesetting of CWEB source
+ \@nobreaktrue % don't add space from first chunk
+ }
+\def\cweb@restore_env{%
+ \global\c@chapter\cweb@save_c@chapter
+ \global\c@part\cweb@save_c@part
+ \global\c@section\cweb@save_c@section
+ \global\c@chunk\cweb@save_c@chunk
+ \global\let\thechapter\cweb@save_thechapter
+ \global\let\thesection\cweb@save_thesection
+ }
+% \endprog
+
+
+% \sect For the moment, we use list spacing around a |cweb| environment.
+% \LaTeX{} supplies the |trivlist| environment for such purposes.
+
+% But the environment content itself must not be typeset in a list
+% context. Two problems exist: There must be at least one |\item|, it
+% will actually add the vertical space and must therefore not be
+% discarded. And list environments redefine |\par|, but the
+% \pkg{cwebbase} macros need a more `normal' definition.
+
+% An item label is stored by |\item| in the |\everypar| register. The
+% label is only typeset if |@inlabel| is true. The flag |@newlist| tells
+% most other \LaTeX{} macros, that there was no item yet; typically
+% they'll add an empty paragraph if necessary. As the content will
+% start with a \LaTeX{} section division, or even with document markup,
+% we don't want that empty paragraph and should skip the empty item
+% label. (Often enough, due to the initialisation of |\everypar| in
+% |\@startsection|, it will be thrown away anyhow). So we establish its
+% postconditions, the next macro won't notice that the item hasn't been
+% set actually.
+
+% CHANGE: Added \CwebNoLists, as I consider it part of the env layout.
+
+% CHANGE: Added paragraph below.
+%
+% Furthermore, section divisions should not start a new page.
+% %
+% \begin{fixme}
+% Currently, sections in the included parts are typeset exactly the
+% same way as they're typeset in the main document. That should be
+% different -- but how do I change it in a general way?
+% \end{fixme}
+
+% \begin{fixme}
+% Experimentally, we reduce the skip in front of a chunk to 1\,pc,
+% otherwise the distance between two chunks is far too wide compared
+% to other vertical distances in the text around the included stuff.
+% \end{fixme}
+
+% \beginprog
+\def\CwebBeginEnvLayout{%
+ %% use spacing of list, but switch off list handling afterwards
+ \trivlist
+ \@inlabelfalse % don't go in horizontal mode
+ \item\relax
+ \global\@newlistfalse
+ \global\@inlabelfalse\the\everypar
+ \@setpar{\@@par}%
+ %% now change cweb base layout, meant for classes
+ \CwebNoLists
+ \let\CwebEjectSection\@gobble
+ \CwebChunkPreSkip 1pc plus 6pt minus 3pt
+ }
+\let\CwebEndEnvLayout=\endtrivlist
+% \endprog
+
+
+% \sect The environment gets trivial now. We supply a hook for a whole
+% document and also environment-specific ones (as an optional
+% environment argument).
+
+% CHANGE: Renamed to `Cweb'. Lusers should actually not call that env
+% directly, it's too dangerous.
+
+% CHANGE: Added forgotten evaluation of \cweb@restore_env.
+
+% \beginprog
+\newenvironment{Cweb}[1][]%
+ {%
+ \CwebBeginEnvLayout
+ \cweb@save_for_env
+ \csname CwebEnvHook\endcsname
+ #1
+ }{%
+ \cweb@restore_env
+ \CwebEndEnvLayout
+ }
+% \endprog
+
+
+% CHANGE: Have to improve the introduction of |cwebInput|. Now it has
+% an additional optional parameter, where instance specific cseqs may
+% be passed (like the cweb env).
+
+% \sect An a convenience tag to include a complete \cweave{} file. Since
+% we want to test the typesetting of our included \cweb{} program alone,
+% we allow but ignore most kinds of document markup. Many of them have
+% two arguments, one is optional.
+
+% In the preambel, only |\documentclass| and |\usepackage| may appear.
+% Titlepage material is handled, table of contents is discarded. The
+% document environment does nothing, i.e., unlike the \cls{cweb} class,
+% material behind |\end{document}| is not ignored.
+
+% An |\end| tag is inserted by \cweave{} and must be redefined, we don't
+% end an environment there. We'll open a group and will use that tag to
+% close it again, as we know that we'll have reached the end of our
+% \cweb{} file. Of course, |\end| must not be redefined during
+% processing the actual program source -- environments are valid there.
+% The redefinition is best be triggered by material inserted from
+% \cweave{} at the very end, e.g., from |\inx|. But |\inx| may not be
+% there (when \cweave{} is called with the `|-x|' option), we'll use
+% |\enddocument| to add a |\inx| tag for safety reason. If the user
+% didn't add an |\end{document}| to his file and used `|-x|', we're
+% going to loose.
+
+% CHANGE: Added handling of empty chunk. (@noskipsec was set after the
+% environment, what messed up later invocations of program code if
+% used with \cwebCodeExample.)
+
+% \TeX{}nical note: |\document| must not have the same binding as
+% |\relax|, as this would be interpreted as a non-existing environment
+% by \LaTeX{}. (That's due to the |\csname| lossage.)
+
+% \TeX{}nical note: Evaluating |\inx| in |\enddocument| will not have
+% any effect, we're inside a group. That evaluation must be postponed
+% after the group has been closed.
+
+% \beginprog
+\providecommand\cweb@doc_markup_gobble[2][]{}
+\let\cwebInput\undefined
+\newcommand*\cwebInput[2][]{%
+ \begin{Cweb}
+ [%
+ \begingroup
+ \let\documentclass\cweb@doc_markup_gobble
+ \let\usepackage\cweb@doc_markup_gobble
+ \let\title\@gobble
+ \let\author\@gobble
+ \let\date\@gobble
+ \let\maketitle\relax
+ \let\tableofcontents\relax
+ \let\document\empty
+ \def\inx{\let\end\endgroup}%
+ \let\fin\relax
+ \let\con\relax
+ \def\enddocument{%
+ \if@noskipsec % empty chunk
+ \global\@noskipsecfalse
+ \global\everypar{}%
+ \fi
+ \aftergroup\inx
+ }%
+ #1%
+ ]%
+ \input{#2}%
+ \end{Cweb}%
+ }
+% \endprog
+
+
+%%% ------------------------------------------------------------
+
+
+%%
+%% New macros for cwebbase.doc.
+%%
+
+
+% For production of code examples in CWEB documents. It's expected
+% that the document consists of only \emph{one} chunk that has no
+% documentation. Just the code is of interest. Neither chunk markers,
+% nor chunk numbers, nor refinement numbers are typeset.
+%
+% But one may change that with the optional parameter.
+
+\providecommand*\cwebCodeExample[2][]{%
+ \cwebInput[%
+ \def\CwebChunkHeading##1{\kern -1em}%
+ \let\cwbb@marginal_chunkno\relax
+ \let\CwebRefNumber\@gobble
+ \CwebChunkPreSkip\z@skip
+ #1%
+ ]{#2}%
+ }
+
+
+% Often one has an explanation for more than one refinement. (E.g.,
+% when there's needed a definition and a declaration.) Then the
+% documentation of the follow-up refinements shall consist solely of
+% |\cwebNoChunk|. Then no chunk marker is set, and the refinement
+% is typeset nearer to the previous refinement. Optically, we create
+% one chunk with many refinements this way.
+
+% \TeX{}nical note: No chunk marker is created by redefining macro
+% |\@svsechd| that creates header for run-in \LaTeX{} section. Every
+% header is followed by 1\,em space, we have to annulate that skip.
+% (The refinement shall start at the beginning of the line.) One must
+% not use |\hskip| to backup, an |\unskip| comes after the header and
+% would discard that skip.
+
+\providecommand\cwebNoChunk{%
+ \vskip -\lastskip \bigskip % discard LaTeX's inter-section skip
+ \def\@svsechd{\kern -1em}%
+ }
+
+
+
+
+\catcode`\_=\CatUsCode
+\endinput
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: cwebx.sty,v $
+% Revision 3.1 1995/11/21 00:42:49 schrod
+% version 1.1
+%
+
+
+
+%%%------------------------------------------------------------
+
+% Local Variables:
+% mode: LaTeX
+% TeX-brace-indent-level: 4
+% indent-tabs-mode: t
+% TeX-auto-untabify: nil
+% page-delimiter: "^%%%"
+% End:
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.dtx b/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.dtx
new file mode 100644
index 00000000000..6873fde7110
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.dtx
@@ -0,0 +1,347 @@
+%% $ITI: keyvald.dtx,v 1.2 1995/08/25 19:00:31 schrod Exp $
+%%------------------------------------------------------------
+%% (history at end)
+%
+% \iffalse
+%% File: keyvald.dtx; changed version of
+%% File: keyval.dtx Copyright (C) 1993-1994 David Carlisle
+%
+%<*dtx>
+ \ProvidesFile{keyvald.dtx}
+%</dtx>
+%<*!plain>
+%<package&!plain>\NeedsTeXFormat{LaTeX2e}
+%<package&!plain>\ProvidesPackage{keyvald}
+%<driver> \ProvidesFile{keyvald.drv}
+% \fi
+% \ProvidesFile{keyvald.dtx}
+ [1994/08/23 v1.08.1 key=value parser (DPC) w/ defaults (js)]
+%
+% \iffalse
+%</!plain>
+%<*driver>
+\documentclass{ltxdoc}
+\usepackage{keyvald}
+\begin{document}
+ \DocInput{keyvald.dtx}
+\end{document}
+%</driver>
+% \fi
+%
+% \GetFileInfo{keyvald.dtx}
+% \title{The \textsf{keyvald} package\thanks{This file
+% has version number \fileversion, last
+% revised \filedate.}}
+% \author{David Carlisle\\carlisle@cs.man.ac.uk
+% \and
+% changed by\\Joachim Schrod\\
+% \texttt{<schrod@iti.informatik.th-darmstadt.de>}
+% }
+% \date{\filedate}
+% \maketitle
+%
+% \begin{abstract}
+% A \LaTeX\ package implementing a system allowing the setting
+% of parameters (or `named arguments' with a
+% \meta{key}${}={}$\meta{value} syntax.
+%
+% Eg: |\foo[height=3in, shadow = true ]{bar}|
+% \end{abstract}
+%
+% \CheckSum{110}
+%
+% \changes{v1.03}
+% {1993/10/13}{Initial version}
+% \changes{v1.04}{1993/11/15}
+% {Upgrade after comments from Timothy van Zandt}
+% \changes{v1.05}{1993/11/17}
+% {Further small improvements}
+% \changes{v1.06}{1994/02/01}
+% {Update to LaTeX2e}
+% \changes{v1.07}{1994/03/15}
+% {New style ltxdoc}
+% \changes{v1.08}{1994/09/12}
+% {Improve docstrip handling}
+% \changes{v1.08.1}{1995/08/23}
+% {Default keys}
+%
+%
+% This package implements a system of defining and using sets of
+% parameters, which are set using the syntax \meta{key}=\meta{value}.
+%
+% For each keyword in such a set, there exists a function which is
+% called whenever the parameter appears in a parameter list. For
+% instance if the set |dpc| is to have the keyword |scale| then I
+% would define.\\
+% | \define@key{dpc}{scale}{scale ({\tt\string#1})\\}|\\
+% The first argument of |\define@key| is the set of keywords being
+% used, the second is the keyword, and the third is the function to
+% call. This function will be given as |#1| the \meta{value} specified
+% by the user.
+%
+% Normally it is an error to omit the `=\meta{value}' however if an
+% optional \meta{value} is supplied when the keyword is defined, then
+% just the keyword need be supplied.\\
+% |\define@key{dpc}{clip}[true]{...}|\\
+% For `|clip|' you can go `|clip = true|' or `|clip = false|' or
+% just `|clip|', which is the same as `|clip = true|'
+%
+% To use these keywords, just call `|\setkeys|' with a comma
+% separated list of settings, each of the form
+% \meta{key}=\meta{value}, or just \meta{key}. Any white space around
+% the `|=|' and `|,|' is ignored.
+%
+% As the \meta{key} is passed as a macro argument, if it consists
+% entirely of a |{ }| group, the outer braces are stripped off. Thus
+% |,key=foo,| and |key=,{foo},| are equivalent. This fact enables one to
+% `hide' any commas or equals signs that must appear in the value. i.e.\
+% in |foo={1,2,3},bar=4|, |foo| gets the value |1,2,3|, the comma after
+% |1| does not terminate the keyval pair, as it is `hidden' by the
+% braces.
+%
+% Empty entries, with nothing between the commas, are silently ignored.
+% This means that it is not an error to have a comma after the last
+% term, or before the first.
+%
+% \section{Example}
+%
+% We may extend the examples above to give a `fake' graphics
+% inclusion macro, with a syntax similar to that used in the psfig
+% macros.
+%
+% \makeatletter
+% \def\dpcgraphics{\@ifnextchar[\@dpcgraphics{\@dpcgraphics[]}}
+%
+% |\dpcgraphics| has one optional argument which is passed through
+% |\setkeys|, and one mandatory argument, the filename. It actually
+% just typesets its arguments, for demonstration.
+%
+% \def\@dpcgraphics[#1]#2{{\setkeys{dpc}{#1}INPUT: #2}}%
+%
+% \define@key{dpc}{scale}{scale ({\tt\string#1\relax})\\}
+% \define@key{dpc}{height}{height ({\tt#1})\\}
+% \define@key{dpc}{width}{width ({\tt#1})\\}
+% \define@key{dpc}{bb}{bounding box ({\tt#1})\\}
+% \define@key{dpc}{clip}[true]{clip ({\tt\string#1\relax})\\}
+% \makeatother
+%
+% The declared keys are: |scale|, |height|, |width|, |bb|,
+% and |clip|. Except for the last, they must all be given a value if
+% used.
+%
+% Note how in the following, any white space arround |=| or |,| is
+% ignored, as are the `empty' arguments caused by extra commas. Note
+% also that each macro receives {\em exactly\/} the tokens that you
+% specify as arguments, no premature expansion is done.
+%
+% \begin{verbatim}
+% \def\dpcgraphics{\@ifnextchar[\@dpcgraphics{\@dpcgraphics[]}}
+% \def\@dpcgraphics[#1]#2{{\setkeys{dpc}{#1}INPUT: #2}}
+%
+% \define@key{dpc}{scale}{scale ({\tt\string#1\relax})\\}
+% \define@key{dpc}{clip}[true]{clip ({\tt\string#1\relax})\\}
+% \end{verbatim}
+%
+% \begin{minipage}{.4\textwidth}
+% \begin{verbatim}
+% \def\scalemacro{9}
+% \dpcgraphics
+% [ height =4in, ,
+% width = 3in,
+% scale = \scalemacro,
+% bb = 20 20 300 400 ,
+% clip,
+% ]{aaa}
+% \end{verbatim}
+% \end{minipage}
+% \hfill
+% \begin{minipage}{.4\textwidth}
+% \def\scalemacro{9}
+%\dpcgraphics
+% [ height =4in, ,
+% width = 3in,
+% scale = \scalemacro ,
+% bb = 20 20 300 400 ,
+% clip ,
+% ]{aaa}
+% \end{minipage}
+%
+%
+% \section{The Internal Interface}
+% A declaration of the form:\\
+% |\define@key{family}{key}{...}|\\
+% Defines a macro |\KV@prefix@key| with one argument. When used in a
+% keyval list, the macro receives the value as its argument.
+%
+% A declaration of the form:\\
+% |\define@key{family}{key}[default]{...}|\\
+% Defines a macro |\KV@family@key| as above, however it also defines the
+% macro |\KV@family@key@default| as a macro with no arguments, and
+% definition\\
+% |\KV@family@key{default}|.
+%
+% Thus if macros are defined using |\define@key|, the use of a key with
+% no value \ldots|,foo,|\ldots\ is always equivalent to the use of the
+% key with some value, \ldots|,foo=default,|\ldots. However a package
+% writer may wish that the `default' behaviour for some key is not
+% directly equivalent to using that key with a value. (In particular, as
+% pointed out to me by Timothy Van Zandt, you may wish to omit error
+% checking on the default value as you know it is correct.) In these
+% cases one simply needs to define the two macros
+% |\KV@|\meta{family}|@key| and |\KV@|\meta{family}|@key@default|
+% directly using |\def| (or |\newcommand|). I do not supply a user
+% interface for this type of definition, but it is supported in the
+% sense that I will try to ensure that any future upgrades of this
+% package do not break styles making use of these `low level'
+% definitions.
+%
+% \StopEventually{}
+%
+% \section{The Macros}
+%
+% From version~1.05, all `internal' macros associated to keys have names
+% of the form:\\
+% |\KV@|\meta{family}|@|\meta{key} or
+% |\KV@|\meta{family}|@|\meta{key}|@|\meta{default}
+%
+% \begin{macrocode}
+%<*package>
+% \end{macrocode}
+%
+% \begin{macro}{\setkeys}
+% The top level macro. |#2| should be a comma separated values of the
+% form \meta{key} |=| \meta{value} or just simply \meta{key}.
+% The macro associated with this key in the `family' |#1| is called with
+% argument \meta{value}. The second form is only allowed if the key was
+% declared with a default value.
+% \begin{macrocode}
+\def\setkeys#1#2{%
+% \end{macrocode}
+% Save the `family' for later. Then begin acting on the comma
+% separated list.
+% \begin{macrocode}
+ \def\KV@prefix{KV@#1@}%
+ \KV@do#2,\relax,}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@do}
+% Iterate down the list of comma separated argument pairs.
+% \begin{macrocode}
+\def\KV@do#1,{%
+ \ifx\relax#1\empty\else
+ \KV@split#1==\relax
+ \expandafter\KV@do\fi}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@split}
+% Split up the keyword and value, and call the appropriate command.
+% This macro was slightly reorganised for version 1.04, after some
+% suggestions from Timothy Van Zandt.
+% \begin{macrocode}
+\def\KV@split#1=#2=#3\relax{%
+ \KV@@sp@def\KV@key{#1}%
+ \ifx\KV@key\@empty\else
+ \expandafter\let\expandafter\@tempc
+ \csname\KV@prefix\KV@key\endcsname
+ \ifx \@tempc\relax
+ \expandafter\let \expandafter\@tempc
+ \csname\KV@prefix*\endcsname
+ \def\@tempa{*}%
+ \else
+ \let\@tempa\KV@key
+ \fi
+ \ifx\@tempc\relax
+ \KV@err{\KV@key\space undefined}%
+ \else
+ \ifx\@empty#3\@empty
+ \KV@default
+ \else
+ \KV@@sp@def\@tempb{#2}%
+ \expandafter\@tempc\expandafter{\@tempb}\relax
+ \fi
+ \fi
+ \fi}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@default}
+% Run the default code, or raise an error.
+% \begin{macrocode}
+\def\KV@default{%
+ \expandafter\let\expandafter\@tempb
+ \csname\KV@prefix\@tempa @default\endcsname
+ \ifx\@tempb\relax
+ \KV@err{No value specified for \KV@key}%
+ \else
+ \@tempb\relax
+ \fi}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@err}
+% Error messages.
+% \begin{macrocode}
+\def\KV@err#1{\PackageError{keyvald}{#1}{}}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@@sp@def}
+% \begin{macro}{\KV@@sp@b}
+% \begin{macro}{\KV@@sp@c}
+% |\KV@@sp@def|\meta{cmd}\meta{token list} is like |\def|, except that
+% a space token at the beginning or end of \meta{token list} is
+% removed before making the assignment. \meta{token list} may not
+% contain the token |\@nil|, unless it is within a brace group.
+% The names of these commands were changed at version~1.05 to ensure
+% that they do not clash with `internal' macros in a key family `sp'.
+%
+% The following three lines are equivalent to\\
+% |\def\KV@@sp@def#1#2{\KV@@sp@b#2\@nil\@nil~\@nil\relax#1}|\\
+% Where |~| is an explicit space token.
+% \begin{macrocode}
+\def\@tempa#1{%
+\def\KV@@sp@def##1##2{\KV@@sp@b##2\@nil\@nil#1\@nil\relax##1}}
+\@tempa{ }
+\def\KV@@sp@b#1#2 \@nil{\KV@@sp@c#1#2}
+\def\KV@@sp@c#1\@nil#2\relax#3{\def#3{#1}}
+% \end{macrocode}
+% \end{macro}
+% \end{macro}
+% \end{macro}
+%
+% \begin{macro}{\define@key}
+% Define the command associated to the key |#2| in the family |#1|.
+% First looks for a default argument (the default value for the
+% key)
+% \begin{macrocode}
+\def\define@key#1#2{%
+ \@ifnextchar[{\KV@def{#1}{#2}}{\@namedef{KV@#1@#2}####1}}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\KV@def}
+% Make the definitions of the command, and the default value.
+% \begin{macrocode}
+\def\KV@def#1#2[#3]{%
+ \@namedef{KV@#1@#2@default\expandafter}\expandafter
+ {\csname KV@#1@#2\endcsname{#3}}%
+ \@namedef{KV@#1@#2}##1}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macrocode}
+%</package>
+% \end{macrocode}
+% \Finale
+%
+\endinput
+
+%------------------------------------------------------------
+%
+% $ITIlog: keyvald.dtx,v $
+% Revision 1.2 1995/08/25 19:00:31 schrod
+% Add enhanced version of keyval package, with ability to specify
+% default action for unknown keyword options.
+%
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.ins b/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.ins
new file mode 100644
index 00000000000..0fe873fd74f
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/keyvald.ins
@@ -0,0 +1,6 @@
+\def\batchfile{keyvald.ins}
+\input docstrip
+
+\keepsilent
+
+\generateFile{keyvald.sty}{f}{\from{keyvald.dtx}{package}}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/style/cweb-doc.el b/Master/texmf-dist/doc/latex/cweb-latex/src/style/cweb-doc.el
new file mode 100644
index 00000000000..eb801786499
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/style/cweb-doc.el
@@ -0,0 +1,22 @@
+;; $Id: cweb-doc.el,v 1.1 1995/08/29 15:15:30 schrod Exp $
+;;----------------------------------------------------------------------
+
+;;;
+;;; AUC-TeX configuration for cweb-doc.sty
+;;;
+
+(TeX-add-style-hook "cweb-doc"
+ (function
+ (lambda ()
+ (TeX-add-symbols "cweb" "ctangle" "cweave"
+ '("arg" 1)
+ '("<" TeX-arg-free ">")
+ '("cls" 1)
+ '("pkg" 1)
+ )
+ (LaTeX-add-environments "fixme" "cseqtab"
+ '("options" LaTeX-env-item)
+ )
+ (setq LaTeX-item-list (cons '("options" . LaTeX-item-argument)
+ LaTeX-item-list))
+ )))
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/Imakefile b/Master/texmf-dist/doc/latex/cweb-latex/src/test/Imakefile
new file mode 100644
index 00000000000..0d32d8301c3
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/Imakefile
@@ -0,0 +1,142 @@
+XCOMM $Id: Imakefile,v 1.9 1995/11/20 22:34:44 schrod Exp $
+/* ---------------------------------------------------------------------- */
+
+/*
+ * Imakefile for test of cweb bundle
+ *
+ * (history at end)
+ */
+
+
+/* needed modules */
+#define idx NO_IDX_CLEAN /* don't delete *.idx on make clean */
+#include <TeX.imk>
+#undef idx
+
+XCOMM ------------------------------------------------------------
+XCOMM continue in Imakefile
+
+.SUFFIXES: .tex .w
+
+.w.tex: @@\
+ if [ -r $*.ch ] ; \ @@\
+ then cweave $* $* ; \ @@\
+ else cweave $* ; \ @@\
+ fi
+
+
+regression:: clean
+ $(MAKE) all
+
+
+/* ============================================================
+ *
+ * tests for cweb class & cwebarray package
+ */
+
+TeXTarget(latex,minimal) /* minimal document */
+TeXTarget(latex,token) /* rendering of tokens */
+TeXTarget(latex,refname) /* refinement names, filename ref. */
+TeXTarget(latex,modes) /* cweb FSA */
+TeXTarget(latex,section) /* hierarchic document structure */
+TeXTarget(latex,flat) /* flat document structure */
+TeXTarget(latex,report) /* report as base class */
+TeXTarget(latex,nolists) /* nolists via CWEAVE option -x */
+TeXTarget(latex,newif) /* idiosyncrasy of old LateX version */
+
+TeXTarget(latex,enddocbegin) /* skip stuff after \end{document} */
+TeXTarget(latex,vbar) /* ruled table [cwebarray] */
+TeXTarget(latex,titlepage) /* titlepage option */
+
+TeXTarget(latex,language-german) /* language option german */
+TeXTarget(latex,german) /* babel-compatible option german */
+
+TeXTarget(latex,change) /* changefile-related output */
+TeXTarget(latex,sup-changes) /* changefile-related suppression */
+
+TeXTarget(latex,sup-lists) /* suppression of index & reflist */
+TeXTarget(latex,sup-format) /* suppression of format directives */
+
+TeXTarget(latex '\nonstopmode \input',badend) /* unknown \end expansion */
+TeXTarget(latex '\nonstopmode \input',badopts) /* unknown & illegal options */
+
+
+/*
+ * Special (additional) rules for targets above
+ */
+
+nolists.tex: nolists.w
+ cweave -x nolists
+
+change.tex: change.ch
+
+sup-changes.tex: sup-changes.w change.ch
+ cweave sup-changes change
+
+
+
+/* ============================================================
+ *
+ * tests for packages
+ */
+
+/* cwebparts */
+
+TeXTarget(latex,parts) /* CWEB parts in LaTeX document */
+
+parts.dvi: parts-code.tex
+
+parts-code.tex: parts-code.w
+ cweave -x parts-code
+
+
+
+/* ============================================================
+ *
+ * Tests for stuff that is not strictly part of the cweb bundle.
+ */
+
+TeXTarget(latex,keyvald) /* keyword-value options w/ defaults */
+
+
+realclean:: clean
+ rm -f *.idx *.scn *.tex
+
+
+
+/* ============================================================
+ *
+ * $Log: Imakefile,v $
+ * Revision 1.9 1995/11/20 22:34:44 schrod
+ * Test language support, actually for German.
+ *
+ * Revision 1.8 1995/11/07 18:28:52 schrod
+ * Add test newif, for old LaTeX version.
+ *
+ * Revision 1.7 1995/09/17 14:21:36 schrod
+ * Test package cwbl-german.
+ *
+ * Revision 1.6 1995/09/16 17:02:09 schrod
+ * Test cwebparts.sty.
+ *
+ * Revision 1.5 1995/09/12 22:32:21 schrod
+ * Add test case for output of CWEAVE with `-x' option.
+ *
+ * Revision 1.4 1995/08/29 15:22:14 schrod
+ * Add test of format directive suppression.
+ *
+ * Revision 1.3 1995/08/27 13:29:38 schrod
+ * Tests of options & changefile-related suppression.
+ *
+ * Revision 1.2 1995/08/25 19:12:39 schrod
+ * Test implementation of hierarchic and flat structure.
+ *
+ * Revision 1.1 1995/08/08 00:14:46 schrod
+ * Updated to \LaTeXe{}, the |cweb| style is now a document class.
+ *
+
+ * ------------------------------------------------------------
+ * Local Variables:
+ * mode: Indented-Text
+ * End:
+ */
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/badend.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/badend.w
new file mode 100644
index 00000000000..51060a65cfd
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/badend.w
@@ -0,0 +1,21 @@
+% $Id: badend.w,v 1.3 1995/08/08 00:14:46 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests unknown \end macro
+% assumes "h" and "x" user reaction
+% note: produces no output!
+
+\makeatletter
+\def\end#1{\csname end#1\endcsname\@@checkend{#1}%
+ \expandafter\endgroup \if@@endpe \@@doendpe \fi
+\relax % <-- added
+ \if@@ignore \global\@@ignorefalse \ignorespaces\fi}
+\makeatother
+
+\documentclass{cweb}
+
+\begin{document}
+
+@ Test
+
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/badopts.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/badopts.w
new file mode 100644
index 00000000000..5fff5f7add7
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/badopts.w
@@ -0,0 +1,13 @@
+% $Id: badopts.w,v 1.1 1995/08/27 13:29:39 schrod Exp $
+%------------------------------------------------------------
+
+% test bad options
+
+\documentclass[undef,structure,structure=chaos]{cweb}
+
+\begin{document}
+
+Text.
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.ch b/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.ch
new file mode 100644
index 00000000000..2796a280e68
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.ch
@@ -0,0 +1,5 @@
+@x
+change here
+@y
+did the change
+@z
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.w
new file mode 100644
index 00000000000..d787065b933
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/change.w
@@ -0,0 +1,20 @@
+% $Id: change.w,v 1.4 1995/08/27 13:29:40 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests changefile output
+
+\documentclass{cweb}
+
+\begin{document}
+
+\title{Test for changefiles}
+\author{Coogar}
+\maketitle
+
+@ test
+
+@
+change here
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/enddocbegin.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/enddocbegin.w
new file mode 100644
index 00000000000..64d77368e1f
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/enddocbegin.w
@@ -0,0 +1,16 @@
+% $Id: enddocbegin.w,v 1.3 1995/08/08 00:14:47 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests skip of stuff after \end{document}
+
+\documentclass{cweb}
+
+\begin{document}
+
+@ Test.
+
+\end{document}
+
+\begin{itemize}
+\item an item
+\end{itemize}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/flat.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/flat.w
new file mode 100644
index 00000000000..993193971f3
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/flat.w
@@ -0,0 +1,47 @@
+% $Id: flat.w,v 1.1 1995/08/25 19:12:40 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests section levels and references for flat structure documents.
+
+\documentclass[structure=flat]{cweb}
+
+\begin{document}
+
+\tableofcontents
+
+\newpage
+
+@** s title.
+\label{main-sec}
+
+Some text.
+
+@* e title.
+
+Some text.
+And some references: Part~\ref{main-sec} and section~\ref{normal-sec}.
+
+@*0 0 title.
+
+Some text.
+
+@*1 1 title.
+
+Some text.
+
+@*2 2 title.
+
+Some text.
+
+@ \label{normal-sec}
+Doc.
+
+@d def foo
+@f def int
+
+@c
+for ( int i=1; i<10; i++ )
+ @< do something @>
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/german.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/german.w
new file mode 100644
index 00000000000..dd6f5d64bbe
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/german.w
@@ -0,0 +1,33 @@
+% $Id: german.w,v 1.3 1995/11/20 22:34:44 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests german language support
+
+
+% check if option after german works as well
+\documentclass[german,suppress=index]{cweb}
+
+\usepackage{german}
+
+\begin{document}
+
+\title{Test der \texttt{german} Option}
+\author{Puma}
+\maketitle
+
+@ Aktives Anf\"uhrungszeichen im String.
+@c
+#include "foo.h"
+
+@ @( foo.h@>=
+@< etwas |code| @>@;
+@< mehr Code @>
+
+@ @< etwas ...@>=
+for ( int i=1; i<10; i++ )
+ @< mehr... @>
+
+@ @< mehr... @>=
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.ch b/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.ch
new file mode 100644
index 00000000000..1e41e64e1e3
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.ch
@@ -0,0 +1,5 @@
+@x
+@< mehr Code @>
+@y
+@< mehr -- ge\"anderter -- Code @>
+@z
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.w
new file mode 100644
index 00000000000..bd145c4f921
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/language-german.w
@@ -0,0 +1,31 @@
+% $Id: language-german.w,v 1.1 1995/11/20 22:34:45 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests german language support
+
+\documentclass[language=german]{cweb}
+
+\usepackage{german}
+
+\begin{document}
+
+\title{Test der \texttt{language} Option}
+\author{Puma}
+\maketitle
+
+@ Aktives Anf\"uhrungszeichen im String.
+@c
+#include "foo.h"
+
+@ @( foo.h@>=
+@< etwas |code| @>@;
+@< mehr Code @>
+
+@ @< etwas ...@>=
+for ( int i=1; i<10; i++ )
+ @< mehr... @>
+
+@ @< mehr... @>=
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/minimal.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/minimal.w
new file mode 100644
index 00000000000..fc0cdaa492a
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/minimal.w
@@ -0,0 +1,12 @@
+% $Id: minimal.w,v 1.4 1995/08/25 19:12:41 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests minimal CWEB w/ LaTeX input file
+
+\documentclass{cweb}
+\begin{document}
+
+Test.
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/modes.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/modes.w
new file mode 100644
index 00000000000..48525ae38d5
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/modes.w
@@ -0,0 +1,16 @@
+% $Id: modes.w,v 1.3 1995/08/27 19:28:49 schrod Exp $
+%----------------------------------------------------------------------
+
+\documentclass{cweb}
+
+\begin{document}
+
+@ program $\to$ program with documentation of |bar(i+1)|.
+
+@d foo bar
+@< refinement @>=
+ stmt;
+
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/newif.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/newif.w
new file mode 100644
index 00000000000..53ae2775145
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/newif.w
@@ -0,0 +1,39 @@
+% $Id: newif.w,v 1.1 1995/11/07 18:28:52 schrod Exp $
+%------------------------------------------------------------
+
+% test that proper definition of \newif is established.
+
+
+% First, restore problematic definition and pretend we're an old LaTeX.
+
+\makeatletter
+
+\outer\def\newif#1{\count@@\escapechar \escapechar\m@@ne
+ \expandafter\expandafter\expandafter
+ \edef\@@if#1{true}{\let\noexpand#1=\noexpand\iftrue}%
+ \expandafter\expandafter\expandafter
+ \edef\@@if#1{false}{\let\noexpand#1=\noexpand\iffalse}%
+ \@@if#1{false}\escapechar\count@@} % the condition starts out false
+
+\let\OldNewif=\newif
+
+\edef\fmtversion{1994/12/01}
+
+\makeatother
+
+
+% Now, start the document and check if \newif got redefined.
+
+\documentclass{cweb}
+
+\begin{document}
+
+\ifx \OldNewif\newif
+ \errmessage{\string\newif\space was not redefined.}
+\else
+ \message{\string\newif\space got redefined.}
+\fi
+
+@ Test.
+
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/nolists.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/nolists.w
new file mode 100644
index 00000000000..190c22c3c21
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/nolists.w
@@ -0,0 +1,16 @@
+% $Id: nolists.w,v 1.1 1995/09/12 22:32:22 schrod Exp $
+%------------------------------------------------------------
+
+% test suppression of index & ref list
+
+\documentclass{cweb}
+
+\begin{document}
+
+@ We need something to add to the lists.
+
+@< refinement @>=
+foo(bar);
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts-code.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts-code.w
new file mode 100644
index 00000000000..ad24b1e2b29
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts-code.w
@@ -0,0 +1,32 @@
+% $Id: parts-code.w,v 1.1 1995/09/16 17:02:10 schrod Exp $
+%------------------------------------------------------------
+
+% test of `cwebparts' package:
+% code parts to be included in a LaTeX document
+% no sections
+
+\documentclass{cweb}
+
+\usepackage{rcs}
+
+\begin{document}
+
+\title{Believe it}
+\author{Coogar}
+
+\maketitle
+
+@ Some explanation of the code below. It's very hairy, isn't it?
+
+@( foo.h@>=
+@< some |code| @>@;
+@< more code @>
+
+@ @< some ...@>=
+for ( int i=1; i<10; i++ )
+ @< more... @>
+
+@ @< more... @>=
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts.tex b/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts.tex
new file mode 100644
index 00000000000..252b2129918
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/parts.tex
@@ -0,0 +1,47 @@
+% $Id: parts.tex,v 1.1 1995/09/16 17:02:10 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% test of inclusion of CWEB program parts
+%
+% [LaTeX]
+% (history at end)
+
+
+
+\documentclass{article}
+
+\usepackage{cwebparts}
+
+
+\begin{document}
+
+
+
+\title{Foobar Algorithm}
+\author{Cougar}
+
+\maketitle
+
+\makeatletter
+
+
+\section{C Implementation}
+
+Let's have a look at the implementation in C:
+
+\cwebInput{parts-code}
+
+\noindent Got it?
+
+
+
+\end{document}
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+$Log: parts.tex,v $
+Revision 1.1 1995/09/16 17:02:10 schrod
+ Test cwebparts.sty.
+
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/refname.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/refname.w
new file mode 100644
index 00000000000..e15fa990ddd
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/refname.w
@@ -0,0 +1,21 @@
+% $Id: refname.w,v 1.3 1995/08/08 00:14:49 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests refinement names, incl. pseudo-refinements w/ file names
+
+\documentclass{cweb}
+
+\begin{document}
+
+@ @( foo.h@>=
+@< some |code| @>@;
+@< more code @>
+
+@ @< some ...@>=
+for ( int i=1; i<10; i++ )
+ @< more... @>
+
+@ @< more... @>=
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/report.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/report.w
new file mode 100644
index 00000000000..15242e7908d
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/report.w
@@ -0,0 +1,48 @@
+% $Id: report.w,v 1.1 1995/08/27 17:25:40 schrod Exp $
+%------------------------------------------------------------
+
+% test baseclass option
+% default is `report'
+
+\documentclass[baseclass]{cweb}
+
+\begin{document}
+
+\tableofcontents
+
+\newpage
+
+@** s title.
+\label{main-sec}
+
+Some text.
+
+@* e title.
+
+Some text.
+And some references: Part~\ref{main-sec} and section~\ref{normal-sec}.
+
+@*0 0 title.
+
+Some text.
+
+@*1 1 title.
+
+Some text.
+
+@*2 2 title.
+
+Some text.
+
+@ \label{normal-sec}
+Doc.
+
+@d def foo
+@f def int
+
+@c
+for ( int i=1; i<10; i++ )
+ @< do something @>
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/section.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/section.w
new file mode 100644
index 00000000000..13159e2fc6a
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/section.w
@@ -0,0 +1,55 @@
+% $Id: section.w,v 1.6 1995/11/07 14:14:25 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests section levels and references.
+
+\documentclass{cweb}
+
+\begin{document}
+
+\tableofcontents
+
+\newpage
+
+@** s title.
+\label{main-sec}
+
+Some text.
+
+@* e title.
+
+Some text.
+And some references: Part~\ref{main-sec} and section~\ref{normal-sec}.
+
+@*0 0 title.
+
+Some text.
+
+@*1 1 title.
+
+Some text.
+
+@*2 2 title.
+
+Some text.
+
+@*1 1 title.
+
+Reset counter.
+
+@*2 2 title.
+
+Some text.
+
+@ \label{normal-sec}
+Doc.
+
+@d def foo
+@f def int
+
+@c
+for ( int i=1; i<10; i++ )
+ @< do something @>
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-changes.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-changes.w
new file mode 100644
index 00000000000..019dc8388e0
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-changes.w
@@ -0,0 +1,20 @@
+% $Id: sup-changes.w,v 1.1 1995/08/27 13:29:40 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests suppress=unchanged
+
+\documentclass[suppress=unchanged]{cweb}
+
+\begin{document}
+
+\title{Test for suppression of unchanged chunks}
+\author{Coogar}
+\maketitle
+
+@ test
+
+@
+change here
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-format.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-format.w
new file mode 100644
index 00000000000..eec493ff4fd
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-format.w
@@ -0,0 +1,17 @@
+% $Id: sup-format.w,v 1.1 1995/08/29 15:22:17 schrod Exp $
+%------------------------------------------------------------
+
+% test suppression of format directives
+
+\documentclass[suppress=format]{cweb}
+
+\begin{document}
+
+@ test
+@f XmMainWindow int
+@f foo int
+@f bar i
+XmMainWindow dir = XtCreateWidget("foobar");
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-lists.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-lists.w
new file mode 100644
index 00000000000..2049f361512
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/sup-lists.w
@@ -0,0 +1,16 @@
+% $Id: sup-lists.w,v 1.1 1995/08/27 19:29:07 schrod Exp $
+%------------------------------------------------------------
+
+% test suppression of index & ref list
+
+\documentclass[suppress={index,reflist}]{cweb}
+
+\begin{document}
+
+@ We need something to add to the lists.
+
+@< refinement @>=
+foo(bar);
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/titlepage.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/titlepage.w
new file mode 100644
index 00000000000..b8230cebeb2
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/titlepage.w
@@ -0,0 +1,19 @@
+% $Id: titlepage.w,v 1.1 1995/08/27 13:29:41 schrod Exp $
+%------------------------------------------------------------
+
+% test titlepage option
+% test also forwarding of option to baseclass
+
+\documentclass[titlepage]{cweb}
+
+\begin{document}
+
+\title{Test title}
+\author{Coogar}
+
+\maketitle
+
+Test text.
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/token.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/token.w
new file mode 100644
index 00000000000..7765f1b9b37
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/token.w
@@ -0,0 +1,29 @@
+% $Id: token.w,v 1.3 1995/08/29 15:21:08 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests: rendering of tokens
+
+\documentclass{cweb}
+
+\begin{document}
+
+@
+@c
+a = 100L;
+a = 100UL;
+
+@ shift operators:
+@c
+cout << "hello, world" << endl;
+cin >> var;
+
+@ formatting that depend on font sizes in Plain \texttt{CWEB}:
+@f foo_bar int
+@c
+foo++; bar--;
+foo_bar obj;
+obj->method();
+
+
+@
+\end{document}
diff --git a/Master/texmf-dist/doc/latex/cweb-latex/src/test/vbar.w b/Master/texmf-dist/doc/latex/cweb-latex/src/test/vbar.w
new file mode 100644
index 00000000000..144b2810b36
--- /dev/null
+++ b/Master/texmf-dist/doc/latex/cweb-latex/src/test/vbar.w
@@ -0,0 +1,30 @@
+% $Id: vbar.w,v 1.2 1995/08/08 00:14:51 schrod Exp $
+%----------------------------------------------------------------------
+
+% tests ways to get ruled tables
+
+\documentclass{cweb}
+
+\usepackage{cwebarray}
+
+\begin{document}
+
+@ Test 1:
+
+\begin{tabular}{^^7cl^^7cl^^7c}
+\hline
+1st column & 2nd column\\
+\hline
+\end{tabular}
+
+
+@ Test 2:
+
+\begin{tabular}{IlIlI}
+\hline
+1st column & 2nd column\\
+\hline
+\end{tabular}
+
+
+\end{document}
diff --git a/Master/texmf-dist/source/latex/cweb-latex/Makefile b/Master/texmf-dist/source/latex/cweb-latex/Makefile
new file mode 100644
index 00000000000..a280ae7cd99
--- /dev/null
+++ b/Master/texmf-dist/source/latex/cweb-latex/Makefile
@@ -0,0 +1,79 @@
+# $Id: Makefile,v 1.5 1995/11/21 00:41:49 schrod Exp $
+#----------------------------------------------------------------------
+
+default:: install
+
+
+#
+# installation
+#
+
+# root of TDS compliant installation
+
+ TEXMF = /usr/local/lib/texmf
+
+# Comment it out if you don't want to install contributed packages.
+# (Not recommended.)
+
+ INSTALL_CONTRIB = install-contrib
+
+# installation target directories
+# (No need to change here for TDS installations.)
+
+ LATEX = $(TEXMF)/tex/latex
+ CWEB = $(LATEX)/cweb
+
+ LATEX_DOC = $(TEXMF)/doc/latex
+ CWEB_DOC = styles/cweb
+
+
+# installation actions
+
+.PHONY: install install-cweb-sty install-contrib
+
+install:: install-cweb-sty $(INSTALL_CONTRIB)
+
+install-cweb-sty:: $(LATEX) $(LATEX_DOC)
+ test -d $(CWEB) || mkdir $(CWEB)
+ cp *.sty *.cls $(CWEB)
+ cd $(LATEX_DOC) ; \
+ if [ ! -d $(CWEB_DOC) ] ; \
+ then d=`echo $(CWEB_DOC) | sed 's:/.*::'` ; \
+ test -d $$d || mkdir $$d ; \
+ test -d $(CWEB_DOC) || mkdir $(CWEB_DOC) ; \
+ fi
+ cp *.dvi $(LATEX_DOC)/$(CWEB_DOC)
+
+install-contrib::
+ cp contrib/*.sty $(CWEB)
+ : cp contrib/*.dvi $(CWEB_DOC)
+
+$(LATEX):
+ @echo "You don't have a LaTeX styles directory (tree)." >&2
+ @exit 1
+
+$(LATEX_DOC):
+ @echo "You don't have a LaTeX documentation directory (tree)." >&2
+ @exit 1
+
+
+#
+# management tasks
+#
+
+.PHONY: distrib collect pack
+
+distrib:: collect pack
+
+pack::
+ cp $(HOME)/Notes/Copying.gpl License
+ pwd=`pwd` ; pwd=`suffix $$pwd /` ; cd .. ; \
+ gtar -czvf $$pwd.tar.gz \
+ --exclude=CVS --exclude=.cvsignore --exclude=.rcsrc $$pwd ; \
+ cp $$pwd/README $$pwd.README
+ /bin/rm -f License
+
+collect::
+ cd src ; imkmf ; make distrib
+ cd examples ; make distrib
+ : 'cd contrib ; make distrib'
diff --git a/Master/texmf-dist/tex/latex/cweb-latex/cwbl-german.sty b/Master/texmf-dist/tex/latex/cweb-latex/cwbl-german.sty
new file mode 100644
index 00000000000..68a7ce6c054
--- /dev/null
+++ b/Master/texmf-dist/tex/latex/cweb-latex/cwbl-german.sty
@@ -0,0 +1,109 @@
+% $Id: cwbl-german.sty,v 1.1 1995/09/17 14:21:25 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% LaTeX2e package cwbl-german
+% adapt CWEB documents to language `german'
+%
+% [LaTeX2e]
+% (history at end)
+
+
+% Identification. The code below is explained in the implementation
+% documentation of the rcs package.
+
+\begingroup
+ \def\RCSPackage#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesPackage{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSPackage{cwbl-german}{German CWEB documents}
+ $Date: 1995/09/17 14:21:25 $: 9999/00/00 \endRCS
+ $Revision: 1.1 $: 0.0 \endRCS
+\endgroup
+
+
+%
+% What's this package for?
+%
+% XXX
+%
+
+
+% This class is supported. Send bug reports, comments and repairs.
+%
+% The reference version may be retrieved via anonymous ftp from
+% |ftp.th-darmstadt.de| [130.83.47.112], directory |pub/tex/latex|. It's
+% placed there as a gzipped tar file. (The information on the
+% IP~number is dated 17 Sep 95. It might have changed, also
+% this is very unlikely. Use your friendly nameserver.)
+
+
+%%% ------------------------------------------------------------
+
+
+% at the end of chunks
+
+\def\CwebCRAlso{\CwebCrossRef{Siehe auch Baustein}}
+\def\CwebCRsAlso{\CwebCrossRef{Siehe auch Bausteine}}
+
+\def\CwebCRCite{\CwebCrossRef{Dies wird zitiert im Baustein}}
+\def\CwebCRsCite{\CwebCrossRef{Dies wird zitiert in den Bausteinen}}
+
+\def\CwebCRUse{\CwebCrossRef{Dies wird benutzt im Baustein}}
+\def\CwebCRsUse{\CwebCrossRef{Dies wird benutzt in den Bausteinen}}
+
+\def\CwebCREt{ und~}
+\def\CwebCRsEt{ und~}
+
+
+% back matter
+
+\def\CwebCRChanged{%
+ \CwebCrossRef{%
+ Die folgenden Bausteine wurden durch ein Changefile ge\"andert:%
+ }%
+ }
+\def\CwebChangedListName{Ge\"anderte Bausteine}
+
+
+\def\CwebRefListName{Verzeichnis der Verfeinerungen}
+
+\def\CwebXRCite{\CwebCrossRef{Zitiert im Baustein}}
+\def\CwebXRCites{\CwebCrossRef{Zitiert in den Bausteinen}}
+\def\CwebXRUse{\CwebCrossRef{Benutzt im Baustein}}
+\def\CwebXRUses{\CwebCrossRef{Benutzt in den Bausteinen}}
+
+
+\def\CwebMacrosHere{%
+ \begingroup
+ \def\CwebRefNumber##1{}%
+ \CwebRefName :Pr\"aprozessor Definitionen\X
+ \endgroup
+ }
+
+
+
+\endinput
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: cwbl-german.sty,v $
+% Revision 1.1 1995/09/17 14:21:25 schrod
+% Add support for German CWEB documents.
+%
+
+
+
+%%%------------------------------------------------------------
+
+% Local Variables:
+% mode: LaTeX
+% TeX-brace-indent-level: 4
+% TeX-parse-self: nil
+% TeX-auto-save: nil
+% TeX-master: t
+% page-delimiter: "^%%%"
+% End:
diff --git a/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls b/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls
new file mode 100644
index 00000000000..ae512c54b15
--- /dev/null
+++ b/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls
@@ -0,0 +1,1417 @@
+% $StyleId: cweb.doc,v 3.6 1995/11/30 15:44:53 schrod Exp $
+%----------------------------------------------------------------------
+% Written by Joachim Schrod <schrod@iti.informatik.th-darmstadt.de>.
+% Copyright conditions see below (GPL).
+
+%
+% LaTeX class cweb
+% support for LaTeX markup in CWEB sources
+%
+% [LaTeX in MAKEPROG]
+% (history at end)
+
+% If you have received this style file without the user manual (in the
+% file cweb-user.dvi or the respective LaTeX file), it's incomplete and
+% near to useless. If it was given to you as something that you shall
+% use as an author -- complain bitterly to your provider. You need the
+% documentation and you have a right on it! (Below you can find info
+% where to get the reference version.)
+
+
+
+
+%%%%
+%%%%
+%%%% These TeX macros were documented with the documentation system
+%%%% MAKEPROG and automatically converted to the current form.
+%%%% If you have MAKEPROG available you may transform it back to
+%%%% the original input: Remove every occurence of three percents
+%%%% and one optional blank from the beginning of a line and remove
+%%%% every line which starts with four percents. The following lex
+%%%% program will do this:
+%%%%
+%%%% %%
+%%%%
+%%%% ^%%%\ ? ;
+%%%% ^%%%%.*\n ;
+%%%%
+%%%% If you just want to print the documentation you may fetch
+%%%% the archive print-makeprog.tar.Z from ftp.th-darmstadt.de (directory
+%%%% pub/tex/latex). It contains *all* used styles -- but beware, they
+%%%% may not be in a documented form...
+%%%%
+%%%%
+%%% \documentclass{progltx}
+
+%%% \usepackage{cweb-doc} % document-specific markup
+%%% \usepackage{fullpage}
+
+
+%%% \RCS $StyleRevision: 3.6 $
+%%% \RCS $StyleDate: 1995/11/30 15:44:53 $
+
+
+%%% \begin{document}
+
+
+%%% \title{The \texttt{cweb} Class\\
+%%% {\large (Implementation)}%
+%%% }
+%%% \author{% % LaTeX does not discard unnecessary glue...
+%%% Joachim Schrod%
+%%% \thanks{%
+%%% \protect\raggedright
+%%% TU~Darmstadt, Computer Science Department, WG Systems Programming,
+%%% Alexanderstr.~10, D-64283~Darmstadt, Germany.
+%%% Email: \texttt{jschrod@acm.org}%
+%%% }%
+%%% }
+%%% \date{%
+%%% Revision \RCSStyleRevision\\
+%%% (as of \RCSStyleDate)%
+%%% }
+
+%%% \maketitle
+
+
+%%% % doesn't work with progltx yet
+%%% %\tableofcontents
+
+
+
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The user interface of cweb.cls
+%%% %
+
+%%% \input{cweb-user}
+
+%%% %
+%%% % ------------------------------------------------------------
+%%% %
+%%% % subdocument: The internal interface
+%%% %
+
+%%% \input{cweb-conf}
+
+%%% %
+%%% % ------------------------------------------------------------
+
+
+
+
+%%% \chap Implementation.
+
+%%% This implementation must typeset complete files output by \cweave{}.
+%%% Large parts of this task are available from the module \pkg{cwebbase}.
+%%% It remains mainly to implement document-level markup and front and
+%%% back matter (table of contents, index, etc.) In particular, the back
+%%% matter material is not trivial; its tags are behind |\end{document}|,
+%%% we have to read carefully behind the end of our \cweb{} document to
+%%% detect if we have to typeset them at all.
+
+
+%%% \sect This module reserves the namespace |cweb|.
+
+%%% \beginprog
+\ifx \cweb@loaded\undefined
+ \def\cweb@loaded{$StyleRevision: 3.6 $}
+\else
+ \PackageWarningNoLine{cweb}%
+ {Some other package already uses namespace `cweb'}
+\fi
+%%% \endprog
+
+
+%%% \sect Let's identify this class.
+
+%%% \noindent The code below is explained in the implementation
+%%% documentation of the \pkg{rcs} package.
+
+%%% \beginprog
+\begingroup
+ \def\RCSClass#1#2 $#3: #4 #5\endRCS $#6: #7 #8\endRCS{%
+ \def\date{#4}\def\id{v#7}%
+ \ProvidesClass{#1}[\date\space\id\space #2]%
+ }
+
+ \RCSClass{cweb}{LaTeX markup for CWEB sources}
+ $StyleDate: 1995/11/30 15:44:53 $: 9999/00/00 \endRCS
+ $StyleRevision: 3.6 $: 0.0 \endRCS
+\endgroup
+%%% \endprog
+
+
+%%% \sect This module is from a supported bundle. Send bug reports,
+%%% comments and repairs.
+
+%%% The reference version may be retrieved via anonymous ftp from
+%%% |ftp.th-darmstadt.de| [130.83.47.112], directory
+%%% |pub/programming/literate-programming/c.c++/|. It's
+%%% placed there as a gzipped tar file. (The information on the IP~number
+%%% is dated August~29,~1995.
+%%% It might have changed, although this is very unlikely. Use
+%%% your friendly nameserver.)
+
+
+%%% \sect This is freely distributable software; you can redistribute it
+%%% and/or modify it under the terms of the GNU General Public License as
+%%% published by the Free Software Foundation; either version~2 of the
+%%% License, or (at your option) any later version.
+
+%%% This software is distributed in the hope that it will be useful, but
+%%% \textbf{without any warranty}; without even the implied warranty of
+%%% \textbf{merchantability} or \textbf{fitness for a particular purpose}.
+%%% See the GNU General Public License for more details.
+
+%%% You should have received a copy of the GNU General Public License in
+%%% the file |License| along with this package; if not, write to the Free
+%%% Software Foundation, Inc., 675~Mass Ave, Cambridge, MA~02139,~USA.
+
+
+%%% \sect Before we start we declare some shorthands for category codes.
+%%% By declaring the underscore~`(|_|)' as letter we can use it in our
+%%% macros. (I agree with \textsc{D.~Knuth} that
+%%% |\identifier_several_words_long| is more readable than
+%%% |\IdentifierSeveralWordsLong| and in every case better than |\p@@@s|.)
+
+%%% As this is a \LaTeX{} style file the at sign is a letter anyhow; so
+%%% we can use the ``private'' \LaTeX{} macros; and with the underscore we
+%%% can make our own macros more readable. But as we have to restore this
+%%% category code at the end of this macro file we store its former value
+%%% in the control sequence |\CatUsCode|. This method is better than to
+%%% use a group because not all macros have to be defined global this way.
+
+%%% Since somebody might use more styles from me, this cseqs might be
+%%% defined already.
+
+%%% \beginprog
+\ifx \CatEscape\undefined
+ \chardef\CatEscape=0
+ \chardef\CatOpen=1
+ \chardef\CatClose=2
+ \chardef\CatIgnore=9
+ \chardef\CatLetter=11
+ \chardef\CatOther=12
+ \chardef\CatActive=13 % is defined in Plain already
+
+ \chardef\CatUsCode=\catcode`\_
+\fi
+
+\catcode`\_=\CatLetter % top level macro file
+%%% \endprog
+
+
+%%% \sect Problems over problems with different \LaTeX{} versions. I have
+%%% decided that I will take the same stand as the \LaTeX{} team: I will
+%%% only support the last two official \LaTeX{} versions, folks who run
+%%% older software might need to upgrade.
+
+%%% The implementation will use `star'-forms of \LaTeX{}'s command
+%%% definition macros. They were introduced with \LaTeX{}
+%%% \mbox{$\langle$1994/12/01$\rangle$}.
+
+%%% Currently (as of 07 Nov 95), 1995/06/01 is the most recent version.
+%%% I want to use a features introduced with that version: |\newif| is not
+%%% outer any more. So I check here for the version and supply the new
+%%% definition if necessary.
+
+%%% \beginprog
+\NeedsTeXFormat{LaTeX2e}[1994/12/01]
+\@ifl@t@r \fmtversion{1995/06/01}{}% else
+ {
+ \@namedef{newif}#1{%
+ \count@\escapechar \escapechar\m@ne
+ \let#1\iffalse
+ \@if#1\iftrue
+ \@if#1\iffalse
+ \escapechar\count@
+ }
+ \def\@if#1#2{%
+ \expandafter\def
+ \csname \expandafter\@gobbletwo\string#1%
+ \expandafter\@gobbletwo\string#2\endcsname {%
+ \let#1#2%
+ }%
+ }
+ }
+%%% \endprog
+
+
+
+
+%%% \chap Options.
+
+%%% Our options are of the ``keyword-value'' category. We realize them by
+%%% the \pkg{keyvald} package, an enhancement of the \pkg{keyval} package
+%%% from the Graphics bundle that supports defaulting for unknown keys.
+
+%%% \begin{fixme}
+%%% Most of the code below should be extracted and moved to a
+%%% \pkg{keyvalx} package (or a new version of \pkg{keyval}). In
+%%% particular, support for package/class option processing, enumeration
+%%% values, and set values for keyword options.
+%%% \end{fixme}
+
+%%% |\ProcessKeyValOptions| may be used instead of |\ProcessOptions|, it
+%%% takes the keyval set identifier as argument. At the end it evaluates
+%%% and resets a hook. That hook may be used at option definition to
+%%% define further dependencies. (E.g., enabling that option implies
+%%% enabling an other option, too.)
+%%% %
+%%% \begin{fixme}
+%%% Like |\ProcessOptions|, I define |\@curroptions|, since I don't know
+%%% if that's needed later. The only place where it could been needed is
+%%% in the check for unprocessed options.
+%%% \end{fixme}
+
+%%% \beginprog
+\RequirePackage{keyvald}
+
+\let\ProcessKeyValOptionsHook\empty
+\providecommand*\ProcessKeyValOptions[1]{%
+ \edef\@curroptions{\@ptionlist{\@currname.\@currext}}%
+ \edef\next{\noexpand\setkeys{#1}{\@curroptions}}%
+ \next
+ \ProcessKeyValOptionsHook
+ \let\ProcessKeyValOptionsHook\empty
+ }
+%%% \endprog
+
+
+%%% \sect The enumeration option |structure| selects hierarchic~(0) or
+%%% flat~(1) structure. The predefined value is `hierarchic', there is no
+%%% default. The option value is bound to |\cweb@structure|.
+
+%%% \beginprog
+%% \define@key@enum{cweb}{structure}{hierarchic,flat}[hierarchic]
+\@namedef{KV@cweb@structure@enum:hierarchic}{0 } % <-- space!
+\@namedef{KV@cweb@structure@enum:flat}{1 } % <-- space!
+\define@key{cweb}{structure}{%
+ \expandafter\let \expandafter\@tempa
+ \csname KV@cweb@structure@enum:#1\endcsname % \relax if undefined
+ \ifx \@tempa\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for option structure%
+ }{%
+Possible valid values are `hierarchic' and `flat'.\MessageBreak
+The predefined value is `hierarchic'.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \let\cweb@structure\@tempa
+ \fi
+ }
+\setkeys{cweb}{structure=hierarchic}
+%%% \endprog
+
+
+%%% \sect With the set option |suppress| one can select suppression of
+%%% different document parts: |changehints| suppresses output of hints
+%%% that a changefile was involved, |unchanged| output of unchanged
+%%% chunks, |index| and |reflist| output of index and reference list, and
+%%% |format| output of `|@f|' directives.
+
+%%% The predefined value is no suppression. There is no default value.
+%%% %
+%%% \begin{fixme}
+%%% Shouldn't I supply a default value? `|{index,reflist}|'?
+%%% \end{fixme}
+
+%%% Set options are represented by flags named
+%%% |@cweb@|\<option>|@|\<value>|@|.
+
+%%% \beginprog
+%% \define@key@set{cweb}{suppress}{changehints,unchanged,index,reflist,format}
+\@for \@tempa :=changehints,unchanged,index,reflist,format\do {%
+ \expandafter\newif \csname if@cweb@suppress@\@tempa @\endcsname
+ \csname @cweb@suppress@\@tempa @false\endcsname
+ }%
+\define@key{cweb}{suppress}{%
+ \@for \@tempa :=#1\do {%
+ \expandafter\let \expandafter\@tempb
+ \csname @cweb@suppress@\@tempa @true\endcsname
+ %% \relax if undefined
+ \ifx \@tempb\relax
+ \PackageError{cweb}%
+ {%
+ Invalid value `#1' for set option suppress%
+ }{%
+Possible valid values are `changehints', `unchanged',\MessageBreak
+or a comma-separated list of these words enclosed in braces.\MessageBreak
+Specified values are turned on, you can't turn them off.\MessageBreak
+No value is on initially.\MessageBreak
+There's no default value, you have to specify one.%
+ }%
+ \else
+ \@tempb
+ \fi
+ }%
+ }
+
+\g@addto@macro\ProcessKeyValOptionsHook{%
+ \if@cweb@suppress@unchanged@
+ \@cweb@suppress@changehints@true
+ \fi
+ }
+%%% \endprog
+
+
+%%% \sect The base document class may be specified by the option
+%%% |baseclass|.
+
+%%% \beginprog
+\def\CwebBaseClass{article}
+\define@key{cweb}{baseclass}[report]{\def\CwebBaseClass{#1}}
+%%% \endprog
+
+
+%%% \sect Language-specific adaptions are stored in files
+%%% \texttt{cwbl-{\it language}.sty}. They may be accessed as packages or
+%%% by the |language| option that takes the \textit{language} as value. We
+%%% have to require the respective package at the end of the class, then
+%%% all cseqs with default names are defined and may be overwritten.
+
+%%% For some languages, we provide Babel-compatible option names, to ease
+%%% the usage of global class options taken up by other packages (like
+%%% \pkg{varioref}), too.
+
+%%% \beginprog
+\def\cweb_set_language#1{%
+ \IfFileExists{cwbl-#1.sty}{%
+ %% true
+ \RequirePackage{cwbl-#1}%
+ }{%
+ %% false
+ \ClassError{cweb}{Language `#1' is unknown.}{%
+Language definitions are stored in files `cwbl-<language>.sty'.\MessageBreak
+I could not locate the file `cwbl-#1.sty'. If you didn't make a%
+\MessageBreak
+typo, please check if there is such a file in a directory searched\MessageBreak
+by LaTeX.%
+ }%
+ }%
+ }
+
+\define@key{cweb}{language}{\AtEndOfClass{\cweb_set_language{#1}}}
+
+\define@key{cweb}{german}[]{\setkeys{cweb}{language=german}}
+\define@key{cweb}{french}[]{\setkeys{cweb}{language=french}}
+\define@key{cweb}{italian}[]{\setkeys{cweb}{language=italian}}
+%%% \endprog
+
+
+
+
+%%% \chap Configuration, inheritance, and aggregation.
+
+%%% This chapter sets up access to other modules, therefore we restore the
+%%% original catcodes for the time being.
+
+%%% \beginprog
+\catcode`\_=\CatUsCode
+%%% \endprog
+
+
+%%% \sect As usual, we provide the possibility to configure this class.
+
+%%% \beginprog
+\InputIfFileExists{cweb.cfg}{%
+ \typeout{******************************************^^J%
+ *^^J%
+ * Using local configuration file cweb.cfg^^J%
+ *^^J%
+ ******************************************%
+ }%
+ }{}
+%%% \endprog
+
+
+%%% \sect Inherit article class.
+
+%%% \beginprog
+\ifx \CwebBaseClass\undefined
+ \def\CwebBaseClass{article}
+\fi
+\define@key{cweb}*[]{\PassOptionsToClass{\KV@key}{\CwebBaseClass}}
+\ProcessKeyValOptions{cweb}
+\LoadClass{\CwebBaseClass}
+%%% \endprog
+
+
+%%% \sect OK, now comes our own code; switch on again our lexical
+%%% conventions.
+
+%%% \beginprog
+\catcode`\_=\CatLetter
+%%% \endprog
+
+
+
+
+%%% \chap Typesetting \cweave{} output.
+
+%%% This task has been out-sourced to an other module, \pkg{cwebbase}. As
+%%% mentioned at the start of the implementation, there remain mainly some
+%%% bigger tasks, they will be implemented in further sections below. For
+%%% the most part, the necessary preconditions are actually our class
+%%% option definitions, we don't have to do anything. (Gosh, what a
+%%% coincidence. :--) We need to tell \pkg{cwebbase}, that it should use
+%%% class warnings on problems.
+
+%%% \beginprog
+\let\cweb_warning=\ClassWarningNoLine
+\input{cwebbase}
+%%% \endprog
+
+
+%%% \sect But there remain also some small things to be done right now:
+%%% For flat structure, the counter |secnumdepth| must be set to~11, then
+%%% we want to number chunks in the whole document.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \or
+ %% flat
+ \setcounter{secnumdepth}{11}
+\fi
+%%% \endprog
+
+
+%%% \sect \pkg{cwebbase} uses |\part|, |\chapter| (if bound), and
+%%% |\section|. We have no use for any other section division that came
+%%% from the base class, let's discard the standard ones.
+
+%%% \beginprog
+\let\subsection\undefined
+\let\subsubsection\undefined
+\let\paragraph\undefined
+\let\subparagraph\undefined
+%%% \endprog
+
+
+
+
+%%% \chap Table of contents.
+
+%%% The table of contents features only sections, chunks don't have
+%%% titles, after all. Titles are indented proportionally to the rank,
+%%% with an basic indentation of |\CwebTocIndent|. An entry on major
+%%% ranks may be differentiated by layout instead of indentation, as it's
+%%% done in standard \LaTeX{} classes, too.
+
+%%% The counter |\CwebTocIndentMaxLevel| constitutes an upper limit for a
+%%% recognized indent (concerning indentation, that is).
+
+%%% \beginprog
+\newdimen\CwebTocIndent
+ \CwebTocIndent=1.5em
+\newcount\CwebTocIndentMaxLevel
+ \CwebTocIndentMaxLevel=6 % increase indentation if <= max level
+%%% \endprog
+
+
+%%% \sect For hierarchic structure, we can ignore chunk entries
+%%% completely, but have to supply a bunch of definitions for entries of
+%%% type `|section|$l$'. But for flat structure, sections are passed as
+%%% chunks, too. Then the current rank is stored in the title, as
+%%% |\cwbbRank{|$r$|}|. These differences are so large that we'll have
+%%% to make a very large |\ifcase|, over many chunks in this macro source
+%%% file. Let's start with the hierarchic stuff.
+
+%%% \beginprog
+\ifcase \cweb@structure
+ %% hierarchic
+ \def\l@chunk#1#2{}
+%%% \endprog
+
+
+%%% \sect Let's assume that we're subclassed from one of \LaTeX{}'s
+%%% standard classes; otherwise this stuff must be redefined, too. They
+%%% indent |\section| entries always by the same amount (1.5\,em), our
+%%% layout must fit to that.
+
+%%% If we have chapters, they are also indented by 1.5\,em and
+%%% distinguished by layout. The first minor section starts on rank~3 and
+%%% is placed on level~2, it should be indented larger. Let's simply add
+%%% 1.5\,em for each rank, that makes 3\,em for sections of type
+%%% |section2|. I.e., we use $f=l$ as a factor for our basic indentation.
+
+%%% If we don't have chapters, the first minor section starts on rank~2
+%%% and is placed also on level~2, it should be indented like sections
+%%% (1.5\,em), as it will be in another layout. I.e., we use $f=l-1$ as
+%%% the indentation factor.
+
+%%% The actual definitions are generated in a loop. `|\l@section|$l$'
+%%% expands to `|\cweb@toc_entry{|$l$|}|'. The indentation will be
+%%% computed within that macro.
+
+%%% \beginprog
+ \count@=2
+ \@whilenum \count@<11 \do {
+ \expandafter\edef \csname l@section\number\count@\endcsname{%
+ \noexpand\cweb@toc_entry{\number\count@}%
+ }
+ \advance \count@ by 1
+ }
+%%% \endprog
+
+
+%%% \sect A toc entry may be produced by |\@dottedtocline|. That cseq
+%%% needs the level (to decide if that entry is produced after all), the
+%%% entry indentation, the inner-entry indentation used for the number,
+%%% entry text, and page number.
+
+%%% As explained above, we use an indentation of $f \cdot
+%%% |\CwebTocIndent|$, if the level is not larger than
+%%% |\CwebTocIndentMaxLevel|. Otherwise we use the indentation of that
+%%% level.
+
+%%% Each entry is kind of an item with the number as the label. We need an
+%%% inner-entry indentation to be used for subsequent lines within this
+%%% entry. If a number is specified, it's tagged by |\numberline| in the
+%%% entry text. With a proper local definition and a pro-forma evaluation
+%%% of the text, we get at the width of the number (with a fitting space
+%%% for separation to text) and can use that as the inner-entry
+%%% indentation.
+
+%%% \beginprog
+ \def\cweb@toc_entry#1#2{% % #3 (page) will follow
+ %% \count@ == factor for basic indentation
+ %% min(#1,TocIndentMaxRank) - correct_level
+ \ifnum #1>\CwebTocIndentMaxLevel
+ \count@\CwebTocIndentMaxLevel
+ \else
+ \count@ #1 % <-- space!
+ \fi
+ \ifx \chapter\undefined
+ \advance \count@\m@ne
+ \fi
+ %% \dimen@ == inner-entry indentation
+ %% width of number & space, if existing
+ \begingroup
+ \global\dimen@ \z@ % assert value
+ \def\numberline##1{\settowidth{\global\dimen@}{##1\enskip}}%
+ \setbox\z@ \hbox{#2}% % evaluates \numberline if specified
+ \endgroup
+ \@dottedtocline
+ {#1}% % level
+ {\count@ \CwebTocIndent}% % basic indent
+ \dimen@ % num width
+ {#2}% % entry; page will follow
+ }
+%%% \endprog
+
+
+%%% \sect We're done with toc configuration for hierarchic structure,
+%%% since entry definitions for |chapter| and |section| types exist
+%%% already. Let's turn over to the flat structure.
+
+%%% \beginprog
+ \or
+ %% flat
+%%% \endprog
+
+
+%%% \sect The section rank determines how they are featured in the table.
+%%% Titles on rank~0 are typeset boldface, other titles are dotted
+%%% contents lines in roman.
+
+%%% The rank is specified within the title, i.e., in the
+%%% first argument to |\l@chunk|, as an argument for the |\cwbbRank| tag.
+%%% With an appropriate definition we might set |\count@| to the
+%%% respective value. If it is not set, an explicite invocation of
+%%% |\section| or |\addcontentsline| is responsible for this entry. Then
+%%% we assume that the rank is~0.
+
+%%% \beginprog
+ \def\l@chunk#1{% % page will be processed later
+ \count@\@ne % default value of group level
+ \begingroup
+ \let\numberline\@gobble % width not known yet
+ \def\cwbbRank##1{\global\count@ ##1\relax}%
+ \setbox\z@ \hbox{#1}%
+ \endgroup
+ \ifnum \count@=\z@
+ \let\next\cweb@bold_toc_line
+ \else
+ \let\next\cweb@normal_toc_line
+ \fi
+ \next{#1}%
+ }
+%%% \endprog
+
+
+%%% \sect A boldface line is typeset similar to the definition of
+%%% |\@dottedtocline|, only with fewer parameters. It takes two
+%%% parameters, the entry and the page number.
+
+%%% \beginprog
+ \def\cweb@bold_toc_line#1#2{%
+ \addpenalty{\@secpenalty}%
+ \addvspace{1em plus\p@}%
+ \begingroup
+ \noindent \bf
+ \hangindent\CwebTocIndent
+ \rightskip\@tocrmarg \parfillskip -\rightskip
+ \interlinepenalty\@M
+ \@tempdima\CwebTocIndent % for \numberline
+ #1\nobreak\hfill \hbox to\@pnumwidth{\hss #2}%
+ \par
+ \endgroup
+ }
+%%% \endprog
+
+
+%%% \sect A normal line uses |\@dottedtocline| like the hierarchic stuff.
+%%% Rank~1 shall still be on the same level as the bold toc line, so we
+%%% have to use $r-1$ as the proportional factor for |\CwebTocIndent|.
+
+%%% \beginprog
+ \def\cweb@normal_toc_line#1{% % page will be processed later
+ \edef\next{\noexpand\@dottedtocline{\the\count@}}% % toc entry level
+ \ifnum \count@>\CwebTocIndentMaxLevel
+ \count@ \CwebTocIndentMaxLevel
+ \fi
+ \advance\count@\m@ne % group level -= 1
+ \next{\count@\CwebTocIndent}% % basic indent
+ \CwebTocIndent % numwidth
+ {#1}% % entry
+ }
+%%% \endprog
+
+
+%%% \sect The first argument of |\@dottedtocline| is used to specify a
+%%% depth of the issued entry. All entries with a depth larger than
+%%% |tocdepth| are discarded. We typeset all entries, our |tocdepth|
+%%% value is the largest possible section level.
+
+%%% \beginprog
+ \setcounter{tocdepth}{10}
+%%% \endprog
+
+
+%%% \sect That's it. Close our distinction between hierarchic and flat
+%%% structure.
+
+%%% \beginprog
+\fi
+%%% \endprog
+
+
+
+
+%%% \chap The end of the document.
+
+%%% The end of a \cweave{} document is a rather complicated thing. It is
+%%% described in detail in the \cweave{} interface specification, we repeat the
+%%% most important points shortly.
+%%% %
+%%% \begin{itemize}
+
+%%% \item The user has issued an |\end{document}| tag in the documentation
+%%% part of the last chunk.
+
+%%% This tag is therefore part of a chunk, the chunk end (i.e., the
+%%% |\fi| cseq) does appear later in the document. Particularly, this
+%%% means that at the accurence of this tag there is still an open |\if|.
+
+%%% \item Between the |\end{document}| and the |\fi| a lot of text may
+%%% come which must be ignored. Within this text no |\fi| is allowed.
+
+%%% \item After the |\fi| the document may be finished by |\end|.
+%%% (Then the |-x| option of \cweave{} was used.)
+
+%%% Or additional information about the whole document is added: An
+%%% optional tag |\ch|, two mandatory tags |\inx\fin|, and another
+%%% optional tag |\con|.
+
+%%% \end{itemize}
+
+%%% Btw, the last chunk is marked as changed if any changefile entry has
+%%% been applied in the document. (In Plain \cweb{} that chunk is
+%%% colloquially used as the introduction to the index, and that shall be
+%%% printed always.) If we suppress unchanged chunks, the last
+%%% chunk will therefore always be processed, independent if it was really
+%%% changed or not. Yet another reason to use a whole chunk just for
+%%% |\end{document}|.
+
+%%% We'll use a the following approach for the implementation of the
+%%% iidentifier index and the refinement list (remember that we do not
+%%% produce a table of contents): We will check if there is any of these
+%%% two lists available. The available ones are typeset. Then we'll
+%%% finish the document. This functionality will be bound to |\fin|.
+%%% |\inx| will be simply a no-op. |\con| will never be reached, so we
+%%% don't have to cope with it.
+
+
+%%% \sect With the introduction of an end-document hook in \LaTeXe{}, one
+%%% might have hoped that we can utilize that hook for our task. But this
+%%% is not possible: We don't want to add some additional actions
+%%% \emph{into} the end-document handling, we want to look ahead and issue
+%%% lots of commands \emph{before} the end-document handling has happened.
+%%% (If you know \TeX{} by heart, you may skip the rest of the explanation
+%%% and go to the next chunk.)
+
+%%% As a macro language, \TeX{}'s cseqs take those tokens as arguments
+%%% that are behind them in the token stream. The token stream is
+%%% initially the document source, evaluation of a macro pushes its
+%%% expansion (a token list) on front of the stream. I.e., the evaluation
+%%% of |\end| will push the expansion on the token stream before
+%%% |\enddocument| is even looked at. And |\enddocument| will also push
+%%% its expansion before the hook is evaluated.
+
+%%% If we would allow |\enddocument| to push its expansion, all these
+%%% tokens have to be evaluated after we have typeset the backmatter --
+%%% after all, we need a proper document end, don't we? One would need to
+%%% skip over them, look at and eventually evaluate tokens that come
+%%% behind (from the initial document stream, the document source), and
+%%% then evaluate the skipped tokens. That's too hairy, we don't know
+%%% exactly how many tokens to skip, this is probable to change in future
+%%% \LaTeX{} revisions, etc.
+
+%%% The pushed tokens from |\end| don't pose so much problems. As
+%%% |\enddocument| will terminate the document processing, they are not
+%%% evaluated anyhow, we can just throw them away. And |\end| has changed
+%%% only once in the last few years, so it seems to be a stable macro --
+%%% necessary as we need to know its expansion for propper skipping.
+
+%%% A very first consequence of the demand to process tokens after
+%%% |\end{document}| is that we have to save the ``standard'' \LaTeX{}
+%%% document termination, since we will execute that later.
+
+%%% \beginprog
+\let\cweb@save@enddocument=\enddocument
+%%% \endprog
+
+
+%%% \sect When we encounter |\end{document}| we have to skip tokens
+%%% until we reach a |\fi|. Then we establish the correct
+%%% environment for the material behind.
+
+%%% The skip itself is actually a bit difficil, so we'll look at it more
+%%% closely. |\enddocument| is executed within |\end|. As mentioned above,
+%%% the original
+%%% definition of |\enddocument| cheats, it terminates \LaTeX{} before
+%%% the rest of |\end| is executed. |\end| first checks if we're in
+%%% the correct current environment, then it establishs the end
+%%% processing for paragraph building environments, closes the
+%%% group---each environment is in its own group, executes the end
+%%% processing established before, and it may ignore following white
+%%% space.
+
+%%% The |document| pseudo-environment is different in that it isn't in a
+%%% group. I.e., we have to open a group which can be closed by |\end|
+%%% afterwards. The environment check will be done in this group. With
+%%% |\aftergroup| we establish a macro which will skip the rest of the
+%%% |\end| processing. This processing is not needed anyhow: |document|
+%%% is no paragraph building environment and the following white space is
+%%% ignored by ourselves. Since |document| is no paragraph building
+%%% environment, the end processing described above is empty, i.e., not
+%%% existant. Then come a few tokens which ignore the white space. One of
+%%% these tokens is a |\fi|, so we have to skip them first separately.
+%%% They are discarded by |\cweb@skip_end|, we can start to look
+%%% for our |\fi| afterwards.
+
+%%% If |\end{document}| is the first (i.e., the only) text in this chunk
+%%% we discard the chunk start marker. This is possible since it isn't set
+%%% until now, it's in the |\everypar| token register. Of course, then we
+%%% have to do the house keeping work stored in |\everypar| ourselves.
+
+%%% \beginprog
+\def\enddocument{%
+ \if@noskipsec
+ \global\everypar{}%
+ \global\@noskipsecfalse
+ \global\@nobreakfalse
+ \global\clubpenalty\@clubpenalty
+ \fi
+ \begingroup
+ \aftergroup\cweb@skip_end
+ }
+%%% \endprog
+
+
+%%% \sect Let's make sure that the expansion of |\end| is the one that we
+%%% expect. Then we can define |\cweb@skip_end| appropriately, we have to
+%%% skip 5~tokens.
+
+%%% \beginprog
+\CheckCommand*\end[1]{%
+ \csname end#1\endcsname\@checkend{#1}%
+ \expandafter\endgroup\if@endpe\@doendpe\fi
+% 1 2 3 4 5
+ \if@ignore\global\@ignorefalse\ignorespaces\fi}
+
+\def\cweb@skip_end#1#2#3#4#5{\cweb@check_fi}
+%%% \endprog
+
+
+%%% \sect |\cweb@check_fi| gets the next token, and invokes
+%%% |\cweb@do_check_fi| who really checks it. If this token is |\fi| we
+%%% can prepare to finish our document. Otherwise we have to check the
+%%% next token.
+
+%%% Of course, first the token we have checked already must be discarded.
+%%% Therefore we insert the followup action by |\afterassignment|. It's
+%%% important that the assignment gets \emph{only} the next token, we
+%%% have to supply both the equals sign and the optional blank which are
+%%% allowed between the both arguments to |\let|.
+
+%%% \beginprog
+\def\cweb@check_fi{%
+ \futurelet\next \cweb@do_check_fi
+ }
+\def\cweb@do_check_fi{%
+ \ifx \next\fi
+ \afterassignment\cweb@finish
+ \else
+ \afterassignment\cweb@check_fi
+ \fi
+ \let\next= % <-- blank!!
+ }
+%%% \endprog
+
+
+%%% \sect When we found the |\fi|, we also discard it. But we're still
+%%% within a chunk, i.e., there is an open |\if|. So we must supply the
+%%% |\fi|.
+
+%%% \TeX{}nical note: |\next| is now equivalent to |\fi|. It must not be
+%%% used within a loop where it denotes the tail recursion's action. I.e.,
+%%% code like
+%%% %
+%%% \begin{verbatim}
+%%% \ifx \foo\bar
+%%% \let\next\relax
+%%% \else
+%%% \let\next\continue
+%%% \fi
+%%% \next
+%%% \end{verbatim}
+%%% %
+%%% will fail. This trap is closed by a redefinition of |\next|.
+
+%%% The output of all following lists may be suppressed by some option.
+%%% That is tested at the place where the respective list shall be
+%%% produced.
+
+%%% All following lists are typeset in CR state.
+
+%%% \beginprog
+\def\cweb@finish{%
+ \fi
+ \let\next\relax
+ \let\ch\cweb@changed_list
+ \let\inx\relax
+ \let\fin\cweb@end_document
+ \let\end\cweb@save@enddocument
+ \cwbb@CR
+ }
+%%% \endprog
+
+
+%%% \sect The following lists are started with unnumbered section
+%%% divisions. They are either chapters or sections.
+
+%%% \beginprog
+\ifx \chapter\undefined
+ \let\CwebListSectionDiv=\section
+\else
+ \let\CwebListSectionDiv=\chapter
+\fi
+\def\CwebListSection{\CwebListSectionDiv*}
+%%% \endprog
+
+
+%%% \sect The list of changed chunks is a cross reference list, (nearly)
+%%% like all others at the end of a chunk. The only difference is that
+%%% we do not show changeflags any more---each chunk number in this list
+%%% carries a change flag by definition.
+
+%%% \TeX{}nical note: The redefinition of |\*| is part of the second
+%%% argument of |\CwebCrossRef|. It is \emph{not} a global redefinition.
+
+%%% \beginprog
+\def\CwebCRChanged{%
+ \CwebCrossRef{The following chunks were changed by the change file:}%
+ }
+\def\CwebChangedListName{Changed Chunks}
+
+\if@cweb@suppress@changehints@
+ \def\cweb@changed_list#1.{}
+\else
+ \def\cweb@changed_list{%
+ \CwebListSection{\CwebChangedListName}%
+ \@mkboth{\uppercase{\CwebChangedListName}}%
+ {\uppercase{\CwebChangedListName}}%
+ \message{\CwebChangedListName}%
+ \let\CwebCRSize\relax
+ \CwebCRChanged \let\*\relax
+ }
+\fi
+%%% \endprog
+
+
+%%% \sect As outlined above, we want to check if there are any entries in
+%%% the identifier index or the refinement list. The former is stored in
+%%% the file |\jobname.idx|, the latter in the file |\jobname.scn|.
+
+%%% We write a macro |\cweb@if_file_not_empty| that's like |\IfFileExists|.
+
+%%% \beginprog
+\def\cweb@end_document{%
+ \cweb@if_file_not_empty{\jobname.idx}{%
+ \if@cweb@suppress@index@ \else
+ \CwebIdIndex
+ \fi
+ }{}%
+ \cweb@if_file_not_empty{\jobname.scn}{%
+ \if@cweb@suppress@reflist@ \else
+ \CwebRefList
+ \fi
+ }{}%
+ \cweb@save@enddocument
+ }
+%%% \endprog
+
+
+%%% \sect The paragraph layout for both the index and the refinement list
+%%% is much the same, it is taken from Plain \cweb{}: Each entry is a
+%%% paragraph, nearly no skip between the paragraphs (just a bit to
+%%% prevent underfull vboxes), no paragraph indentation, ragged right.
+%%% Overfull hboxes in the lists don't make sense, so we prevent them. And
+%%% we don't allow hyphenation.
+
+%%% \beginprog
+\def\cweb@list_par_layout{%
+ \parskip \z@ plus .5\p@
+ \parindent\z@
+ \rightskip \z@ plus 2.5em
+ \tolerance\@M \hyphenpenalty\@M
+ }
+%%% \endprog
+
+
+%%% \sect We consider a file as ``not empty'' if (1)~the file exists,
+%%% (2)~is not empty, and (3)~has a non-empty line at the very front. (In
+%%% fact, an empty line at the front means that there should not be
+%%% anything behind it---but we can't test this portably.)
+
+%%% |\cweb@if_file_not_empty| tests the property, it's parametrized by the
+%%% file name and gets passed actions for the true and the false case.
+
+%%% We open the file first. Then we check if it doesn't exist or if it's
+%%% empty, both conditions deliver true on |\ifeof|; in this case we
+%%% pretend that there was an empty line. (The \LaTeX{} kernel already
+%%% once provided a macro for an empty line, |\@defpar|, we repeat that
+%%% definition here since it was discarded there.) Otherwise we read the
+%%% first line. At this state, the emptiness of the first line is
+%%% equivalent to the non-availability of entries, we can easily construct
+%%% an appropriate macro call to set the flag.
+
+%%% \beginprog
+\def\@defpar{\par}
+\def\cweb@if_file_not_empty#1#2#3{%
+ \openin\@inputcheck #1\relax
+ \ifeof \@inputcheck
+ \let\next\@defpar
+ \else
+ \read\@inputcheck to \next
+ \fi
+ \closein\@inputcheck
+ \ifx \next\@defpar
+ \def\reserved@a{#3}%
+ \else
+ \def\reserved@a{#2}%
+ \fi
+ \reserved@a
+ }
+%%% \endprog
+
+
+%%% \sect The identifier index is available in the file |\jobname.idx|.
+%%% The setup for the index is a mixture of the |theindex| environment of
+%%% the \cls{article} class and DEK's index macros. It's typeset in two
+%%% columns; the user may specify an introductionary text for the index
+%%% by |\cwebIndexIntro|. If there is any introductionary text we add a
+%%% medium skip below.
+
+%%% For the index, |\parfillskip| set in such a way that almost empty
+%%% lines are avoided.
+
+%%% Before we read in this file, we have to bind the special cseqs used
+%%% therein.
+
+%%% \begin{fixme}
+%%% \LaTeX{} feature or bug? The optional argument of |\twocolumn| is
+%%% typeset as a vbox, therefore the space at top of the section heading
+%%% is not discarded at the top of a page. But we know that |\twocolumn|
+%%% will start a page in any case, we must prevent the addition of this
+%%% space. Looking in the source, we see that setting the flag
+%%% |@nobreak| will prevent this addition, let's do that as a
+%%% workaround. File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \begin{fixme}
+%%% Surprise: If |@nobreak| is set \emph{and} if there are no index
+%%% entries, |\twocolumn| will drop its optional argument. (That
+%%% happened to be noticed when there was an error in
+%%% |\cweb@if_file_not_empty|.) File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \begin{fixme}
+%%% If |\CwebListSection| is a chapter tag of the standard classes, that
+%%% macro will check if it's in twocolumn mode and will try to set
+%%% itself at the top of the current page in full width, as it would
+%%% have been given in an optional argument of |\twocolumn|. As the
+%%% method that optional argument uses (a float) must not be nested, we
+%%% must prevent this to happen. We set the |@twocolumn| flag to false,
+%%% chapter macros will then think they're evaluated in a one-column
+%%% document. It's debatable if that is a kludge or even \emph{the}
+%%% solution -- after all, within the optional argument we aren't in
+%%% twocolumn mode any more. File a \LaTeX{} bug report?
+%%% \end{fixme}
+
+%%% \beginprog
+\newtoks\cwebIndexIntro
+ \cwebIndexIntro={}
+\def\CwebIdIndex{%
+ %% two column format
+ \columnseprule\z@
+ \columnsep 35\p@ % value is from article.sty
+ %% introduction of index
+ \twocolumn[%
+ \cwbb@tex
+ \@nobreaktrue % don't add space in front of section
+ \@twocolumnfalse % \CwebListSection might be \chapter
+ \CwebListSection{\indexname}%
+ \edef\intro{\the\cwebIndexIntro}% % is a local def
+ \ifx \intro\empty
+ \else
+ \the\cwebIndexIntro\unskip
+ \par\medskip
+ \fi
+ ]%
+ \@mkboth{\uppercase{\indexname}}{\uppercase{\indexname}}%
+ \message{\indexname:}% % tell the user what we're doing
+ %% paragraph layout
+ \cweb@list_par_layout
+ \parfillskip \z@ plus .6\hsize
+ %% bindings
+ \let\I\CwebIndexEntry
+ \let\[\CwebIndexDeclared
+ \let\*\cwbb@change_flag
+ %% read the index
+ \CwebReadIndex
+ }
+\def\CwebReadIndex{\@input{\jobname.idx}}
+%%% \endprog
+
+
+%%% \sect An index entry is typeset with the same hanging indentation like
+%%% a cross reference list.
+
+%%% The entry is tagged with |\.|, if it was entered by the \cweb{}
+%%% operator~`|@.|'. Then it shall be typeset as a string. But an indexed
+%%% name may also want to use |\.| as an accent. This is the same
+%%% situation as we had at the refinement names (in module
+%%% \pkg{cwebbase}), where we introduced |\cwbb@check_dot| to handle this
+%%% case. The same minor restriction as there holds here, a refinement
+%%% name may not consist of a single dot-accented expression. (|@:|~helps
+%%% in this singular case.)
+
+%%% \beginprog
+\def\CwebIndexEntry#1, {%
+ \par
+ \hangindent\CwebNumberListHangindent
+ \leavevmode
+ \cwbb@check_dot{#1}:\quad
+ }
+%%% \endprog
+
+
+%%% \sect The chunks where identifiers are declared are noted with
+%%% underlined numbers. We also must not forget the default declaration of
+%%% |\9|, the tag for the user definable index layout.
+
+%%% \beginprog
+\def\CwebIndexDeclared#1]{$\underline{#1}$}
+\def\9#1{}
+%%% \endprog
+
+
+%%% \sect The list of the refinement names is available in the file
+%%% |\jobname.scn|. The layout is taken from the plain version: ragged
+%%% right, each entry is a paragraph, the different cross reference
+%%% categories are separated by a quad.
+
+%%% We must initialize our whole paragraph layout, as the index might have
+%%% been suppressed and no initialization has happened there then. If
+%%% there was an index, it changed |\parfillskip| and |\*|, let's restore it.
+
+%%% \beginprog
+\def\CwebRefListName{List of Refinements}
+
+\def\CwebRefList{%
+ %% paragraph layout: like in index, but
+ \onecolumn
+ \cweb@list_par_layout
+ \parfillskip\@flushglue % must be reset before section heading
+ \CwebListSection{\CwebRefListName}%
+ \@mkboth{\uppercase{\CwebRefListName}}%
+ {\uppercase{\CwebRefListName}}%
+ \message{\CwebRefListName:}%
+ %% different bindings
+ \def\I{\par \hangindent\CwebNumberListHangindent}%
+ \def\CwebCrossRef##1##2.{\quad {\reset@font\footnotesize ##1~##2.}}%
+ \let\Q\CwebXRCite
+ \let\Qs\CwebXRCites
+ \let\U\CwebXRUse
+ \let\Us\CwebXRUses
+ \let\*\cwbb@change_flag
+ %% Read the list, allow configuration first.
+ \csname CwebRefListHook\endcsname
+ \CwebReadPPList
+ }
+\def\CwebReadPPList{\@input{\jobname.scn}}
+%%% \endprog
+
+
+%%% \sect Extra definitions for the sake of internationalization.
+
+%%% \beginprog
+\def\CwebXRCite{\CwebCrossRef{Cited in chunk}}
+\def\CwebXRCites{\CwebCrossRef{Cited in chunks}}
+\def\CwebXRUse{\CwebCrossRef{Used in chunk}}
+\def\CwebXRUses{\CwebCrossRef{Used in chunks}}
+%%% \endprog
+
+
+
+%%% \chap The end.
+
+%%% Well, after all we're finished with this class. We must not forget to
+%%% restore the underscore catcode.
+
+%%% \beginprog
+\catcode`\_=\CatUsCode
+
+\endinput
+%%% \endprog
+
+
+%%% \sect I would like to thank those who helped me to improve this module.
+%%% % In particular, XXX provided XXXsubstantial parts of the code.
+
+%%% \textsc{Michael M\"uller} and \textsc{Zden\v{e}k Wagner} did thorough
+%%% checks that helped me to improve the alpha test version.
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+%%% \vskip \PltxPreSectSkip
+
+%%% \begin{rcslog}
+%%% $StyleLog: cweb.doc,v $
+%%% \Revision 3.6 1995/11/30 15:44:53 schrod
+%%% Still didn't work with version \mbox{$\langle$1994/12/01$\rangle$},
+%%% the copy of |\newif| was not complete.\\
+%%% Problem reported by Michael Seel \path|<seel@mpi-sb.mpg.de>|.
+
+%%% \Revision 3.5 1995/11/20 22:34:18 schrod
+%%% Add language support.\\
+%%% Triggered by Christian Kumpf \path|<smurf@igd.fhg.de>| and Andreas
+%%% Scherer \path|<scherer@physik.rwth-aachen.de>|.
+
+%%% Use my `official' (ACM) email address.
+
+%%% \Revision 3.4 1995/11/07 17:55:18 schrod
+%%% \LaTeX{} \cweb{} should work at least with the last two \LaTeX{}
+%%% versions; make it work with the previous-to-last one, version
+%%% \mbox{$\langle$1994/12/01$\rangle$}. For that, one has to install the
+%%% \mbox{$\langle$1995/06/01$\rangle$} (non-outer) definition of
+%%% |\newif|, and |\hb@xt@| must not be used.\\
+%%% Problem reported by Laurent Desnogues
+%%% \path|<laurent.desnogues@aiguemarine.unice.fr>| and somebody else (XXX
+%%% -- add name).
+
+%%% \Revision 3.3 1995/09/17 14:15:36 schrod
+%%% Initialize list paragraph layout in refinement list, too. With the
+%%% introduction of index suppression, the initialization -- located in
+%%% the index code -- might have been skipped.
+
+%%% \Revision 3.2 1995/09/15 10:22:59 schrod
+%%% The \cls{cweb} class needs at least \LaTeX{} version
+%%% \mbox{$\langle$1994/12/01$\rangle$}.\\
+%%% Problem reported by John S. Robinson \path|jsrobin@umiacs.umd.edu|.
+
+%%% \Revision 3.1 1995/09/12 23:00:48 schrod
+%%% Moved all code that does the actual typesetting of \cweave{} tags and
+%%% is therefore also needed for a |cweb| environment. It's now an own
+%%% module named \pkg{cwebbase}.
+
+%%% \Revision 2.8 1995/09/12 11:44:33 schrod
+%%% Standard base classes with chapters (i.e., report \& book) produced an
+%%% error at the start of the index: |\chapter| may not be called in the
+%%% optional argument of |\twocolumn|. Add a workaround to |\twocolumn|.\\
+%%% Problem reported by Christopher Higgins \path|<C.Higgins@cit.gu.edu.au>|.
+
+%%% \Revision 2.7 1995/08/29 17:32:36 schrod
+%%% Add hook for refinement list.
+
+%%% Allow internationalization of refinement list.\\
+%%% (Problem reported by Christian Kumpf \path|<kumpf@igd.fhg.de>|.)
+
+%%% Cweb bundle, version 1.0.
+
+%%% \Revision 2.6 1995/08/29 15:19:50 schrod
+%%% Added (hint to non-existence of) description of the internal interface.
+
+%%% \Revision 2.5 1995/08/29 02:07:26 schrod
+%%% Discard dependencies on 10\,pt fonts.
+
+%%% Support suppression of format directives.
+
+%%% \Revision 2.4 1995/08/27 19:31:43 schrod
+%%% Discard superfluous space at the top of index. Spacing in refinement
+%%% heading was wrong. Repair test on empty auxilliary files. Make
+%%% section heading for index and refinement list divisions (more)
+%%% configurable. Suppression of index and reference list is supported.
+
+%%% Make configuration of change flag easier.
+
+%%% Put section title of changed chunks list in marks, and tell the user
+%%% that it got typeset.
+
+%%% \Revision 2.3 1995/08/27 17:24:46 schrod
+%%% Make usage of baseclass with chapters work.
+
+%%% \Revision 2.2 1995/08/27 13:26:22 schrod
+%%% Add possibility to suppress change hints. Suppression of unchanged
+%%% chunks suppress change hints as well, they are meaningless as all
+%%% printed chunks are changed by definition.
+
+%%% Realize dependencies in options by |\ProcessKeyValOptionsHook|.
+%%% Illegal values for enumeration options don't alter the current value
+%%% any more.
+
+%%% List of changed chunks is an unnumbered section now.
+
+%%% \Revision 2.1 1995/08/25 19:11:18 schrod
+%%% Add keyword-value option style, with new \pkg{keyvald} package.
+
+%%% Hierarchic strucutures are supported now, in addition to the flat
+%%% structure of the beta-test version. One can choose with an option. For
+%%% that step, the terminology was cleaned up, too: Chunks are not named
+%%% sections any more. (That change involved reimplementation of almost
+%%% all the structure and toc stuff.)
+
+%%% The chunk number supplied by \cweave{} is used now, not some computed
+%%% number. Change flags are printed, too.
+
+%%% One can suppress output of unchanged sections.
+
+%%% One can select the baseclass with an option. That may be used to use
+%%% \cls{report} or \cls{book} to get chapter-style layout. Of course,
+%%% using an arbitrary baseclass is dangerous, it must conform to the
+%%% conventions of \LaTeX{} standard classes.
+
+%%% \Revision 1.13 1995/08/08 00:14:29 schrod
+%%% Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+%%% my standard templates for that, no changes in functionality.
+
+%%% \Revision 1.12 1993/08/10 14:15:43 schrod
+%%% New page on main section only if group level $<$ |\cwebSecNoEject|.
+%%% Default for the latter is 3.
+
+%%% Document that logos will not be defined in this style file.
+
+%%% Copy of plain macros for |\CwebNumber| does not work. Repaired the
+%%% most important one (subscript must be accessed via |\sb|).
+%%% Incompatibility to NFSS will be addressed later.\\
+%%% (Problems reported by Zden\v{e}k Wagner \path|<wagner@csearn.bitnet>|.)
+
+%%% \Revision 1.11 1993/08/10 11:21:07 schrod
+%%% Reference to section number does not render a period after the
+%%% number any more.
+
+%%% \Revision 1.10 1993/08/09 20:08:20 schrod
+%%% |\cweb@cweave_bindings| is now a no-op if \cweave{} bindings are in
+%%% effect already.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.9 1993/08/09 18:05:28 schrod
+%%% Left shift operator wasn't defined correctly.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% \Revision 1.8 1993/06/15 15:22:30 schrod
+%%% \textbf{Version 0.2:} Moved from |pub/incoming| to official place, in
+%%% Literate Programming Archive.
+
+%%% \Revision 1.7 1993/06/15 13:25:48 schrod
+%%% First attempt to make it a style: |article| is imported. Warns the
+%%% user if he tries to use |cweb| as a style option.
+
+%%% \Revision 1.6 1993/06/15 08:49:23 schrod
+%%% |\cweb@check_dot| must not evaluate its argument in an |\edef|, this
+%%% causes problems if a |\PB| is within. Now I try hard not to evaluate any
+%%% tokens outside of my control.
+
+%%% Can use |\@defpar| for an empty line in |\cweb@has_entries|, don't
+%%% need an own macro.
+
+%%% \Revision 1.5 1993/06/14 17:50:30 schrod
+%%% Handle a missing |\con|: |\inx| and |\con| are now never looked at.
+%%% Instead |\fin| checks for the existence of the lists and typeset them
+%%% if there are entries available.
+
+%%% \Revision 1.4 1993/06/14 15:54:18 schrod
+%%% Made it work with older versions of \LaTeX{}, too. The |\enddocument|
+%%% implementation of cweb.sty depends on the |\end| implementation; check
+%%% if the current implementation is one we know of (currently: 91-01-14
+%%% and 92-03-25).\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Add FSA diagram about processing states. CR state is also switched
+%%% to from \TeX{} state (that happens with |\ch| at the document end).
+
+%%% \Revision 1.3 1993/05/13 17:51:21 schrod
+%%% Refinements may also be filenames (`|@(|'). Then the complete name
+%%% consists of a |\.| macro call, which is handled now.\\
+%%% (Problem reported by Michael M\"uller \path|<mimu@mpi-sb.mpg.de>|.)
+
+%%% Made the detection of `|@.|' index entries more robust.
+
+%%% \Revision 1.2 1993/05/12 18:28:59 schrod
+%%% Adapted to recent changes of \cweave{} (of April 93):
+
+%%% Main sections have a group level, represented in the table of
+%%% contents. This changed the complete implementation of section tags.
+
+%%% New C token cseqs: |\Z| and |\MRL|, implemented as |\CwebLe| and
+%%% |\CwebCombinedOp|.
+
+%%% \Revision 1.1 1993/04/09 15:00:37 schrod
+%%% Initial revision
+
+%%% \end{rcslog}
+
+
+
+%%% \end{document}
+
+
+%%%
+%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%%% Local Variables:
+%%% mode: LaTeX
+%%% TeX-brace-indent-level: 4
+%%% indent-tabs-mode: t
+%%% TeX-auto-untabify: nil
+%%% TeX-auto-regexp-list: LaTeX-auto-regexp-list
+%%% compile-command: "make cweb.cls"
+%%% End:
diff --git a/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls.patch b/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls.patch
new file mode 100644
index 00000000000..07f20ff7fa3
--- /dev/null
+++ b/Master/texmf-dist/tex/latex/cweb-latex/cweb.cls.patch
@@ -0,0 +1,36 @@
+This patch fixes cweb.cls so that it works with new versions of the
+LaTeX format. (Any format later than 1997/06/30 or thereabouts is
+affected.)
+
+This patch may be freely redistributed.
+
+To use it, run the command "patch -p0 < cweb.cls.patch" in the
+directory containing cweb.cls if you have a UNIX-type system with the
+patch command. If not, just make the necessary changes by hand to
+lines 846 and 848 as shown in the patch below.
+
+ Julian Gilbey <jdg@debian.org>, November 1999
+
+--- cweb.cls.orig
++++ cweb.cls
+@@ -836,16 +836,16 @@
+
+ %%% \sect Let's make sure that the expansion of |\end| is the one that we
+ %%% expect. Then we can define |\cweb@skip_end| appropriately, we have to
+-%%% skip 5~tokens.
++%%% skip 4~tokens.
+
+ %%% \beginprog
+ \CheckCommand*\end[1]{%
+ \csname end#1\endcsname\@checkend{#1}%
+ \expandafter\endgroup\if@endpe\@doendpe\fi
+-% 1 2 3 4 5
+- \if@ignore\global\@ignorefalse\ignorespaces\fi}
++% 1 2 3 4
++ \if@ignore\@ignorefalse\ignorespaces\fi}
+
+-\def\cweb@skip_end#1#2#3#4#5{\cweb@check_fi}
++\def\cweb@skip_end#1#2#3#4{\cweb@check_fi}
+ %%% \endprog
+
+
diff --git a/Master/texmf-dist/tex/latex/cweb-latex/cwebarray.sty b/Master/texmf-dist/tex/latex/cweb-latex/cwebarray.sty
new file mode 100644
index 00000000000..1eb73a0d48b
--- /dev/null
+++ b/Master/texmf-dist/tex/latex/cweb-latex/cwebarray.sty
@@ -0,0 +1,50 @@
+% $Id: cwebarray.sty,v 1.2 1995/08/08 00:14:31 schrod Exp $
+%----------------------------------------------------------------------
+
+%
+% LaTeX style option cwebarray
+% depends on array style option
+%
+% [LaTeX]
+% (history at end)
+
+\ProvidesPackage{cwebarray}
+
+
+%
+% What's this style option for?
+%
+% If you use CWEB, you cannot use vertical bars (`|') as rule specifiers in
+% tables any more; CWEAVE processes them already. This style option
+% defines `I' (that's an uppercase i) as a replacement for the
+% vertical bar.
+%
+% I.e., write
+%
+% \begin{tabular}{lIl}
+%
+% instead of \begin{tabular}{l|l}.
+
+% This option won't work without the array style.
+
+\ifx \d@llarbegin\undefined
+ \RequirePackage{array}
+\fi
+
+\newcolumntype{I}{|}
+
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+%
+% $Log: cwebarray.sty,v $
+% Revision 1.2 1995/08/08 00:14:31 schrod
+% Updated to \LaTeXe{}, the |cweb| style is now a document class. Used
+% my standard templates for that, no changes in functionality.
+%
+% Revision 1.1 1993/08/09 18:05:55 schrod
+% Mentioned that `|' cannot be used for LaTeX purposes, in
+% particular, not for ruled tables. Described workarounds, one of them
+% is the new style option cwebarray.
+% [Reported by Felix G\"artner]
+%
diff --git a/Master/texmf-dist/tex/latex/cweb-latex/keyvald.sty b/Master/texmf-dist/tex/latex/cweb-latex/keyvald.sty
new file mode 100644
index 00000000000..2f6e49cef8a
--- /dev/null
+++ b/Master/texmf-dist/tex/latex/cweb-latex/keyvald.sty
@@ -0,0 +1,74 @@
+%%
+%% This is file `keyvald.sty', generated on <1995/11/30>
+%% with the docstrip utility (2.2i).
+%%
+%% The original source files were:
+%%
+%% keyvald.dtx (with options: `package')
+%%
+%% IMPORTANT NOTICE:
+%% You are not allowed to distribute this file.
+%% For distribution of the original source see
+%% the copyright notice in the file keyvald.dtx .
+%%
+%% $ITI: keyvald.dtx,v 1.2 1995/08/25 19:00:31 schrod Exp $
+%%------------------------------------------------------------
+%% (history at end)
+%% File: keyvald.dtx; changed version of
+%% File: keyval.dtx Copyright (C) 1993-1994 David Carlisle
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesPackage{keyvald}
+ [1994/08/23 v1.08.1 key=value parser (DPC) w/ defaults (js)]
+\def\setkeys#1#2{%
+ \def\KV@prefix{KV@#1@}%
+ \KV@do#2,\relax,}
+\def\KV@do#1,{%
+ \ifx\relax#1\empty\else
+ \KV@split#1==\relax
+ \expandafter\KV@do\fi}
+\def\KV@split#1=#2=#3\relax{%
+ \KV@@sp@def\KV@key{#1}%
+ \ifx\KV@key\@empty\else
+ \expandafter\let\expandafter\@tempc
+ \csname\KV@prefix\KV@key\endcsname
+ \ifx \@tempc\relax
+ \expandafter\let \expandafter\@tempc
+ \csname\KV@prefix*\endcsname
+ \def\@tempa{*}%
+ \else
+ \let\@tempa\KV@key
+ \fi
+ \ifx\@tempc\relax
+ \KV@err{\KV@key\space undefined}%
+ \else
+ \ifx\@empty#3\@empty
+ \KV@default
+ \else
+ \KV@@sp@def\@tempb{#2}%
+ \expandafter\@tempc\expandafter{\@tempb}\relax
+ \fi
+ \fi
+ \fi}
+\def\KV@default{%
+ \expandafter\let\expandafter\@tempb
+ \csname\KV@prefix\@tempa @default\endcsname
+ \ifx\@tempb\relax
+ \KV@err{No value specified for \KV@key}%
+ \else
+ \@tempb\relax
+ \fi}
+\def\KV@err#1{\PackageError{keyvald}{#1}{}}
+\def\@tempa#1{%
+\def\KV@@sp@def##1##2{\KV@@sp@b##2\@nil\@nil#1\@nil\relax##1}}
+\@tempa{ }
+\def\KV@@sp@b#1#2 \@nil{\KV@@sp@c#1#2}
+\def\KV@@sp@c#1\@nil#2\relax#3{\def#3{#1}}
+\def\define@key#1#2{%
+ \@ifnextchar[{\KV@def{#1}{#2}}{\@namedef{KV@#1@#2}####1}}
+\def\KV@def#1#2[#3]{%
+ \@namedef{KV@#1@#2@default\expandafter}\expandafter
+ {\csname KV@#1@#2\endcsname{#3}}%
+ \@namedef{KV@#1@#2}##1}
+\endinput
+%%
+%% End of file `keyvald.sty'.
diff --git a/Master/tlpkg/tlpsrc/collection-latexextra.tlpsrc b/Master/tlpkg/tlpsrc/collection-latexextra.tlpsrc
index a420c790957..5ac63f5ac88 100644
--- a/Master/tlpkg/tlpsrc/collection-latexextra.tlpsrc
+++ b/Master/tlpkg/tlpsrc/collection-latexextra.tlpsrc
@@ -88,6 +88,7 @@ depend cuisine
depend currvita
depend cursor
depend cv
+depend cweb-latex
depend cwpuzzle
depend dashbox
depend dashrule
diff --git a/Master/tlpkg/tlpsrc/cweb-latex.tlpsrc b/Master/tlpkg/tlpsrc/cweb-latex.tlpsrc
new file mode 100644
index 00000000000..a4a399825f8
--- /dev/null
+++ b/Master/tlpkg/tlpsrc/cweb-latex.tlpsrc
@@ -0,0 +1,2 @@
+name cweb-latex
+category Package