diff options
Diffstat (limited to 'Master/texmf-dist/source/latex/base/ltfntcmd.dtx')
| -rw-r--r-- | Master/texmf-dist/source/latex/base/ltfntcmd.dtx | 777 |
1 files changed, 777 insertions, 0 deletions
diff --git a/Master/texmf-dist/source/latex/base/ltfntcmd.dtx b/Master/texmf-dist/source/latex/base/ltfntcmd.dtx new file mode 100644 index 00000000000..f6850793662 --- /dev/null +++ b/Master/texmf-dist/source/latex/base/ltfntcmd.dtx @@ -0,0 +1,777 @@ +% \iffalse meta-comment +% +% Copyright 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 +% The LaTeX3 Project and any individual authors listed elsewhere +% in this file. +% +% This file is part of the LaTeX base system. +% ------------------------------------------- +% +% It may be distributed and/or modified under the +% conditions of the LaTeX Project Public License, either version 1.3 +% of this license or (at your option) any later version. +% The latest version of this license is in +% http://www.latex-project.org/lppl.txt +% and version 1.3 or later is part of all distributions of LaTeX +% version 2003/12/01 or later. +% +% This file has the LPPL maintenance status "maintained". +% +% The list of all files belonging to the LaTeX base distribution is +% given in the file `manifest.txt'. See also `legal.txt' for additional +% information. +% +% The list of derived (unpacked) files belonging to the distribution +% and covered by LPPL is defined by the unpacking scripts (with +% extension .ins) which are part of the distribution. +% +% \fi +% \iffalse +%%% From File: ltfntcmd.dtx +%% +%<*driver> +% \fi +\ProvidesFile{ltfntcmd.dtx} + [2003/01/01 v3.3z LaTeX Kernel (Font commands)] +% \iffalse +\documentclass{ltxdoc} +\GetFileInfo{ltfntcmd.dtx} +\title{\filename} +\date{\filedate} +\author{Frank Mittelbach, Chris Rowley} +\begin{document} +\maketitle + \DocInput{\filename} +\end{document} +%</driver> +% \fi +% +% \CheckSum{256} +% +%% \CharacterTable +%% {Upper-case \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z +%% Lower-case \a\b\c\d\e\f\g\h\i\j\k\l\m\n\o\p\q\r\s\t\u\v\w\x\y\z +%% Digits \0\1\2\3\4\5\6\7\8\9 +%% Exclamation \! Double quote \" Hash (number) \# +%% Dollar \$ Percent \% Ampersand \& +%% Acute accent \' Left paren \( Right paren \) +%% Asterisk \* Plus \+ Comma \, +%% Minus \- Point \. Solidus \/ +%% Colon \: Semicolon \; Less than \< +%% Equals \= Greater than \> Question mark \? +%% Commercial at \@ Left bracket \[ Backslash \\ +%% Right bracket \] Circumflex \^ Underscore \_ +%% Grave accent \` Left brace \{ Vertical bar \| +%% Right brace \} Tilde \~} +% +% \iffalse +% Copyright (C) 1994-1995 Frank Mittelbach and Chris Rowley +% all rights reserved. +% \fi +% +% \changes{v2.0b}{1993/05/05}{Removed all LaTeX related cmds} +% \changes{v3.0a}{1993/12/11}{Complete reworking of all text commands, +% using just one creator function} +% \changes{v3.0a}{1993/12/11}{newfontswitch command corrected and +% changed} +% \changes{v3.0a}{1993/12/11}{newcommands replaced by defs} +% \changes{v3.0a}{1993/12/11}{italic correction now put in front of +% penalty before glue} +% \changes{v3.0b}{1993/12/19}{\cs{@pdef} comand added} +% \changes{v3.0b}{1993/12/19}{Other tidying} +% \changes{v3.0b}{1993/12/19}{Made \cs{@newfontswitch} produce an +% error if command already exists, and added \cs{@renewfontswitch}, +% ASAJ} +% \changes{v3.0b}{1993/12/19}{Untidying added, so this is now a +% TEMPORARY version.} +% \changes{v3.0b}{1993/12/19}{Some more tidying done} +% \changes{v3.1b}{1994/01/31}{\cs{@normalsize} no longer defined} +% \changes{v3.2a}{1994/03/15}{Adapted to mass formatting} +% \changes{v3.2a}{1994/03/15}{Removed defs of short-forms and all +% sizes except \cs{normalize}} +% \changes{v3.2a}{1994/03/15}{Removed \cs{@renewfontswitch}} +% \changes{v3.2a}{1994/03/15}{Changed \cs{/} to \cs{@@italiccorr}} +% \changes{v3.2b}{1994/04/14}{Macros renamed to non-private forms, JB} +% \changes{v3.3a}{1994/04/20}{New implementation of \cs{\nocorr}} +% \changes{v3.3a}{1994/04/20}{Documentation up-dated} +% \changes{v3.3b}{1994/04/30}{Documentation up-dated and tidied} +% \changes{v3.3b}{1994/04/30}{Title changed} +% \changes{v3.3b}{1994/04/30}{Prefix frag@ changed to frag in +% \cs{@protecteddef}} +% \changes{v3.3b}{1994/04/30}{Warning changed to info message in +% \cs{@protecteddef}} +% \changes{v3.3c}{1994/05/05}{Corrected \cs{@fontswitch}} +% \changes{v3.3d}{1994/05/08}{Removed \cs{@undefinedfonterror}} +% \changes{v3.3f}{1994/05/09}{Replaced all \cs{next} by +% \cs{@let@token} and undo change 3.3e, +% whatever that was.} +% \changes{v3.3g}{1994/05/13}{Replaced \cs{@protecteddef} by +% \cs{DeclareRobustCommand}} +% \changes{v3.3h}{1994/05/20}{Use new error commands} +% \changes{v3.3i}{1994/05/24}{Tidying and typos fixed} +% \changes{v3.3j}{1994/05/25}{Insertion of \cs{aftergroup}s to +% implement \cs{nocorr} moved to the end of the group} +% \changes{v3.3k}{1994/06/09}{Tidying and typos fixed in documentation} +% \changes{v3.3l}{1994/06/18}{Added check for empty text} +% \changes{v3.3m}{1994/06/22}{Removed space from \cs{nfss@text}} +% \changes{v3.3m}{1994/06/22}{Renamed \cs{check@nocorr}} +% \changes{v3.3p}{1994/11/17} +% {\cs{@tempa} to \cs{reserved@a}} +% \changes{v3.3v}{1997/06/16}{Fix typo in documentation.} +% \changes{v3.3x}{1998/08/17}{(RmS) Minor documentation fixes.} +% +% \def \ie {i.e.~} +% \def \eg {e.g.~} +% +% \def \dst {{\normalfont\scshape docstrip}} +% \def \NFSS {\textsf{NFSS}} +% +% +% \begin{abstract} +% The commands defined in this file \texttt{ltfntcmd} are +% part of the kernel code for \LaTeXe/NFSS2. +% +% It is also meant to serve as documentation for package writers since +% it demonstrates how to define high-level font changing commands +% using a small number of creator functions. +% \end{abstract} +% +% \section{Introduction} +% +% Font changes such as |\bfseries|, |\sffamily|, etc.\ are +% declarations; this means that their scope is delimited by the +% grouping structure, either by the next |\end| of some environment or +% by explicitly using a group, e.g., writing something like +% |{\bfseries...}| in the source. If you make the mistake of writing +% |\bfseries{...}| (thinking of |\bfseries| as a command with one +% argument) then the result is rather striking. +% +% Font declarations are an artifact of the \TeX{} system and for +% several reasons it is better to avoid them on the user level +% whenever possible. In \LaTeX3 they will probably all be +% replaced by environments and by font commands taking one argument. +% +% This file defines a creator function for such declarative font +% switches. This function creates commands which can be used in +% both math and text. +% +% This file also defines a number of high-level commands (all +% starting with |\text..|) that have one argument and typeset this +% argument in the requested way. Thus these commands are for +% typesetting short pieces of text in a specific family, series or +% shape. These are all produced as examples of the use of a creator +% function which is itself also defined in this file. +% +% Table~\ref{tab:fontcmds} shows all these high-level commands in +% action. A further advantage of using these commands is that they +% automatically take care of any necessary italic correction on either +% side of their argument. +% \begin{table}[t] +% \begin{center} +% \begin{tabular}{lll} +% \textit{Command} & \textit{Corresponds to }& \textit{Action} \\[4pt] +% |\textrm{..}| & |\rmfamily| & +% Typeset argument in roman family \\ +% |\textsf{..}| & |\sffamily| & +% Typeset argument in \textsf{sans serif} family \\ +% |\texttt{..}| & |\ttfamily| & +% Typeset argument in \texttt{typewriter} family +% \\[2pt] +% |\textmd{..}| & |\mdseries| & +% Typeset argument in medium series \\ +% |\textbf{..}| & |\bfseries| & +% Typeset argument in \textbf{bold} series \\[2pt] +% |\textup{..}| & |\upshape| & +% Typeset argument in normal shape \\ +% |\textit{..}| & |\itshape| & +% Typeset argument in \textit{italic} shape \\ +% |\textsl{..}| & |\slshape| & +% Typeset argument in \textsl{slanted} shape \\ +% |\textsc{..}| & |\scshape| & +% Typeset argument in \textsc{small caps} shape +% \\[2pt] +% |\emph{..}| & |\em| & +% Typeset argument \emph{emphasized} +% \end{tabular} +% \end{center} +% \caption{Font-change commands with arguments} +% \label{tab:fontcmds} +% \begin{quote} +% The font change commands provided here +% all start with |\text..| to emphasize that they +% are for use in normal text and to be easily memorable. They +% automatically take care of any necessary italic correction on either +% side of the argument. +% \end{quote} +% \end{table} +% +% Thus, when using such commands, one does not have to worry about +% forgetting the italic correction when changing fonts. Only in very +% few situations is this additional space wrong but, for example, most +% typographers recommend omitting the italic correction if a small +% punctuation character, like a comma, directly follows the font +% change. Since the amount of correction required is partly a matter +% of taste, you can define in what situations the italic correction +% should be suppressed. This is done by putting the characters that +% should cancel a preceding italic correction in the list +% |\nocorrlist|.\footnote{Any package that changes the +% \texttt{\string\catcode} of a character inside +% \texttt{\string\nocorrlist} must then explicitly reset the list. +% Otherwise the changed character will no longer be recognized +% by the suppression algorithm.} +% The default definition for this list is produced by the following. +% \begin{verbatim} +% \newcommand \nocorrlist {,.} +% \end{verbatim} +% It is best to declare the most often used characters first, because +% this will make the processing slightly faster. For example, +% \begin{verbatim} +% \emph{When using the \NFSS{} high-level commands, +% the \emph{proper} use of italic corrections is +% automatically taken care of}. Only +% \emph{sometimes} one has to help \LaTeX{} by +% adding a \verb=\nocorr= command. +% \end{verbatim} +% which results in: +% \begin{quote} +% \emph{When using the \NFSS{} high-level commands, +% the \emph{proper} use of italic corrections is +% automatically taken care of}. Only +% \emph{sometimes} one has to help \LaTeX{} by +% adding a \verb=\nocorr= command. +% \end{quote} +% +% In contrast, the use of the declaration forms is often more +% appropriate when you define your own commands or environments. +% \begin{verbatim} +% \newenvironment{bfitemize}{\begin{itemize}\normalfont\bfseries} +% {\end{itemize}} +% \begin{bfitemize} +% \item This environment produces boldface items. +% \item It is defined in terms of \LaTeX's +% \texttt{itemize} environment and NFSS +% declarations. +% \end{bfitemize} +% \end{verbatim} +% This gives: +% \begin{quote} +% \newenvironment{bfitemize} +% {\begin{itemize}\normalfont\bfseries} +% {\end{itemize}} +% \begin{bfitemize} +% \item This environment produces boldface items. +% \item It is defined in terms of \LaTeX's +% \texttt{itemize} environment and NFSS +% declarations. +% \end{bfitemize} +% \end{quote} +% +% In addition to global customization of when to insert the italic +% correction, it is of course sometimes necessary to explicitly insert +% one with |\/|. +% +% It is also possible to suppress the italic correction +% in individual instances. For this, the command |\nocorr| is provided. +% +% The |\nocorr| must appear as the first or last token inside the +% braces of the argument of the |\text...| commands, at that end of +% the text where you wish to suppress the italic correction. +% +% +% It is worth pointing out here that inserting a |\/| in places where +% it can have no function (\ie anywhere except immediately after a +% slanted letter) is not an error---it will just be silently ignored. +% Unfortunately this is not true if the redefinition of |\/| in {\tt +% amstex.sty} is used as this version can cause space to be removed +% immediately before the |\/|. +% +% \StopEventually{} +% +% +% \section{The implementation} +% +% \changes{v3.3z}{2003/01/01}{Code checked and documentation extended +% by Chris} +% +% \begin{macro}{\DeclareTextFontCommand} +% \changes{v3.0a}{1993/12/11}{Macro changed} +% \changes{v3.0a}{1993/12/11}{Macro changed} +% This is the creator function for |\text..| commands. +% It gives a warning if |\foo| or |\fragfoo| is already defined. +% \changes{v3.0b}{1993/12/19}{Corrected and tidied} +% +% In math mode it simply puts the font declaration and text into a +% box (possibly an automagically sized one). +% +% Otherwise it first scans the text to see where |\nocorr| occurs +% within it. This sets the |\check@ic| commands to do what is +% necessary concerning the italic correction at both ends. +% +% The algorithm for deciding whether to put in an italic correction +% is not very subtle: one is added whenever the newly current font +% is not itself positively sloped, unless the next token is a +% character in the `nocorr' list. At the end of the text this +% is done after closing the group so as to check the `outer font'. +% Note that this +% will often result in adding an italic correction token after a +% character in an unsloped font; we believe (in early 2003) that +% this is perhaps inefficient but not dangerous. +% +% It also now checks for empty contents of the text command and optimises +% this case. Some care is also taken to check that doing dangerous +% things in vertical mode is avoided. +% +% The italic correction token is added to the horizontal list +% before (in the list) an immediately prceding non-zero glob of +% glue (skip) and any non-zero penalty preceding that since, in +% the typical case, this puts it immediately +% after the last character in the preceding word. +% +% +% \changes{v3.3j}{1994/05/25}{Insertion of \cs{aftergroup}s moved} +% \changes{v3.3j}{1994/05/25}{\cs{expandafter} inserted} +% \changes{v3.3m}{1994/06/22}{Removed space from \cs{nfss@text}} +% \changes{v3.3u}{1996/10/08}{Removed \cs{check@icr} when in vmode +% since it causes various errors (see pr/2157)} +% \changes{v3.3w}{1997/10/17}{Reinstalled \cs{check@icr} as check is +% now done in \cs{check@nocorr@} (see PR/2646).} +% \changes{v3.3y}{2000/01/30}{Use \cs{hmode@bgroup} now (pr/3160)} +% +% Note that it is necessary to put in the |\aftergroup\maybe@ic| at +% the end of the group so that it comes after any other aftergroup +% tokens and immediately before the following tokens. +% It is also necessary to remove the |\fi| from the token list before +% the group ends; this is done by adding an |\expandafter| just +% before the closing brace. +% +% \begin{macrocode} +%<*2ekernel> +\def \DeclareTextFontCommand #1#2{% + \DeclareRobustCommand#1[1]{% + \ifmmode + \nfss@text{#2##1}% + \else + \hmode@bgroup + \text@command{##1}% + #2\check@icl ##1\check@icr + \expandafter + \egroup + \fi + }% +} +% \end{macrocode} +% \end{macro} +% +% \begin{macro}{\textrm} +% \begin{macro}{\textsf} +% \begin{macro}{\texttt} +% \begin{macro}{\textnormal} +% Now we define the |\text|\meta{family} commands in terms of the +% above; |\texttt| does not look very nice! +% \begin{macrocode} +\DeclareTextFontCommand{\textrm}{\rmfamily} +\DeclareTextFontCommand{\textsf}{\sffamily} +\DeclareTextFontCommand{\texttt}{\ttfamily} +\DeclareTextFontCommand{\textnormal}{\normalfont} +% \end{macrocode} +% \end{macro} +% \end{macro} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\textbf} +% \begin{macro}{\textmd} +% For the series attribute: +% \begin{macrocode} +\DeclareTextFontCommand{\textbf}{\bfseries} +\DeclareTextFontCommand{\textmd}{\mdseries} +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\textit} +% \begin{macro}{\textsl} +% \begin{macro}{\textsc} +% \begin{macro}{\textup} +% \changes{v3.0a}{1993/12/11}{Macros changed} +% And for the shapes: +% \begin{macrocode} +\DeclareTextFontCommand{\textit}{\itshape} +\DeclareTextFontCommand{\textsl}{\slshape} +\DeclareTextFontCommand{\textsc}{\scshape} +\DeclareTextFontCommand{\textup}{\upshape} +% \end{macrocode} +% \end{macro} +% \end{macro} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\emph} +% \changes{v3.0a}{1993/12/11}{Macro changed} +% Finally we have the |\em| font change declaration of \LaTeX. The +% corresponding definition with argument is +% \begin{macrocode} +\DeclareTextFontCommand{\emph}{\em} +% \end{macrocode} +% \end{macro} +% +% \begin{macro}{\nocorr} +% This is just a label, so it does nothing; it should also be +% unexpandable. +% \begin{macrocode} +\let \nocorr \relax +% \end{macrocode} +% \end{macro} +% +% \begin{macro}{\check@icl} +% \begin{macro}{\check@icr} +% \changes{v3.3j}{1994/05/25}{Macros added} +% \changes{v3.3t}{1996/05/09}{Default definitions added} +% We define these defaults in case some error causes them to be +% expanded at the wrong time. +% \begin{macrocode} +\let \check@icl \@empty +\let \check@icr \@empty +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\text@command} +% \begin{macro}{\check@nocorr@} +% \changes{v3.3a}{1994/04/20}{Macros added} +% \changes{v3.3m}{1994/06/22}{Renamed \cs{check@nocorr} to +% \cs{text@command} to improve \cs{long} error message} +% \changes{v3.3j}{1994/05/25}{Insertion of \cs{aftergroup}s moved +% and defaults set up for efficiency} +% This checks for a |\nocorr| as the first token in its argument +% and also for one in any other position not protected within +% braces (the latter is treated as if it were at the end of the +% argument). +% +% Is this the correct action in the `empty' case? It is efficient but +% typographically it is, strictly, incorrect! +% \changes{v3.3l}{1994/06/18}{Added check for empty text} +% \begin{macrocode} +\def \text@command #1{% + \def \reserved@a {#1}% + \ifx \reserved@a \@empty + \let \check@icl \@empty + \let \check@icr \@empty + \else +% \end{macrocode} +% |\space| is a reserved word in \LaTeX{} or actually already in +% plain \TeX. If somebody really redefines it so many things will +% break that I don't see any reason to make this routine here +% slower than necessary. +% \changes{v3.3q}{1994/12/10}{Use \cs{space} command for comparison} +% \begin{macrocode} +% \def \reserved@b { }% +% \ifx \reserved@a \reserved@b + \ifx \reserved@a \space + \let \check@icl \@empty + \let \check@icr \@empty + \else + \check@nocorr@ #1\nocorr\@nil + \fi + \fi +} +\def \check@nocorr@ #1#2\nocorr#3\@nil {% +% \end{macrocode} +% The two checks are initialised here to their values in +% the normal case. +% \changes{v3.3w}{1997/10/17}{Check for vertical mode moved here, from +% \cs{DeclareTextFontCommand} (see PR/2646).} +% \begin{macrocode} + \let \check@icl \maybe@ic + \def \check@icr {\ifvmode \else \aftergroup \maybe@ic \fi}% + \def \reserved@a {\nocorr}% + \def \reserved@b {#1}% + \def \reserved@c {#3}% + \ifx \reserved@a \reserved@b + \ifx \reserved@c \@empty +% \end{macrocode} +% In this case there is a |\nocorr| at the start but not at the +% end, so |\check@icl| should be empty. +% \begin{macrocode} + \let \check@icl \@empty + \else +% \end{macrocode} +% Otherwise there is a |\nocorr| both at the start and +% elsewhere, so no italic corrections should be added. +% \begin{macrocode} + \let \check@icl \@empty + \let \check@icr \@empty + \fi + \else + \ifx \reserved@c \@empty +% \end{macrocode} +% In this case there is no |\nocorr| anywhere, so we need to check +% for an italic correction at both the beginning and the end. This +% has been set up as the default so no code is needed here. +% \begin{macrocode} + \else +% \end{macrocode} +% In this case there is no |\nocorr| at the start but there is one +% elsewhere, so no |\aftergroup| is needed. +% \begin{macrocode} + \let \check@icr \@empty + \fi + \fi +} +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\maybe@ic} +% \changes{v3.0a}{1993/12/11}{Macro name changed} +% \begin{macro}{\maybe@ic@} +% \changes{v3.0a}{1993/12/11}{Macro and name changed} +% \task{CAR}{Make this more efficient by checking for non-characters +% first and using \cs{if} test?} +% These macros implement the italic correction. +% \begin{macrocode} +\def \maybe@ic {\futurelet\@let@token\maybe@ic@} +\def \maybe@ic@ {% +% \end{macrocode} +% We first check to see if the current font is positively sloped. +% (But do not forget the message Rainer sent about an upright font +% with non-zero slope! Or is this an urban myth?) It has been +% suggested that this should test against a small positive value, +% but what? +% \begin{macrocode} + \ifdim \fontdimen\@ne\font>\z@ + \else + \@tempswatrue +% \end{macrocode} +% It would be possible, but probably not worthwhile, to continue the +% forward scan beyond any closing braces. +% +% \changes{v3.3a}{1994/04/20}{\cs{nocorr} etc removed from list of +% tokens to check, leaving only punctuation characters} +% \begin{macrocode} + \expandafter\@tfor\expandafter\reserved@a\expandafter:\expandafter=% + \nocorrlist +% \end{macrocode} +% We have to hide the |\@let@token| in the macro |\t@st@ic| rather +% than testing it directly in the loop since it might be |\let| +% to a |\fi| or |\else|, which would result in chaos. +% \changes{v2.1a}{1993/11/24}{Use \cs{t@st@ic}} +% \begin{macrocode} + \do \t@st@ic +% \end{macrocode} +% Frank thinks that the next bit it is inefficient if done after +% the second change. Chris thinks that most all of this is +% inefficient for the commonest cases: but that is the price of a +% cleverer algorithm. It is certainly needed to deal with the use +% of |\nolinebreak|. +% +% \begin{macrocode} + \if@tempswa \sw@slant \fi + \fi +} +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\t@st@ic} +% \changes{v2.1a}{1993/11/24}{Macro added} +% \changes{v3.3s}{1995/07/05}{Renamed from \cs{test@next}} +% The next token in the input stream is stored in |\@let@token| via a +% |\let|, the current token from |\nocorrlist| is stored via |\def| +% in |\reserved@a|. To compare them we have to fiddle around a bit. +% +% If the only things to check were characters then this could be +% done via an |\if| thus their catcodes would not matter; but this +% will not work whilst |\futurelet| is used above. +% \begin{macrocode} +\def \t@st@ic {% + \expandafter\let\expandafter\reserved@b\expandafter=\reserved@a\relax + \ifx\reserved@b\@let@token +% \end{macrocode} +% If they are the same we record the fact and jump out of the loop. +% \changes{v3.3s}{1995/07/03}{Use clean interface for jump} +% \begin{macrocode} + \@tempswafalse + \@break@tfor + \fi +} +% \end{macrocode} +% \end{macro} +% +% \begin{macro}{\sw@slant} +% \changes{v3.0a}{1993/12/11}{Macro changed} +% \begin{macro}{\fix@penalty} +% \changes{v3.0a}{1993/12/11}{Macro added} +% The definition of the mysterious |\sw@slant| command is as follows. +% \begin{macrocode} +\def \sw@slant {% +% \end{macrocode} +% It is surely correct to put in an italic correction when there +% is no skip. If the last thing on the list is actually a zero +% skip (including things whose dimension part is zero, such as +% |\hfill|), or anything other than a character, then the +% italic correction will have no effect. +% +% In order to work correctly with unbreakable spaces from |~| +% (and other common forms of line-breaking control) we also move +% back across a penalty before the glue. +% \begin{macrocode} + \ifdim \lastskip=\z@ + \fix@penalty + \else + \skip@ \lastskip + \unskip + \fix@penalty + \hskip \skip@ + \fi +} +% \end{macrocode} +% The above code means: ``If there is a non-zero space just before +% the current position (|\ifdim...|) save the amount of that space +% (|\skip@\lastskip|), remove it (|\unskip|), then do a similar +% thing if there is a penalty just before the skip, +% and finally put the space back in.'' +% +% Since zero glue cannot be distinguished in this context from no +% glue, we dare not put in an |\hskip| in this case as this may +% produce an unwanted breakpoint. This is not satisfactory. +% +% The penalty before the glue is handled similarly, with the same +% caveats concerning the zero case. Is this the first recorded +% use of |\unpenalty| in standard \LaTeX{} code? +% \begin{macrocode} +\def \fix@penalty {% + \ifnum \lastpenalty=\z@ + \@@italiccorr + \else + \count@ \lastpenalty + \unpenalty + \@@italiccorr + \penalty \count@ + \fi +} +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}{\nocorrlist} +% This holds the list of characters that should prevent italic +% correction. They should be ordered by decreasing frequency of +% use. If any such character is made active later on one needs +% to redefine the list so that the active character becomes part +% of it. +% \changes{v3.1a}{1994/01/27}{Only ., used as default for cm fonts} +% \begin{macrocode} +\def \nocorrlist {,.} +% \end{macrocode} +% \end{macro} +% +% +% \begin{macro}{\nfss@text} +% This command will by default behave like a \LaTeX{} |\mbox| but +% may be redefined by packages such as |amstext.sty| to be a bit +% cleverer. +% \begin{macrocode} +\ifx \nfss@text\@undefined + \def \nfss@text {\leavevmode\hbox} +\fi +% \end{macrocode} +% \end{macro} +% +% +% \begin{macro}{\DeclareOldFontCommand} +% \changes{v3.2b}{1994/04/14}{Renamed from \cs{@newfontswitch}} +% This is the function used to create declarative font-changing +% commands that can also be used to change alphabets in math-mode. +% +% Usage: |\DeclareOldFontCommand \fn{|\meta{font-change decls}|}| +% \meta{math-alphabet} +% +% Here |\fn| is the font-declaration command being defined, +% \meta{font-change decls} is the declaration it will expand to in +% text-mode, and \meta{math-alphabet} is the (single) math alphabet +% specifier which is to be used in math-mode. +% +% It does not care whether the command being defined already +% exists but it does give a warning if it redefines anything. +% +% Here are some typical examples of its use in conjunction with +% more basic NFSS2 font commands. +% \begin{verbatim} +% \DeclareOldFontCommand{\rm}{\normalfont\rmfamily}{\mathrm} +% \DeclareOldFontCommand{\sf}{\normalfont\sffamily}{\mathsf} +% \DeclareOldFontCommand{\tt}{\normalfont\ttfamily}{\mathtt} +% \end{verbatim} +% +% \changes{v3.0b}{1993/12/19}{Corrected and tidied} +% \begin{macrocode} +\def \DeclareOldFontCommand #1#2#3{% + \DeclareRobustCommand #1{\@fontswitch {#2}{#3}}% +} +% \end{macrocode} +% \end{macro} +% +% \begin{macro}{\@fontswitch} +% \begin{macro}{\@@math@egroup} +% \begin{macro}{\@@math@egroup} +% These two commands actually do the necessary tests and +% declarative \mbox{font-} or alphabet-changing. +% +% \changes{v3.3c}{1994/05/05}{Corrected \cs{@fontswitch} and added +% saved versions} +% \changes{v3.3q}{1994/12/10}{Don't read arguments} +% \changes{v3.3r}{1995/04/02}{Read them again to be able to add \cs{relax}.} +% \begin{macrocode} +\def \@fontswitch #1#2{% + \ifmmode + \let \math@bgroup \relax + \def \math@egroup {\let \math@bgroup \@@math@bgroup + \let \math@egroup \@@math@egroup}% +% \end{macrocode} +% We need to have a |\relax| in the following line in case the |#2| +% is something like |\mathsf| grabbing the next token as an +% argument. For this reason the code also uses explicit arguments +% again (see pr/1275). +% \begin{macrocode} + #2\relax + \else + #1% + \fi +} +\let \@@math@bgroup \math@bgroup +\let \@@math@egroup \math@egroup +% \end{macrocode} +% \end{macro} +% \end{macro} +% \end{macro} +% +% \changes{v2.1b}{1993/12/08}{Macros \cs{rm}, \cs{bf} and \cs{sf} +% moved to classes.dtx} +% +% These commands are available only in the preamble. +% \begin{macrocode} +\@onlypreamble \DeclareTextFontCommand +\@onlypreamble \DeclareOldFontCommand +% \end{macrocode} +% +% \section{Initialization} +% \changes{v3.0b}{1993/12/19}{Added by ASAJ.} +% \changes{v3.0b}{1993/12/19}{Wording changes by CAR.} +% +% \begin{macro}{\normalsize} +% \changes{v3.3d}{1994/05/08}{Removed \cs{@undefinedfonterror}} +% \changes{v3.3o}{1994/11/12}{Added \cs{MessageBreak}} +% This is defined to produce an error. +% \begin{macrocode} +\def\normalsize{% + \@latex@error {The font size command \protect\normalsize\space + is not defined:\MessageBreak + there is probably something wrong with + the class file}\@eha +} +%</2ekernel> +% \end{macrocode} +% \end{macro} +% +% \Finale +\endinput |