diff options
Diffstat (limited to 'support/crossrefware')
| -rw-r--r-- | support/crossrefware/Makefile | 5 | ||||
| -rw-r--r-- | support/crossrefware/README | 9 | ||||
| -rw-r--r-- | support/crossrefware/bibdoiadd.1 | 34 | ||||
| -rwxr-xr-x | support/crossrefware/bibdoiadd.pl | 38 | ||||
| -rw-r--r-- | support/crossrefware/bibmradd.1 | 13 | ||||
| -rwxr-xr-x | support/crossrefware/bibmradd.pl | 14 | ||||
| -rw-r--r-- | support/crossrefware/bibzbladd.1 | 11 | ||||
| -rwxr-xr-x | support/crossrefware/bibzbladd.pl | 13 | ||||
| -rw-r--r-- | support/crossrefware/crossrefware.pdf | bin | 234836 -> 236994 bytes | |||
| -rw-r--r-- | support/crossrefware/head.ltx | 17 | ||||
| -rw-r--r-- | support/crossrefware/ltx2crossrefxml.1 | 57 | ||||
| -rwxr-xr-x | support/crossrefware/ltx2crossrefxml.pl | 416 |
12 files changed, 394 insertions, 233 deletions
diff --git a/support/crossrefware/Makefile b/support/crossrefware/Makefile index c8caf35d60..fb54525cba 100644 --- a/support/crossrefware/Makefile +++ b/support/crossrefware/Makefile @@ -25,6 +25,11 @@ check: pod2man -c "LATEX CROSSREFWARE" -n $* -s 1 -r "" $< > $@ +lastrel = /home/ftp/tex-archive/support/crossrefware +reldiff: + diff -u0r $(lastrel) . +gitdiff: + git diff clean: $(RM) *.aux *.toc *.log *.tex *.idx *.ilg *.ind *.out *.zip *.tgz *~ diff --git a/support/crossrefware/README b/support/crossrefware/README index 05d375253e..d2de384b79 100644 --- a/support/crossrefware/README +++ b/support/crossrefware/README @@ -1,5 +1,5 @@ Crossrefware Bundle - version 2024-01-31 + version 2024-08-27 Scripts useful for working with Crossref, MathSciNet and Zentralblatt MATH. @@ -30,9 +30,14 @@ Princeton University (Mathematics Department). Thank you! Changes: +2024-08-27 - output crossref schema 5.3.1: + https://www.crossref.org/documentation/schema-library/schema-versions/ + (and update various doc urls that crossref broke.) + - more documentation on the bib*add scripts. + 2024-01-31 - exit with bad status if no \end{thebibliography}. -2022-09-11 - add "***" prefix to bib warnings. +2022-09-11 - add "*** " prefix to bib warnings. 2022-07-28 - find dev checkout of bibtexperllibs. diff --git a/support/crossrefware/bibdoiadd.1 b/support/crossrefware/bibdoiadd.1 index 737f4f2a4b..1425b1e08d 100644 --- a/support/crossrefware/bibdoiadd.1 +++ b/support/crossrefware/bibdoiadd.1 @@ -55,7 +55,7 @@ .\" ======================================================================== .\" .IX Title "bibdoiadd 1" -.TH bibdoiadd 1 2023-08-20 "" "LATEX CROSSREFWARE" +.TH bibdoiadd 1 2024-09-02 "" "LATEX CROSSREFWARE" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l @@ -73,7 +73,7 @@ Configuration file. If this file is absent, some defaults are used. See below for its format. .IP "\fB\-C\fR 1|0" 4 .IX Item "-C 1|0" -Whether to canonize names in the output (1) or not (0). By default, 1. +Whether to canonicalize names in the output (1) or not (0). By default, 1. .IP \fB\-e\fR 4 .IX Item "-e" If 1 (default), add empty doi if a doi cannot be found. This prevents @@ -96,19 +96,29 @@ corresponding DOI. The result is a BibTeX file with the fields The name of the output file is either set by the \fB\-o\fR option or is derived by adding the suffix \f(CW\*(C`_doi\*(C'\fR to the output file. .PP -There are two options for making queries with Crossref: free account -and paid membership. In the first case you still must register with -Crossref and are limited to a small number of queries, see the +Every BibTeX record in the input is parsed, using BibTeX::Parser, but +only the ones that do not have the \f(CW\*(C`doi\*(C'\fR field (or \f(CW\*(C`mrnumber\*(C'\fR or +\&\f(CW\*(C`zblnumber\*(C'\fR for the sibling scripts) are processed. These entries +without the requested field are written back, as described in +BibTeX::Parser::Entry. +.PP +The bib records that are not processed (because they already have the +requested field) are written back as-is, without any reformatting. +.PP +There are (were?) two options for making queries with Crossref: free +account and paid membership. In the first case you still must register +with Crossref and are limited to a small number of queries, see the agreement at -\&\f(CW\*(C`http://www.crossref.org/01company/free_services_agreement.html\*(C'\fR. In +\&\f(CW\*(C`http://www.crossref.org/01company/free_services_agreement.html\*(C'\fR. In the second case you have a username and password, and can use them for -automatic queries. I am not sure whether the use of this script is -allowed for the free account holders. Anyway if you try to add DOI -to a large number of entries, you should register as a paid member. +automatic queries. I am not sure whether the use of this script is +allowed for the free account holders. At any rate, if you want to add +DOIs to a large number of entries, you should register as a paid member. .SH "CONFIGURATION FILE" .IX Header "CONFIGURATION FILE" -The configuration file is mostly self-explanatory: it has comments -(starting with \f(CW\*(C`#\*(C'\fR) and assginments in the form +The configuration file relates to the Crossref queries, and is mostly +self-explanatory: it has comments (starting with \f(CW\*(C`#\*(C'\fR) and assginments +in the form .PP .Vb 1 \& $field = value ; @@ -128,7 +138,7 @@ members. Boris Veytsman .SH "COPYRIGHT AND LICENSE" .IX Header "COPYRIGHT AND LICENSE" -Copyright (C) 2014\-2021 Boris Veytsman +Copyright (C) 2014\-2024 Boris Veytsman .PP This is free software. You may redistribute copies of it under the terms of the GNU General Public License diff --git a/support/crossrefware/bibdoiadd.pl b/support/crossrefware/bibdoiadd.pl index bace84c000..91222af008 100755 --- a/support/crossrefware/bibdoiadd.pl +++ b/support/crossrefware/bibdoiadd.pl @@ -21,7 +21,7 @@ See below for its format. =item B<-C> 1|0 -Whether to canonize names in the output (1) or not (0). By default, 1. +Whether to canonicalize names in the output (1) or not (0). By default, 1. =item B<-e> @@ -50,22 +50,31 @@ C<doi=...> added. The name of the output file is either set by the B<-o> option or is derived by adding the suffix C<_doi> to the output file. -There are two options for making queries with Crossref: free account -and paid membership. In the first case you still must register with -Crossref and are limited to a small number of queries, see the +Every BibTeX record in the input is parsed, using BibTeX::Parser, but +only the ones that do not have the C<doi> field (or C<mrnumber> or +C<zblnumber> for the sibling scripts) are processed. These entries +without the requested field are written back, as described in +BibTeX::Parser::Entry. + +The bib records that are not processed (because they already have the +requested field) are written back as-is, without any reformatting. + +There are (were?) two options for making queries with Crossref: free +account and paid membership. In the first case you still must register +with Crossref and are limited to a small number of queries, see the agreement at -C<http://www.crossref.org/01company/free_services_agreement.html>. In +C<http://www.crossref.org/01company/free_services_agreement.html>. In the second case you have a username and password, and can use them for -automatic queries. I am not sure whether the use of this script is -allowed for the free account holders. Anyway if you try to add DOI -to a large number of entries, you should register as a paid member. - +automatic queries. I am not sure whether the use of this script is +allowed for the free account holders. At any rate, if you want to add +DOIs to a large number of entries, you should register as a paid member. =head1 CONFIGURATION FILE -The configuration file is mostly self-explanatory: it has comments -(starting with C<#>) and assginments in the form +The configuration file relates to the Crossref queries, and is mostly +self-explanatory: it has comments (starting with C<#>) and assginments +in the form $field = value ; @@ -85,7 +94,7 @@ Boris Veytsman =head1 COPYRIGHT AND LICENSE -Copyright (C) 2014-2021 Boris Veytsman +Copyright (C) 2014-2024 Boris Veytsman This is free software. You may redistribute copies of it under the terms of the GNU General Public License @@ -113,7 +122,7 @@ $ENV{PERL_LWP_SSL_VERIFY_HOSTNAME}=0; my $USAGE="USAGE: $0 [-c config] [-C 1|0] [-e 1|0] [-f] [-o output] file\n"; my $VERSION = <<END; -bibdoiadd v2.2 +bibdoiadd v2.3 This is free software. You may redistribute copies of it under the terms of the GNU General Public License http://www.gnu.org/licenses/gpl.html. There is NO WARRANTY, to the @@ -258,7 +267,7 @@ sub GetDoi { if ($entry->has('pages')) { my $pages=$entry->field('pages'); $pages =~ s/-.*$//; - $url .= "&spage=".uri_escape_utf8($pages); + $url .= "&spage=".uri_escape_utf8($pages); } if ($entry->has('year')) { $url .= "&date=".uri_escape_utf8($entry->field('year')); @@ -296,4 +305,3 @@ sub SanitizeText { $string =~ s/[\{\}]//g; return $string; } - diff --git a/support/crossrefware/bibmradd.1 b/support/crossrefware/bibmradd.1 index ee49fd5238..54e5c43095 100644 --- a/support/crossrefware/bibmradd.1 +++ b/support/crossrefware/bibmradd.1 @@ -55,7 +55,7 @@ .\" ======================================================================== .\" .IX Title "bibmradd 1" -.TH bibmradd 1 2023-08-20 "" "LATEX CROSSREFWARE" +.TH bibmradd 1 2024-09-02 "" "LATEX CROSSREFWARE" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l @@ -84,19 +84,20 @@ Output file. If this option is not used, the name for the output file is formed by adding \f(CW\*(C`_mr\*(C'\fR to the input file .SH DESCRIPTION .IX Header "DESCRIPTION" -The script reads a BibTeX file. It checks whether the entries have -mrnumberss. If not, tries to contact internet to get the numbers. The -result is a BibTeX file with the fields -\&\f(CW\*(C`mrnumber=...\*(C'\fR added. +The script reads a BibTeX file. It checks whether the entries have +mrnumbers. If not, it tries to find the numbers from Internet sites. The +result is a BibTeX file with \f(CW\*(C`mrnumber=...\*(C'\fR fields added. .PP The name of the output file is either set by the \fB\-o\fR option or is derived by adding the suffix \f(CW\*(C`_mr\*(C'\fR to the output file. +.PP +See the \f(CW\*(C`bibdoiadd\*(C'\fR script for more details on the processing. .SH AUTHOR .IX Header "AUTHOR" Boris Veytsman .SH "COPYRIGHT AND LICENSE" .IX Header "COPYRIGHT AND LICENSE" -Copyright (C) 2014\-2022 Boris Veytsman +Copyright (C) 2014\-2024 Boris Veytsman .PP This is free software. You may redistribute copies of it under the terms of the GNU General Public License diff --git a/support/crossrefware/bibmradd.pl b/support/crossrefware/bibmradd.pl index 3fea62e3c5..5a0b5a30f7 100755 --- a/support/crossrefware/bibmradd.pl +++ b/support/crossrefware/bibmradd.pl @@ -38,21 +38,22 @@ output file is formed by adding C<_mr> to the input file =head1 DESCRIPTION -The script reads a BibTeX file. It checks whether the entries have -mrnumberss. If not, tries to contact internet to get the numbers. The -result is a BibTeX file with the fields -C<mrnumber=...> added. +The script reads a BibTeX file. It checks whether the entries have +mrnumbers. If not, it tries to find the numbers from Internet sites. The +result is a BibTeX file with C<mrnumber=...> fields added. The name of the output file is either set by the B<-o> option or is derived by adding the suffix C<_mr> to the output file. +See the C<bibdoiadd> script for more details on the processing. + =head1 AUTHOR Boris Veytsman =head1 COPYRIGHT AND LICENSE -Copyright (C) 2014-2022 Boris Veytsman +Copyright (C) 2014-2024 Boris Veytsman This is free software. You may redistribute copies of it under the terms of the GNU General Public License @@ -91,7 +92,7 @@ $ENV{PERL_LWP_SSL_VERIFY_HOSTNAME}=0; my $USAGE="USAGE: $0 [-d] [-e 1|0] [-f] [-o output] file\n"; my $VERSION = <<END; -bibmradd v2.2 +bibmradd v2.3 This is free software. You may redistribute copies of it under the terms of the GNU General Public License http://www.gnu.org/licenses/gpl.html. There is NO WARRANTY, to the @@ -220,4 +221,3 @@ sub GetMr { } } - diff --git a/support/crossrefware/bibzbladd.1 b/support/crossrefware/bibzbladd.1 index 755536c73b..ce4240144e 100644 --- a/support/crossrefware/bibzbladd.1 +++ b/support/crossrefware/bibzbladd.1 @@ -55,7 +55,7 @@ .\" ======================================================================== .\" .IX Title "bibzbladd 1" -.TH bibzbladd 1 2023-08-20 "" "LATEX CROSSREFWARE" +.TH bibzbladd 1 2024-09-02 "" "LATEX CROSSREFWARE" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l @@ -85,18 +85,19 @@ output file is formed by adding \f(CW\*(C`_zbl\*(C'\fR to the input file .SH DESCRIPTION .IX Header "DESCRIPTION" The script reads a BibTeX file. It checks whether the entries have -Zbls. If not, tries to contact internet to get the numbers. The -result is a BibTeX file with the fields -\&\f(CW\*(C`zblnumber=...\*(C'\fR added. +Zbls. If not, it tries to find the numbers from Internet sites. The +result is a BibTeX file with \f(CW\*(C`zblnumber=...\*(C'\fR fields added. .PP The name of the output file is either set by the \fB\-o\fR option or is derived by adding the suffix \f(CW\*(C`_zbl\*(C'\fR to the output file. +.PP +See the \f(CW\*(C`bibdoiadd\*(C'\fR script for more details on the processing. .SH AUTHOR .IX Header "AUTHOR" Boris Veytsman .SH "COPYRIGHT AND LICENSE" .IX Header "COPYRIGHT AND LICENSE" -Copyright (C) 2014\-2021 Boris Veytsman +Copyright (C) 2014\-2024 Boris Veytsman .PP This is free software. You may redistribute copies of it under the terms of the GNU General Public License diff --git a/support/crossrefware/bibzbladd.pl b/support/crossrefware/bibzbladd.pl index c05a1137f7..549d1cb0db 100755 --- a/support/crossrefware/bibzbladd.pl +++ b/support/crossrefware/bibzbladd.pl @@ -24,7 +24,6 @@ If 1 (default), add an empty zblnumber if a zbl cannot be found. This prevents repeated searches for the same entries if you add new entries to the file. Calling C<-e 0> suppresses this behavior. - =item B<-f> Force searching for Zbl numbers even if the entry already has one. @@ -39,20 +38,21 @@ output file is formed by adding C<_zbl> to the input file =head1 DESCRIPTION The script reads a BibTeX file. It checks whether the entries have -Zbls. If not, tries to contact internet to get the numbers. The -result is a BibTeX file with the fields -C<zblnumber=...> added. +Zbls. If not, it tries to find the numbers from Internet sites. The +result is a BibTeX file with C<zblnumber=...> fields added. The name of the output file is either set by the B<-o> option or is derived by adding the suffix C<_zbl> to the output file. +See the C<bibdoiadd> script for more details on the processing. + =head1 AUTHOR Boris Veytsman =head1 COPYRIGHT AND LICENSE -Copyright (C) 2014-2021 Boris Veytsman +Copyright (C) 2014-2024 Boris Veytsman This is free software. You may redistribute copies of it under the terms of the GNU General Public License @@ -78,7 +78,7 @@ $ENV{PERL_LWP_SSL_VERIFY_HOSTNAME}=0; my $USAGE="USAGE: $0 [-d] [-e 1|0] [-f] [-o output] file\n"; my $VERSION = <<END; -bibzbladd v2.2 +bibzbladd v2.3 This is free software. You may redistribute copies of it under the terms of the GNU General Public License http://www.gnu.org/licenses/gpl.html. There is NO WARRANTY, to the @@ -202,4 +202,3 @@ sub GetZbl { } } - diff --git a/support/crossrefware/crossrefware.pdf b/support/crossrefware/crossrefware.pdf Binary files differindex 83b8d3e8a3..18c39074d5 100644 --- a/support/crossrefware/crossrefware.pdf +++ b/support/crossrefware/crossrefware.pdf diff --git a/support/crossrefware/head.ltx b/support/crossrefware/head.ltx index 72ae89028e..21d2b2d9bb 100644 --- a/support/crossrefware/head.ltx +++ b/support/crossrefware/head.ltx @@ -30,13 +30,15 @@ upload the submission, just outputs XML. This \path{.rpi} file is a plain text representation of the metadata for one article. It is written by the \path{resphilosophica} package -(\url{https://ctan.org/pkg/resphilosophica}). It can also be created by -hand. +(\url{https://ctan.org/pkg/resphilosophica}) and the TUGboat publication +procedure (\url{https://tug.org/TUGboat/repository.html}). It can also +be created by hand. Several scripts, \path{bibdoiadd}, \path{bibmradd} and \path{bibzbladd} take a \path{bib} file, and add to each entry a DOI, MR or ZBL number correspondingly, if they can find this entry in the corresponding -database. +database. The output of these scripts reformats the BibTeX entries +where the respective fields were not already present. The \path{bbl2bib} script tries to reconstruct a \path{bib} file from the corresponding \path{thebibliography} environment. One can argue that @@ -44,7 +46,10 @@ this operation is akin to reconstructing the cow from a steak. The way the script does it is by searching for the entry in the MR database, and creating the corresponding Bib\TeX\ fields. -I am grateful to Josko Plazonic from Princeton Math Dept whose -(unpublished) Python script was an inspiration for this suite. +I am grateful to Josko Plazonic from the Princeton mathematics +department whose (unpublished) Python script was an inspiration for this +suite. -Following are manual pages for these scripts. +Following are manual pages for these scripts. See also the +\texttt{BibTeX::Parser} package +(\url{https://ctan.org/pkg/bibtexperllibs}). diff --git a/support/crossrefware/ltx2crossrefxml.1 b/support/crossrefware/ltx2crossrefxml.1 index 9f6800d8b7..5c9fc681e2 100644 --- a/support/crossrefware/ltx2crossrefxml.1 +++ b/support/crossrefware/ltx2crossrefxml.1 @@ -55,7 +55,7 @@ .\" ======================================================================== .\" .IX Title "ltx2crossrefxml 1" -.TH ltx2crossrefxml 1 2024-02-03 "" "LATEX CROSSREFWARE" +.TH ltx2crossrefxml 1 2024-09-02 "" "LATEX CROSSREFWARE" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l @@ -91,8 +91,10 @@ ignored, and \fIlatex_file\fR itself is not read (and need not even exist). Each \f(CW\*(C`.rpi\*(C'\fR file specifies the metadata for a single article to be uploaded to Crossref (a \f(CW\*(C`journal_article\*(C'\fR element in their schema); an example is below. These files are output by the \f(CW\*(C`resphilosophica\*(C'\fR -package (<https://ctan.org/pkg/resphilosophica>), but (as always) can -also be created by hand or by whatever other method you implement. +package (<https://ctan.org/pkg/resphilosophica>) and the TUGboat +publication procedure (<https://tug.org/TUGboat/repository.html>), but +(as always) can also be created by hand or by whatever other method you +implement. .PP Any \f(CW\*(C`.bbl\*(C'\fR files present are used for the citation information in the output XML. See the CITATIONS section below. @@ -102,18 +104,25 @@ citations), standard TeX control sequences are replaced with plain text or UTF\-8 or eliminated, as appropriate. The \f(CW\*(C`LaTeX::ToUnicode::convert\*(C'\fR routine is used for this (<https://ctan.org/pkg/bibtexperllibs>). Tricky TeX control sequences will almost surely not be handled -correctly. If \f(CW\*(C`\-\-rpi\-is\-xml\*(C'\fR is given, the author and title strings -from the rpi files are output as-is, assuming they are valid XML; no -checking is done. Citation text from \f(CW\*(C`.bbl\*(C'\fR files is always converted -from LaTeX to plain text. +correctly. .PP -This script just writes an XML file. It's up to you to actually do the +If \f(CW\*(C`\-\-rpi\-is\-xml\*(C'\fR is given, the author and title strings from the rpi +files are output as-is, assuming they are valid XML; no checking is +done. +.PP +Citation text from \f(CW\*(C`.bbl\*(C'\fR files is always converted from LaTeX to plain +text. +.PP +This script just writes an XML file. It's up to you to do the uploading to Crossref; for example, you can use their Java tool \&\f(CW\*(C`crossref\-upload\-tool.jar\*(C'\fR (<https://www.crossref.org/education/member\-setup/direct\-deposit\-xml/https\-post>). -For the definition of their schema, see -<https://data.crossref.org/reports/help/schema_doc/4.4.2/index.html> -(this is the schema version currently followed by this script). +.PP +For the definition of the Crossref schema currently output by this +script, see +<https://data.crossref.org/reports/help/schema_doc/5.3.1/index.html> +with additional links and information at +<https://www.crossref.org/documentation/schema\-library/metadata\-deposit\-schema\-5\-3\-1/>. .SH "CONFIGURATION FILE FORMAT" .IX Header "CONFIGURATION FILE FORMAT" The configuration file is read as Perl code. Thus, comment lines @@ -188,7 +197,8 @@ Any other journal must always specify this. The \f(CW%authors\fR field is split at \f(CW\*(C`\eand\*(C'\fR (ignoring whitespace before and after), and output as the \f(CW\*(C`contributors\*(C'\fR element, using \&\f(CW\*(C`sequence="first"\*(C'\fR for the first listed, \f(CW\*(C`sequence="additional"\*(C'\fR for -the remainder. +the remainder. The authors are parsed using \f(CW\*(C`BibTeX::Parser::Author\*(C'\fR +(<https://ctan.org/pkg/bibtexperllibs>). .PP If the \f(CW%publicationType\fR is not specified, it defaults to \&\f(CW\*(C`full_text\*(C'\fR, since that has historically been the case; \f(CW\*(C`full_text\*(C'\fR @@ -200,8 +210,7 @@ from the given \f(CW\*(C`journal_article\*(C'\fR element. Each \f(CW\*(C`.rpi\*(C'\fR must contain information for only one article, but multiple files can be read in a single run. It would not be difficult to support multiple articles in a single \f(CW\*(C`.rpi\*(C'\fR file, but it makes debugging and -error correction easier when each uploaded XML contains a single -article. +error correction easier to keep the input to one article per file. .SS "MORE ABOUT AUTHOR NAMES" .IX Subsection "MORE ABOUT AUTHOR NAMES" The three formats for names recognized are (not coincidentally) the same @@ -220,7 +229,8 @@ separators are not supported, unlike BibTeX. In short, you may almost always use the first form; you shouldn't if either there's a Jr part, or the Last part has multiple tokens but there's no von part. See the \f(CW\*(C`btxdoc\*(C'\fR (``BibTeXing'' by Oren Patashnik) -document for details. +document for details. The authors are parsed using +\&\f(CW\*(C`BibTeX::Parser::Author\*(C'\fR (<https://ctan.org/pkg/bibtexperllibs>). .PP In the \f(CW%authors\fR line of a \f(CW\*(C`.rpi\*(C'\fR file, some secondary directives are recognized, indicated by \f(CW\*(C`|\*(C'\fR characters. Easiest to explain with an @@ -284,12 +294,17 @@ this same format. .PP Feature request: if anyone is interested in figuring out how to generate structured citations -(<https://data.crossref.org/reports/help/schema_doc/5.3.1/schema_5_3_1.html#citation>) -instead of these flat text dumps, that would be great. Except the schema -seems to support much less than described at -<https://www.crossref.org/documentation/principles\-practices/best\-practices/bibliographic/>? -Anyway, the most viable approach is probably to change tugboat.bst to -output no-op TeX commands like \etubibauthor, \etubibtitle, etc. (a la +(<https://data.crossref.org/reports/help/schema_doc/5.3.1/common5_3_1_xsd.html#citation>), +that would be great. The schema does not support many useful fields, so +we also want to keep the unstructured text output. +.PP +Norman Gray's beastie program (<https://heptapod.host/nxg/beastie>) +supports this, via \f(CW\*(C`beastie extract\-bib.scm \-O crossref $(doc).aux\*(C'\fR, +as invoked in the TUGboat \f(CW\*(C`Common.mak\*(C'\fR file. Work in progress. +.PP +By the way, if for some reason we have to switch away from using +beastie, the most viable approach is probably to change \f(CW\*(C`tugboat.bst\*(C'\fR +to output no-op TeX commands like \etubibauthor, \etubibtitle, etc. (a la biblatex), and use those commands to discern the various crossref field values. We can't start from the .bib because then we'd have to reimplement Bib(La)TeX. diff --git a/support/crossrefware/ltx2crossrefxml.pl b/support/crossrefware/ltx2crossrefxml.pl index 5d8e16b383..80c7218758 100755 --- a/support/crossrefware/ltx2crossrefxml.pl +++ b/support/crossrefware/ltx2crossrefxml.pl @@ -43,8 +43,10 @@ ignored, and I<latex_file> itself is not read (and need not even exist). Each C<.rpi> file specifies the metadata for a single article to be uploaded to Crossref (a C<journal_article> element in their schema); an example is below. These files are output by the C<resphilosophica> -package (L<https://ctan.org/pkg/resphilosophica>), but (as always) can -also be created by hand or by whatever other method you implement. +package (L<https://ctan.org/pkg/resphilosophica>) and the TUGboat +publication procedure (L<https://tug.org/TUGboat/repository.html>), but +(as always) can also be created by hand or by whatever other method you +implement. Any C<.bbl> files present are used for the citation information in the output XML. See the L<CITATIONS> section below. @@ -54,18 +56,25 @@ citations), standard TeX control sequences are replaced with plain text or UTF-8 or eliminated, as appropriate. The C<LaTeX::ToUnicode::convert> routine is used for this (L<https://ctan.org/pkg/bibtexperllibs>). Tricky TeX control sequences will almost surely not be handled -correctly. If C<--rpi-is-xml> is given, the author and title strings -from the rpi files are output as-is, assuming they are valid XML; no -checking is done. Citation text from C<.bbl> files is always converted -from LaTeX to plain text. +correctly. -This script just writes an XML file. It's up to you to actually do the +If C<--rpi-is-xml> is given, the author and title strings from the rpi +files are output as-is, assuming they are valid XML; no checking is +done. + +Citation text from C<.bbl> files is always converted from LaTeX to plain +text. + +This script just writes an XML file. It's up to you to do the uploading to Crossref; for example, you can use their Java tool C<crossref-upload-tool.jar> (L<https://www.crossref.org/education/member-setup/direct-deposit-xml/https-post>). -For the definition of their schema, see -L<https://data.crossref.org/reports/help/schema_doc/4.4.2/index.html> -(this is the schema version currently followed by this script). + +For the definition of the Crossref schema currently output by this +script, see +L<https://data.crossref.org/reports/help/schema_doc/5.3.1/index.html> +with additional links and information at +L<https://www.crossref.org/documentation/schema-library/metadata-deposit-schema-5-3-1/>. =head1 CONFIGURATION FILE FORMAT @@ -136,7 +145,8 @@ Any other journal must always specify this. The C<%authors> field is split at C<\and> (ignoring whitespace before and after), and output as the C<contributors> element, using C<sequence="first"> for the first listed, C<sequence="additional"> for -the remainder. +the remainder. The authors are parsed using C<BibTeX::Parser::Author> +(L<https://ctan.org/pkg/bibtexperllibs>). If the C<%publicationType> is not specified, it defaults to C<full_text>, since that has historically been the case; C<full_text> @@ -148,8 +158,7 @@ from the given C<journal_article> element. Each C<.rpi> must contain information for only one article, but multiple files can be read in a single run. It would not be difficult to support multiple articles in a single C<.rpi> file, but it makes debugging and -error correction easier when each uploaded XML contains a single -article. +error correction easier to keep the input to one article per file. =head2 MORE ABOUT AUTHOR NAMES @@ -167,7 +176,8 @@ separators are not supported, unlike BibTeX. In short, you may almost always use the first form; you shouldn't if either there's a Jr part, or the Last part has multiple tokens but there's no von part. See the C<btxdoc> (``BibTeXing'' by Oren Patashnik) -document for details. +document for details. The authors are parsed using +C<BibTeX::Parser::Author> (L<https://ctan.org/pkg/bibtexperllibs>). In the C<%authors> line of a C<.rpi> file, some secondary directives are recognized, indicated by C<|> characters. Easiest to explain with an @@ -230,12 +240,17 @@ this same format. Feature request: if anyone is interested in figuring out how to generate structured citations -(L<https://data.crossref.org/reports/help/schema_doc/5.3.1/schema_5_3_1.html#citation>) -instead of these flat text dumps, that would be great. Except the schema -seems to support much less than described at -L<https://www.crossref.org/documentation/principles-practices/best-practices/bibliographic/>? -Anyway, the most viable approach is probably to change tugboat.bst to -output no-op TeX commands like \tubibauthor, \tubibtitle, etc. (a la +(L<https://data.crossref.org/reports/help/schema_doc/5.3.1/common5_3_1_xsd.html#citation>), +that would be great. The schema does not support many useful fields, so +we also want to keep the unstructured text output. + +Norman Gray's beastie program (L<https://heptapod.host/nxg/beastie>) +supports this, via C<beastie extract-bib.scm -O crossref $(doc).aux>, +as invoked in the TUGboat C<Common.mak> file. Work in progress. + +By the way, if for some reason we have to switch away from using +beastie, the most viable approach is probably to change C<tugboat.bst> +to output no-op TeX commands like \tubibauthor, \tubibtitle, etc. (a la biblatex), and use those commands to discern the various crossref field values. We can't start from the .bib because then we'd have to reimplement Bib(La)TeX. @@ -273,7 +288,7 @@ extent permitted by law. # find files relative to our installed location within TeX Live chomp(my $TLMaster = `kpsewhich -var-value=TEXMFROOT`); # TL root if (length($TLMaster)) { - unshift @INC, "$TLMaster/texmf-dist/scripts/bibtexperllibs"; + unshift @INC, "$TLMaster/texmf-dist/scripts/bibtexperllibs"; } # find development bibtexperllibs in sibling checkout to this script, # even if $0 is a symlink. All irrelevant when using from an installation. @@ -292,15 +307,17 @@ extent permitted by law. my $USAGE = <<END; Usage: $0 [-c CONFIG] [-o OUTPUT] [--rpi-is-xml] LTXFILE... -Convert .rpi and (if any are present) .bbl files corresponding to each -LTXFILE to xml, for submitting to crossref.org. The LTXFILE is not read -(and need not even exist); any extension it has is replaced by .rpi and -.bbl. +Convert .rpi and (if any are present) .bbl and .crbib files +corresponding to each LTXFILE to xml, for submitting to crossref.org. +The LTXFILE is not read, and need not even exist; any extension given is +replaced by .rpi, .bbl, .crbib. The .rpi files are plain text, with values on lines beginning with %, as output by (for example) the resphilosophica LaTeX package. The .bbl -files are as output by BibTeX. Both are also commonly created by hand. -The documentation for this script has examples. +files are as output by BibTeX. The .crbib files are xml files ready for +incorporation in the final xml, as output by the beastie program. All +may also be created by other methods. The documentation for this script +has examples. The xml is written to standard output by default; the -o (--output) option overrides this. @@ -314,6 +331,8 @@ For an example of using this script and associatd code, see the TUGboat processing at https://github.com/TeXUsersGroup/tugboat/tree/trunk/capsules/crossref. +This script depends on https://github.com/borisveytsman/bibtexperllibs. + Development sources, bug tracker: https://github.com/borisveytsman/crossrefware Releases: https://ctan.org/pkg/crossrefware END @@ -343,6 +362,7 @@ END use utf8; binmode(STDOUT, ":utf8"); + ################################################################ # Defaults and parameters ################################################################ @@ -365,7 +385,7 @@ END our $timestamp = strftime("%Y%m%d%H%M%S", gmtime); # use timestamp in batchid, since the value is supposed to be unique # for every submission to crossref by a given publisher. - # https://data.crossref.org/reports/help/schema_doc/4.4.2/schema_4_4_2.html#doi_batch_id + # https://data.crossref.org/reports/help/schema_doc/5.3.1/common5_3_1_xsd.html#doi_batch_id our $batchId="ltx2crossref-$timestamp-$$"; if ($opts{c}) { @@ -373,9 +393,9 @@ END # if config arg is absolute, fine; if not, prepend "./" as slightly # less troublesome than putting "." in the @INC path. my $rel = (File::Spec->file_name_is_absolute($opts{c}) ? "" : "./"); - require "$rel$opts{c}"; + require "$rel$opts{c}"; } else { - die "Cannot read config file $opts{c}. Goodbye."; + die "Cannot read config file $opts{c}. Goodbye."; } } @@ -386,30 +406,32 @@ END # my %papers; + # Read the papers. foreach my $file (@ARGV) { AddPaper($file); } + # Write the papers. foreach my $year (keys %papers) { foreach my $volume (keys %{$papers{$year}}) { - foreach my $issue (keys %{$papers{$year}->{$volume}}) { - PrintIssueHead($year, $volume, $issue); - my $paperList = $papers{$year}->{$volume}->{$issue}; + foreach my $issue (keys %{$papers{$year}->{$volume}}) { + PrintIssueHead($year, $volume, $issue); + my $paperList = $papers{$year}->{$volume}->{$issue}; #warn "papers for year=$year, volume=$volume, issue=$issue\n"; # Nice to have the issue.xml in some stable order, so sort # by starting page. Doesn't matter if it's not perfect. - foreach my $paper (sort { $a->{startpage} <=> $b->{startpage} } - @{$paperList}) { - PrintPaper($paper); - } - } + foreach my $paper (sort { $a->{startpage} <=> $b->{startpage} } + @{$paperList}) { + PrintPaper($paper); + } + } } } PrintTail(); exit($ERROR_COUNT); - + ##################################################### # Printing the head and the tail ##################################################### @@ -422,12 +444,10 @@ sub PrintHead { ? "\n$indent<abbrev_title>$abbrevTitle</abbrev_title>" : ""; - # as of schema version 4.3.4, crossref renamed the <name> element - # inside <depositor> to <depositor_name>. Sigh. Something to take - # into account with older schemas. - # https://www.crossref.org/education/content-registration/crossrefs-metadata-deposit-schema/schema-versions/ + # Crossref schema info: + # https://www.crossref.org/documentation/schema-library/schema-versions/ print OUT <<END; -<doi_batch xmlns="http://www.crossref.org/schema/4.4.2" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="4.4.2" xsi:schemaLocation="http://www.crossref.org/schema/4.4.2 http://www.crossref.org/schema/deposit/crossref4.4.2.xsd"> +<doi_batch xmlns="http://www.crossref.org/schema/5.3.1" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="5.3.1" xsi:schemaLocation="http://www.crossref.org/schema/5.3.1 http://www.crossref.org/schema/deposit/crossref5.3.1.xsd"> <head> <doi_batch_id>$batchId</doi_batch_id> <timestamp>$timestamp</timestamp> @@ -440,7 +460,7 @@ sub PrintHead { <body><journal> <journal_metadata language="en"> <full_title>$fullTitle</full_title>$abbrev_title_out - <issn>$issn</issn>$coden_out + <issn>$issn</issn>$coden_out </journal_metadata> END } @@ -454,20 +474,20 @@ END return; } - + ####################################################### -# Adding one paper from $file.rpi and .bbl to global %papers. +# Adding one paper from $file.rpi and .bbl and .crbib to global %papers. ####################################################### sub AddPaper { my $file = shift; my ($name,$path,$suffix) = fileparse($file, '\.[^\.]*$'); my $rpifile = File::Spec->catfile($path, "$name.rpi"); open (RPI, $rpifile) - or die "open($rpifile) failed: $! (did you process $file?)\n"; + or die "$0: open($rpifile) failed: $! (did you process $file?)\n"; my %data; #warn "reading rpi file: $rpifile\n"; while (<RPI>) { - chomp; + chomp; if (/^%([^=]*)\s*=\s*(.*)\s*$/) { if (exists $data{$1}) { warn "$rpifile:$.: already saw data{$1}=$data{$1};" @@ -479,12 +499,15 @@ sub AddPaper { } close RPI; - # look for bibliographies in both the .rpi and any .bbl file. + # also look for bibliographies in FILE.bbl and FILE.crbib files. my @bibliography; - foreach my $bibfile ($file, File::Spec->catfile($path, "$name.bbl")) { - @bibliography = (@bibliography, AddBibliography($bibfile)); + foreach my $bblfile ($rpifile, File::Spec->catfile($path, "$name.bbl")) { + push (@bibliography, AddBibliography($bblfile)); } $data{'bibliography'} = \@bibliography; + # + $data{'crbib'} + = AddCrossrefBib (File::Spec->catfile($path, "$name.crbib")); # Die if the fields we use unconditionally are empty. Not all of # them are required by the schema, but we can wait to generalize. @@ -500,13 +523,15 @@ sub AddPaper { push @{$papers{$data{year}}->{$data{volume}}->{$data{issue}}}, \%data; } + ############################################################## -# Reading a list of papers from BIBFILE and adding it to the +# Read a list of references from BIBFILE and adding it to the # bibliography. Each item is assumed to start with # \bibitem{KEY} and the whole bib to end with \end{thebibliography}. # # We return a list of hashes, each hash with a single key, the citation -# key, and its value a flat string of the entry. +# key with an integer (starting at 1), and its value a flat string of +# the entry. # # No conversion of the text is done here. ############################################################## @@ -515,59 +540,59 @@ sub AddBibliography { open (BIB, $bibfile) or return; my $insidebibliography = 0; - my $currpaper = ""; # that is, the current bib entry + my $currpaper = ""; # the current bib entry my $bibno = 0; my @result; my $key; while (<BIB>) { - chomp; - next if /^\s*%/; # TeX comment line - s/[ \t]%.*//; # remove TeX comment - # - # allow empty \bibitem key for the sake of handwritten bbls. - # Similarly, might be more stuff on the line when handwritten. - # Ignore a TeX %comment following. - if (s/^\s*\\bibitem(?:\[.*?\])?+\s*\{(.*?)\}\s*(%.*$)?//) { - my $newkey = $1; - if ($insidebibliography) { - if ($currpaper) { + chomp; + next if /^\s*%/; # TeX comment line + s/[ \t]%.*//; # remove TeX comment + # + # allow empty \bibitem key for the sake of handwritten bbls. + # Similarly, might be more stuff on the line when handwritten. + # Ignore a TeX %comment following. + if (s/^\s*\\bibitem(?:\[.*?\])?+\s*\{(.*?)\}\s*(%.*$)?//) { + my $newkey = $1; + if ($insidebibliography) { + if ($currpaper) { # Append the current sequence number for this citation, # since that's what Crossref recommends (sort of). # For prettiness, if the key is otherwise empty, # don't include a dash beforehand. - $bibno++; + $bibno++; $key .= ($key ? "-" : "") . $bibno; # my %paperhash; - $paperhash{$key} = $currpaper; - push @result, \%paperhash; - } - } - # The citation key (required by schema) starts as the bibitem key. - $key = $newkey; - - $currpaper = $_; - $insidebibliography = 1; - next; - } - if (/^\s*\\end\{thebibliography\}/) { - if ($currpaper) { - $bibno++; + $paperhash{$key} = $currpaper; + push @result, \%paperhash; + } + } + # The citation key (required by schema) starts as the bibitem key. + $key = $newkey; + + $currpaper = $_; + $insidebibliography = 1; + next; + } + if (/^\s*\\end\{thebibliography\}/) { + if ($currpaper) { + $bibno++; $key .= ($key ? "-" : "") . $bibno; # - my %paperhash; - $paperhash{$key} = $currpaper; - push @result, \%paperhash; - } - $currpaper = ""; - $insidebibliography = 0; - next; - } - if ($insidebibliography) { - $currpaper .= " $_"; - } + my %paperhash; + $paperhash{$key} = $currpaper; + push @result, \%paperhash; + } + $currpaper = ""; + $insidebibliography = 0; + next; + } + if ($insidebibliography) { + $currpaper .= " $_"; + } } - close BIB; + close BIB or warn "close($bibfile) failed: $!"; # We look in the .rpi files too, which will generally have none. if (@result == 0 && $bibfile =~ /\.bbl$/) { @@ -580,6 +605,75 @@ sub AddBibliography { return @result; } + +############################################################## +# Read an XML <citation_list> element from CRBIBFILE, if it exists. +# No error if it doesn't exist; it often won't, even if there is a bbl file. +# +# Return a hash reference, with each element's key being the citation +# key plus an integer, the same keys as in AddBibliography from the .bbl +# file.# Each value is a flat string, the structured citation items for +# that element. +# +# We ignore any <unstructured_citation> element, since we generate our +# own (which we prefer). +# +# We don't parse XML, just extract the pieces with regexps. +# This is generated by Norman Gray's beastie program. Example: +# <citation_list> +# <citation key="bookshelf"> +# <author>Peter Flynn</author> +# <volume_title>The bookshelf package</volume_title> +# <cYear>2020</cYear> +# <unstructured_citation>Flynn, Peter (manual): The bookshelf package[...] +# </citation> +# <citation key="Calibre"> +# <author>Kovid Goyal</author> +# <volume_title>calibre User Manual</volume_title> +# <cYear>2024</cYear> +# <unstructured_citation>Kovid Goyal (manual): calibre User Manual[...] +# </citation> +# </citation_list> +############################################################## +sub AddCrossrefBib { + my ($crbibfile,$refs) = @_; + my %result; + + #warn "crbibfile=$crbibfile\n"; + open (CRBIB, $crbibfile) or return; + + # read whole file. + my $crbib_as_string = join("", <CRBIB>); + #warn "doing crbib $crbibfile; $crbib_as_string\n"; + close (CRBIB) or warn "close($crbibfile) failed: $!"; + + my $bibno = 0; + + # We're matching each <citation> here by virtue of .*? to be a + # non-greedy match, the /s modifier to treat the whole thing as one + # string, and the /g modifier to return an array of all matches. + my @crbib = ($crbib_as_string =~ m,<citation\s+(key=.*?)</citation>,sg); + for my $crb (@crbib) { + $bibno++; + + # wipe out the unstructured text. + $crb =~ s,\s*<unstructured_citation>.*</unstructured_citation>\s*,,; + + $crb = SanitizeTextNoEntities($crb); + + # qqq undone - must save by key, then write by key into the xml. + # need to be able to clean the text, beastie removes braces. + warn "crb $bibno: $crb\n"; + } + + if ($bibno == 0) { + warn "$0: *** no crossref cites found in: $crbibfile; check if ok\n"; + } + + return %result; +} + + ################################################################# # Printing information about one issue ################################################################# @@ -596,8 +690,9 @@ sub PrintIssueHead { END } + ############################################################### -# Printing information about one paper +# Printing information about one paper ############################################################### sub PrintPaper { my $paper = shift; @@ -617,7 +712,7 @@ END my @authors = split /\s*\\and\s*/, $paper->{authors}; my $seq = 'first'; foreach my $author (@authors) { - PrintAuthor($author, $seq); + PrintAuthor($author, $seq); $seq = 'additional'; } @@ -638,15 +733,7 @@ END END if (scalar(@{$paper->{bibliography}})) { - print OUT <<END; - <citation_list> -END - foreach my $citation (@{$paper->{bibliography}}) { - PrintCitation($citation); - } - print OUT <<END; - </citation_list> -END + PrintCitationList($paper->{bibliography}, $paper->{crbib}); } print OUT <<END; @@ -654,16 +741,15 @@ END END } - + ############################################################### # Crossref <title> strings can contain a few so-called "face" HTML # commands. Complain if they have anything anything else. -# schema doc: https://data.crossref.org/reports/help/schema_doc/4.4.2/schema_4_4_2.html#title -# face doc: https://www.crossref.org/education/content-registration/crossrefs-metadata-deposit-schema/face-markup/ -# mathml doc: https://www.crossref.org/education/content-registration/crossrefs-metadata-deposit-schema/including-mathml-in-deposits/ +# schema doc: https://data.crossref.org/reports/help/schema_doc/5.3.1/crossref5_3_1_xsd.html#title +# face doc: https://www.crossref.org/documentation/schema-library/markup-guide-metadata-segments/face-markup/ # # We don't technically validate the string, e.g., mismatched tags will -# go unnoticed here. The real validator at Crossref will catch whatever. +# go unnoticed here. The real validator at Crossref should catch everything. ############################################################### sub TitleCheck { my $title = shift; @@ -686,33 +772,7 @@ sub TitleCheck { } } -############################################################### -# Simplistic TeX-to-html -# (no-op for rpi text if --input-is-xml was given). -############################################################### -sub SanitizeText { - my $string = shift; - return $string if $opts{xi}; # do nothing if --rpi-is-xml - return SanitizeTextAlways($string); -} - -# Split into two functions so we can sanitize bbl but not rpi. -sub SanitizeTextAlways { - my $string = shift; - - # pass user hook subroutine if defined. - my @hook = (defined(&{"LaTeX_ToUnicode_convert_hook"})) - ? ("hook" => \&LaTeX_ToUnicode_convert_hook) - : (); - - # conversion of accented control sequences to characters, etc. - # Let's use &#uuuu; entities instead of literal UTF-8; Crossref - # recommends it, and it's easier for postprocessing. - $string = LaTeX::ToUnicode::convert($string, entities => 1, @hook); - - return $string; -} - + ################################################################ # Printing one author in arg ORIG_AUTHOR, in sequence SEQ. ################################################################ @@ -809,27 +869,45 @@ END END } + ############################################################# -# Printing citations +# Print citations in order from BIBLIOGRAPHY, a list reference, and +# CRBIB, a hash reference. Each element in BIBLIOGRAPHY is a +# one-element hash, with the key being the citation key and the value +# the (original) bbl text. We sanitize (de-texify) the text. +# Each element in CRBIB has key the citation key (from the same set) +# and value the structured citation string from any .crbib file. +# ############################################################# -sub PrintCitation { - my $paperhash=shift; - - foreach my $key (keys (%{$paperhash})) { - my $citation = $paperhash->{$key}; - $citation = SanitizeTextAlways($citation); - - print OUT <<END; - <citation key="$key"><unstructured_citation> - $citation - </unstructured_citation></citation> +sub PrintCitationList { + my ($bibliography,$crbib) = shift; + + print OUT " <citation_list>\n"; + foreach my $citation_hash (@$bibliography) { + foreach my $citekey (keys (%{$citation_hash})) { # only one key + my $citation_text = $citation_hash->{$citekey}; + $citation_text = SanitizeTextAlways($citation_text); + + #warn " printing citation $citekey: $citation_text\n"; + my $structured_citation = ""; + if ($crbib->{$citekey}) { + $structured_citation = "\n" . " "x10 . $crbib->{$citekey}; + warn " with structured citation: $structured_citation\n"; + } + print OUT <<END; + <citation key="$citekey">$structured_citation + <unstructured_citation>$citation_text</unstructured_citation> + </citation> END + } } + print OUT " </citation_list>\n"; } + ############################################################## # Return publication_type attribute for <journal_article>, given $PUBTYPE. -# https://data.crossref.org/reports/help/schema_doc/4.4.2/schema_4_4_2.html#publication_type.atts +# https://data.crossref.org/reports/help/schema_doc/5.3.1/crossref5_3_1_xsd.html#publication_type.atts_publication_type # # If not specified in input, return " publication_type=full_text" since # it was hardwired that way before. If set to "omit", return empty @@ -862,11 +940,11 @@ sub GetURL { my $result; if ($paper->{paperUrl}) { - $result = $paper->{paperUrl} + $result = $paper->{paperUrl} } elsif ($paper->{doi} =~ m,^10\.11612/resphil,) { - my $doi = $paper->{doi}; - $result = 'http://www.pdcnet.org/oom/service?url_ver=Z39.88-2004&rft_val_fmt=&rft.imuse_synonym=resphilosophica&rft.DOI='.$doi.'&svc_id=info:www.pdcnet.org/collection'; + my $doi = $paper->{doi}; + $result = 'http://www.pdcnet.org/oom/service?url_ver=Z39.88-2004&rft_val_fmt=&rft.imuse_synonym=resphilosophica&rft.DOI='.$doi.'&svc_id=info:www.pdcnet.org/collection'; } else { die ("$0: paperUrl field is required\n " @@ -877,7 +955,41 @@ sub GetURL { return $result; } + +############################################################### +# Simplistic TeX-to-html +# (no-op for rpi text if --input-is-xml was given). +############################################################### +sub SanitizeText { + my $string = shift; + return $string if $opts{xi}; # do nothing if --rpi-is-xml + return SanitizeTextEntities($string); +} + +# Conversion of accented control sequences to characters, etc. +# This uses &#uuuu; entities instead of literal UTF-8; Crossref +# recommends it, and it's easier for postprocessing. +# +sub SanitizeTextEntities { + my $string = shift; + return SanitizeTextNoEntities($string, entities => 1, @_); +} + +# Generic sanitize text. +sub SanitizeTextNoEntities { + my $string = shift; + + # pass user hook subroutine if defined. + my @hook = (defined(&{"LaTeX_ToUnicode_convert_hook"})) + ? ("hook" => \&LaTeX_ToUnicode_convert_hook) + : (); + + $string = LaTeX::ToUnicode::convert($string, @hook, @_); + + return $string; +} + ############################################################## # debug_hash_as_string($LABEL, HASH) # |