diff options
| -rw-r--r-- | doc/LinuxDocTools.pm | 637 | ||||
| -rw-r--r-- | doc/Makefile | 10 | ||||
| -rwxr-xr-x | doc/sgml2html | 53 | ||||
| -rwxr-xr-x | doc/sgml2latex | 53 | ||||
| -rwxr-xr-x | doc/sgml2txt | 53 | ||||
| -rwxr-xr-x | tools/linuxdoc | 75 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools.pm | 668 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/BackEnd.pm | 185 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/CharEnts.pm | 176 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/FixRef.pm | 76 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/Html2Html.pm | 583 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/InfoUtils.pm | 357 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/Lang.pm | 238 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/Utils.pm | 392 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/LinuxDocTools/Vars.pm | 22 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/Text/EntityMap.pm | 121 | ||||
| -rw-r--r-- | tools/linuxdoc-tools/copyright | 85 |
17 files changed, 2983 insertions, 801 deletions
diff --git a/doc/LinuxDocTools.pm b/doc/LinuxDocTools.pm deleted file mode 100644 index d32f3171..00000000 --- a/doc/LinuxDocTools.pm +++ /dev/null @@ -1,637 +0,0 @@ -#! /usr/bin/perl -# -# LinuxDocTools.pm -# -# $Id$ -# -# LinuxDoc-Tools driver core. This contains all the basic functionality -# we need to control all other components. -# -# © Copyright 1996, Cees de Groot. -# © Copyright 2000, Taketoshi Sano -# -# THIS VERSION HAS BEEN HACKED FOR BIRD BY MARTIN MARES -# -package LinuxDocTools; - -require 5.004; -use strict; - -=head1 NAME - -LinuxDocTools - SGML conversion utilities for LinuxDoc DTD. - -=head1 SYNOPSIS - - use LinuxDocTools; - LinuxDocTools::init; - @files = LinuxDocTools::process_options ($0, @ARGV); - for $curfile (@files) { - LinuxDocTools::process_file ($curfile); - } - -=head1 DESCRIPTION - -The LinuxDocTools package encapsulates all the functionality offered by -LinuxDoc-Tools. It is used, of course, by LinuxDoc-Tools; -but the encapsulation should provide for a simple interface for other users as well. - -=head1 FUNCTIONS - -=over 4 - -=cut - -use DirHandle; -use File::Basename; -use File::Find; -use File::Copy; -use FileHandle; -use IPC::Open2; -use Cwd; -use LinuxDocTools::Lang; -use LinuxDocTools::Utils qw(process_options usage cleanup trap_signals remove_tmpfiles create_temp); -use LinuxDocTools::Vars; - -sub BEGIN -{ - # - # Make sure we're always looking here. Note that "use lib" adds - # on the front of the search path, so we first push dist, then - # site, so that site is searched first. - # - use lib "$main::DataDir/dist"; - use lib "$main::DataDir/site"; -} - -=item LinuxDocTools::init - -Takes care of initialization of package-global variables (which are actually -defined in L<LinuxDocTools::Vars>). The package-global variables are I<$global>, -a reference to a hash containing numerous settings, I<%Formats>, a hash -containing all the formats, and I<%FmtList>, a hash containing the currently -active formats for help texts. - -Apart from this, C<LinuxDocTools::init> also finds all distributed and site-local -formatting backends and C<require>s them. - -=cut - -sub init -{ - trap_signals; - - # - # Register the ``global'' pseudoformat. Apart from the global settings, - # we also use $global to keep the global variable name space clean; - # everything that we need to provide to other modules is stuffed - # into $global. - # - $global = {}; - $global->{NAME} = "global"; - $global->{HELP} = ""; - $global->{OPTIONS} = [ - { option => "backend", type => "l", - 'values' => [ "html", "info", "latex", - "lyx", "rtf", "txt", "check" ], - short => "B" }, - { option => "papersize", type => "l", - 'values' => [ "a4", "letter" ], short => "p" }, - { option => "language", type => "l", - 'values' => [ @LinuxDocTools::Lang::Languages ], short => "l" }, - { option => "charset", type => "l", - 'values' => [ "latin", "ascii", "nippon", "euc-kr" ], short => "c" }, - { option => "style", type => "s", short => "S" }, - { option => "tabsize", type => "i", short => "t" }, -# { option => "verbose", type => "f", short => "v" }, - { option => "debug", type => "f", short => "d" }, - { option => "define", type => "s", short => "D" }, - { option => "include", type => "s", short => "i" }, - { option => "pass", type => "s", short => "P" } - ]; - $global->{backend} = "linuxdoc"; - $global->{papersize} = "a4"; - $global->{language} = "en"; - $global->{charset} = "ascii"; - $global->{style} = ""; - $global->{tabsize} = 8; - $global->{verbose} = 0; - $global->{define} = ""; - $global->{debug} = 0; - $global->{include} = ""; - $global->{pass} = ""; - $global->{InFiles} = []; - $Formats{$global->{NAME}} = $global; # All formats we know. - $FmtList{$global->{NAME}} = $global; # List of formats for help msgs. - - # automatic language detection: disabled by default - # { - # my $lang; - # foreach $lang (@LinuxDocTools::Lang::Languages) - # { - # if (($ENV{"LC_ALL"} =~ /^$lang/i) || - # ($ENV{"LC_CTYPE"} =~ /^$lang/i) || - # ($ENV{"LANG"} =~ /^$lang/i)) { - # $global->{language} = Any2ISO($lang); - # } - # } - # } - - # - # Used when the format is "global" (from sgmlcheck). - # - $global->{preNSGMLS} = sub { - $global->{NsgmlsOpts} .= " -s "; - $global->{NsgmlsPrePipe} = "cat $global->{file}"; - }; - - # - # Build up the list of formatters. - # - my $savdir = cwd; - my %Locs; - chdir "$main::DataDir/dist"; - my $dir = new DirHandle("."); - die "Unable to read directory $main::DataDir/dist: $!" unless defined($dir); - foreach my $fmt (grep(/^fmt_.*\.pl$/, $dir->read())) - { - $Locs{$fmt} = "dist"; - } - $dir->close(); - chdir "$main::DataDir/site"; - $dir = new DirHandle("."); - die "Unable to read directory $main::DataDir/site: $!" unless defined($dir); - foreach my $fmt (grep(/^fmt_.*\.pl$/, $dir->read())) - { - $Locs{$fmt} = "site"; - } - $dir->close(); - foreach my $fmt (keys %Locs) - { - require $fmt; - } - chdir $savdir; -} - -=item LinuxDocTools::process_options ($0, @ARGV) - -This function contains all initialization that is bound to the current -invocation of LinuxDocTools. It looks in C<$0> to deduce the backend that -should be used (ld2txt activates the I<txt> backend) and parses the -options array. It returns an array of filenames it encountered during -option processing. - -As a side effect, the environment variables I<SGMLDECL> and -I<SGML_CATALOG_FILES> are modified. - -=cut - -sub process_options -{ - my $progname = shift; - my @args = @_; - - # - # Deduce the format from the caller's file name - # - my ($format, $dummy1, $dummy2) = fileparse ($progname, ""); - $global->{myname} = $format; - $format =~ s/sgml2*(.*)/$1/; - - # - # check the option "--backend / -B" - # - if ($format eq "linuxdoc") { - my @backends = @args; - my $arg; - while (@backends) { - $arg = shift @backends; - if ($arg eq "-B") { - $arg = shift @backends; - $format = $arg; - last; - } - if ( $arg =~ s/--backend=(.*)/$1/ ) { - $format = $arg; - last; - } - } - } - - $format = "global" if $format eq "check"; - usage ("") if $format eq "linuxdoc"; - $format = "latex2e" if $format eq "latex"; - $FmtList{$format} = $Formats{$format} or - usage ("$global->{myname}: unknown format"); - $global->{format} = $format; - - # - # Parse all the options. - # - my @files = LinuxDocTools::Utils::process_options (@args); - $global->{language} = Any2ISO ($global->{language}); - # - # check the number of given files - $#files > -1 || usage ("no filenames given"); - - # - # Setup the SGML environment. - # (Note that Debian package rewrite path to catalog of - # iso-entities using debian/rules so that it can use - # entities from sgml-data pacakge. debian/rules also - # removes iso-entites sub directory after doing make install.) - # - $ENV{SGML_CATALOG_FILES} .= (defined $ENV{SGML_CATALOG_FILES} ? ":" : "") . - "$main::prefix/share/sgml/sgml-iso-entities-8879.1986/catalog:" . - "$main::prefix/share/sgml/entities/sgml-iso-entities-8879.1986/catalog"; - $ENV{SGML_CATALOG_FILES} .= ":$main::DataDir/linuxdoc-tools.catalog"; - $ENV{SGML_CATALOG_FILES} .= ":$main::/etc/sgml.catalog"; - if (-f "$main::DataDir/dtd/$format.dcl") - { - $ENV{SGMLDECL} = "$main::DataDir/dtd/$format.dcl"; - } - elsif (-f "$main::DataDir/dtd/$global->{style}.dcl") - { - $ENV{SGMLDECL} = "$main::DataDir/dtd/$global->{style}.dcl"; - } - elsif (-f "$main::DataDir/dtd/sgml.dcl") - { - $ENV{SGMLDECL} = "$main::DataDir/dtd/sgml.dcl"; - } - - # - # OK. Give the list of files we distilled from the options - # back to the caller. - # - return @files; -} - -=item LinuxDocTools::process_file - -With all the configuration done, this routine will take a single filename -and convert it to the currently active backend format. The conversion is -done in a number of steps in tight interaction with the currently active -backend (see also L<LinuxDocTools::BackEnd>): - -=over - -=item 1. Backend: set NSGMLS options and optionally create a pre-NSGMLS pipe. - -=item 2. Here: Run the preprocessor to handle conditionals. - -=item 3. Here: Run NSGMLS. - -=item 4. Backend: run pre-ASP conversion. - -=item 5. Here: Run SGMLSASP. - -=item 6. Backend: run post-ASP conversion, generating the output. - -=back - -All stages are influenced by command-line settings, currently active format, -etcetera. See the code for details. - -=cut - -sub process_file -{ - my $file = shift (@_); - my $saved_umask = umask; - - print "Processing file $file\n"; - umask 0077; - - my ($filename, $filepath, $filesuffix) = fileparse ($file, "\.sgml"); - my $tmpnam = $filepath . '/' . $filename; - $file = $tmpnam . $filesuffix; - -f $file || $file =~ /.*.sgml$/ || ($file .= '.sgml'); - -f $file || ($file = $tmpnam . '.SGML'); - -f $file || die "Cannot find $file\n"; - $global->{filename} = $filename; - $global->{file} = $file; - $global->{filepath} = $filepath; - - my $tmp = new FileHandle "<$file"; - my $dtd; - while ( <$tmp> ) - { - tr/A-Z/a-z/; - # check for [<!doctype ... system] type definition - if ( /<!doctype\s*(\w*)\s*system/ ) - { - $dtd = $1; - last; - } - # check for <!doctype ... PUBLIC ... DTD ... - if ( /<!doctype\s*\w*\s*public\s*.*\/\/dtd\s*(\w*)/mi ) - { - $dtd = $1; - last; - } - # check for <!doctype ... - # PUBLIC ... DTD ... - # (multi-line version) - if ( /<!doctype\s*(\w*)/ ) - { - $dtd = "precheck"; - next; - } - if ( /\s*public\s*.*\/\/dtd\s*(\w*)/ && $dtd eq "precheck" ) - { - $dtd = $1; - last; - } - } - $tmp->close; - if ( $global->{debug} ) - { - print "DTD: " . $dtd . "\n"; - } - $global->{dtd} = $dtd; - - # prepare temporary directory - my $tmpdir = $ENV{'TMPDIR'} || '/tmp'; - $tmpdir = $tmpdir . '/' . 'linuxdoc-dir-' . $$; - mkdir ($tmpdir, 0700) || - die " - temporary files can not be created, aborted - \n"; - - my $tmpbase = $global->{tmpbase} = $tmpdir . '/sgmltmp.' . $filename; - $ENV{"SGML_SEARCH_PATH"} .= ":$filepath"; - - # - # Set up the preprocessing command. Conditionals have to be - # handled here until they can be moved into the DTD, otherwise - # a validating SGML parser will choke on them. - # - # check if output option for latex is pdf or not - if ($global->{format} eq "latex2e") - { - if ($Formats{$global->{format}}{output} eq "pdf") - { - $global->{define} .= " pdflatex=yes"; - } - } - # - - local $ENV{PATH} = "$ENV{PATH}:/usr/lib/linuxdoc-tools"; - my($precmd) = "|sgmlpre output=$global->{format} $global->{define}"; - - # - # You can hack $NsgmlsOpts here, etcetera. - # - $global->{NsgmlsOpts} .= "-D $main::prefix/share/sgml -D $main::DataDir"; - $global->{NsgmlsOpts} .= "-i$global->{include}" if ($global->{include}); - $global->{NsgmlsPrePipe} = "NOTHING"; - if ( defined $Formats{$global->{format}}{preNSGMLS} ) - { - $global->{NsgmlsPrePipe} = &{$Formats{$global->{format}}{preNSGMLS}}; - } - - # - # Run the prepocessor and nsgmls. - # - my ($ifile, $writensgmls); - - if ($global->{NsgmlsPrePipe} eq "NOTHING") - { - $ifile = new FileHandle $file; - } - else - { - $ifile = new FileHandle "$global->{NsgmlsPrePipe}|"; - } - - create_temp("$tmpbase.1"); - $writensgmls = new FileHandle - "$precmd|$main::progs->{NSGMLS} $global->{NsgmlsOpts} $ENV{SGMLDECL} >\"$tmpbase.1\""; - if ($global->{charset} eq "latin") - { - while (<$ifile>) - { - # Outline these commands later on - CdG - #change latin1 characters to SGML - #by Farzad Farid, adapted by Greg Hankins - s/À/\À/g; - s/Á/\Á/g; - s/Â/\Â/g; - s/Ã/\Ã/g; - s/Ä/\Ä/g; - s/Å/\Å/g; - s/Æ/\Æ/g; - s/Ç/\Ç/g; - s/È/\È/g; - s/É/\É/g; - s/Ê/\Ê/g; - s/Ë/\Ë/g; - s/Ì/\Ì/g; - s/Í/\Í/g; - s/Î/\Î/g; - s/Ï/\Ï/g; - s/Ñ/\Ñ/g; - s/Ò/\Ò/g; - s/Ó/\Ó/g; - s/Ô/\Ô/g; - s/Õ/\Õ/g; - s/Ö/\Ö/g; - s/Ø/\Ø/g; - s/Ù/\Ù/g; - s/Ú/\Ú/g; - s/Û/\Û/g; - s/Ü/\Ü/g; - s/Ý/\Ý/g; - s/Þ/\Þ/g; - s/ß/\ß/g; - s/à/\à/g; - s/á/\á/g; - s/â/\â/g; - s/ã/\ã/g; - s/ä/\ä/g; - s/å/\å/g; - s/æ/\æ/g; - s/ç/\ç/g; - s/è/\è/g; - s/é/\é/g; - s/ê/\ê/g; - s/ë/\ë/g; - s/ì/\ì/g; - s/í/\í/g; - s/î/\î/g; - s/ï/\ï/g; - s/µ/\μ/g; - s/ð/\ð/g; - s/ñ/\ñ/g; - s/ò/\ò/g; - s/ó/\ó/g; - s/ô/\ô/g; - s/õ/\õ/g; - s/ö/\ö/g; - s/ø/\ø/g; - s/ù/\ù/g; - s/ú/\ú/g; - s/û/\û/g; - s/ü/\ü/g; - s/ý/\ý/g; - s/þ/\þ/g; - s/ÿ/\ÿ/g; - print $writensgmls $_; - } - } - else - { - while (<$ifile>) - { - print $writensgmls $_; - } - } - $ifile->close; - $writensgmls->close; - - # - # Special case: if format is global, we're just checking. - # - $global->{format} eq "global" && cleanup; - - # - # If the output file is empty, something went wrong. - # - ! -e "$tmpbase.1" and die "can't create file - exiting"; - -z "$tmpbase.1" and die "SGML parsing error - exiting"; - if ( $global->{debug} ) - { - print "Nsgmls stage finished.\n"; - } - - # - # If a preASP stage is defined, let the format handle it. - # - # preASP ($inhandle, $outhandle); - # - my $inpreasp = new FileHandle "<$tmpbase.1"; - my $outpreasp = new FileHandle "$tmpbase.2",O_WRONLY|O_CREAT|O_EXCL,0600; - if (defined $Formats{$global->{format}}{preASP}) - { - &{$Formats{$global->{format}}{preASP}}($inpreasp, $outpreasp) == 0 or - die "error pre-processing $global->{format}.\n"; - } - else - { - copy ($inpreasp, $outpreasp); - } - $inpreasp->close; - $outpreasp->close; - ! -e "$tmpbase.2" and die "can't create file - exiting"; - - if ( $global->{debug} ) - { - print "PreASP stage finished.\n"; - } - - # - # Run sgmlsasp, with an optional style if specified. - # - # Search order: - # - datadir/site/<dtd>/<format> - # - datadir/dist/<dtd>/<format> - # So we need to fetch the doctype from the intermediate. - # - # Note: this is a very simplistic check - but as far as I know, - # it is correct. Am I right? - # - my $tmp = new FileHandle "<$tmpbase.2"; - my $dtd; - while ( ($dtd = <$tmp>) && ! ( $dtd =~ /^\(/) ) { }; - $tmp->close; - $dtd =~ s/^\(//; - $dtd =~ tr/A-Z/a-z/; - chop $dtd; - $global->{dtd} = $dtd; - - my $style = ""; - if ($global->{style}) - { - $style = "$main::DataDir/site/$dtd/$global->{format}/$global->{style}mapping"; - -r $style or - $style = "$main::DataDir/dist/$dtd/$global->{format}/$global->{style}mapping"; - } - my $mapping = "$main::DataDir/site/$dtd/$global->{format}/mapping"; - -r $mapping or $mapping = "$main::DataDir/dist/$dtd/$global->{format}/mapping"; - - $global->{charset} = "nippon" if ($global->{language} eq "ja"); - # - # we don't have Korean groff so charset should be latin1. - # - if ($global->{language} eq "ko") - { - if ($global->{format} eq "groff") - { - $global->{charset} = "latin1"; - } - else - { - $global->{charset} = "euc-kr"; - } - } - - if ($global->{format} eq "groff" or $global->{format} eq "latex2e") - { - if ($dtd eq "linuxdoctr") - { - $mapping = "$main::DataDir/dist/$dtd/$global->{format}/tr-mapping"; - } - } - - create_temp("$tmpbase.3"); - system ("$main::progs->{SGMLSASP} $style $mapping <\"$tmpbase.2\" | - expand -$global->{tabsize} >\"$tmpbase.3\""); - ! -e "$tmpbase.3" and die "can't create file - exiting"; - - - if ( $global->{debug} ) - { - print "ASP stage finished.\n"; - } - - # - # If a postASP stage is defined, let the format handle it. - # It should leave whatever it thinks is right based on $file. - # - # postASP ($inhandle) - # - umask $saved_umask; - my $inpostasp = new FileHandle "<$tmpbase.3"; - if (defined $Formats{$global->{format}}{postASP}) - { - &{$Formats{$global->{format}}{postASP}}($inpostasp) == 0 or - die "error post-processing $global->{format}.\n"; - } - $inpostasp->close; - - if ( $global->{debug} ) - { - print "postASP stage finished.\n"; - } - - # - # All done, remove the temporaries. - # - if( !$global->{debug} ) { - remove_tmpfiles($tmpbase); - } -} - -=pod - -=back - -=head1 SEE ALSO - -Documentation for various sub-packages of LinuxDocTools. - -=head1 AUTHOR -SGMLTools are written by Cees de Groot, C<E<lt>cg@cdegroot.comE<gt>>, -and various SGML-Tools contributors as listed in C<CONTRIBUTORS>. -Taketoshi Sano C<E<lt>sano@debian.org<gt>> rename to LinuxDocTools. - -=cut -1; diff --git a/doc/Makefile b/doc/Makefile index f36642be..0d1deb8e 100644 --- a/doc/Makefile +++ b/doc/Makefile @@ -1,8 +1,8 @@ # Force rebuilds .PHONY: progspell docs progdocs userdocs -doc-srcdir := $(shell cd $(s) && pwd) -sgml2 := $(doc-srcdir)/sgml2 +doc-srcdir := $(realpath $(srcdir)/doc) +toolsdir := $(realpath $(srcdir)/tools) docs: progdocs userdocs @@ -23,10 +23,10 @@ $(o)%.sgml: $(s)%.sgml $(objdir)/.dir-stamp cp $< $@ $(o)%.html: $(o)%.sgml - cd $(dir $@) && $(sgml2)html $(notdir $<) + cd $(dir $@) && $(toolsdir)/linuxdoc -B html $(notdir $<) $(o)%.tex: $(o)%.sgml - cd $(dir $@) && $(sgml2)latex --output=tex $(notdir $<) + cd $(dir $@) && $(toolsdir)/linuxdoc -B latex --output=tex $(notdir $<) $(o)%.dvi: $(o)%.tex cd $(dir $@) && TEXINPUTS=$(TEXINPUTS):$(doc-srcdir)/tex latex $(notdir $<) @@ -40,7 +40,7 @@ $(o)%.pdf: $(o)%.tex TEXINPUTS=$(TEXINPUTS):$(doc-srcdir)/tex pdflatex -output-directory=$(dir $@) $< $(o)%.txt: $(o)%.sgml - cd $(dir $@) && $(sgml2)txt $(notdir $<) + cd $(dir $@) && $(toolsdir)/linuxdoc -B txt $(notdir $<) $(o)prog.spell: $(o)prog.sgml $(s)prog-spell.sed sed -f $(lastword $^) <$< >$@ diff --git a/doc/sgml2html b/doc/sgml2html deleted file mode 100755 index ea8e8c92..00000000 --- a/doc/sgml2html +++ /dev/null @@ -1,53 +0,0 @@ -#!/usr/bin/perl -# -# sgmltools.in -# -# $Id$ -# -# SGML-Tools driver. Calls all other SGML-Tools components, contains -# configuration information, etcetera. -# -package main; - -sub BEGIN -{ - require 5.004; -} -use strict; - -use vars qw($prefix $DataDir $BinDir $progs); - -use FindBin; - -$prefix = "/usr"; -$DataDir = "$FindBin::Bin/sbase"; -$BinDir = "/usr/bin"; - -use lib "/usr/share/linuxdoc-tools"; -use lib "/usr/perl5"; -use lib "/usr/lib/perl5"; -use lib "/usr/share/perl5"; -$progs = { - "NSGMLS" => "/usr/bin/nsgmls", - "SGMLSASP" => "/usr/bin/sgmlsasp", - "GROFF" => "/usr/bin/groff", - "GROFFMACRO" => "-ms", - "AWK" => "/usr/share/linuxdoc-tools/awkwhich" -}; - -if (! -x $progs->{"NSGMLS"}) - { $progs->{"NSGMLS"} = "/usr/bin/onsgmls"; } - -$ENV{"SGML_CATALOG_FILES"} = "$DataDir/dtd/catalog" . - (defined $ENV{SGML_CATALOG_FILES} ? ":$ENV{SGML_CATALOG_FILES}" : ""); - -require "$FindBin::Bin/LinuxDocTools.pm"; -&LinuxDocTools::init; - -my @FileList = LinuxDocTools::process_options ("html", @ARGV); -for my $curfile (@FileList) - { - LinuxDocTools::process_file ($curfile); - } - -exit 0; diff --git a/doc/sgml2latex b/doc/sgml2latex deleted file mode 100755 index 79c6df03..00000000 --- a/doc/sgml2latex +++ /dev/null @@ -1,53 +0,0 @@ -#!/usr/bin/perl -# -# sgmltools.in -# -# $Id$ -# -# SGML-Tools driver. Calls all other SGML-Tools components, contains -# configuration information, etcetera. -# -package main; - -sub BEGIN -{ - require 5.004; -} -use strict; - -use vars qw($prefix $DataDir $BinDir $progs); - -use FindBin; - -$prefix = "/usr"; -$DataDir = "$FindBin::Bin/sbase"; -$BinDir = "/usr/bin"; - -use lib "/usr/share/linuxdoc-tools"; -use lib "/usr/perl5"; -use lib "/usr/lib/perl5"; -use lib "/usr/share/perl5"; -$progs = { - "NSGMLS" => "/usr/bin/nsgmls", - "SGMLSASP" => "/usr/bin/sgmlsasp", - "GROFF" => "/usr/bin/groff", - "GROFFMACRO" => "-ms", - "AWK" => "/usr/share/linuxdoc-tools/awkwhich" -}; - -if (! -x $progs->{"NSGMLS"}) - { $progs->{"NSGMLS"} = "/usr/bin/onsgmls"; } - -$ENV{"SGML_CATALOG_FILES"} = "$DataDir/dtd/catalog" . - (defined $ENV{SGML_CATALOG_FILES} ? ":$ENV{SGML_CATALOG_FILES}" : ""); - -require "$FindBin::Bin/LinuxDocTools.pm"; -&LinuxDocTools::init; - -my @FileList = LinuxDocTools::process_options ("latex", @ARGV); -for my $curfile (@FileList) - { - LinuxDocTools::process_file ($curfile); - } - -exit 0; diff --git a/doc/sgml2txt b/doc/sgml2txt deleted file mode 100755 index 013479fe..00000000 --- a/doc/sgml2txt +++ /dev/null @@ -1,53 +0,0 @@ -#!/usr/bin/perl -# -# sgmltools.in -# -# $Id$ -# -# SGML-Tools driver. Calls all other SGML-Tools components, contains -# configuration information, etcetera. -# -package main; - -sub BEGIN -{ - require 5.004; -} -use strict; - -use vars qw($prefix $DataDir $BinDir $progs); - -use FindBin; - -$prefix = "/usr"; -$DataDir = "$FindBin::Bin/sbase"; -$BinDir = "/usr/bin"; - -use lib "/usr/share/linuxdoc-tools"; -use lib "/usr/perl5"; -use lib "/usr/lib/perl5"; -use lib "/usr/share/perl5"; -$progs = { - "NSGMLS" => "/usr/bin/nsgmls", - "SGMLSASP" => "/usr/bin/sgmlsasp", - "GROFF" => "/usr/bin/groff", - "GROFFMACRO" => "-ms", - "AWK" => "/usr/share/linuxdoc-tools/awkwhich" -}; - -if (! -x $progs->{"NSGMLS"}) - { $progs->{"NSGMLS"} = "/usr/bin/onsgmls"; } - -$ENV{"SGML_CATALOG_FILES"} = "$DataDir/dtd/catalog" . - (defined $ENV{SGML_CATALOG_FILES} ? ":$ENV{SGML_CATALOG_FILES}" : ""); - -require "$FindBin::Bin/LinuxDocTools.pm"; -&LinuxDocTools::init; - -my @FileList = LinuxDocTools::process_options ("txt", @ARGV); -for my $curfile (@FileList) - { - LinuxDocTools::process_file ($curfile); - } - -exit 0; diff --git a/tools/linuxdoc b/tools/linuxdoc new file mode 100755 index 00000000..51110e79 --- /dev/null +++ b/tools/linuxdoc @@ -0,0 +1,75 @@ +#!/usr/bin/perl +# +# linuxdoc.in +# +# LinuxDoc-Tools driver. Calls all other LinuxDoc-Tools components, +# contains configuration information, etcetera. +# ------------------------------------------------------------------- + +package main; + +sub BEGIN { + require 5.004; +} + +use strict; + +use vars qw($prefix + $isoentities_prefix + $DataDir + $AuxBinDir + $progs); + +use FindBin; + +$prefix = "/usr"; +$isoentities_prefix = "/usr"; +$DataDir = "$FindBin::Bin/../doc/sbase"; +$AuxBinDir = "/usr/lib/linuxdoc-tools"; + +use lib "$FindBin::Bin/linuxdoc-tools"; + +# --------------------------------------------------------------------- +sub ldt_which { +# --------------------------------------------------------------------- +# --------------------------------------------------------------------- + die "ldt_which: No filename(s) array given. Aborting ...\n" unless scalar @_; + + foreach my $file ( @_ ){ + if ( $file =~ m/\// ) { + return $file if -x $file; + } else { + foreach my $path ( split(':',$ENV{'PATH'}) ){ + $path =~ s/\/+$//; + return $file if -x "$path/$file"; + } + } + } + die "No executable found in path for (", join(' ',@_) ,"). Aborting ...\n"; +} + +$progs = { + "SGMLSASP" => ldt_which("sgmlsasp"), + "NSGMLS" => ldt_which("nsgmls","onsgmls"), + "GROFF" => ldt_which("groff"), + "GROFFMACRO" => "-ms", +# "NKF" => "@NKF@" +}; + +$ENV{"SGML_CATALOG_FILES"} = "$DataDir/dtd/catalog" . + (defined $ENV{SGML_CATALOG_FILES} ? ":$ENV{SGML_CATALOG_FILES}" : ""); + +require LinuxDocTools; +&LinuxDocTools::init; + +my @FileList = LinuxDocTools::process_options ($0, @ARGV); + +foreach my $curfile (@FileList) { + &LinuxDocTools::process_file ($curfile); +} + +exit 0; + +# Local Variables: +# mode: perl +# End: diff --git a/tools/linuxdoc-tools/LinuxDocTools.pm b/tools/linuxdoc-tools/LinuxDocTools.pm new file mode 100644 index 00000000..12b77319 --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools.pm @@ -0,0 +1,668 @@ +#! /usr/bin/perl +# +# LinuxDocTools.pm +# +# LinuxDoc-Tools driver core. This contains all the basic functionality +# we need to control all other components. +# +# Copyright © 1996, Cees de Groot. +# Copyright © 2000, Taketoshi Sano +# Copyright © 2006-2018, Agustin Martin +# -------------------------------------------------------------------------------- + +package LinuxDocTools; + +require 5.006; +use strict; + +=head1 NAME + +LinuxDocTools - SGML conversion utilities for LinuxDoc DTD. + +=head1 SYNOPSIS + + use LinuxDocTools; + LinuxDocTools::init; + @files = LinuxDocTools::process_options ($0, @ARGV); + for $curfile (@files) { + LinuxDocTools::process_file ($curfile); + } + +=head1 DESCRIPTION + +The LinuxDocTools package encapsulates all the functionality offered by +LinuxDoc-Tools. It is used, of course, by LinuxDoc-Tools; +but the encapsulation should provide for a simple interface for other users as well. + +=head1 FUNCTIONS + +=over 4 + +=cut + +use File::Copy; +use File::Temp qw(tempdir); +use File::Basename qw(fileparse); +use LinuxDocTools::Lang; +use LinuxDocTools::Utils qw(usage cleanup trap_signals remove_tmpfiles create_temp); +use LinuxDocTools::Vars; + +sub BEGIN +{ + # + # Make sure we're always looking here. Note that "use lib" adds + # on the front of the search path, so we first push dist, then + # site, so that site is searched first. + # + use lib "$main::DataDir/dist"; + use lib "$main::DataDir/site"; +} + +# ----------------------------------------------------------------------------------- +sub ldt_searchfile { +# ----------------------------------------------------------------------------------- +# Look for a readable file in the locations. Return first math. +# ----------------------------------------------------------------------------------- + my $files = shift; + foreach my $file ( @$files ){ + return $file if -r $file; + } +} + +# ----------------------------------------------------------------------------------- +sub ldt_getdtd_v1 { +# ----------------------------------------------------------------------------------- +# Get the dtd +# ----------------------------------------------------------------------------------- + my $file = shift; + my $error_header = "LinuxdocTools::ldt_getdtd_v1"; + my $dtd; + + open ( my $FILE, "< $file") + or die "$error_header: Could not open \"$file\" for reading. Aborting ...\n"; + + while ( <$FILE> ) { + tr/A-Z/a-z/; + # check for [<!doctype ... system] type definition + if ( /<!doctype\s*(\w*)\s*system/ ) { + $dtd = $1; + last; + # check for <!doctype ... PUBLIC ... DTD ... + } elsif ( /<!doctype\s*\w*\s*public\s*.*\/\/dtd\s*(\w*)/mi ) { + $dtd = $1; + last; + # check for <!doctype ... + # PUBLIC ... DTD ... + # (multi-line version) + } elsif ( /<!doctype\s*(\w*)/ ) { + $dtd = "precheck"; + next; + } elsif ( /\s*public\s*.*\/\/dtd\s*(\w*)/ && $dtd eq "precheck" ) { + $dtd = $1; + last; + } + } + close $FILE; + + return $dtd; +} + +# ----------------------------------------------------------------------------------- +sub ldt_getdtd_v2 { +# ----------------------------------------------------------------------------------- +# Second way of getting dtd, fron nsgmls output. +# ----------------------------------------------------------------------------------- + my $preaspout = shift; + my $error_header = "LinuxdocTools::ldt_getdtd_v2"; + my $dtd2; + + open (my $TMP,"< $preaspout") + or die "%error_header: Could not open $preaspout for reading. Aborting ...\n"; + while ( defined ($dtd2 = <$TMP>) && ! ( $dtd2 =~ /^\(/) ) { }; + close $TMP; + $dtd2 =~ s/^\(//; + $dtd2 =~ tr/A-Z/a-z/; + chomp $dtd2; + return $dtd2; +} + +# ----------------------------------------------------------------------------------- +sub ldt_latin1tosgml { +# ----------------------------------------------------------------------------------- +# Convert latin1 chars in input filehandle to sgml entities in the returned string +# ----------------------------------------------------------------------------------- + my $FILE = shift; + my $sgmlout; + + while (<$FILE>){ + # Outline these commands later on - CdG + #change latin1 characters to SGML + #by Farzad Farid, adapted by Greg Hankins + s/À/\À/g; + s/Á/\Á/g; + s/Â/\Â/g; + s/Ã/\Ã/g; + s/Ä/\Ä/g; + s/Å/\Å/g; + s/Æ/\Æ/g; + s/Ç/\Ç/g; + s/È/\È/g; + s/É/\É/g; + s/Ê/\Ê/g; + s/Ë/\Ë/g; + s/Ì/\Ì/g; + s/Í/\Í/g; + s/Î/\Î/g; + s/Ï/\Ï/g; + s/Ñ/\Ñ/g; + s/Ò/\Ò/g; + s/Ó/\Ó/g; + s/Ô/\Ô/g; + s/Õ/\Õ/g; + s/Ö/\Ö/g; + s/Ø/\Ø/g; + s/Ù/\Ù/g; + s/Ú/\Ú/g; + s/Û/\Û/g; + s/Ü/\Ü/g; + s/Ý/\Ý/g; + s/Þ/\Þ/g; + s/ß/\ß/g; + s/à/\à/g; + s/á/\á/g; + s/â/\â/g; + s/ã/\ã/g; + s/ä/\ä/g; + s/å/\å/g; + s/æ/\æ/g; + s/ç/\ç/g; + s/è/\è/g; + s/é/\é/g; + s/ê/\ê/g; + s/ë/\ë/g; + s/ì/\ì/g; + s/í/\í/g; + s/î/\î/g; + s/ï/\ï/g; + s/µ/\μ/g; + s/ð/\ð/g; + s/ñ/\ñ/g; + s/ò/\ò/g; + s/ó/\ó/g; + s/ô/\ô/g; + s/õ/\õ/g; + s/ö/\ö/g; + s/ø/\ø/g; + s/ù/\ù/g; + s/ú/\ú/g; + s/û/\û/g; + s/ü/\ü/g; + s/ý/\ý/g; + s/þ/\þ/g; + s/ÿ/\ÿ/g; + $sgmlout .= $_; + } + return $sgmlout; +} + +# ------------------------------------------------------------------------ + +=item LinuxDocTools::init + +Takes care of initialization of package-global variables (which are actually +defined in L<LinuxDocTools::Vars>). The package-global variables are I<$global>, +a reference to a hash containing numerous settings, I<%Formats>, a hash +containing all the formats, and I<%FmtList>, a hash containing the currently +active formats for help texts. + +Apart from this, C<LinuxDocTools::init> also finds all distributed and site-local +formatting backends and C<require>s them. + +=cut + +# ----------------------------------------------------------------------------------- +sub init { +# ----------------------------------------------------------------------------------- + trap_signals; + + # Register the ``global'' pseudoformat. Apart from the global settings, we + # also use $global to keep the global variable name space clean everything + # that we need to provide to other modules is stuffed into $global. + $global = {}; + $global->{NAME} = "global"; + $global->{HELP} = ""; + $global->{OPTIONS} = [ + { option => "backend", + type => "l", + 'values' => [ "html", "info", "latex", "lyx", "rtf", "txt", "check" ], + short => "B" }, + { option => "papersize", + type => "l", + 'values' => [ "a4", "letter" ], + short => "p" }, + { option => "language", + type => "l", + 'values' => [ @LinuxDocTools::Lang::Languages ], + short => "l" }, + { option => "charset", type => "l", + 'values' => [ "latin", "ascii", "nippon", "euc-kr" ], short => "c" }, + { option => "style", type => "s", short => "S" }, + { option => "tabsize", type => "i", short => "t" }, + # { option => "verbose", type => "f", short => "v" }, + { option => "debug", type => "f", short => "d" }, + { option => "define", type => "s", short => "D" }, + { option => "include", type => "s", short => "i" }, + { option => "pass", type => "s", short => "P" } + ]; + $global->{backend} = "linuxdoc"; + $global->{papersize} = "a4"; + $global->{language} = "en"; + $global->{charset} = "ascii"; + $global->{style} = ""; + $global->{tabsize} = 8; + $global->{verbose} = 0; + $global->{define} = ""; + $global->{debug} = 0; + $global->{include} = ""; + $global->{pass} = ""; + $global->{InFiles} = []; + $global->{fmtlist} = ""; # List of loaded fmt files + $Formats{$global->{NAME}} = $global; # All formats we know. + $FmtList{$global->{NAME}} = $global; # List of formats for help msgs. + + $global->{sgmlpre} = "$main::AuxBinDir/sgmlpre"; + my $error_header = "LinuxdocTools::init"; + + if ( -e "/etc/papersize" ){ + open (my $PAPERSIZE,"< /etc/papersize") || + die "$error_header: Count not open \"/etc/papersize\" for reading\n"; + chomp (my $paper = <$PAPERSIZE>); + $global->{papersize} = "letter" if ( $paper eq "letter"); + close $PAPERSIZE; + } + + # automatic language detection: disabled by default + # { + # my $lang; + # foreach $lang (@LinuxDocTools::Lang::Languages) + # { + # if (($ENV{"LC_ALL"} =~ /^$lang/i) || + # ($ENV{"LC_CTYPE"} =~ /^$lang/i) || + # ($ENV{"LANG"} =~ /^$lang/i)) { + # $global->{language} = Any2ISO($lang); + # } + # } + # } + + # -------------------------------------------------------------------------------- + $global->{preNSGMLS} = sub { + # ------------------------------------------------------------------------------ + # Define a fallback preNSGMLS. Used when the format is "global" (from sgmlcheck). + # ------------------------------------------------------------------------------ + $global->{NsgmlsOpts} .= " -s "; + $global->{NsgmlsPrePipe} = "cat $global->{file}"; + }; + + # We need to load all fmt files here, so the allowed options for all + # format are put into $global and a complete usage message is built, + # including options for all formats. + my %locations = (); + foreach my $path ("$main::DataDir/site", + "$main::DataDir/dist", + "$main::DataDir/fmt"){ + foreach my $location (<$path/fmt_*.pl>){ + my $fmt = $location; + $fmt =~ s/^.*_//; + $fmt =~ s/\.pl$//; + $locations{$fmt} = $location unless defined $locations{$fmt}; + } + } + + foreach my $fmt ( keys %locations ){ + $global->{fmtlist} .= " Loading $locations{$fmt}\n"; + require $locations{$fmt}; + } +} + +# ------------------------------------------------------------------------ + +=item LinuxDocTools::process_options ($0, @ARGV) + +This function contains all initialization that is bound to the current +invocation of LinuxDocTools. It looks in C<$0> to deduce the backend that +should be used (ld2txt activates the I<txt> backend) and parses the +options array. It returns an array of filenames it encountered during +option processing. + +As a side effect, the environment variable I<SGML_CATALOG_FILES> is +modified and, once I<$global->{format}> is known, I<SGMLDECL> is set. + +=cut + +# ------------------------------------------------------------------------ +sub process_options { +# ------------------------------------------------------------------------ + my $progname = shift; + my @tmpargs = @_; + my @args = (); + my $format = ''; + + # Try getting the format. We need to do this here so process_options + # knows which is the format and which format options are allowed + + # First, see if we have an explicit backend option by looping over command line. + # Do not shift in the while condition itself, 0 in options like '-s 0' will + # otherwise stop looping + while ( @tmpargs ){ + $_ = shift @tmpargs; + if ( s/--backend=// ){ + $format = $_; + } elsif ( $_ eq "-B" ){ + $format = shift @tmpargs; + } else { + push @args, $_; + } + } + + unless ( $format ){ + my ($tmpfmt, $dummy1, $dummy2) = fileparse($progname, ""); + if ( $tmpfmt =~ s/^sgml2// ) { # Calling program through sgml2xx symlinks + $format = $tmpfmt; + } elsif ( $tmpfmt eq "sgmlcheck" ) { # Calling program through sgmlcheck symlink + $format = "global"; + } + } + + if ( $format ) { + if ( $format eq "check" ){ + $format = "global"; + } elsif ( $format eq "latex" ){ + $format = "latex2e"; + } + $FmtList{$format} = $Formats{$format} or + usage("$format: Unknown format"); + $global->{format} = $format; + } else { + usage(""); + } + + # Parse all the options from @args, and return files. + my @files = LinuxDocTools::Utils::process_options(@args); + + # Check the number of given files + $#files > -1 || usage("No filenames given"); + + # Normalize language string + $global->{language} = Any2ISO($global->{language}); + + # Setup the SGML environment. + my @sgmlcatalogs = + (# SGML iso-entities catalog location in Debian sgml-data package + "$main::isoentities_prefix/share/sgml/entities/sgml-iso-entities-8879.1986/catalog", + # SGML iso-entities catalog location in ArchLinux, Fedora and Gentoo + "$main::isoentities_prefix/share/sgml/sgml-iso-entities-8879.1986/catalog", + # SGML iso-entities catalog location when installed from linuxdoc-tools + "$main::isoentities_prefix/share/sgml/iso-entities-8879.1986/iso-entities.cat", + # dtd/catalog for SGML-Tools + "$main::DataDir/linuxdoc-tools.catalog", + # The super catalog + "/etc/sgml/catalog"); + + @sgmlcatalogs = ($ENV{SGML_CATALOG_FILES}, @sgmlcatalogs) if defined $ENV{SGML_CATALOG_FILES}; + + $ENV{SGML_CATALOG_FILES} = join(':', @sgmlcatalogs); + + # Set to one of these if readable, nil otherwise + $ENV{SGMLDECL} = ldt_searchfile(["$main::DataDir/dtd/$global->{format}.dcl", + "$main::DataDir/dtd/$global->{style}.dcl", + "$main::DataDir/dtd/sgml.dcl"]); + + # Show the list of loaded fmt_*.pl files if debugging + print STDERR $global->{fmtlist} if $global->{debug}; + + # Return the list of files to be processed + return @files; +} + +# ------------------------------------------------------------------------ + +=item LinuxDocTools::process_file + +With all the configuration done, this routine will take a single filename +and convert it to the currently active backend format. The conversion is +done in a number of steps in tight interaction with the currently active +backend (see also L<LinuxDocTools::BackEnd>): + +=over + +=item 1. Backend: set NSGMLS options and optionally create a pre-NSGMLS pipe. + +=item 2. Here: Run the preprocessor to handle conditionals. + +=item 3. Here: Run NSGMLS. + +=item 4. Backend: run pre-ASP conversion. + +=item 5. Here: Run SGMLSASP. + +=item 6. Backend: run post-ASP conversion, generating the output. + +=back + +All stages are influenced by command-line settings, currently active format, +etcetera. See the code for details. + +=cut + +# ------------------------------------------------------------------------ +sub process_file { +# ------------------------------------------------------------------------ + my $file = $global->{origfile} = shift (@_); + my $saved_umask = umask; + my $error_header = "LinuxdocTools::process_file"; + + print "Processing file $file\n"; + umask 0077; + + my ($filename, $filepath, $filesuffix) = fileparse($file, "\.sgml"); + $global->{filename} = $filename; + $global->{filepath} = $filepath; + $global->{file} = ldt_searchfile(["$filepath/$filename.sgml", + "$filepath/$filename.SGML"]) + or die "$error_header: Cannot find $file. Aborting ...\n"; + + my $dtd = ldt_getdtd_v1("$global->{file}"); + print STDERR "DTD: " . $dtd . "\n" if $global->{debug}; + + # Prepare temporary directory + my $tmpdir = $ENV{'TMPDIR'} || '/tmp'; + $tmpdir = tempdir("linuxdoc-tools.XXXXXXXXXX", DIR => "$tmpdir"); + + # Set common base name for temp files and temp file names + my $tmpbase = $global->{tmpbase} = $tmpdir . '/sgmltmp.' . $filename; + my $precmdout = "$tmpbase.01.precmdout"; + my $nsgmlsout = "$tmpbase.02.nsgmlsout"; # Was $tmpbase.1 + my $preaspout = "$tmpbase.03.preaspout"; # Was $tmpbase.2 + my $aspout = "$tmpbase.04.aspout"; # Was $tmpbase.3 + + # Set up the preprocessing command. Conditionals have to be + # handled here until they can be moved into the DTD, otherwise + # a validating SGML parser will choke on them. + + # Check if output option for latex is pdf or not + if ($global->{format} eq "latex2e") { + if ($Formats{$global->{format}}{output} eq "pdf") { + $global->{define} .= " pdflatex=yes"; + } + } + + # Set the actual pre-processing command + my($precmd) = "| $global->{sgmlpre} output=$global->{format} $global->{define}"; + + # Make sure path of file to be processed is in SGML_SEARCH_PATH + $ENV{"SGML_SEARCH_PATH"} .= ":$filepath"; + + # You can hack $NsgmlsOpts here, etcetera. + $global->{NsgmlsOpts} .= "-D $main::prefix/share/sgml -D $main::DataDir"; + $global->{NsgmlsOpts} .= "-i$global->{include}" if ($global->{include}); + + # If a preNSGMLS function is defined in the fmt file, pipe its output to $FILE, + # otherwise just open $global->{file} as $IFILE + # ---------------------------------------------------------------------------- + my $IFILE; + if ( defined $Formats{$global->{format}}{preNSGMLS} ) { + $global->{NsgmlsPrePipe} = &{$Formats{$global->{format}}{preNSGMLS}}; + open ($IFILE,"$global->{NsgmlsPrePipe} |") + || die "$error_header: Could not open pipe from $global->{NsgmlsPrePipe}. Aborting ...\n"; + } else { + open ($IFILE,"< $global->{file}") + || die "$error_header: Could not open $global->{file} for reading. Aborting ...\n"; + } + + # Create a temp file with $precmd output + my $precmd_command = "$precmd > $precmdout"; + + open (my $PRECMDOUT, "$precmd_command") + or die "$error_header: Could not open pipe to $precmdout. Aborting ...\n"; + + if ($global->{charset} eq "latin") { + print $PRECMDOUT ldt_latin1tosgml($IFILE); + } else { + copy($IFILE,$PRECMDOUT); + } + + close $IFILE; + close $PRECMDOUT; + + # Process with nsgmls. + my $nsgmls_command = "$main::progs->{NSGMLS} $global->{NsgmlsOpts} $ENV{SGMLDECL} $precmdout > $nsgmlsout"; + system($nsgmls_command) == 0 + or die "Error: \"$nsgmls_command\" failed with exit status: ",$? >> 8,"\n"; + + # Special case: if format is global, we're just checking. + cleanup if ( $global->{format} eq "global"); + + # If output file does not exists or is empty, something went wrong. + if ( ! -e "$nsgmlsout" ) { + die "$error_header: Can't create file $nsgmlsout. Aborting ...\n"; + } elsif ( -z "$nsgmlsout" ){ + die "$error_header: $nsgmlsout empty, SGML parsing error. Aborting ...\n"; + } + + print "- Nsgmls stage finished.\n" if $global->{debug}; + + # If a preASP stage is defined, let the format handle it. + # -------------------------------------------------------- + open (my $PREASP_IN, "< $nsgmlsout") + or die "$error_header: Could not open $nsgmlsout for reading. Aborting ...\n"; + open (my $PREASP_OUT, "> $preaspout") + or die "$error_header: Could not open $preaspout for writing. Aborting ...\n"; + + if (defined $Formats{$global->{format}}{preASP}) { + # Usage: preASP ($INHANDLE, $OUTHANDLE); + &{$Formats{$global->{format}}{preASP}}($PREASP_IN, $PREASP_OUT) == 0 + or die "$error_header: Error pre-processing $global->{format}.\n"; + } else { + copy ($PREASP_IN, $PREASP_OUT); + } + + close $PREASP_IN; + close $PREASP_OUT; + + die "$error_header: Can't create $preaspout file. Aborting ...\n" + unless -e "$preaspout"; + + print "- PreASP stage finished.\n" if ( $global->{debug} ); + + # Run sgmlsasp, with an optional style if specified. + # ----------------------------------------------------------- + my $dtd2 = ldt_getdtd_v2($preaspout) + or die "$error_header: Could not read dtd from $preaspout. Aborting ...\n"; + + unless ( $dtd eq $dtd2 ){ + print STDERR "Warning: Two different values for dtd, dtd1: $dtd, dtd2: $dtd2\n"; + $dtd = $dtd2; + } + + $global->{'dtd'} = $dtd; + + # Search order: + # - datadir/site/<dtd>/<format> + # - datadir/dist/<dtd>/<format> + + my $style = ($global->{style}) ? + ldt_searchfile(["$main::DataDir/site/$dtd/$global->{format}/$global->{style}mapping", + "$main::DataDir/dist/$dtd/$global->{format}/$global->{style}mapping", + "$main::DataDir/mappings/$global->{format}/$global->{style}mapping"]) + : + ''; + + my $mapping = ldt_searchfile(["$main::DataDir/site/$dtd/$global->{format}/mapping", + "$main::DataDir/dist/$dtd/$global->{format}/mapping", + "$main::DataDir/mappings/$global->{format}/mapping"]) + or die "$error_header: Could not find mapping file for $dtd/$global->{format}. Aborting ...\n"; + + $mapping = "$style $mapping" if $style; + + $global->{charset} = "nippon" if ($global->{language} eq "ja"); + + # We don't have Korean groff so charset should be latin1. + if ($global->{language} eq "ko") { + if ($global->{format} eq "groff") { + $global->{charset} = "latin1"; + } else { + $global->{charset} = "euc-kr"; + } + } + + if ($global->{format} eq "groff"){ + if ($dtd eq "linuxdoctr") { + $mapping = "$main::DataDir/mappings/$global->{format}/tr-mapping"; + } + } + + my $sgmlsasp_command = "$main::progs->{SGMLSASP} $mapping < $preaspout | + expand -t $global->{tabsize} > $aspout"; + system ($sgmlsasp_command) == 0 + or die "$error_header: Error running $sgmlsasp_command. Aborting ...\n"; + + die "$error_header: Can't create $aspout file. Aborting ...\n" + unless -e "$aspout"; + + print "- ASP stage finished.\n" if ( $global->{debug} ); + + # If a postASP stage is defined, let the format handle it. + # ---------------------------------------------------------------- + umask $saved_umask; + + open (my $INPOSTASP, "< $aspout" ) + or die "$error_header: Could not open $aspout for reading. Aborting ...\n"; + if (defined $Formats{$global->{format}}{postASP}) { + # Usage: postASP ($INHANDLE) + # Should leave whatever it thinks is right based on $INHANDLE. + &{$Formats{$global->{format}}{postASP}}($INPOSTASP) == 0 + or die "$error_header: Error post-processing $global->{format}. Aborting ...\n"; + } + close $INPOSTASP; + + print "- postASP stage finished.\n" if ( $global->{debug} ); + + # All done, remove the temporaries. + remove_tmpfiles($tmpbase) unless ( $global->{debug} ); +} + +=pod + +=back + +=head1 SEE ALSO + +Documentation for various sub-packages of LinuxDocTools. + +=head1 AUTHOR +SGMLTools are written by Cees de Groot, C<E<lt>cg@cdegroot.comE<gt>>, +and various SGML-Tools contributors as listed in C<CONTRIBUTORS>. +Taketoshi Sano C<E<lt>sano@debian.org<gt>> rename to LinuxDocTools. + +=cut +1; diff --git a/tools/linuxdoc-tools/LinuxDocTools/BackEnd.pm b/tools/linuxdoc-tools/LinuxDocTools/BackEnd.pm new file mode 100644 index 00000000..e402cc5d --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/BackEnd.pm @@ -0,0 +1,185 @@ +# +# BackEnd.pm +# +# $Id: BackEnd.pm,v 1.1.1.1 2001/05/24 15:57:41 sano Exp $ +# +# Dummy module containing backend specification. +# +# © Copyright 1997, Cees de Groot +# +package LinuxDocTools::BackEnd; + +die "This is a documentation package only!"; + +=head1 NAME + +LinuxDocTools::BackEnd - LinuxDocTools back-end specification + +=head1 SYNOPSIS + + require LinuxDocTools::BackEnd; + $BackEnd->{...}; + +=head1 DESCRIPTION + +LinuxDoc-Tools backend modules need to conform to a certain interface which is +detailed in this document. The interface makes sure that new backend modules +(or customer overrides) are compatible with what the main B<LinuxDocTools> +package expects. Note that this interface is still subject to change, you +should check this document on new releases of LinuxDoc-Tools. + +=head1 INTERFACE + +The interface between the main package and individual backends is very +minimal - only one global variable is modified, everything else is local. It +relies heavily on references and complex datatypes, so you want to make +sure that you're up-to-date with Perl5. + +Every backend creates a reference to a hash and stores this reference in +the global I<%Formats> hash: + + my $BackEnd = {}; + $Formats{"BackEnd"} = $BackEnd; + +The rest of this document will deal with the entries in the local hash +referenced by I<$BackEnd>. + +=head1 HASH ENTRIES + +=over 4 + +=item NAME + +Specify the name of the backend, for help messages etcetera. + + $BackEnd->{NAME} = "BackEnd"; + +=item HELP + +Specify an optional extra help message printed when the default usage +function is executed (see L<LinuxDocTools::Utils>). + + $BackEnd->{HELP} = "This is just and example message"; + +=item OPTIONS + +This specifies the local set of options, which is added to the global set +of options (available in I<$global>). The options are specified as an +array of hashes containing a number of keys: + +=over 4 + +=item option + +The long option name + +=item type + +The type of the option, one of B<f> (flag), B<l> (list of allowed values), +B<s> (string), or B<i> (integer). + +=item values + +An array of allowed values, in case the option is of the list type. + +=item short + +A short (single-letter) version of the option name. + +=back + +Options can be specified as long options: + + --papersize=a4 + +or as short options: + + -p a4 + +Note that both the long options as the short options must not conflict with +the global options (an override is not - yet - possible) and should not +conflict with other backends. + + $BackEnd->{OPTIONS} = [ + { option => "split", type => "l", + 'values' => [ "0", "1", "2" ], short => "s" }, + { option => "dosnames", type => "f", short => "D" }, + { option => "imagebuttons", type => "f", short => "I"} + ]; + +The long names themselves function as hash keys; a default can be given +here and the option processing function will store any values found +at the same place: + + $BackEnd->{'split'} = 1; + $BackEnd->{dosnames} = 0; + $BackEnd->{imagebuttons} = 0; + +=item preNSGMLS + +If defined, this should contain a subroutine that normally does two things: it +can modify the global value C<$global-E<gt>{NsgmlsOpts}> and it can set the +global value C<$global-E<gt>{NsgmlsPrePipe}>. The first variable contains +the option string passed to B<nsgmls>, and the second variable can contain +a command that generates the input for B<nsgmls>, presumably using the +current input file in some way (the current input file can be found +in C<$global-E<gt>{file}>). + + $BackEnd->{preNSGMLS} = sub { + $global->{NsgmlsOpts} .= " -ifmtBackEnd "; + $global->{NsgmlsPrePipe} = "sed 's/\@/\@\@/g' $global->{file}"; + }; + +=item preASP + +If defined, this should contain a subroutine accepting an input and an output +file descriptor. The input file descriptor contains the raw output from +B<nsgmls>, and the output file descriptor should be filled with input +to B<sgmlsasp>. This stage is often used to munch character entities +before they're fed to B<sgmlsasp>, see L<LinuxDocTools::CharEnts>. If the routine +doesn't return C<0>, LinuxDocTools aborts. + + $BackEnd->{preASP} = sub + { + my ($infile, $outfile) = @_; + + while (<$infile>) + { + s/([^\\])\\n/$1 \\n/g; + print $outfile $_; + } + return 0; + }; + +=item postASP + +This entry should always be defined, because it needs to contain a routine +that receives the output from B<sgmlsasp> which normally needs finalization. +LinuxDocTools itself doesn't know about file-naming conventions, etcetera, of +the backend so writing the final file is left to the backend. The subroutine +receives a reference to a filehandle (containing B<sgmlsasp> output) and +should do whatever it likes with this datastream. + + $BackEnd->{postASP} = sub + { + my $infile = shift; + + copy ($infile, "$global->{filename}.ext"); + return 0; + }; + +=back + +=head1 SEE ALSO + +L<LinuxDocTools> and subpackages. + +=head1 AUTHOR + +SGML-Tools are written by Cees de Groot, C<E<lt>cg@cdegroot.comE<gt>>, +and various SGML-Tools contributors as listed in C<CONTRIBUTORS>. +Taketoshi Sano C<E<lt>sano@debian.org<gt>> rename it to LinuxDocTools, +and do some bug-fixes and updates on it. + +=cut +1; diff --git a/tools/linuxdoc-tools/LinuxDocTools/CharEnts.pm b/tools/linuxdoc-tools/LinuxDocTools/CharEnts.pm new file mode 100644 index 00000000..b0bcd532 --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/CharEnts.pm @@ -0,0 +1,176 @@ +# +# CharEnts.pm +# +# $Id: CharEnts.pm,v 1.1.1.1 2001/05/24 15:57:41 sano Exp $ +# +# SGML Character Entity utilities -- interface to Perl module +# Text::EntityMap. +# +package LinuxDocTools::CharEnts; +use strict; + +=head1 NAME + +LinuxDocTools::CharEnts - Interface to Text::EntityMap + +=head1 SYNOPSIS + + my $char_maps = load_char_maps ('.2ext', [ Text::EntityMap::sdata_dirs() ]); + + $value = parse_data ($value, $char_maps, $escape_sub); + +=head1 DESCRIPTION + +This module provides a simple interface to the entity map handling provided by +B<Text::EntityMap>. + +=head1 FUNCTIONS + +=over 4 + +=cut + +use Text::EntityMap; +use Exporter; + +use vars qw(@ISA @EXPORT $VERSION); +@ISA = qw(Exporter); +@EXPORT = qw(load_char_maps parse_data); +$VERSION = sprintf("%d.%02d", q$Revision: 1.1.1.1 $ =~ /(\d+)\.(\d+)/); + +# `%warn_map' tracks entities that were not able to be mapped so they +# are only warned once. +my %warn_map = (); + +=item parse_data ($data, $char_map, $escape_sub) + +B<parse_data> takes a string of I<$data> in the output format of +B<nsgmls> (see SP's C<sgmlsout.htm> document) without the leading dash. +B<parse_data> calls I<$char_map>'s lookup method for each sdata +entity reference. If the entity reference is undefined, it is +left alone (without the (n)sgmls C<\|>). For all remaining data, +B<parse_data> calls back into I<$escape_sub> to properly escape +characters for the backend formatter. Strings returned from the +lookup method are assumed to be already escaped. + +This routine is derived from David Megginson's SGMLSpm. + +=cut + +sub parse_data { + my ($data, $char_map, $escape_sub) = @_; + my ($result) = ""; + + my $sdata_flag = 0; + my $out = ''; + + while ($data =~ /\\(\\|n|\||[0-7]{1,3})/) { + $out .= $`; + $data = $'; + + if ($1 eq '|') { + # beginning or end of SDATA + if ("$out" ne '') { + if ($sdata_flag) { + my ($mapping) = $char_map->lookup ($out); + if (defined $mapping) { + # escape `\' in mapping for ASP + $mapping =~ s/\\/\\\\/g; + $result .= $mapping; + } else { + if (!$warn_map{$out}) { + warn "parse_data: no entity map for \`$out'\n"; + $warn_map{$out} = 1; + } + # output the entity reference inside of `{}' + $result .= &$escape_sub ("{" . $out . "}"); + } + } else { + $result .= &$escape_sub ($out); + } + $out = ''; + } + $sdata_flag = !$sdata_flag; + + } elsif ($1 eq 'n') { + # record end + + # pass '\\n' through to ASP + $result .= &$escape_sub ($out) . '\\n'; + $out = ''; + } elsif ($1 eq '\\') { + # backslash + + $result .= &$escape_sub ($out); + + $out = '[bsol ]'; # bsol == entity name for backslash + my ($mapping) = $char_map->lookup ($out); + if (defined $mapping) { + # escape `\' in mapping for ASP + $mapping =~ s/\\/\\\\/g; + $result .= $mapping; + } else { + if (!$warn_map{$out}) { + warn "parse_data: no entity map for \`$out'\n"; + $warn_map{$out} = 1; + } + # output the entity reference inside of `{}' + $result .= &$escape_sub ("{" . $out . "}"); + } + $out = ''; + } else { + # other octal character + $result .= &$escape_sub ($out . chr(oct($1))); + $out = ''; + } + } + $out .= $data; + if ("$out" ne '') { + $result .= &$escape_sub ($out); + } + + return ($result); +} + +=item load_char_maps ($format, $paths) + +B<load_char_maps> takes an EntityMap format suffix and loads all of the +character entity replacement sets for that suffix into an EntityMapGroup. +It searches every directory in I<@{$path}>. + +=cut + +sub load_char_maps { + my ($format, $paths) = @_; + + my (@char_maps) = (); + my ($path, $file_name, $char_map); + + foreach $path (@{$paths}) { + if (-d $path) { + opendir (SDATADIR, $path) + || die "load_char_map: opening directory \`$path' for reading: $!\n"; + foreach $file_name (readdir (SDATADIR)) { + next if ($file_name !~ /$format$/); + eval {$char_map = Text::EntityMap->load ("$path/$file_name")} + || die "load_char_map: loading \`$path/$file_name'\n$@\n"; + push (@char_maps, $char_map); + } + closedir (SDATADIR); + } + } + + warn "load_char_maps: no entity maps found\n" + if ($#char_maps == -1); + + return (Text::EntityMap->group (@char_maps)); +} + +=back + +=head1 AUTHOR + +Ken MacLeod, C<E<lt>ken@bitsko.slc.ut.usE<gt>> + +=cut +1; diff --git a/tools/linuxdoc-tools/LinuxDocTools/FixRef.pm b/tools/linuxdoc-tools/LinuxDocTools/FixRef.pm new file mode 100644 index 00000000..d2549857 --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/FixRef.pm @@ -0,0 +1,76 @@ +# +# FixRef.pm +# +# $Id: FixRef.pm,v 1.1.1.1 2001/05/24 15:57:41 sano Exp $ +# +# Start conversion from parsed linuxdoc-sgml to html. +# - Identify references and file count +# +# Rules based on fixref.l +# +package LinuxDocTools::FixRef; + +# Externally visible variables +$fixref = {}; + +# Initialize: set splitlevel before using rules +# Usage: &{$fixref->{init}}(<split level>); + # 0 - super page mode + # 1 - big page mode + # 2 - small page mode +$fixref->{init} = sub { + $splitlevel = shift; +}; + +# Outputs: Read after using rules +$fixref->{filenum} = 0; # Count of files we will create +$fixref->{lrec} = {}; # label -> filenum + +# Package variables +$chapter_mode = 0; # <report> vs. <article> +$splitlevel = 0; # See $fixref->{init} above; + # Automatically reduced by 1 for chapter mode + +# Finalize parsing +$fixref->{finish} = sub { }; # Do nothing when we're done + +# Ruleset +$fixref->{rules} = {}; # Individual parsing rules +$fixref->{defaultrule} = sub { }; # If line does not match any rules + +# Set the rules +# <@@ssect> - split file if necessary +$fixref->{rules}->{'^<@@ssect>.*$'} = sub { &splitfile(2); }; + +# <@@sect> - split file if necessary +$fixref->{rules}->{'^<@@sect>.*$'} = sub { &splitfile(1); }; + +# <@@chapt> - set chapter mode; reduce splitlevel if needed; split file +$fixref->{rules}->{'^<@@chapt>.*$'} = sub { + $splitlevel-- if (!$chapter_mode); + $chapter_mode = 1; &splitfile(0); +}; + +# <@@label> - Identify label location +$fixref->{rules}->{'^<@@label>(.*)$'} = sub { + $fixref->{lrec}->{$1} = $fixref->{filenum}; +}; + +#============================== +# Split the file (-split option; level in parentheses): +# non-chapter mode: -0 -> don't split +# -1 -> split at sect (1) +# -2 -> split at sect (1) and ssect (2) +# chapter mode: -0 -> split at chapt (0) +# -1 -> split at chapt (0) +# -2 -> split at chapt (0) and sect (1) +sub splitfile +{ + my ($level) = @_; + if (($level == 0) || ($splitlevel >= $level)) { + $fixref->{filenum}++; + } +} + +1; + diff --git a/tools/linuxdoc-tools/LinuxDocTools/Html2Html.pm b/tools/linuxdoc-tools/LinuxDocTools/Html2Html.pm new file mode 100644 index 00000000..9ff2e4cc --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/Html2Html.pm @@ -0,0 +1,583 @@ +# +# Html2Html.pm +# +# $Id: Html2Html.pm,v 1.4 2001/08/31 23:09:10 sano Exp $ +# +# Convert parsed linuxdoc-sgml to html. +# - Split files; match references, generate TOC and navigation +# aids, etc. +# +# Rules based on html2html.l +# +package LinuxDocTools::Html2Html; + +use FileHandle; +use LinuxDocTools::Lang; + +# Externally visible variables +$html2html = {}; + +# Initialize: set splitlevel, extension, images, filename, +# filenumber, label, header, footer, toclevel, +# tmpbase, debug. +# Usage: +# &{$html2html->{init}}(split,ext,img,filename,filenum,label,hdr,ftr,toc,tmpbase, debug); +# split level: 0 - super page mode +# 1 - big page mode +# 2 - small page mode +$html2html->{init} = sub { + $splitlevel = shift; + SWITCH: { + $super_page_mode = 0, $big_page_mode = 1, last SWITCH + if ($splitlevel == 1); + $super_page_mode = 0, $big_page_mode = 0, last SWITCH + if ($splitlevel == 2); + } + + $fileext = shift; + $use_imgs = shift; + $firstname = shift; + $filecount = 1 + shift; + $lprec = shift; + + $header = shift; + $footer = shift; + + $toclevel = shift; + if ($toclevel == -1) { + if ($splitlevel == 0) { + $toclevel = 0; + } else { + $toclevel = 2; + } + } + + $tmpbase = shift; + $content_file = $tmpbase . ".content"; + + $debug = shift; + + $nextlabel = Xlat ("Next"); + $prevlabel = Xlat ("Previous"); + $toclabel = Xlat ("Contents"); +}; + +# Package variables +$big_page_mode = 0; # '-2' subsection splitting +$super_page_mode = 1; # One page vs. page/section +$chapter_mode = 0; # <article> vs. <report> +$current = ""; # State of section/subsection/etc. +$filenum = 1; # Current output file number +$filecount = 1; +$firstname = "$$"; # Base name for file +$headbuf = ""; # Buffer for URL's +$fileext = "html"; # "html" vs. "htm" for 8.3 +$in_appendix = 0; # n.n vs. a.n section numbers +$in_section_list = 0; # List of sections flag +$language = ""; # Default English; use '-Lname' +# $lprec{label} # Label record +$nextlabel = ""; # Link string +$outfh = STDOUT; # Output filehandle +$outname = ""; # Output file name +$prevlabel = ""; # Link string +$refname = ""; # Ref string +$sectname = ""; # Section name +$secnr = 0; # Section count +$ssectname = ""; # Subsection name +$ssecnr = 0; # Subsection count +$skipnewline = 0; # Flag to ignore new line +$toclabel = ""; # Link string +$titlename = ""; # Title of document +$use_imgs = 0; # '-img' pictorial links +$urlname = ""; # Name for url links +$header = ""; +$footer = ""; +$toclevel = -1; +$tmpbase = "/tmp/sgmltmp" . $$; +$debug = 0; +$content_file = $tmpbase . ".content.init"; + +# Ruleset +$html2html->{rules} = {}; # Individual parsing rules + +$html2html->{rules}->{'^<@@appendix>.*$'} = sub { + $in_appendix = 1; $secnr = 0; $ssecnr = 0; +}; + +$html2html->{rules}->{'^<@@url>(.*)$'} = sub { + $skipnewline = 1; $urlname = $1; $headbuf = qq(<A HREF="$1">); +}; + +$html2html->{rules}->{'^<@@urlnam>(.*)$'} = sub { + $headbuf = $headbuf . "$urlname</A>"; +}; + +$html2html->{rules}->{'^<@@endurl>.*$'} = sub { + $skipnewline = -1; $outfh->print($headbuf); $headbuf = ""; +}; + +$html2html->{rules}->{'^<@@title>(.*)$'} = sub { + $titlename = $1; &heading(STDOUT); print(STDOUT "<H1>$1</H1>\n\n"); +}; + +$html2html->{rules}->{'^<@@head>(.*)$'} = sub { + $skipnewline = 1; $headbuf = $1; +}; + +$html2html->{rules}->{'^<@@part>.*$'} = sub { $current = "PART"; }; + +$html2html->{rules}->{'^<@@endhead>.*$'} = sub { + SWITCH: { + $outfh->print("<H1>$headbuf</H1>\n\n"), last SWITCH + if ($current eq "PART"); + $outfh->print("<H1>$headbuf</H1>\n\n"), last SWITCH + if ($current eq "CHAPTER"); + $outfh->print("<H2>$headbuf</H2>\n\n"), last SWITCH + if ($current eq "SECTION"); + $outfh->print("<H2>$headbuf</H2>\n\n"), last SWITCH + if ($current eq "SUBSECT"); + $outfh->print("<H3>$headbuf</H3>\n\n"), last SWITCH; + } + $current = ""; $headbuf = ""; $skipnewline = 0; +}; + +$html2html->{rules}->{'^<@@chapt>(.*)$'} = sub { + $chapter_mode = 1; $skipnewline = 1; $sectname = $1; + &start_chapter($sectname); +}; + +$html2html->{rules}->{'^<@@sect>(.*)$'} = sub { + $skipnewline = 1; $ssectname = $1; + if ($chapter_mode) { + &start_section($ssectname); + } else { + $sectname = $ssectname; &start_chapter($ssectname); + } +}; + +$html2html->{rules}->{'^<@@ssect>(.*)$'} = sub { + $skipnewline = 1; $ssectname = $1; + if (!$chapter_mode) { + &start_section($ssectname); + } else { + $current = ""; $headbuf = $ssectname; + } +}; + +$html2html->{rules}->{'^<@@endchapt>.*$'} = sub { + STDOUT->print("</UL>\n") if ($in_section_list); + if ($outfh->fileno != STDOUT->fileno) { + &footing($outfh) if (!$super_page_mode); + $outfh->close; $outfh = STDOUT; + } +}; + +$html2html->{rules}->{'^<@@endsect>.*$'} = sub { + STDOUT->print("</UL>\n") if (!$chapter_mode && $in_section_list); + if (($outfh->fileno != STDOUT->fileno) + && ((!$chapter_mode) || (!$big_page_mode))) { + &footing($outfh) if (!$super_page_mode); + $outfh->close; $outfh = STDOUT; + } +}; + +$html2html->{rules}->{'^<@@endssect>.*$'} = sub { + if (($outfh->fileno != STDOUT->fileno) + && (!$chapter_mode) && (!$big_page_mode) && (!$super_page_mode)) { + &footing($outfh); $outfh->close; $outfh = STDOUT; + } +}; + +$html2html->{rules}->{'^<@@enddoc>.*$'} = sub { }; + +$html2html->{rules}->{'^<@@label>(.*)$'} = sub { + if (!defined($lprec->{$1})) { + STDERR->print(qq(html2html: Problem with label "$1"\n)); next; + } + if ($skipnewline) { + $headbuf = sprintf(qq(<A NAME="%s"></A> %s), $1, $headbuf); + } else { + $outfh->print(qq(<A NAME="$1"></A> )); + } +}; + +$html2html->{rules}->{'^<@@ref>(.*)$'} = sub { + my $tmp; + + $refname = $1; + if (!defined($lprec->{$1})) { + STDERR->print(qq(html2html: Problem with ref "$1"\n)); + $skipnewline++; next; + } + SWITCH: { + $tmp = qq(<A HREF="#$1">), last SWITCH + if ($lprec->{$1} == $filenum - 1); + $tmp = qq(<A HREF="$firstname.$fileext#$1">), last SWITCH + if ($lprec->{$1} == 0); + $tmp = qq(<A HREF="$firstname-$lprec->{$1}.$fileext#$1">), + last SWITCH; + } + if ($skipnewline) { + $headbuf = "$headbuf$tmp"; + } else { + $headbuf = $tmp; + } + $skipnewline++; +}; + +$html2html->{rules}->{'^<@@refnam>.*$'} = sub { + $headbuf = "$headbuf$refname</A>\n"; +}; + +$html2html->{rules}->{'^<@@endref>.*$'} = sub { + if ($skipnewline == 1) { + $outfh->print($headbuf); $skipnewline = -1; + } elsif ($skipnewline == 2) { + $skipnewline--; + } else { + STDERR->print("html2html: Problem with endref\n"); + $skipnewline--; + } +}; + +# Default parsing rule +$html2html->{defaultrule} = sub { + $skipnewline++ if ($skipnewline < 0); + if ($skipnewline) { + chop; $headbuf = "$headbuf$_"; + } else { + $outfh->print($_); + } +}; + +# Finalize parsing process +$html2html->{finish} = sub { + # Finish footers + if ($outfh->fileno != STDOUT->fileno) { + if (!$super_page_mode) { + &footing($outfh); + $outfh->close; + } + } + # + if ($super_page_mode) { + if ($toclevel > 0) { STDOUT->print("\n<HR>\n"); } + $outfh->close if ($outfh->fileno != STDOUT->fileno); + if ( -r $content_file ) { + open CONTENT, "<$content_file" + or die "Can't open content file\n"; + while (<CONTENT>) { + STDOUT->print($_); + } + close CONTENT; + unlink $content_file if (! $debug); + } + } + # Finish the TOC; ensure "next" points to the first page. + &browse_links(STDOUT, 1, 0) if (!$super_page_mode); + # + # add Footer + if ( -r "$footer" ) { + open FTRFILE, "<$footer" or die "Cannot open footer file\n"; + while (<FTRFILE>) { + STDOUT->print($_); + } + close FTRFILE; + } else { + STDOUT->print("</BODY>\n</HTML>\n"); + } +}; + + +################################################################### +# Secondary Functions +################################################################### + +# Print standard links +sub browse_links { + my ($outfh, $myfilenum, $top) = @_; + + return if ($super_page_mode); + + $outfh->print("<HR>\n") unless ($top); + + # NOTE: For pages where a next or prev button isn't appropriate, include + # the graphic anyway - just don't make it a link. That way, the mouse + # position of each button is unchanged from page to page. + # Use the passed myfilenum since filenum may already be incremented + + # Next link (first) + my $next = $use_imgs + ? qq(<IMG SRC="next.png" ALT="$nextlabel">) + : qq($nextlabel); + $next = qq(<A HREF="$firstname-$myfilenum.$fileext">$next</A>) + if ($myfilenum < $filecount); + $next = join "", $next, "\n"; + $outfh->print($next); + + # Previous link + my $prev = $use_imgs + ? qq(<IMG SRC="prev.png" ALT="$prevlabel">) + : qq($prevlabel); + $prev = join "", qq(<A HREF="$firstname-), ($myfilenum - 2), + qq(.$fileext">$prev</A>) + if ($myfilenum >= 3); + $prev = join "", $prev, "\n"; + $outfh->print($prev); + + # Table of contents link + my $toc = $use_imgs + ? qq(<IMG SRC="toc.png" ALT="$toclabel">) + : qq($toclabel); + $toc = join "", qq(<A HREF="$firstname.$fileext#toc), + §ion_num($secnr, 0), qq(">$toc</A>) + if ($outfh->fileno != STDOUT->fileno); + $toc = join "", $toc, "\n"; + $outfh->print($toc); + + print($outfh "<HR>\n") if ($top); +} + +# Print end-of-file markup +sub footing { + my $outfh = shift; + &browse_links($outfh, $filenum, 0); + if ( -r "$footer" ) { + open FTRFILE, "<$footer" or die "Cannot open footer file\n"; + while (<FTRFILE>) { + $outfh->print($_); + } + close FTRFILE; + } else { + $outfh->print("</BODY>\n</HTML>\n"); + } +} + +# Print top-of-file markup +sub heading { + my $outfh = shift; my $match; + + # Emit 3.2 HTML until somebody comes up with a better idea - CdG + if ( -r "$header" ) { + open HDRFILE, "<$header" or die "Cannot open header file\n"; + while (<HDRFILE>) { + $outfh->print($_); + } + close HDRFILE; + } else { + $outfh->print( + qq(<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">\n)); + $outfh->print("<HTML>\n<HEAD>\n"); + } + open VERSFILE, "<$main::DataDir/VERSION" or die "Cannot open version file\n"; + $version = <VERSFILE>; + close VERSFILE; + chop $version; + $outfh->print( + " <META NAME=\"GENERATOR\" CONTENT=\"LinuxDoc-Tools $version\">\n"); + + $outfh->print(" <TITLE>"); + $match = $titlename; + $match =~ s/<[^>]*>//g; + $outfh->print($match); + if ($secnr > 0) { + $match = $sectname; + $match =~ s/<[^>]*>//g; + $outfh->print(": $match"); + } + if ($ssecnr > 0) { + $match = $ssectname; + $match =~ s/<[^>]*>//g; + $outfh->print(": $match"); + } + $outfh->print("</TITLE>\n"); + + if (!$super_page_mode) { + # + # <LINK> Information for next, previous, contents, etc... + # + $outfh->print(qq( <LINK HREF="$firstname-$filenum.$fileext" REL=next>),"\n") + if ($filenum < $filecount); + my $prev; + $prev = join "", qq( <LINK HREF="$firstname-), ($filenum - 2), + qq(.$fileext" REL=previous>) + if ($filenum >= 3); + $outfh->print($prev,"\n"); + + # + # Table of contents link + # + my $toc ; + $toc = join "", qq( <LINK HREF="$firstname.$fileext#toc), + §ion_num($secnr, 0), qq(" REL=contents>) + if ($outfh->fileno != STDOUT->fileno); + $outfh->print($toc,"\n"); + } # (!$super_page_mode) + + $outfh->print("</HEAD>\n<BODY>\n"); + &browse_links($outfh, $filenum, 1); +} + +# Return the section and subsection as a dotted string +sub section_num { + my ($sec, $ssec) = @_; + my $l = "A"; + + if ($in_appendix) { + $sec--; + while ($sec) { $l++; $sec--; } + return("$l.$ssec") if ($ssec > 0); + return("$l"); + } else { + return("$sec.$ssec") if ($ssec > 0); + return("$sec"); + } +} + +# Create a chapter head; start a new file, etc. +sub start_chapter { + my $sectname = shift; + + if (!$super_page_mode && $outfh->fileno != STDOUT->fileno) { + &footing($outfh); $outfh->close; + } + $current = "SECTION"; $secnr++; $ssecnr = 0; + if ($super_page_mode) { + $outname = $content_file; + $outfh = new FileHandle ">>$outname" + or die qq(html2html: Fatal: Could not open file "$outname"\n); + if ($toclevel > 0) { + $headbuf = sprintf( + qq(<A NAME="s%s">%s.</A> <A HREF="#toc%s">%s</A>), + §ion_num($secnr, 0), §ion_num($secnr, 0), + §ion_num($secnr, 0), + $sectname); + STDOUT->printf( + qq(<P>\n<H2><A NAME="toc%s">%s.</A> <A HREF="%s#s%s">%s</A></H2>\n\n), + §ion_num($secnr, 0), §ion_num($secnr, 0), + "$firstname.$fileext", §ion_num($secnr, 0), $sectname); + } else { + $headbuf = sprintf( + qq(<A NAME="s%s">%s. %s</A>), + §ion_num($secnr, 0), §ion_num($secnr, 0), + $sectname); + } + } else { + $outname = "$firstname-$filenum.$fileext"; $filenum++; + $outfh = new FileHandle ">$outname" + or die qq(html2html: Fatal: Could not open file "$outname"\n); + &heading($outfh); + if ($toclevel > 0) { + $headbuf = sprintf( + qq(<A NAME="s%s">%s.</A> <A HREF="%s#toc%s">%s</A>), + §ion_num($secnr, 0), §ion_num($secnr, 0), + "$firstname.$fileext", §ion_num($secnr, 0), + $sectname); + STDOUT->printf( + qq(<P>\n<H2><A NAME="toc%s">%s.</A> <A HREF="%s">%s</A></H2>\n\n), + §ion_num($secnr, 0), §ion_num($secnr, 0), + $outname, $sectname); + } else { + $headbuf = sprintf( + qq(<A NAME="s%s">%s. %s</A>), + §ion_num($secnr, 0), §ion_num($secnr, 0), + $sectname); + } + } + $in_section_list = 0; +} + +# Create a section; start a new file, etc. +sub start_section { + my $ssectname = shift; + + $current = "SUBSECT"; $ssecnr++; + if ($toclevel > 1) { + if (!$in_section_list) { + STDOUT->print("<UL>\n"); $in_section_list = 1; + } + } + if ($super_page_mode) { + if ($outfh->fileno != STDOUT->fileno && !$chapter_mode) { + $outfh->close; + } + $outname = $content_file; + $outfh = new FileHandle ">>$outname" + or die qq(html2html: Fatal: Could not open file "$outname"\n); + if ($toclevel > 1) { + $headbuf = sprintf(qq(<A NAME="ss%s">%s</A> <A HREF="#toc%s">%s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + $ssectname); + STDOUT->printf( + qq(<LI><A NAME="toc%s">%s</A> <A HREF="%s#ss%s">%s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + "$firstname.$fileext", + §ion_num($secnr, $ssecnr), + $ssectname); + } else { + $headbuf = sprintf(qq(<A NAME="ss%s">%s %s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + $ssectname); + } + } else { + if (!$big_page_mode) { + if ($outfh->fileno != STDOUT->fileno) { + &footing($outfh); $outfh->close; + } + $outname = "$firstname-$filenum.$fileext"; $filenum++; + $outfh = new FileHandle ">$outname" + or die qq(html2html: Fatal: Could not open file "$outname"\n); + heading($outfh); + + # Since only one section is on any page, + # don't use # so that when we + # jump to this page, we see the browse + # links at the top of the page. + if ($toclevel > 1) { + $headbuf = sprintf("%s <A HREF=\"%s#toc%s\">%s</A>", + §ion_num($secnr, $ssecnr), + "$firstname.$fileext", + §ion_num($secnr, $ssecnr), + $ssectname); + STDOUT->printf( + qq(<LI><A NAME="toc%s">%s</A> <A HREF="%s">%s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + $outname, $ssectname); + } else { + $headbuf = sprintf("%s %s</A>", + §ion_num($secnr, $ssecnr), + $ssectname); + } + } else { + # Since many sections are on one page, we need to use # + if ($toclevel > 1) { + $headbuf = sprintf( + qq(<A NAME="ss%s">%s</A> <A HREF="%s#toc%s">%s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + "$firstname.$fileext", + §ion_num($secnr, $ssecnr), + $ssectname); + STDOUT->printf( + qq(<LI><A NAME="toc%s">%s</A> <A HREF="%s#ss%s">%s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + $outname, + §ion_num($secnr, $ssecnr), + $ssectname); + } else { + $headbuf = sprintf( + qq(<A NAME="ss%s">%s %s</A>\n), + §ion_num($secnr, $ssecnr), + §ion_num($secnr, $ssecnr), + $ssectname); + } + } + } +} + diff --git a/tools/linuxdoc-tools/LinuxDocTools/InfoUtils.pm b/tools/linuxdoc-tools/LinuxDocTools/InfoUtils.pm new file mode 100644 index 00000000..b4bd50bd --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/InfoUtils.pm @@ -0,0 +1,357 @@ +# InfoUtils.pm +# +# Some utils for the linuxdoc info backend. +# +# * Create menus +# * Normalize node names and associated text +# * Point references to the associated node as needed +# +# Copyright (C) 2009 Agustín Martín Domingo, agmartin at debian org +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# -------------------------------------------------------------------- + + +package LinuxDocTools::InfoUtils; + +use base qw(Exporter); + +# List all exported symbols here. +our @EXPORT_OK = qw(info_process_texi); + +# Import :all to get everything. +our %EXPORT_TAGS = (all => [@EXPORT_OK]); + +=head1 NAME + + InfoUtils - Some utils for the linuxdoc info backend. + +=head1 SYNOPSIS + +use InfoUtils q{:all}; + +info_process_texi($infile, $outfile, $infoname) + +=head1 DESCRIPTION + +This module contains some utils to process the raw texinfo file +creating menus, normalizing node names and associated text and +pointing references to the associated node as needed. + +=head1 FUNCTIONS + +=over 4 + +=cut + +# ------------------------------------------------------------------------- +sub info_normalize_node_text { +# ------------------------------------------------------------------------- +# Filter characters not allowed in section names +# ------------------------------------------------------------------------- + my $text = shift; + + $text =~ s/\s+/ /g; + $text =~ s/\@[A-Za-z][A-Za-z0-9]*//g; + $text =~ s/(\{|\})//g; + $text =~ s/\,//g; +# $text =~ s/\.+$//g; + $text =~ s/\./-/g; + $text =~ s/\s+$//g; + + return $text; +} + +# ------------------------------------------------------------------------- +sub info_normalize_node_name { +# ------------------------------------------------------------------------- +# Filter characters not allowed in node names. Previous filtering of +# characters not allowed in section names is supposed. +# ------------------------------------------------------------------------- + my $text = shift; +# my $tmpnodedata = shift; + + $text =~ s/\://g; + $text =~ s/\;//g; + +# die "Error: Reference \"$text\" already used" +# if defined $tmpnodedata->{$text}; + + return $text; +} + +# ------------------------------------------------------------------------- +sub info_parse_raw_file { +# ------------------------------------------------------------------------- +# Parse raw texinfo file. It does not yet contain section names, menus, +# correct references or title. +# ------------------------------------------------------------------------- + my $inputfile = shift; + my $INPUT; + + my @inputtext = (); # Copy of input file with some preprocessing + my %nodedata = # A hash of hashes with all node info + ( 'Top' => + { 'text' => "Top", + 'depth' => 0, + 'up' => "", + 'next' => '', + 'previous' => "", + 'sort' => 0, + 'debug' => "", + 'menu' => []} + ); + + my %levellast = (0 => "Top"); + my %labels = (); + my %docdata = # Some misc data for the document + ( 'title' => "", + 'author' => "", + 'subtitle' => "" + ); + + my $depth = my $lastdepth = 0; + my $lastnode = ""; + my $sort = 0; + + my $inauthor; + my $authorline; + + open ($INPUT, "< $inputfile") + or die "info-postASP: Could not open $inputfile for read. Aborting ...\n"; + + while (<$INPUT>){ + chomp; + if ( s/^\@SUB\s+// ){ + my $updepth = $depth; + my $uppernode = $levellast{$updepth}; + $depth++; + $sort++; + + my @levelmenu = (); + + if ( defined $nodedata{$uppernode}->{'menu'} ){ + @levelmenu = @{ $nodedata{$uppernode}->{'menu'} }; + } + + my $nodetext = info_normalize_node_text($_); + my $nodename = info_normalize_node_name($nodetext,\%nodedata); + + # Make first appearing node the next node for top node + $nodedata{'Top'}->{'next'} = $nodename if ( $lastdepth eq 0); + + # Fill info for current node (and 'next' for last one in level) + $nodedata{$nodename}->{'orig'} = $_; + $nodedata{$nodename}->{'text'} = $nodetext; + $nodedata{$nodename}->{'depth'} = $depth; + $nodedata{$nodename}->{'previous'} = + defined $levellast{$depth} ? $levellast{$depth} : ""; + $nodedata{$levellast{$depth}}->{'next'} = $nodename + if defined $levellast{$depth}; + $nodedata{$nodename}->{'up'} = $uppernode; + $nodedata{$nodename}->{'sort'} = $sort; + $nodedata{$nodename}->{'debug'} = + "updepth: $updepth, lastdepth: $lastdepth, up: $uppernode"; + + # Keep this defined in case tbere is no next node in the same level. + $nodedata{$nodename}->{'next'} = ""; + + push @inputtext, "\@SUB $nodename"; # Rewrite @SUB with the new name + push @levelmenu, $nodename; # Add $nodename to the level menu list + + # Prepare things for next @SUB entry found + $levellast{$depth} = $lastnode = $nodename; + $lastdepth = $depth; + $nodedata{$uppernode}->{'menu'} = \@levelmenu; + + } elsif ( s/^\@ENDSUB// ){ + $depth--; + push @inputtext, $_; + } elsif (s/^\@LABEL\s+//){ + # Keep record of node labels vs nodenames. Will use the last. + $labels{$_} = $lastnode; + } elsif (s/^\@title\s+//){ + $docdata{'title'} = $_; + } elsif (/^\@ldt_endauthor/){ + $inauthor = ''; + my @authors; + if ( @$docdata{'authors'} ){ + @authors = @$docdata{'authors'}; + } + push @authors, $authorline; + $docdata{'authors'} = \@authors; + $authorline = ""; + } elsif ( s/^\@author\s+// ){ + $inauthor = 1; + $authorline = $_; + } elsif ( $inauthor ){ + next if m/^\s*$/; + s/^\s+//; + $authorline .= " $_ "; + } elsif (s/^\@subtitle\s+//){ + $docdata{'subtitle'} = $_; + } elsif (s/^\@ldt_translator\s+//){ + $docdata{'translator'} = $_; + } elsif (s/^\@ldt_tdate\s+//){ + $docdata{'tdate'} = $_; + } else { + push @inputtext, $_; + } + } + close $INPUT; + + $docdata{'nodedata'} = \%nodedata; + $docdata{'labels'} = \%labels; + $docdata{'inputtext'} = \@inputtext; + + return \%docdata; +} + +# ------------------------------------------------------------------------- +sub info_write_preprocessed_file { +# ------------------------------------------------------------------------- +# Write processed texinfo file. Add section names, menus, correct +# references and title. +# ------------------------------------------------------------------------- + my $docdata = shift; + my $infoname = shift; + my $texiout = shift; + + die "InfoUtils.pm: No info file name $infoname.\n" unless $infoname; + die "InfoUtils.pm: No output texi file $texiout\n" unless $texiout; + + my $nodedata = $docdata->{'nodedata'}; + my $labels = $docdata->{'labels'}; + my $inputtext = $docdata->{'inputtext'}; + + my $OUTFILE; + + # info_check_parsed_data($nodedata); + + my %sections = ( 1 => "\@chapter", + 2 => "\@section", + 3 => "\@subsection", + 4 => "\@subsubsection"); + + my $lastdepth = 0; + my $lastnode = "Top"; + my $texinfo = "\@c %** START OF HEADER +\@setfilename $infoname +\@c %** END OF HEADER\n"; + + foreach ( @$inputtext ) { + if ( s/^\@SUB\s+// ){ + my $key = $_; + my $depth = $nodedata->{$key}->{'depth'}; + my $name = $nodedata->{$key}->{'text'}; + + if ( $depth le 4 ){ + my $next = $nodedata->{$key}->{'next'}; + my $previous = $nodedata->{$key}->{'previous'}; + my $up = $nodedata->{$key}->{'up'}; + # my $txt = "\@comment nodename, next, previous, up\n"; + my $txt = ""; + + # $txt .= "\@node $key, $previous, $next, $up\n"; + $txt .= "\@node $key\n"; + $txt .= "$sections{$depth} $name\n"; + + if ( $depth gt $lastdepth && defined $nodedata->{$lastnode}->{'menu'}){ + $txt = "\n\@menu\n\* " + . join("::\n\* ",@{$nodedata->{$lastnode}->{'menu'}}) + . "::\n\@end menu\n" + . "\n$txt"; + } + + $texinfo .= $txt; + $lastdepth = $depth; + $lastnode = $key; + } elsif ( $depth eq 5 ){ + $texinfo .= "\@subsubheading $nodedata->{$key}->{'text'}\n"; + } else { + die "info-postASP: Entry \"$key\" has wrong depth $depth\n"; + } + } elsif (s/^\@REF\s+//){ + if ( defined $labels->{$_} ){ + # If this reference is to a node, use its nodename + $texinfo .= "\@ref{" . $labels->{$_} . "}\n"; + } else { + $texinfo .= "\@ref{$_}\n"; + } + } elsif (s/^\@TOP//){ + $texinfo .= "\@node top\n" + . "\@top " . $docdata->{'title'} . "\n" + . "\@example\n"; + + $texinfo .= join(' and ',@{$docdata->{'authors'}}) . "\n" + if ( @{$docdata->{'authors'}} ); + + $texinfo .= $docdata->{'subtitle'} . "\n" + if ( defined $docdata->{'subtitle'} ); + + $texinfo .= $docdata->{'translator'} . "\n" + if ( defined $docdata->{'translator'} ); + + $texinfo .= $docdata->{'tdate'} . "\n" + if ( defined $docdata->{'tdate'} ); + + $texinfo .= "\@end example\n"; + } else { + $texinfo .= "$_\n"; + } + } + + open ($OUTFILE, "> $texiout") + or die "Could not open \"$texiout\" for write. Aborting ...\n"; + print $OUTFILE $texinfo; + close $OUTFILE; +} + +# ------------------------------------------------------------------------- +sub info_check_parsed_data { +# ------------------------------------------------------------------------- +# ------------------------------------------------------------------------- + my $tmpnodedata = shift; + my @sections = sort { + $tmpnodedata->{$a}->{'sort'} <=> $tmpnodedata->{$b}->{'sort'} + } keys %$tmpnodedata; + + foreach ( @sections ){ + my $ref = $tmpnodedata->{$_}; + print STDERR "Node: $_\n"; + print STDERR " orig: $ref->{'orig'}\n"; + print STDERR " text: $ref->{'text'}\n"; + print STDERR " debug: $ref->{'debug'}\n"; + print STDERR " up: $ref->{'up'}\n"; + print STDERR " depth: $ref->{'depth'}\n"; + print STDERR " previous: $ref->{'previous'}\n"; + print STDERR " next: $ref->{'next'}\n"; + print STDERR " sort: $ref->{'sort'}\n"; + print STDERR " menu:\n * " . join("\n * ",@{$ref->{'menu'}}) . "\n" if defined $ref->{'menu'}; + } +} + +# ------------------------------------------------------------------------- +sub info_process_texi { +# ------------------------------------------------------------------------- +# info_process_texi($infile, $outfile, $infoname) +# +# Call the other functions. +# ------------------------------------------------------------------------- + my $infile = shift; + my $outfile = shift; + my $infoname = shift; + + info_write_preprocessed_file(info_parse_raw_file($infile),$infoname,$outfile); +} diff --git a/tools/linuxdoc-tools/LinuxDocTools/Lang.pm b/tools/linuxdoc-tools/LinuxDocTools/Lang.pm new file mode 100644 index 00000000..2b0e99d6 --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/Lang.pm @@ -0,0 +1,238 @@ +# +# Lang.pm +# +# $Id: Lang.pm,v 1.1.1.1 2001/05/24 15:57:41 sano Exp $ +# +# Language support. +# +# © Copyright 1997, Cees de Groot +# + +package LinuxDocTools::Lang; + +use strict; +use vars qw($VERSION @ISA @EXPORT @Languages $translations); + +require 5.0004; +use Exporter; +use LinuxDocTools::Vars; + +$VERSION = sprintf("%d.%02d", q$Revision: 1.1.1.1 $ =~ /(\d+)\.(\d+)/); +@ISA = qw(Exporter); +@EXPORT = qw(Any2ISO ISO2Native ISO2English Xlat); + +=head1 NAME + +LinuxDocTools::Lang - language name and translation functions + +=head1 SYNOPSIS + + $isoname = Any2ISO ('deutsch'); + $native = ISO2Native ('de'); + $engname = ISO2English ('nederlands'); + + $global->{language} = 'nl'; + $dutch = Xlat ('Table of Contents'); + +=head1 DESCRIPTION + +B<LinuxDocTools::Lang> gives a simple interface to various forms of language +names, and provides a translation service. Languages can be specified in +three different ways: by their native name, by their english name, and +by their 2-letter ISO code. For example, you can specify the German +language as C<deutsch>, as C<german> or as C<de>. + +=head1 FUNCTIONS + +=over 4 + +=cut + +@Languages = qw( + en english english + de deutsch german + nl nederlands dutch + fr français french + es español spanish + da dansk danish + no norsk norwegian + se svenska swedish + pt portuges portuguese + ca català catalan + it italiano italian + ro românã romanian + ja japanese japanese + pl polski polish + ko korean korean + fi suomi finnish +); + + +=item Any2ISO + +Maps any of the three forms of languages to the ISO name. So either of +these invocations: + + Any2ISO ('dutch'); + Any2ISO ('nederlands'); + Any2ISO ('nl'); + +will return the string C<"nl">. + +=cut + +sub Any2ISO +{ + my $lang = shift (@_); + + my $i = 0; + foreach my $l (@Languages) + { + ($l eq $lang) && last; + $i++; + } + return $Languages[(int $i / 3) * 3]; +} + + +=item ISO2Native + +Maps the ISO code to the native name of the language. + +=cut + +sub ISO2Native +{ + my $iso = shift (@_); + + my $i = 0; + foreach my $l (@Languages) + { + ($l eq $iso) && last; + $i++; + } + return $Languages[$i + 1]; + +} + + +=item ISO2English + +Maps the ISO code to the english name of the language. + +=cut + +sub ISO2English +{ + my $iso = shift (@_); + + my $i = 0; + foreach my $l (@Languages) + { + ($l eq $iso) && last; + $i++; + } + return $Languages[$i + 2]; +} + +=item Xlat + +Translates its (English) argument to the language specified by the +current value of C<$gobal-E<gt>{language}>. The module, in its source +file, contains a data structure, indexed by the English strings, that +has all available translations. + +=cut + +sub Xlat +{ + my ($txt) = @_; + + return $txt if ($global->{language} eq "en"); + return $translations->{$txt}{$global->{language}}; +}; + + +# +# By the time this grows big, we'll make up something else. +# +$translations = { + "Previous" => { + "nl" => "Terug", + "de" => "Zurück", + "es" => "Página anterior", + "fr" => "Page précédente", + "da" => "Forrige", + "no" => "Forrige", + "se" => "Föregående", + "pt" => "Página anterior", + "ca" => "Pàgina anterior", + "it" => "Indietro", + "ro" => "Înapoi", + "ja" => "Á°¤Î¥Ú¡¼¥¸", + "pl" => "Poprzedni", + "ko" => "ÀÌÀü", + "fi" => "Edellinen" + }, + "Next" => { + "nl" => "Verder", + "de" => "Weiter", + "es" => "Página siguiente", + "fr" => "Page suivante", + "da" => "Næste", + "no" => "Neste", + "se" => "Nästa", + "pt" => "Página seguinte", + "ca" => "Pàgina següent", + "it" => "Avanti", + "ro" => "Înainte", + "ja" => "¼¡¤Î¥Ú¡¼¥¸", + "pl" => "Nastny", + "ko" => "´ÙÀ½", + "fi" => "Seuraava" + }, + "Contents" => { + "nl" => "Inhoud", + "de" => "Inhalt", + "es" => "Índice general", + "fr" => "Table des matières", + "da" => "Indhold", + "no" => "Innhold", + "se" => "Innehållsförteckning", + "pt" => "Índice", + "ca" => "Índex", + "it" => "Indice", + "ro" => "Cuprins", + "ja" => "Ìܼ¡¤Ø", + "pl" => "Spis Trei", + "ko" => "Â÷·Ê", + "fi" => "Sisällys" + }, + "Table of Contents" => { + "nl" => "Inhoudsopgave", + "de" => "Inhaltsverzeichnis", + "es" => "Índice general", + "fr" => "Table des matières", + "da" => "Indholdsfortegnelse", + "no" => "Innholdsfortegnelse", + "se" => "Innehållsförteckning", + "pt" => "Índice geral", + "ca" => "Índex general", + "it" => "Indice Generale", + "ro" => "Cuprins", + "ja" => "Ìܼ¡", + "pl" => "Spis Trei", + "ko" => "Â÷·Ê", + "fi" => "Sisällysluettelo" + } +}; + +=back + +=head1 AUTHOR + +Cees de Groot, C<E<lt>cg@pobox.comE<gt>> + +=cut + +1; diff --git a/tools/linuxdoc-tools/LinuxDocTools/Utils.pm b/tools/linuxdoc-tools/LinuxDocTools/Utils.pm new file mode 100644 index 00000000..63fe5f91 --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/Utils.pm @@ -0,0 +1,392 @@ +# +# Utils.pm +# +# $Id: Utils.pm,v 1.2 2001/08/31 22:39:44 sano Exp $ +# +# Utilities, split off from other modules in order to cut down file size. +# +# © Copyright 1996, 1997, Cees de Groot +# +package LinuxDocTools::Utils; +use strict; + +=head1 NAME + +LinuxDocTools::Utils - various supporting routines + +=head1 SYNOPSIS + + @files = process_options (@args); + + usage ($msg); + + trap_signals; + + cleanup; + + create_temp($tempfile) + +=head1 DESCRIPTION + +The B<LinuxDocTools::Utils> module contains a number of generic routines, mainly +split off from the main module in order to keep file size down. + +=head1 FUNCTIONS + +=over 4 + +=cut + +use DirHandle; +use FileHandle; +use Cwd; +use File::Basename; +use Exporter; +use LinuxDocTools::Vars; + +use vars qw($VERSION @ISA @EXPORT @EXPORT_OK $in_signal); +@ISA = qw(Exporter); +@EXPORT = qw(usage process_options); +@EXPORT_OK = qw(cleanup trap_signals remove_tmpfiles create_temp); +$VERSION = sprintf("%d.%02d", q$Revision: 1.2 $ =~ /(\d+)\.(\d+)/); + +use subs qw(usage); + +# check whether options are unique +sub check_option_consistency +{ + my $owner = {}; + my ($fmt, $opt); + foreach $fmt (keys %FmtList) + { + my $add = sub { # add to options of $fmt + my $str = shift; + if ($owner->{$str}) { + push(@{$owner->{$str}}, $fmt); + } + else { + $owner->{$str} = [$fmt]; + } + }; + foreach $opt (@{$Formats{$fmt}{OPTIONS}}) + { + &$add("--$opt->{option}"); + &$add("-$opt->{short}"); + } + } + my $error = 0; + foreach $opt (keys %$owner) + { + if (scalar @{$owner->{$opt}} > 1) + { + warn "duplicate option: $opt in " . + join(', ', @{$owner->{$opt}}) . "\n"; + $error = 1; + } + } + die "Internal error detected" if $error; +} + + +=item process_options + +This function processes the command line, and sets the variables associated +with the options along the way. When successful, it returns the arguments +on the command line it didn't interpret. Normally, this will be a list of +filenames. + +=cut + +sub process_options +{ + my @args = @_; + my @retval; + + OPTPROC: while ($args[0]) + { + my $long; + my $curarg = $args[0]; + if ($curarg =~ /^--.*/) + { + # + # Long option, --opt[==value] + # + $long = 1; + } + elsif ($curarg =~ /^-.*/) + { + # + # Short option, -o value + # + $long = 0; + } + else + { + # + # Filename + # + push @retval, $curarg; + next OPTPROC; + } + + # + # Start looking for the option + # + foreach my $fmt (keys %FmtList) + { + foreach my $opt (@{$Formats{$fmt}{OPTIONS}}) + { + if (($long && $curarg =~ /^--$opt->{option}.*/) || + $curarg =~ /^-$opt->{short}/) + { + # + # Found it! Get the argument and see whether all is OK + # with the option. + # + my $optval = ""; + if ($long) + { + if ($curarg =~ /^--$opt->{option}=.*/) + { + $optval = $curarg; + $optval =~ s/[^=]*=(.*)/$1/; + } + } + else + { + if ($args[1] =~ /^[^-].*/) + { + $optval = $args[1]; + } + } + $opt->{type} eq "f" && do + { + # + # "f" -> flag. Increment, so '-v -v' can work. + # + $Formats{$fmt}{$opt->{option}} += 1; + next OPTPROC; + }; + # + # All other types require a value (for now). + # + shift @args unless $long; + if ($optval eq "") + { + usage "Option $curarg: value required"; + } + ($opt->{type} eq "i" || $opt->{type} eq "s") && do + { + # + # "i" -> numeric value. + # "s" -> string value. + # + # No type checking yet... + # + if ($opt->{option} eq "define") + { + $Formats{$fmt}{$opt->{option}} .= " " . $optval; + } + else + { + $Formats{$fmt}{$opt->{option}} = $optval; + } + next OPTPROC; + }; + $opt->{type} eq "l" && do + { + # + # "l" -> list of values. + # + foreach my $val (@{$opt->{'values'}}) + { + if ($val eq $optval) + { + $Formats{$fmt}{$opt->{option}} = $optval; + next OPTPROC; + } + } + usage "Invalid value '$optval' for '--$opt->{option}'"; + }; + usage "Unknown option type $opt->{type} in $fmt/$opt"; + } + } + } + usage "Unknown option $curarg"; + } + continue + { + shift @args; + } + return @retval; +} + + +=item usage + +Prints out a generated help message about calling convention and allowed +options, then the argument string, and finally exits. + +=cut + +sub usage +{ + my ($msg) = @_; + + print "LinuxDoc-Tools version " . `cat $main::DataDir/VERSION` . "\n"; + check_option_consistency; + print "Usage:\n"; + print " " . $global->{myname} . " [options] <infile>\n\n"; + my @helplist = sort(keys %Formats); + @helplist = sort (keys %FmtList) if ($global->{format}); + foreach my $fmt (@helplist) + { + if ($fmt eq "global") + { + print "General options:\n"; + } + else + { + print "Format: " . $fmt . "\n"; + } + print $Formats{$fmt}{HELP}; + for my $opt (@{$Formats{$fmt}{OPTIONS}}) + { + my $value = ''; + if ($opt->{type} eq "i") + { + $value = "number"; + } + elsif ($opt->{type} eq "l") + { + $value = "{"; + my $first = 1; + for my $val (@{$opt->{'values'}}) + { + $first || ($value .= ","); + $first = 0; + $value .= $val; + } + $value .= "}"; + } + elsif ($opt->{type} eq "s") + { + $value = "string"; + } + print " --$opt->{option}"; print "=$value" if $value; + print " -$opt->{short}"; print " $value" if $value; + print "\n"; + } + print "\n"; + } + + $msg && print "Error: $msg\n\n"; + exit 1; +} + + +=item cleanup + +This function cleans out all temporary files and exits. The unlink step +is skipped if debugging is turned on. + +=cut + +sub cleanup +{ + my ($signame) = @_; + + if( $signame ) { + if ( $in_signal ) { + if( $global->{debug} ) { + print STDERR "Caught SIG$signame during cleanup -- aborting\n"; + } + exit -1; + } + else { + if( $global->{debug} ) { + print STDERR "Caught SIG$signame -- cleaning up\n"; + } + $in_signal = 1; + } + } + + if( !$global->{debug} && $global->{tmpbase} ) { + remove_tmpfiles($global->{tmpbase}); + } + exit 0; +} + +=item remove_tmpfiles( $tmpbase ) + +This function cleans out all temporary files, using the argument $tmpbase to +determine the directory and pattern to use to find the temporary files. + +=cut + +sub remove_tmpfiles($) { + my $tmpbase = shift; + my ($name,$tmpdir) = fileparse($tmpbase,""); + my $namelength = length $name; + my $savdir = cwd; + + chdir($tmpdir); + my $dir = new DirHandle("."); + + if (!defined($dir) ) { + warn "Couldn't open temp directory $tmpdir: $!\n"; + } else { + foreach my $tmpfile ($dir->read()) { + if (substr ($tmpfile, 0, $namelength) eq $name) { + unlink ($tmpfile) || warn "Couldn't unlink $tmpfile: $! \n"; + } + } + $dir->close(); + } + + chdir($savdir); + rmdir($tmpdir) || return -1; +} + +=item trap_signals + +This function traps all known signals, making sure that the B<cleanup> +function is executed on them. It should be called once at initialization +time. + +=cut + +sub trap_signals +{ + foreach my $sig ( 'HUP', 'INT', 'QUIT', 'ILL', + 'TRAP', 'IOT', 'BUS', 'FPE', + 'USR1', 'SEGV', 'USR2', + 'PIPE', 'ALRM', 'TERM', ) + { + $SIG{$sig} = \&cleanup; + } +} + +=item create_temp ( $tmpfile ) + +This function creates an empty temporary file with the required +permission for security reasons. + +=cut + +sub create_temp($) { + my $tmpnam = shift; + my $fh = new FileHandle($tmpnam,O_CREAT|O_EXCL|O_WRONLY,0600); + $fh or die "$0: failed to create temporary file: $!"; + $fh->close; +} + +=back + +=head1 AUTHOR + +Cees de Groot, C<E<lt>cg@pobox.comE<gt>>. + +=cut + +1; diff --git a/tools/linuxdoc-tools/LinuxDocTools/Vars.pm b/tools/linuxdoc-tools/LinuxDocTools/Vars.pm new file mode 100644 index 00000000..49cf630b --- /dev/null +++ b/tools/linuxdoc-tools/LinuxDocTools/Vars.pm @@ -0,0 +1,22 @@ +# +# Vars.pm +# +# $Id: Vars.pm,v 1.1.1.1 2001/05/24 15:57:41 sano Exp $ +# +# Shared variables. +# +# © Copyright 1996, 1997, Cees de Groot +# +package LinuxDocTools::Vars; +use strict; + +use Exporter; + +use vars qw($VERSION @ISA @EXPORT); +@ISA = qw(Exporter); +@EXPORT = qw(%Formats $global %FmtList); +$VERSION = sprintf("%d.%02d", q$Revision: 1.1.1.1 $ =~ /(\d+)\.(\d+)/); + +use vars @EXPORT; + +1; diff --git a/tools/linuxdoc-tools/Text/EntityMap.pm b/tools/linuxdoc-tools/Text/EntityMap.pm new file mode 100644 index 00000000..d878fa3c --- /dev/null +++ b/tools/linuxdoc-tools/Text/EntityMap.pm @@ -0,0 +1,121 @@ +# -*- perl -*- +# +# Copyright (C) 1996 Ken MacLeod +# See the file COPYING for distribution terms. +# +# This file is preprocessed during the build to fix-up the references +# in `sdata_dirs'. +# +# $Id: EntityMap.pm.in,v 1.1.1.1 2001/05/24 15:57:40 sano Exp $ +# + +package Text::EntityMap; + +use strict; + +=head1 NAME + +Text::EntityMap - map character entities to output formats + +=head1 SYNOPSIS + +use Text::EntityMap; + +$tex_iso_lat1 = Text::EntityMap->load ("ISOlat1.2tex"); +$tex_iso_lat2 = Text::EntityMap->load ("ISOlat2.2tex"); +$ent_group = Text::EntityMap->group ($tex_iso_lat1, $tex_iso_lat2); + +$ent_group->lookup ('[copy ]'); + +@dirs = Text::EntityMap->sdata_dirs (); + +=head1 DESCRIPTION + +Text::EntityMap is a module that can look-up an output-format +equivalent for special character or other entities. This was inspired +by SGML character entities but can be used in any scenario where +output formatting codes are different for special characters. + +The C<load()> function takes a file name of a mapping table and +returns an Text::EntityMap object. + +The C<group()> function takes a ordered list of Text::EntityMap and +returns an Text::EntityMapGroup object. Looking up entities in a +group object returns the entity replacement returned by the first +EntityMap object. This can be used both to group sets of mapping +files into one object as well as overriding entity replacements. A +EntityMapGroup may contain other EntityMapGroup's. + +The C<lookup()> function can be used with either a EntityMap or +EntityMapGroup object. It takes an entity name and returns the +output-format equivalent. + +C<sdata_dirs()> returns an array containing the local site directory +and ``this'' version of EntityMap's installed directory that contain +the entity maps. Callers can use these paths when looking for tables +to pass to C<load()>. + +=head1 AUTHOR + +Ken MacLeod E<lt>ken@bitsko.slc.ut.usE<gt> + +=cut + +sub sdata_dirs { + return ("/usr/share/entity-map", "/usr/share/entity-map/0.1.0"); +} + +sub load { + my ($type, $file_name) = @_; + + my ($self) = {}; + bless ($self, $type); + + open (FILE, "$file_name") + || die "Can't open \`$file_name' for reading: $!\n"; + while (<FILE>) { + chop; + m/(^[^\t]+)\t(.*)/; + $self->{"$1"} = $2; + } + close (FILE); + + return ($self); +} + +sub group { + my ($type) = shift; + + my ($self) = [{}, @_]; + bless ($self, 'Text::EntityMapGroup'); + + return ($self); +} + +sub lookup { + my ($self, $entity) = @_; + + return ($self->{$entity}); +} + +package Text::EntityMapGroup; + +sub lookup { + my ($self, $entity) = @_; + + my ($replacement) = $self->[0]{$entity}; + return $replacement if defined $replacement; + + my ($ii); + for ($ii = 1; $ii <= $#{$self}; $ii ++) { + $replacement = $self->[$ii]->lookup($entity); + if (defined $replacement) { + $self->[0]{$entity} = $replacement; + return ($replacement); + } + } + + return (undef); +} + +1; diff --git a/tools/linuxdoc-tools/copyright b/tools/linuxdoc-tools/copyright new file mode 100644 index 00000000..a89b6ada --- /dev/null +++ b/tools/linuxdoc-tools/copyright @@ -0,0 +1,85 @@ + +This is `linuxdoc-tools', a series of tools to implement the Linux +Documentation Project HOWTO and book styles in SGML. + +This copy was modified in order to work with birddoc DTD. It is +based on version 0.9.73-2 from Debian 10. + + +---------------------------------- + The linuxdoc-tools license +---------------------------------- + +linuxdoc-tools is derived from linuxdoc-SGML, originally written by +Matt Welsh and later maintained by Cees de Groot. Linuxdoc-SGML is +based on James Clark's sgmls parser, and the QWERTZ DTD by Tom +Gordon. Magnus Alvestad provided the current HTML support. For +the rest of linuxdoc-SGML, + + Copyright (C) 1994-1996 Matt Welsh <mdw@cs.cornell.edu> + Copyright (C) 1996-1998 Cees de Groot <cg@pobox.com> + +Original Linuxdoc-SGML itself does not have any limitations. +Everything not having explicit additional conditions can be freely +used, modified, and redistributed, under the usual fair use clauses: + + * No warranty. Use at your own risk. + * Do not pretend to have written what you did not (Preserve credits + and Copyright notices of the different elements if present). + +Since then, lots of smaller and bigger changes resulted in a rename +to SGML-Tools (and then to SGMLtools, the hyphen caused confusion) +to indicate that it wasn't just for Linux anymore. See files +CHANGES.old-v1 and CONTRIBUTORS.old-v1 for changelog and list of +contributors to old linuxdoc-sgml and sgmltools-v1. + +When sgml-tools dropped support for the linuxdoc DTD, Taketoshi Sano +<sano@debian.org> forked the code to linuxdoc-tools. See README file. + +Changes after the fork + + Copyright (C) 1999-2002 Taketoshi Sano <sano@debian.org> + Copyright (C) 2000 Juan Jose Amor + Copyright (C) 2007-2018 Agustin Martin Domingo <agmartin@debian.org> + +Unless conflicting with other licenses, changes by Agustin Martin +Domingo are free software: you can redistribute and/or modify them +under the terms of the GNU General Public License as published by +the Free Software Foundation, either version 3 of the License, or +(at your option) any later version. Otherwise they honour previous +license. + +You should have received a copy of the GNU General Public License +along with this program. If not, see <http://www.gnu.org/licenses/>. +In Debian systems you can find a copy under /usr/share/common-licenses. + + +---------------------------------- + The entity-map license +---------------------------------- + +Copyright (C) 1997 Ken MacLeod + +Permission is hereby granted, free of charge, to any person obtaining +a copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be +included in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL KEN MACLEOD BE LIABLE FOR ANY CLAIM, DAMAGES OR +OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE +OR OTHER DEALINGS IN THE SOFTWARE. + +Except as contained in this notice, the name of Ken MacLeod shall not +be used in advertising or otherwise to promote the sale, use or other +dealings in this Software without prior written authorization from +Ken MacLeod. |