summaryrefslogtreecommitdiff
path: root/Master
diff options
context:
space:
mode:
authorNorbert Preining <preining@logic.at>2010-10-14 14:53:33 +0000
committerNorbert Preining <preining@logic.at>2010-10-14 14:53:33 +0000
commitbd292953db91776fbac47fd2b31c226ff8bf864f (patch)
tree429ea24fb1f7f9bc93be7e3cf64cc39782ca2897 /Master
parent10cd8e4c2005b7c7a1e8b5216c2d1811dac0933a (diff)
add parse-texcatalogue-keywords for new .xml dump parsing
git-svn-id: svn://tug.org/texlive/trunk@20094 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Master')
-rw-r--r--Master/tlpkg/bin/parse-texcatalogue-keywords159
1 files changed, 159 insertions, 0 deletions
diff --git a/Master/tlpkg/bin/parse-texcatalogue-keywords b/Master/tlpkg/bin/parse-texcatalogue-keywords
new file mode 100644
index 00000000000..0dc78558681
--- /dev/null
+++ b/Master/tlpkg/bin/parse-texcatalogue-keywords
@@ -0,0 +1,159 @@
+#!/usr/bin/env perl
+# $Id$
+# parse a TeXCatalogue v2 .xml dump containing keywords and
+# characterizations
+#
+# Copyright 2010 Norbert Preining
+# This file is licensed under the GNU General Public License version 2
+# or any later version.
+
+$^W = 1;
+use strict;
+use XML::Parser;
+use XML::XPath;
+use XML::XPath::XMLParser;
+use Text::Unidecode;
+
+my $tcfile;
+if (@ARGV) {
+ $tcfile = shift @ARGV;
+} else {
+ $tcfile = "texcatalogue.xml";
+}
+
+my %seikaku;
+my %kw_pkg;
+my %cz_pkg;
+
+parse_texcatalogue($tcfile);
+print_keywords();
+print "\n===================\nprimary characterizations\n";
+walk_cz_tree($cz_pkg{'primary'}, "PRIM");
+print "\n===================\nsecondary characterizations\n";
+walk_cz_tree($cz_pkg{'secondary'}, "SEC");
+print "\n===================\nby-function characterizations\n";
+walk_cz_tree($cz_pkg{'by-function'}, "BFUNC");
+
+##### only subs from here ##########
+
+sub parse_texcatalogue {
+ my $tcfile = shift;
+ my $_parser = XML::Parser->new(
+ ErrorContext => 2,
+ ParseParamEnt => 1,
+ NoLWP => 1
+ );
+ my $io;
+ open($io, "<", $tcfile) or die "Cannot open texcatalogue.xml: $!";
+ my $parser = new XML::XPath->new(ioref => $io, parser => $_parser) ||
+ die "Failed to parse texcatalogue.xml: $!";
+
+ # format of the data
+ # $seikaku{'keyword'}{$package} = [ $kwi, $kwii,...];
+ # $seikaku{'primary'}{$package} = $primchar;
+ # $seikaku{'secondary'}{$package'} = $secchar;
+ # $seikaku{'by-function'}{$package} = [ $bfi, $bfii, ...];
+ #
+ # post processing gives
+ # $kw_pkg{$keyword} = [ $pkg1, $pkg2, ...];
+ # $cz_pkg{'primary'}{$level1}{$level2}...{'_packages_'} = [ $pkg1, $pkg2, ...];
+ # $cz_pkg{'secondary'}{$level1}{$level2}...{'_packages_'} = [ $pkg1, $pkg2, ...];
+ # $cz_pkg{'by-function'}{$level1}{$level2}...{'_packages_'} = [ $pkg1, $pkg2, ...];
+
+
+ foreach my $e ($parser->find('/fullcat/catalogue/entry')->get_nodelist) {
+ my $pkg = $parser->findvalue('./name',$e);
+ #print "FOUND: $pkg\n";
+ my $n = $parser->find('./keyword',$e);
+ for my $kw ($n->get_nodelist) {
+ my $kwval = $parser->find('./@keyword',$kw);
+ push @{$seikaku{'keyword'}{$pkg}}, $kwval;
+ #print "keyword = $kwval\n";
+ }
+ $n = $parser->find('./characterization',$e);
+ for my $cz ($n->get_nodelist) {
+ my $czdimnl = $parser->find('./@dimension',$cz);
+ my $czdim = $czdimnl->string_value();
+ my $czvalnl = $parser->findvalue('.',$cz);
+ my $czval = $czvalnl->value();
+ if (($czdim eq "primary") || ($czdim eq "secondary")) {
+ # assume only one primary and one secondary function
+ $seikaku{$czdim}{$pkg} = $czval;
+ } else {
+ # assume that it is always "by-function"
+ push @{$seikaku{'by-function'}{$pkg}}, $czval;
+ }
+ #print "char dim = $czdim val=$czval\n";
+ }
+ }
+
+
+ #
+ # do the keyword reshuffling
+ for my $pkg (keys %{$seikaku{'keyword'}}) {
+ for my $kw (@{$seikaku{'keyword'}{$pkg}}) {
+ push @{$kw_pkg{$kw}}, $pkg;
+ }
+ }
+
+ parse_characterizations('primary');
+ parse_characterizations('secondary');
+ parse_characterizations('by-function');
+}
+
+sub parse_characterizations {
+ my $what = shift;
+ $cz_pkg{$what} = {};
+ for my $pkg (keys %{$seikaku{$what}}) {
+ my $value = $seikaku{$what}{$pkg};
+ my @charlist;
+ if (!ref($value)) {
+ @charlist = ($value);
+ } else {
+ @charlist = @$value;
+ }
+ for my $prim (@charlist) {
+ # split the primary into levels sep by >
+ my @levels = split(' > ', $prim);
+ my $currentpointer;
+ $currentpointer = $cz_pkg{$what};
+ for my $l (@levels) {
+ if (!defined($currentpointer->{$l})) {
+ $currentpointer->{$l} = {};
+ }
+ $currentpointer = $currentpointer->{$l}
+ }
+ push @{$currentpointer->{'_packages_'}}, $pkg;
+ }
+ }
+}
+
+
+sub print_keywords {
+ for my $k (keys %kw_pkg) {
+ my @pkgl = @{$kw_pkg{$k}};
+ if (defined(@pkgl)) {
+ print "keyword = $k\n package = @pkgl\n";
+ } else {
+ print "keyword = $k\n package = NO PACKAGE FOUND!\n";
+ }
+ }
+}
+
+sub walk_cz_tree {
+ my $cp = shift;
+ my $prestring = shift;
+ if (defined($cp->{'_packages_'})) {
+ my @pkgs = sort @{$cp->{'_packages_'}};
+ print "$prestring\n";
+ print "--> @pkgs\n";
+ }
+ for my $cz (keys %$cp) {
+ if ($cz ne '_packages_') {
+ my $nextstring = "$prestring > $cz";
+ my $np = $cp->{$cz};
+ &walk_cz_tree($np,$nextstring);
+ }
+ }
+}
+