diff options
author | Ismaël Bouya <ismael.bouya@normalesup.org> | 2015-01-13 23:35:48 +0100 |
---|---|---|
committer | Ismaël Bouya <ismael.bouya@normalesup.org> | 2015-01-13 23:35:48 +0100 |
commit | 5172ecf855078938b00e9ac7f8dee2c2676f6e53 (patch) | |
tree | 586d6285d6d2aeb66a7a2458e01541ed78c8e174 /parse_bibtex_html | |
download | Public-5172ecf855078938b00e9ac7f8dee2c2676f6e53.tar.gz Public-5172ecf855078938b00e9ac7f8dee2c2676f6e53.tar.zst Public-5172ecf855078938b00e9ac7f8dee2c2676f6e53.zip |
Commit initial
Diffstat (limited to 'parse_bibtex_html')
-rwxr-xr-x | parse_bibtex_html | 172 |
1 files changed, 172 insertions, 0 deletions
diff --git a/parse_bibtex_html b/parse_bibtex_html new file mode 100755 index 0000000..5e54729 --- /dev/null +++ b/parse_bibtex_html | |||
@@ -0,0 +1,172 @@ | |||
1 | #!/usr/bin/env perl | ||
2 | # The MIT License (MIT) | ||
3 | # | ||
4 | # Copyright (c) 2011-2015 Ismaël Bouya http://www.normalesup.org/~bouya/ | ||
5 | # | ||
6 | # Permission is hereby granted, free of charge, to any person obtaining a copy | ||
7 | # of this software and associated documentation files (the "Software"), to deal | ||
8 | # in the Software without restriction, including without limitation the rights | ||
9 | # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell | ||
10 | # copies of the Software, and to permit persons to whom the Software is | ||
11 | # furnished to do so, subject to the following conditions: | ||
12 | # | ||
13 | # The above copyright notice and this permission notice shall be included in | ||
14 | # all copies or substantial portions of the Software. | ||
15 | # | ||
16 | # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR | ||
17 | # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, | ||
18 | # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE | ||
19 | # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER | ||
20 | # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, | ||
21 | # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN | ||
22 | # THE SOFTWARE. | ||
23 | |||
24 | |||
25 | use BibTeX::Parser; | ||
26 | use IO::File; | ||
27 | use utf8; | ||
28 | use strict; | ||
29 | use open ':utf8'; | ||
30 | |||
31 | sub parse_config_file { | ||
32 | my ($config_line, $Name, $Value, $Config); | ||
33 | (my $File, $Config) = @_; | ||
34 | if (!open (CONFIG, "$File")) { | ||
35 | print "ERROR: Config file not found : $File"; | ||
36 | exit(0); | ||
37 | } | ||
38 | my $multiline = 0; | ||
39 | |||
40 | while (<CONFIG>) { | ||
41 | $config_line=$_; | ||
42 | chop ($config_line); | ||
43 | $config_line =~ s/^\s*//; | ||
44 | $config_line =~ s/\s*$//; | ||
45 | if ( ($config_line !~ /^#/) && ($config_line ne "") ){ | ||
46 | if ($multiline) { | ||
47 | $$Config{$Name} =~ s/\\$//; | ||
48 | $$Config{$Name} .= $config_line; | ||
49 | } else { | ||
50 | ($Name, $Value) = split (/\s*=\s*/, $config_line); | ||
51 | $Value =~ s/^~/$ENV{"HOME"}/; | ||
52 | $$Config{$Name} = $Value; | ||
53 | } | ||
54 | $multiline = ($$Config{$Name} =~ /\\$/); | ||
55 | } | ||
56 | } | ||
57 | close(CONFIG); | ||
58 | } | ||
59 | |||
60 | my %Config; | ||
61 | &parse_config_file ($ENV{"HOME"}."/.parse_bibtex_html.rc", \%Config); | ||
62 | |||
63 | my $biblio = $Config{"biblio"}; | ||
64 | my $entete = $Config{"entete"}; | ||
65 | my $avant = $Config{"avant"}; | ||
66 | my $milieu = $Config{"milieu"}; | ||
67 | my $apres = $Config{"apres"}; | ||
68 | my $html = $Config{"html"}; | ||
69 | |||
70 | my $dossier = $Config{"dossier"}; | ||
71 | my $dossierweb = $Config{"dossier_web"}; | ||
72 | |||
73 | # http://webdesign.about.com/library/bl_htmlcodes.htm | ||
74 | sub echap { | ||
75 | my $t = shift or return; | ||
76 | $t =~ s/&/&/g; | ||
77 | $t =~ s/</</g; | ||
78 | $t =~ s/>/>/g; | ||
79 | |||
80 | $t =~ s/--/—/g; | ||
81 | |||
82 | $t =~ s/{?\\'a}?/á/g; | ||
83 | $t =~ s/{?\\`a}?/à/g; | ||
84 | $t =~ s/{?\\"a}?/ä/g; | ||
85 | |||
86 | $t =~ s/{?\\r A}?/Å/g; | ||
87 | |||
88 | $t =~ s/{?\\'e}?/é/g; | ||
89 | $t =~ s/{?\\`e}?/è/g; | ||
90 | $t =~ s/{?\\'E}?/É/g; | ||
91 | $t =~ s/{?\\"e}?/ë/g; | ||
92 | |||
93 | $t =~ s/{?\\\^i}?/î/g; | ||
94 | $t =~ s/{?\\"i}?/ï/g; | ||
95 | |||
96 | $t =~ s/{?\\"o}?/ö/g; | ||
97 | $t =~ s/{?\\"o}?/ö/g; | ||
98 | $t =~ s/{?\\=o}?/ō/g; | ||
99 | $t =~ s/{?\\o}?/ø/g; | ||
100 | |||
101 | $t =~ s/{?\\"u}?/ü/g; | ||
102 | $t =~ s/{?\\'u}?/ú/g; | ||
103 | |||
104 | $t =~ s/{?\\~n}?/ñ/g; | ||
105 | |||
106 | $t =~ s/{?\\c{?c}?}? ?/ç/g; | ||
107 | $t =~ s/{?\\'{?c}?}? ?/ć/g; | ||
108 | |||
109 | $t =~ s/{?\\v{? ?s}?}? ?/š/g; | ||
110 | |||
111 | $t =~ s/{?\^({[^}]+}|.)}?/<sup>$1<\/sup>/g; | ||
112 | $t =~ s/{(.*)}/$1/g; | ||
113 | return $t; | ||
114 | } | ||
115 | |||
116 | open F, ">".$milieu; | ||
117 | opendir(DIR, $dossier); | ||
118 | my @FILES = readdir(DIR); | ||
119 | my $fh = IO::File->new($biblio); | ||
120 | my $parser = BibTeX::Parser->new($fh); | ||
121 | print F "\t<ul>\n"; | ||
122 | my %liste = (); | ||
123 | |||
124 | while (my $entry = $parser->next ) { | ||
125 | if ($entry->parse_ok) { | ||
126 | my $type = $entry->type; | ||
127 | my $title = $entry->field("title"); | ||
128 | my $key = $entry->key; | ||
129 | my @authors = $entry->author; | ||
130 | # my @editors = $entry->editor; | ||
131 | my $auth = ""; | ||
132 | my @authors_sort = (); | ||
133 | foreach my $author (@authors) { | ||
134 | $auth .= (($author->first)?$author->first. " ":"") .(($author->von)?$author->von." ":"") . (($author->last)?$author->last:"") . ", "; | ||
135 | push(@authors_sort,$author->last); | ||
136 | } | ||
137 | @authors_sort = sort {lc $a cmp lc $b} @authors_sort; | ||
138 | my $cle_sort = shift(@authors_sort); | ||
139 | $auth = substr $auth, 0 , -2; | ||
140 | my $suffix = '(\.|_)'; | ||
141 | my @match = grep(/^$key$suffix/,@FILES); | ||
142 | my $i = 1; | ||
143 | my $chaine = "\t\t<li>"; | ||
144 | $auth = echap $auth; | ||
145 | $title = echap $title; | ||
146 | if($auth =~ m/\\/ || $title =~ m/\\/) { | ||
147 | warn "Unparsed item : $auth, $title"; | ||
148 | } | ||
149 | $chaine .= "<span class='biblio_titre'>".$title."</span><br />".$auth."<br />\n"; | ||
150 | @match = sort {lc $a cmp lc $b} @match; | ||
151 | foreach my $item (@match) { | ||
152 | $chaine .= "\t\t\t<a href='".$dossierweb.$item."'>fichier ".$i++."</a> \n"; | ||
153 | } | ||
154 | $chaine .="\t\t\t<a id='".$key."' href='#".$key."' class='bibtex'>BibTeX</a>\n"; | ||
155 | my $raw = $entry->raw_bibtex; | ||
156 | $raw =~ s/&/&/g; | ||
157 | $chaine .= "\t\t\t<pre class='bibtex'>".$raw."</pre>\n"; | ||
158 | $chaine .= "\t\t\t</li>\n"; | ||
159 | $liste{$cle_sort." ".$key} = $chaine; | ||
160 | } else { | ||
161 | warn "Error parsing file: " . $entry->error; | ||
162 | } | ||
163 | } | ||
164 | |||
165 | foreach my $key (sort keys %liste) { | ||
166 | print F $liste{$key}; | ||
167 | } | ||
168 | |||
169 | print F "\t\t</ul>"; | ||
170 | close F; | ||
171 | |||
172 | exec "cat $entete $avant $milieu $apres 1> $html" or die "$!\n"; | ||