source: gsdl/trunk/perllib/plugins/MediaWikiPlug.pm@ 14337

Last change on this file since 14337 was 14337, checked in by anna, 15 years ago

Fixed a bug in extracting search box.

  • Property svn:keywords set to Author Date Id Revision
File size: 26.0 KB
Line 
1###########################################################################
2#
3# MediaWikiPlug.pm -- html plugin with extra facilities for wiki page
4#
5# A component of the Greenstone digital library software
6# from the New Zealand Digital Library Project at the
7# University of Waikato, New Zealand.
8#
9# Copyright (C) 1999 New Zealand Digital Library Project
10#
11# This program is free software; you can redistribute it and/or modify
12# it under the terms of the GNU General Public License as published by
13# the Free Software Foundation; either version 2 of the License, or
14# (at your option) any later version.
15#
16# This program is distributed in the hope that it will be useful,
17# but WITHOUT ANY WARRANTY; without even the implied warranty of
18# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19# GNU General Public License for more details.
20#
21# You should have received a copy of the GNU General Public License
22# along with this program; if not, write to the Free Software
23# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
24#
25###########################################################################
26# This plugin is to process an HTML file from a MediaWiki website which downloaded by
27# the MediaWikiDownload plug. This plugin will trim MediaWiki functional sections like
28# login, discussion, history, etc. Only the navigation and search section could be preserved.
29# Searchbox will be modified to search the Greenstone collection instead of the website.
30# It also can automatically add the table of contents on the website's Main_Page to the
31# collection's Home page.
32
33package MediaWikiPlug;
34
35use HTMLPlug;
36# use ImagePlug;
37# use File::Copy;
38use unicode;
39
40
41#use strict; # every perl program should have this!
42#no strict 'refs'; # make an exception so we can use variables as filehandles
43
44sub BEGIN {
45 @MediaWikiPlug::ISA = ('HTMLPlug');
46}
47
48my $arguments =
49 [
50 # show the table of contents on collection's home page
51 { 'name' => "show_toc",
52 'desc' => "{MediaWikiPlug.show_toc}",
53 'type' => "flag",
54 'reqd' => "no"},
55 # set to delete the table of contents section on each MediaWiki page
56 { 'name' => "delete_toc",
57 'desc' => "{MediaWikiPlug.delete_toc}",
58 'type' => "flag",
59 'reqd' => "no"},
60 # regexp to match the table of contents
61 { 'name' => "toc_exp",
62 'desc' => "{MediaWikiPlug.toc_exp}",
63 'type' => "regexp",
64 'reqd' => "no",
65 'deft' => "<table([^>]*)id=(\\\"|')toc(\\\"|')(.|\\n)*</table>\\n" },
66 # set to delete the navigation section
67 { 'name' => "delete_nav",
68 'desc' => "{MediaWikiPlug.delete_nav}",
69 'type' => "flag",
70 'reqd' => "no",
71 'deft' => ""},
72 # regexp to match the navigation section
73 { 'name' => "nav_div_exp",
74 'desc' => "{MediaWikiPlug.nav_div_exp}",
75 'type' => "regexp",
76 'reqd' => "no",
77 'deft' => "<div([^>]*)id=(\\\"|')p-navigation(\\\"|')(.|\\n)*?<\/div>" },
78 # set to delete the searchbox section
79 { 'name' => "delete_searchbox",
80 'desc' => "{MediaWikiPlug.delete_searchbox}",
81 'type' => "flag",
82 'reqd' => "no",
83 'deft' => ""},
84 # regexp to match the searchbox section
85 { 'name' => "searchbox_div_exp",
86 'desc' => "{MediaWikiPlug.searchbox_div_exp}",
87 'type' => "regexp",
88 'reqd' => "no",
89 'deft' => "<div([^>]*)id=(\\\"|')p-search(\\\"|')(.|\\n)*?<\/div>"},
90 # regexp to match title suffix
91 # can't use the title_sub option in HTMLPlug instead
92 # because title_sub always matches from the begining
93 { 'name' => "remove_title_suffix_exp",
94 'desc' => "{MediaWikiPlug.remove_title_suffix_exp}",
95 'type' => "regexp",
96 'reqd' => "no",
97 'deft' => ""}
98 ];
99
100my $options = { 'name' => "MediaWikiPlug",
101 'desc' => "{MediaWikiPlug.desc}",
102 'abstract' => "no",
103 'inherits' => "yes",
104 'args' => $arguments };
105
106sub new {
107 my ($class) = shift (@_);
108 my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
109 push(@$pluginlist, $class);
110
111 if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
112 if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
113
114 my $self = new HTMLPlug($pluginlist, $inputargs, $hashArgOptLists);
115 return bless $self, $class;
116}
117
118
119
120sub process {
121 my $self = shift (@_);
122 my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
123 my $outhandle = $self->{'outhandle'};
124
125 print $outhandle "MediaWikiPlug: processing $file\n" if $self->{'verbosity'} > 1;
126
127 my @head_and_body = split(/<body/i,$$textref);
128 my $head = shift(@head_and_body);
129 my $body_text = join("<body", @head_and_body);
130
131 $head =~ m/<title>(.+)<\/title>/i;
132 my $doctitle = $1 if defined $1;
133
134 if (defined $self->{'metadata_fields'} && $self->{'metadata_fields'}=~ /\S/) {
135 my @doc_properties = split(/<xml>/i,$head);
136 my $doc_heading = shift(@doc_properties);
137 my $rest_doc_properties = join(" ", @doc_properties);
138
139 my @extracted_metadata = split(/<\/xml>/i, $rest_doc_properties);
140 my $extracted_metadata = shift (@extracted_metadata);
141 $self->extract_metadata($extracted_metadata, $metadata, $doc_obj);
142 }
143
144 # set the title here if we haven't found it yet
145 if (!defined $doc_obj->get_metadata_element ($doc_obj->get_top_section(), "Title")) {
146 if (defined $doctitle && $doctitle =~ /\S/) {
147 # remove suffix in title if required
148 my $remove_suffix_exp = $self->{'remove_title_suffix_exp'};
149 if (defined $remove_suffix_exp && $remove_suffix_exp =~ /\S/){
150 $doctitle =~ s/$remove_suffix_exp//i;
151 }
152 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Title", $doctitle);
153 } else {
154 $self->title_fallback($doc_obj,$doc_obj->get_top_section(),$file);
155 }
156 }
157
158 # we are only interested in the column-contents div <div id="column-content">
159 # remove header section, it may contain header images or additional search boxes
160 my $header_exp = "<div([^>]*)id=(\"|')container(\"|')([^>]*)>(.|\\n)*<div([^>]*)id=(\"|')column-content";
161 $body_text =~ s/$header_exp/<div$1id='container'$4><div$6id='column-content/isg;
162
163 # remove timeline
164 $body_text =~ s/<div([^>]*)class=("|')smwtimeline("|')[\s\S]*?<\/div>//mg;
165
166 # remove extra bits
167 my $extra_bits = "Retrieved from(.+)</a>\"";
168 $body_text =~ s/$extra_bits//isg;
169
170 $body_text =~ s/(<p[^>]*><span[^>]*><o:p>&nbsp;<\/o:p><\/span><\/p>)//isg;
171 $body_text =~ s/(<p[^>]*><o:p>&nbsp;<\/o:p><\/p>)//isg;
172 $body_text =~ s/<!\[if !vml\]>/<![if vml]>/g;
173 $body_text =~ s/(&nbsp;)+/&nbsp;/sg;
174
175 # get rid of the [edit] buttons
176 $body_text =~ s/\[<a([^>]*)>edit<\/a>]//g;
177 # get rid of the last time edit information at the bottom
178 $body_text =~ s/<a href="([^>]*)edit([^>]*)"([^>]*?)>(\w+)<\/a> \d\d:\d\d,([\s|\w]*?)\(PST\)//g;
179 # get rid of the (Redirected from ...)
180 $body_text =~ s/\(Redirected from <a ([^>]*)>(\w|\s)*?<\/a>\)//isg;
181
182 # escape texts macros
183 $body_text =~ s/_([^\s]*)_/_<span>$1<\/span>_/isg;
184 # may change the links, like Greenstone_Documentation_All.html, then change back
185 $body_text =~ s/<a([^>]*)_<span>([^>]*)<\/span>_/<a$1_$2_/isg;
186
187 # define file delimiter for different platforms
188 my $file_delimiter;
189 if ($ENV{'GSDLOS'} =~ /^windows$/i) {
190 $file_delimiter = "\\";
191 } else {
192 $file_delimiter = "/";
193 }
194
195 # IMPORTANT: different delimiter for $base_dir and $file
196 # $base_dir use forward slash for both windows and linux
197 # print "\nbase_dir : $base_dir\n\n"; # windows: C:/Program Files/Greenstone2.73/collect/wiki/import
198 # linux: /research/lh92/greenstone/greenstone2.73/collect/wiki/import
199 # $file use different delimiters : forward slash for linux; backward slash for windows
200 # print "\nfile : $file\n\n"; # windows: greenstone.sourceforge.net\wiki\index.php\Access_Processing_using_DBPlug.html
201 # linux: greenstone.sourceforge.net/wiki/index.php/Using_GreenstoneWiki.html
202
203 # get the base url for the MediaWiki website
204 my $safe_delimiter = &safe_escape_regexp($file_delimiter);
205 my @url_dirs=split($safe_delimiter, $file);
206 my $url_base = $url_dirs[0];
207
208 # Re-check css files associated with MediaWiki pages
209 if(defined $base_dir && $base_dir ne ""){
210 my @css_files;
211 my $css_file_count = 0;
212
213 # find all the stylesheets imported with @import statement
214 while($head =~ m"<style type=\"text/css\"(.+)import \"(.+)\""ig){
215 $css_files[$css_file_count++] = $2 if defined $2;
216 }
217
218 # download the stylesheets if we haven't downloaded them yet
219 # add prefix to each style elmement, comment out the body element
220 # and copy the files to collection's images folder
221 for ($css_file_count = 0; $css_file_count < scalar(@css_files); $css_file_count++) {
222
223 my $css_file = $css_files[$css_file_count];
224
225 # remove prefix gli/cache directory
226 $css_file =~ s/^(.+)gli(\\|\/)cache(\\|\/)//i;
227
228 # change the \ delimiter in $css_file to / for consistency
229 $css_file =~ s/\\/\//isg;
230 if($css_file !~ /$url_base/) {
231 $css_file = $url_base . $css_file;
232 }
233
234 # trim the ? mark append to the end of a stylesheet
235 $css_file =~ s/\?(.+)$//isg;
236
237 my $css_file_path = &util::filename_cat($base_dir, $css_file);
238
239 # do nothing if we have already downloaded the css files
240 if (! -e $css_file_path) {
241
242 # check the stylesheet's directory in the import folder
243 # if the directory doesn't exist, create one
244 my @dirs = split(/\//i,$css_file);
245 my $path_check = "$base_dir/";
246 for (my $i = 0; $i < (scalar(@dirs)-1); $i++) {
247 $path_check .= $dirs[$i] . "/";
248 mkdir($path_check) if (! -d $path_check );
249 }
250
251 # NOTE: wget needs configuration to directly access Internet
252 # These files should already downloaded if we used the MediaWikiDownload
253 # downloading
254 $css_file = "http://$css_file";
255 print "\ndownloading : " . $css_file . "\n\n";
256 system("wget", "--non-verbose", "$css_file", "--output-document=$css_file_path");
257 if ($? != 0) {
258 print "[ERROR] Download Failed! Make sure WGet connects to Internet directly \n";
259 print "[ERROR] OR ues the MediaWikiDownload in the GLI DownloadPanel to download from a MediaWiki website\n";
260 unlink("$css_file_path");
261 }
262 } # done with download
263
264 # add a prefix "#wikispecificstyle" to each element
265 # because we want to preserve this website's formats and don't want to mess up with Greenstone formats
266 # so we will wrap the web page with a div with id = wikispecificstyle
267 my $css_content;
268 if(open(INPUT, "<$css_file_path")){
269 while(my $line = <INPUT>){
270 # comment out the body element because we change the body to div
271 $line =~ s/^(\s*)body(\s*){(\s*)$/$1\/*body$2*\/{$3/isg;
272
273 if($line =~ m/^(.+)\{/i || $line =~ m/^(\s)*#/i){
274 $line = "#wikispecificstyle " . $line;
275 }
276 $css_content .= $line;
277 }
278 close(INPUT);
279 open(OUTPUT, ">$css_file_path");
280 print OUTPUT $css_content;
281 close(OUTPUT);
282 }
283
284 # Copy the modified stylesheets to collection's images folder
285 # for future customization
286 my $images_dir = $base_dir;
287 $images_dir =~ s/import$/images/;
288 $css_file =~ m/(.*)\/(.*)$/;
289 $images_dir = &util::filename_cat($images_dir, $2);
290
291 if(open(OUTPUT, ">$images_dir")){
292 print OUTPUT $css_content;
293 close(OUTPUT);
294 }
295 }
296 }
297
298
299 # by default, only preserve navigation box and search box
300 # others like toolbox, interaction, languages box, will be removed
301
302 # extract the larger part -- footer section
303 my $print_footer = "<div class=\"printfooter\">(.|\n)+</body>";
304 $body_text =~ /$print_footer/;
305 my $footer = "";
306 $footer = $& if defined $&;
307 $footer =~ s/<\/body>//isg;
308
309 # trim the comments first
310 $footer =~ s/<!--[\s\S]*?--[ \t\n\r]*>//isg;
311
312 # contain sections that are to be preserved
313 my $preserve_sections = "";
314
315 # process the navigation section
316 my $nav_match_exp = "<div([^>]*)id=(\"|')p-navigation(\"|')(.|\n)*?<\/div>";
317 if (defined $self->{'nav_div_exp'}) {
318 $nav_match_exp = $self->{'nav_div_exp'} if ($self->{'nav_div_exp'} =~ /\S/) ;
319 }
320
321 if (defined $self->{'delete_nav'} && ($self->{'delete_nav'} eq "1")) {
322 # do nothing
323 } else {
324 if ($footer =~ m/$nav_match_exp/ig) {
325 $preserve_sections = $& ;
326 } else {
327 print $outhandle "Can't find the navigation section with : $nav_match_exp\n";
328 }
329 # if($preserve_sections =~/\S/){
330 # $preserve_sections .= "</div>";
331 # }
332 }
333
334 # process the searchbox section
335 my $searchbox_exp = "<div([^>]*)id=(\"|')p-search(\"|')(.|\\n)*?<\/div>";
336 if(defined $self->{'searchbox_div_exp'}) {
337 $searchbox_exp = $self->{'searchbox_div_exp'} if ($self->{'searchbox_div_exp'} =~ /\S/);
338 }
339
340 my $searchbox_section = "";
341 $footer =~ m/$searchbox_exp/ig;
342 $searchbox_section = $& if defined $&;
343
344 # make the searchbox form work in Greenstone
345 if($searchbox_section =~ /\S/){
346 # replace action
347 $searchbox_section =~ s/action="([^>]*)"/action="_gwcgi_"/isg;
348
349 # remove buttons
350 $searchbox_section =~ s/name="search"/name="q"/isg;
351 $searchbox_section =~ s/name="go"//isg;
352 $searchbox_section =~ s/name="fulltext"//isg;
353
354 # get collection name from $base_dir for c param
355 $base_dir =~ m/\/collect\/(.+)\//i;
356 my $collection_name = "";
357 $collection_name = $1 if defined $1;
358
359 # add Greenstone search params
360 my $hidden_params = "<input type=\"hidden\" name=\"a\" value=\"q\"/>\n"
361 ."<input type=\"hidden\" name=\"c\" value=\"$collection_name\"/>\n";
362 # ."<input type=\"hidden\" name=\"fqf\" value=\"TX\"/>\n"
363 # ."<input type=\"hidden\" name=\"r\" value=\"1\">\n";
364
365 $searchbox_section =~ s/<form([^>]*)>/<form$1>\n$hidden_params/isg;
366
367 # $searchbox_section .= "</div>";
368 } else {
369 print $outhandle "Can't find the searchbox section with : $searchbox_section\n";
370 }
371
372 # either delete or replace the searchbox
373 if(defined $self->{'delete_searchbox'} && $self->{'delete_searchbox'} eq "1") {
374 # do nothing
375 } else {
376 $preserve_sections .= "\n$searchbox_section\n";
377 }
378
379 if($preserve_sections ne ""){
380 $preserve_sections = "<div id=\"column-one\">\n" . $preserve_sections . "\n</div>\n";
381 }
382 $preserve_sections = "</div></div></div>\n" . $preserve_sections . "\n</body>";
383
384 $body_text =~ s/$print_footer/$preserve_sections/isg;
385
386
387 # delete other forms in the page
388 my @forms;
389 my $form_count = 0;
390 while($body_text =~ m/<form([^>]*)name=("|')([^>"']*)?("|')/isg){
391 next if($3 eq "searchform");
392 $forms[$form_count++] = $&;
393 }
394 foreach my $form (@forms) {
395 $body_text =~ s/$form[\s\S]*?<\/form>//m;
396 }
397
398 # process links.
399 # because current WGET 1.10 the -k and -E option doesn't work together
400 # need to 'manually' convert the links to relative links
401 # Dealing with 3 types of links:
402 # -- outgoing links
403 # -- if we have downloaded the target files, link to the internal version (relative link)
404 # -- otherwise, link to the external version (absolute links)
405 # -- in-page links (relative link)
406
407 # NOTE: (important)
408 # must use the MediaWikiDownload in GLI Download Panel to download files from a MediaWiki website
409 # otherwise, the internal links may have problems
410
411 # remove the title attribute of <a> tag
412 $body_text =~ s/<a([^>]*)title="(.*?)"/<a$1/isg;
413
414 # extract all the links
415 my @links;
416 my $link_count = 0;
417 while($body_text =~ m/(href|src)="([^>\s]*)$url_base\/([^>\s]*)"/ig){
418 $links[$link_count++] = "$1=\"$2$url_base/$3\"";
419 }
420
421 foreach my $cur_link (@links) {
422 # escape greedy match + character
423 $cur_link =~ s/\+/\\+/isg;
424
425 $cur_link =~ m/(.+)"([^>]*)$url_base\/([^>\s]*)"/;
426 my $external_file_path = "$1\"http://$url_base/$3\"";
427
428 $body_text =~ s/$cur_link/$external_file_path/i;
429 }
430
431 # tag links to new wiki pages as red
432 $body_text =~ s/<a([^>]*)class="new"([^>]*)>/<a$1style="color:red"$2)>/gi;
433
434 # tag links to pages external of the MediaWiki website as blue
435 $body_text =~ s/<a([^>]*)class='external text'([^>]*)>/<a$1style="color:blue"$2)>/gi;
436
437
438 # process the table-of-contents section
439 # if 'show_toc' is set, add Main_Page's toc to the collection's About page, change extra.dm file
440 # 1. read _content_ macro from about.dm
441 # 2. append the toc, change all links to the Greenstone internal format for relative links
442 # 3. write to the extra.dm
443 # TODO: we assume the _about:content_ hasn't been specified before
444 # so needs to add function to handle when the macro is already in the extra.dm
445 if($self->{'show_toc'}==1 && $file =~ m/Main_Page.(html|htm)$/){
446
447 # extract toc of the Main_Page
448 my $mainpage_toc = "";
449 my $toc_exp = "<table([^>]*)id=(\"|')toc(\"|')(.|\\n)*</table>\\n";
450 if($self->{'toc_exp'} =~ /\S/){
451 $toc_exp = $self->{'toc_exp'};
452 }
453 if($body_text =~ /$toc_exp/){
454 $mainpage_toc = $&;
455 }
456
457 if($mainpage_toc =~ /\S/) {
458
459 # change the in-page links to relative links, for example, change <a href="#section1"> to
460 # <a href="_httpquery_&a=extlink&rl=1&href=http://www.mediawikisite.com/Main_Page.html#section1">
461 my $file_url_format = $file;
462 $file_url_format =~ s/\\/\//isg;
463 $file_url_format = "http://" . $file_url_format;
464
465 # encode as URL, otherwise doesn't work on Windows
466 $file_url_format =~ s/([^A-Za-z0-9])/sprintf("%%%02X", ord($1))/seg;
467 $mainpage_toc =~ s/<a href="([^>"#]*)#([^>"]*)"/<a href="_httpquery_&a=extlink&rl=1&href=$file_url_format#$2"/isg;
468
469
470 # read the collection's extra.dm
471 my $macro_path = $base_dir;
472 $macro_path =~ s/import$/macros/;
473 my $extradm_file = &util::filename_cat($macro_path, "extra.dm");
474
475 my $extra_dm = "";
476 if(open(INPUT, "<$extradm_file")){
477 while(my $line = <INPUT>){
478 $extra_dm .= $line;
479 }
480 } else {
481 print $outhandle "can't open file $extradm_file\n";
482 }
483 close(INPUT);
484
485 # check whether we have changed the macros
486 my @packages = split("package ", $extra_dm);
487 my $about_package = "";
488 foreach my $package (@packages) {
489 $about_package = "package " . $package if($package =~ /^about/);
490 }
491
492 my $update_extra_dm = 0;
493
494 if( $about_package =~ /\S/ && $about_package =~ m/_content_(\s*){/ && $about_package =~ m/$mainpage_toc/){
495 print $outhandle "_content_ macro already changed!!!!\n";
496 }
497 # if extra.dm doesn't have an "about package"
498 elsif ($about_package !~ /\S/) {
499 # read _content_ macro from $GSDLHOME/macros/about.dm file
500 my $global_about_package = &read_content_from_about_dm();
501
502 # create the extra _content_ macro for this collection
503 # add the original content of the _content_ macro
504 $global_about_package =~ m/{(.|\n)*<\/div>\n\n/;
505
506 # append the new about package to extra.dm
507 $extra_dm .= "\n\npackage about\n_content_$&\n\n";
508 $extra_dm .= "<div class=\"section\">\n$mainpage_toc\n</div>\n</div>\n}";
509
510 $update_extra_dm = 1;
511 }
512 # the about package exists, but either doesn't have the _content_ macro or
513 # the _content_ macro doesn't contain the toc
514 else {
515 # check if there is a content macro
516 my $content_macro_existed = 0;
517 $content_macro_existed = ($about_package =~ /(\s*|\n)_content_(\s*){/);
518
519 # if there is one
520 # append a new section div for toc to the end of the document section
521 if($content_macro_existed ==1) {
522 $about_package =~ /(\s*|\n)_content_(\s*){(.|\n)*?}/;
523 my $content_macro = $&;
524 my $new_content_macro = $content_macro;
525 $new_content_macro =~ s/<div[^>]*class="document">(.|\n)*<\/div>/<div$1class="document">$2\n\n<div class="section">\n$mainpage_toc\n<\/div>\n<\/div>/;
526 $extra_dm =~ s/$content_macro/$new_content_macro/mg;
527 }
528 # otherwise, append _content_ macro to the about package
529 else {
530 my $new_about_package = $about_package;
531 $content_macro = &read_content_from_about_dm();
532 $content_macro =~ m/{(.|\n)*<\/div>\n\n/;
533
534 $new_about_package .= "\n\n_content_$&\n\n";
535 $new_about_package .= "<div class=\"section\">\n$mainpage_toc\n</div>\n</div>\n}";
536 $extra_dm =~ s/$about_package/$new_about_package/mg;
537 }
538
539 # either the case, we need to update the extra.dm
540 $update_extra_dm = 1;
541 }
542
543 if($update_extra_dm==1){
544 # write to the extra.dm file of the collection
545 if (open(OUTPUT, ">$extradm_file")) {
546 print OUTPUT $extra_dm;
547 } else {
548 print "can't open $extradm_file\n";
549 }
550 close(OUTPUT);
551 }
552 } else {
553 print $outhandle "Main_Page doesn't have a table-of-contents section\n";
554 }
555 }
556
557 # If delete_toc is set, remove toc and tof contents.
558 if (defined $self->{'delete_toc'} && ($self->{'delete_toc'} == 1)){
559 if (defined $self->{'toc_exp'} && $self->{'toc_exp'} =~ /\S/){
560 # print "\nit matches toc_exp!!\n" if $body_text =~ /$self->{'toc_exp'}/;
561 if ($body_text =~ /$self->{'toc_exp'}/) {
562 $body_text =~ s/$self->{'toc_exp'}//i;
563 }
564 }
565 }
566
567 $$textref = "<body" . $body_text;
568
569 # Wrap the whole page with <div id="wikispecificstyle"></div>
570 # keep the style of this website and don't mess up with the Greenstone styles
571 $$textref =~ s/<body([^>]*)>/$&\n<div id="wikispecificstyle">\n/is;
572 $$textref =~ s/<\/body>/<\/div><\/body>/is;
573
574 $self->SUPER::process(@_);
575
576 return 1;
577}
578
579
580sub extract_metadata
581{
582 my $self = shift (@_);
583 my ($textref, $metadata, $doc_obj) = @_;
584 my $outhandle = $self->{'outhandle'};
585
586 return if (!defined $textref);
587
588 # metadata fields to extract/save. 'key' is the (lowercase) name of the
589 # html meta, 'value' is the metadata name for greenstone to use
590 my %find_fields = ();
591 my ($tag,$value);
592
593 my $orig_field = "";
594 foreach my $field (split /,/, $self->{'metadata_fields'}) {
595 # support tag<tagname>
596 if ($field =~ /^(.*?)<(.*?)>$/) {
597 # "$2" is the user's preferred gs metadata name
598 $find_fields{lc($1)}=$2; # lc = lowercase
599 $orig_field = $1;
600 } else { # no <tagname> for mapping
601 # "$field" is the user's preferred gs metadata name
602 $find_fields{lc($field)}=$field; # lc = lowercase
603 $orig_field = $field;
604 }
605
606 if ($textref =~ m/<o:$orig_field>(.*)<\/o:$orig_field>/i){
607 $tag = $orig_field;
608 $value = $1;
609 if (!defined $value || !defined $tag){
610 #print $outhandle "StructuredHTMLPlug: can't find VALUE in \"$tag\"\n";
611 next;
612 } else {
613 # clean up and add
614 chomp($value); # remove trailing \n, if any
615 $tag = $find_fields{lc($tag)};
616 #print $outhandle " extracted \"$tag\" metadata \"$value\"\n"
617 # if ($self->{'verbosity'} > 2);
618 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), $tag, $value);
619 }
620 }
621 }
622}
623
624sub safe_escape_regexp
625{
626 my $regexp = shift (@_);
627
628 # if ($ENV{'GSDLOS'} =~ /^windows$/i) {
629 $regexp =~ s/\\/\\\\/isg;
630 #} else {
631 $regexp =~ s/\//\\\//isg;
632 #}
633 return $regexp;
634}
635
636sub read_content_from_about_dm
637{
638 my $about_macro_file = &util::filename_cat($ENV{'GSDLHOME'}, "macros", "about.dm");
639 my $about_page_content = "";
640 if (open(INPUT, "<$about_macro_file")){
641 while (my $line=<INPUT>){
642 $about_page_content .= $line;
643 }
644 } else {
645 print $outhandle "can't open file $about_macro_file\n";
646 }
647 close(INPUT);
648
649 # extract the _content_ macro
650 $about_page_content =~ m/_content_ {(.|\n)*<\/div>\n\n<\/div>\n}/i;
651 $about_page_content = $&;
652
653 return $about_page_content;
654}
655
6561;
Note: See TracBrowser for help on using the repository browser.