1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # HTMLPlug.pm -- basic html plugin
|
---|
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 |
|
---|
27 | #
|
---|
28 | # Note that this plugin handles frames only in a very simple way
|
---|
29 | # i.e. each frame is treated as a separate document. This means
|
---|
30 | # search results will contain links to individual frames rather
|
---|
31 | # than linking to the top level frameset.
|
---|
32 | # There may also be some problems caused by the _parent target
|
---|
33 | # (it's removed by this plugin)
|
---|
34 | # To use frames properly you'll need to use the WebPlug plugin.
|
---|
35 | #
|
---|
36 |
|
---|
37 |
|
---|
38 | package HTMLPlug;
|
---|
39 |
|
---|
40 | use BasPlug;
|
---|
41 | use ghtml;
|
---|
42 | use unicode;
|
---|
43 | use util;
|
---|
44 | use parsargv;
|
---|
45 |
|
---|
46 | sub BEGIN {
|
---|
47 | @ISA = ('BasPlug');
|
---|
48 | }
|
---|
49 |
|
---|
50 | sub print_usage {
|
---|
51 | print STDERR "\n usage: plugin HTMLPlug [options]\n\n";
|
---|
52 | print STDERR " options:\n";
|
---|
53 | print STDERR " -nolinks Don't make any attempt to trap links (setting this flag may\n";
|
---|
54 | print STDERR " improve speed of building/importing but any relative links within\n";
|
---|
55 | print STDERR " documents will be broken).\n";
|
---|
56 | print STDERR " -keep_head Don't remove headers from html files.\n";
|
---|
57 | print STDERR " -no_metadata Don't attempt to extract any metadata from files.\n";
|
---|
58 | print STDERR " -metadata_fields Comma separated list of metadata fields to attempt to extract.\n";
|
---|
59 | print STDERR " Defaults to 'Title'.\n";
|
---|
60 | print STDERR " Use 'tag<tagname>' to have the contents of the first <tagname>\n";
|
---|
61 | print STDERR " pair put in a metadata element called 'tagname' Capitalise \n";
|
---|
62 | print STDERR " 'tagname' as you want the metadata capitalised in the GML \n";
|
---|
63 | print STDERR " file, since the tag extraction is case insensitive.\n";
|
---|
64 | print STDERR " -hunt_creator_metadata Find as much metadata as possible on authorship and place it in the\n ";
|
---|
65 | print STDERR " 'Creator' field. Requires the -metadata_fields flag.\n ";
|
---|
66 | print STDERR " -w3mir Set if w3mir was used to generate input file structure.\n";
|
---|
67 | print STDERR " -assoc_files Perl regular expression of file extensions to associate with\n";
|
---|
68 | print STDERR " html documents. Defaults to '(?i)\.(jpe?g|gif|png|css)\$'\n";
|
---|
69 | print STDERR " -rename_assoc_files Renames files associated with documents (e.g. images). Also\n";
|
---|
70 | print STDERR " creates much shallower directory structure (useful when creating\n";
|
---|
71 | print STDERR " collections to go on cd-rom).\n\n";
|
---|
72 | print STDERR " -title_sub Substitution expression to modify string stored as Title.\n";
|
---|
73 | print STDERR " Used by, for example, PDFHtml to remove Page 1 etc from text\n";
|
---|
74 | print STDERR " chosen to be used as the title.\n";
|
---|
75 | }
|
---|
76 |
|
---|
77 | sub new {
|
---|
78 | my $class = shift (@_);
|
---|
79 | my $self = new BasPlug ($class, @_);
|
---|
80 |
|
---|
81 | if (!parsargv::parse(\@_,
|
---|
82 | q^nolinks^, \$self->{'nolinks'},
|
---|
83 | q^keep_head^, \$self->{'keep_head'},
|
---|
84 | q^no_metadata^, \$self->{'no_metadata'},
|
---|
85 | q^metadata_fields/.*/Title^, \$self->{'metadata_fields'},
|
---|
86 | q^hunt_creator_metadata^, \$self->{'hunt_creator_metadata'},
|
---|
87 | q^w3mir^, \$self->{'w3mir'},
|
---|
88 | q^assoc_files/.*/(?i)\.(jpe?g|gif|png|css)$^, \$self->{'assoc_files'},
|
---|
89 | q^rename_assoc_files^, \$self->{'rename_assoc_files'},
|
---|
90 | q^title_sub/.*/^, \$self->{'title_sub'},
|
---|
91 | "allow_extra_options")) {
|
---|
92 |
|
---|
93 | print STDERR "\nIncorrect options passed to HTMLPlug, check your collect.cfg configuration file\n";
|
---|
94 | &print_usage();
|
---|
95 | die "\n";
|
---|
96 | }
|
---|
97 |
|
---|
98 | $self->{'aux_files'} = {};
|
---|
99 | $self->{'dir_num'} = 0;
|
---|
100 | $self->{'file_num'} = 0;
|
---|
101 |
|
---|
102 | return bless $self, $class;
|
---|
103 | }
|
---|
104 |
|
---|
105 |
|
---|
106 | sub get_default_block_exp {
|
---|
107 | my $self = shift (@_);
|
---|
108 |
|
---|
109 | return q^(?i)\.(gif|jpe?g|png|rtf|css)$^;
|
---|
110 | }
|
---|
111 |
|
---|
112 | sub get_default_process_exp {
|
---|
113 | my $self = shift (@_);
|
---|
114 |
|
---|
115 | # the last option is an attempt to encode the concept of an html query ...
|
---|
116 | return q^(?i)(\.html?|\.shtml|\.shm|\.asp|\.php|\.cgi|.+\?.+=.*)$^;
|
---|
117 | }
|
---|
118 |
|
---|
119 |
|
---|
120 | # do plugin specific processing of doc_obj
|
---|
121 | sub process {
|
---|
122 | my $self = shift (@_);
|
---|
123 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
|
---|
124 | my $outhandle = $self->{'outhandle'};
|
---|
125 |
|
---|
126 | print $outhandle "HTMLPlug: processing $file\n"
|
---|
127 | if $self->{'verbosity'} > 1;
|
---|
128 |
|
---|
129 | my $cursection = $doc_obj->get_top_section();
|
---|
130 |
|
---|
131 | $self->extract_metadata ($textref, $metadata, $doc_obj, $cursection)
|
---|
132 | unless $self->{'no_metadata'};
|
---|
133 |
|
---|
134 | # Store URL for page as metadata - this can be used for an
|
---|
135 | # altavista style search interface. The URL won't be valid
|
---|
136 | # unless the file structure contains the domain name (i.e.
|
---|
137 | # like when w3mir is used to download a website).
|
---|
138 | my $web_url = "http://$file";
|
---|
139 | $web_url =~ s/\\/\//g; # for windows
|
---|
140 | $doc_obj->add_utf8_metadata($cursection, "URL", $web_url);
|
---|
141 |
|
---|
142 | # remove header and footer
|
---|
143 | if (!$self->{'keep_head'}) {
|
---|
144 | $$textref =~ s/^.*?<body[^>]*>//is;
|
---|
145 | $$textref =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
|
---|
146 | }
|
---|
147 |
|
---|
148 | # trap links
|
---|
149 | if (!$self->{'nolinks'}) {
|
---|
150 |
|
---|
151 | # usemap="./#index" not handled correctly => change to "#index"
|
---|
152 | $$textref =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
---|
153 | $self->replace_usemap_links($1, $2, $3)/isge;
|
---|
154 |
|
---|
155 | $$textref =~ s/(<(?:a|area|frame|link)\s+[^>]*?\s*(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
---|
156 | $self->replace_href_links ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
|
---|
157 | }
|
---|
158 |
|
---|
159 | # trap images
|
---|
160 |
|
---|
161 | $$textref =~ s/(<img[^>]*?src\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
---|
162 | $self->replace_images ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
|
---|
163 |
|
---|
164 | # add text to document object
|
---|
165 | $doc_obj->add_utf8_text($cursection, $$textref);
|
---|
166 |
|
---|
167 | return 1;
|
---|
168 | }
|
---|
169 |
|
---|
170 |
|
---|
171 |
|
---|
172 |
|
---|
173 | sub replace_images {
|
---|
174 | my $self = shift (@_);
|
---|
175 | my ($front, $link, $back, $base_dir,
|
---|
176 | $file, $doc_obj, $section) = @_;
|
---|
177 |
|
---|
178 | $link =~ s/\n/ /g;
|
---|
179 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
---|
180 | return $front . $self->add_file ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) . $back;
|
---|
181 | }
|
---|
182 |
|
---|
183 | sub replace_href_links {
|
---|
184 | my $self = shift (@_);
|
---|
185 | my ($front, $link, $back, $base_dir, $file, $doc_obj, $section) = @_;
|
---|
186 |
|
---|
187 | # attempt to sort out targets - frames are not handled
|
---|
188 | # well in this plugin and some cases will screw things
|
---|
189 | # up - e.g. the _parent target (so we'll just remove
|
---|
190 | # them all ;-)
|
---|
191 | $front =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
---|
192 | $back =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
---|
193 | $front =~ s/target=\"?_parent\"?//is;
|
---|
194 | $back =~ s/target=\"?_parent\"?//is;
|
---|
195 |
|
---|
196 | return $front . $link . $back if $link =~ /^\#/s;
|
---|
197 | $link =~ s/\n/ /g;
|
---|
198 |
|
---|
199 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
---|
200 | # href may use '\'s where '/'s should be on Windows
|
---|
201 | $href =~ s/\\/\//g;
|
---|
202 |
|
---|
203 | my ($filename) = $href =~ /^(?:.*?):(?:\/\/)?(.*)/;
|
---|
204 |
|
---|
205 | ##### leave all these links alone (they won't be picked up by intermediate
|
---|
206 | ##### pages). I think that's safest when dealing with frames, targets etc.
|
---|
207 | ##### (at least until I think of a better way to do it). Problems occur with
|
---|
208 | ##### mailto links from within small frames, the intermediate page is displayed
|
---|
209 | ##### within that frame and can't be seen. There is still potential for this to
|
---|
210 | ##### happen even with html pages - the solution seems to be to somehow tell
|
---|
211 | ##### the browser from the server side to display the page being sent (i.e.
|
---|
212 | ##### the intermediate page) in the top level window - I'm not sure if that's
|
---|
213 | ##### possible - the following line should probably be deleted if that can be done
|
---|
214 | return $front . $link . $back if $href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/is;
|
---|
215 |
|
---|
216 |
|
---|
217 | if (($rl == 0) || ($filename =~ /$self->{'process_exp'}/) ||
|
---|
218 | ($href =~ /\/$/) || ($href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/i)) {
|
---|
219 | &ghtml::urlsafe ($href);
|
---|
220 | return $front . "_httpextlink_&rl=" . $rl . "&href=" . $href . $hash_part . $back;
|
---|
221 |
|
---|
222 | } else {
|
---|
223 | # link is to some other type of file (eg image) so we'll
|
---|
224 | # need to associate that file
|
---|
225 | return $front . $self->add_file ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) . $back;
|
---|
226 | }
|
---|
227 | }
|
---|
228 |
|
---|
229 | sub add_file {
|
---|
230 | my $self = shift (@_);
|
---|
231 | my ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) = @_;
|
---|
232 | my ($newname);
|
---|
233 |
|
---|
234 | my $filename = $href;
|
---|
235 | $filename =~ s/^[^:]*:\/\///;
|
---|
236 | $filename = &util::filename_cat($base_dir, $filename);
|
---|
237 |
|
---|
238 | my ($ext) = $filename =~ /(\.[^\.]*)$/;
|
---|
239 |
|
---|
240 | if ((!defined $ext) || ($ext !~ /$self->{'assoc_files'}/)) {
|
---|
241 | return "_httpextlink_&rl=" . $rl . "&href=" . $href . $hash_part;
|
---|
242 | }
|
---|
243 |
|
---|
244 | if ($self->{'rename_assoc_files'}) {
|
---|
245 | if (defined $self->{'aux_files'}->{$href}) {
|
---|
246 | $newname = $self->{'aux_files'}->{$href}->{'dir_num'} . "/" .
|
---|
247 | $self->{'aux_files'}->{$href}->{'file_num'} . $ext;
|
---|
248 | } else {
|
---|
249 | $newname = $self->{'dir_num'} . "/" . $self->{'file_num'} . $ext;
|
---|
250 | $self->{'aux_files'}->{$href} = {'dir_num' => $self->{'dir_num'}, 'file_num' => $self->{'file_num'}};
|
---|
251 | $self->inc_filecount ();
|
---|
252 | }
|
---|
253 | $doc_obj->associate_file($filename, $newname, undef, $section);
|
---|
254 | return "_httpcollimg_/$newname";
|
---|
255 |
|
---|
256 | } else {
|
---|
257 | ($newname) = $filename =~ /([^\/\\]*)$/;
|
---|
258 | $doc_obj->associate_file($filename, $newname, undef, $section);
|
---|
259 | return "_httpdocimg_/$newname";
|
---|
260 | }
|
---|
261 | }
|
---|
262 |
|
---|
263 |
|
---|
264 | sub format_link {
|
---|
265 | my $self = shift (@_);
|
---|
266 | my ($link, $base_dir, $file) = @_;
|
---|
267 |
|
---|
268 | my ($before_hash, $hash_part) = $link =~ /^([^\#]*)(\#?.*)$/;
|
---|
269 |
|
---|
270 | $hash_part = "" if !defined $hash_part;
|
---|
271 | if (!defined $before_hash || $before_hash !~ /[\w\.\/]/) {
|
---|
272 | my $outhandle = $self->{'outhandle'};
|
---|
273 | print $outhandle "HTMLPlug: ERROR - badly formatted tag ignored ($link)\n"
|
---|
274 | if $self->{'verbosity'};
|
---|
275 | return ($link, "", 0);
|
---|
276 | }
|
---|
277 |
|
---|
278 | if ($before_hash =~ s/^((?:http|ftp|file):\/\/)//i) {
|
---|
279 | my $type = $1;
|
---|
280 |
|
---|
281 | if ($link =~ /^(http|ftp):/i) {
|
---|
282 | # Turn url (using /) into file name (possibly using \ on windows)
|
---|
283 | my @http_dir_split = split('/', $before_hash);
|
---|
284 | $before_hash = &util::filename_cat(@http_dir_split);
|
---|
285 | }
|
---|
286 |
|
---|
287 | $before_hash = $self->eval_dir_dots($before_hash);
|
---|
288 |
|
---|
289 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
---|
290 |
|
---|
291 | my $rl = 0;
|
---|
292 | $rl = 1 if (-e $linkfilename);
|
---|
293 |
|
---|
294 | # make sure there's a slash on the end if it's a directory
|
---|
295 | if ($before_hash !~ /\/$/) {
|
---|
296 | $before_hash .= "/" if (-d $linkfilename);
|
---|
297 | }
|
---|
298 |
|
---|
299 | return ($type . $before_hash, $hash_part, $rl);
|
---|
300 |
|
---|
301 | } elsif ($link !~ /^(mailto|news|gopher|nntp|telnet|javascript):/i) {
|
---|
302 | if ($before_hash =~ s/^\/// || $before_hash =~ /\\/) {
|
---|
303 |
|
---|
304 | # the first directory will be the domain name if w3mir was used
|
---|
305 | # to generate archives, otherwise we'll assume all files are
|
---|
306 | # from the same site and base_dir is the root
|
---|
307 | # This is not called in Windows
|
---|
308 | if ($self->{'w3mir'}) {
|
---|
309 | my @dirs = split /[\/\\]/, $file;
|
---|
310 | my $domname = shift (@dirs);
|
---|
311 | $before_hash = &util::filename_cat($domname, $before_hash);
|
---|
312 | $before_hash =~ s/\\/\//g; # for windows
|
---|
313 | }
|
---|
314 | else
|
---|
315 | {
|
---|
316 | # see if link shares directory with source document
|
---|
317 | # => turn into relative link if this is so!
|
---|
318 | my $before_hash_fix = "";
|
---|
319 |
|
---|
320 | # filename_cat appends / or \ to the beginning of the
|
---|
321 | # file pathname but in Windows we don't want to do that
|
---|
322 | if ($ENV{'GSDLOS'} =~ /^windows/i) {
|
---|
323 | $before_hash_fix = $before_hash;
|
---|
324 | }
|
---|
325 | else {
|
---|
326 | $before_hash_fix = &util::filename_cat("",$before_hash);
|
---|
327 | }
|
---|
328 |
|
---|
329 | my $base_dir_win_match = $base_dir;
|
---|
330 | $base_dir_win_match =~ s/\\/\\\\/g;
|
---|
331 |
|
---|
332 | $before_hash_fix =~ s/^$base_dir_win_match(\\|\/)//;
|
---|
333 | $before_hash = $before_hash_fix;
|
---|
334 |
|
---|
335 | }
|
---|
336 |
|
---|
337 | } else {
|
---|
338 | # Turn relative file path into full path
|
---|
339 | my $dirname = &File::Basename::dirname($file);
|
---|
340 | $before_hash = &util::filename_cat($dirname, $before_hash);
|
---|
341 | $before_hash = $self->eval_dir_dots($before_hash);
|
---|
342 | }
|
---|
343 |
|
---|
344 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
---|
345 |
|
---|
346 | # make sure there's a slash on the end if it's a directory
|
---|
347 | if ($before_hash !~ /\/$/) {
|
---|
348 | $before_hash .= "/" if (-d $linkfilename);
|
---|
349 | }
|
---|
350 |
|
---|
351 | return ("http://" . $before_hash, $hash_part, 1);
|
---|
352 |
|
---|
353 | } else {
|
---|
354 | # mailto, news, nntp, telnet, javascript or gopher link
|
---|
355 | return ($before_hash, "", 0);
|
---|
356 | }
|
---|
357 | }
|
---|
358 |
|
---|
359 | sub extract_first_NNNN_characters {
|
---|
360 | my $self = shift (@_);
|
---|
361 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
362 |
|
---|
363 | foreach my $size (split /,/, $self->{'first'}) {
|
---|
364 | my $tmptext = $$textref;
|
---|
365 | $tmptext =~ s/.*<body[^>]*>//i;
|
---|
366 | $tmptext =~ s/$self->{'title_sub'}// if (defined $self->{'title_sub'});
|
---|
367 | $tmptext =~ s/<[^>]*>/ /g;
|
---|
368 | $tmptext =~ s/ / /g;
|
---|
369 | $tmptext =~ s/^\s+//;
|
---|
370 | $tmptext =~ s/\s+$//;
|
---|
371 | $tmptext =~ s/\s+/ /gs;
|
---|
372 | $tmptext = substr ($tmptext, 0, $size);
|
---|
373 | $tmptext =~ s/\s\S*$/…/;
|
---|
374 | $doc_obj->add_utf8_metadata ($thissection, "First$size", $tmptext);
|
---|
375 | }
|
---|
376 | }
|
---|
377 |
|
---|
378 | sub extract_metadata {
|
---|
379 | my $self = shift (@_);
|
---|
380 | my ($textref, $metadata, $doc_obj, $section) = @_;
|
---|
381 | my $outhandle = $self->{'outhandle'};
|
---|
382 |
|
---|
383 | # if we don't want metadata, we may as well not be here ...
|
---|
384 | return if (!defined $self->{'metadata_fields'});
|
---|
385 |
|
---|
386 | # hunt for an author look in the metadata elements:
|
---|
387 | if (defined $self->{'hunt_creator_metadata'}) {
|
---|
388 | for my $name (split /,/, "AUTHOR,AUTHOR.EMAIL,CREATOR,DC.CREATOR,DC.CREATOR.CORPORATENAME") {
|
---|
389 | if ($$textref =~ /<meta(\s*?)(?:name|http-equiv)\s*=\s*\"?$name\"?([^>]*)/is) {
|
---|
390 | my $content = $1 . $2;
|
---|
391 | if ($content =~ /content\s*=\s*\"?(.*)\"?/is) {
|
---|
392 | if (defined $1) {
|
---|
393 | my $value = $1;
|
---|
394 | $value =~ s/\"$//;
|
---|
395 | $value =~ s/\s+/ /gs;
|
---|
396 | $doc_obj->add_utf8_metadata($section, "Creator", $value);
|
---|
397 | print $outhandle " extracted Creator metadata \"$value\"\n"
|
---|
398 | if ($self->{'verbosity'} > 2);
|
---|
399 | next;
|
---|
400 | }
|
---|
401 | }
|
---|
402 | }
|
---|
403 | }
|
---|
404 | }
|
---|
405 |
|
---|
406 | foreach my $field (split /,/, $self->{'metadata_fields'}) {
|
---|
407 |
|
---|
408 | # don't need to extract field if it was passed in from a previous
|
---|
409 | # (recursive) plugin
|
---|
410 | next if defined $metadata->{$field};
|
---|
411 |
|
---|
412 | # see if there's a <meta> tag for this field
|
---|
413 | if ($$textref =~ /<meta(\s*?)(?:name|http-equiv)\s*=\s*\"?$field\"?([^>]*)/is) {
|
---|
414 | my $content = $1 . $2;
|
---|
415 | if ($content =~ /content\s*=\s*\"?(.*)\"?/is) {
|
---|
416 | if (defined $1) {
|
---|
417 | my $value = $1;
|
---|
418 | $value =~ s/\"$//;
|
---|
419 | $value =~ s/\s+/ /gs;
|
---|
420 | $value =~ s/\".*//gs;
|
---|
421 | $doc_obj->add_utf8_metadata($section, $field, $value);
|
---|
422 | print $outhandle " extracted \"$field\" metadata \"$value\"\n"
|
---|
423 | if ($self->{'verbosity'} > 2);
|
---|
424 | next;
|
---|
425 | }
|
---|
426 | }
|
---|
427 | }
|
---|
428 |
|
---|
429 | # TITLE: extract the document title
|
---|
430 |
|
---|
431 | if ($field =~ /^title$/i) {
|
---|
432 |
|
---|
433 | # see if there's a <title> tag
|
---|
434 | if ($$textref =~ /<title[^>]*>([^<]*)<\/title[^>]*>/is) {
|
---|
435 | if (defined $1) {
|
---|
436 | my $title = $1;
|
---|
437 | if ($title =~ /\w/) {
|
---|
438 | $title =~ s/<[^>]*>/ /g;
|
---|
439 | $title =~ s/ / /g;
|
---|
440 | $title =~ s/\s+/ /gs;
|
---|
441 | $title =~ s/^\s+//;
|
---|
442 | $title =~ s/\s+$//;
|
---|
443 | $doc_obj->add_utf8_metadata ($section, $field, $title);
|
---|
444 | print $outhandle " extracted \"$field\" metadata \"$title\"\n"
|
---|
445 | if ($self->{'verbosity'} > 2);
|
---|
446 | next;
|
---|
447 | }
|
---|
448 | }
|
---|
449 | }
|
---|
450 |
|
---|
451 | # if no title use first 100 characters
|
---|
452 | my $tmptext = $$textref;
|
---|
453 | $tmptext =~ s/$self->{'title_sub'}// if (defined $self->{'title_sub'});
|
---|
454 | $tmptext =~ s/<[^>]*>/ /g;
|
---|
455 | $tmptext =~ s/ / /g;
|
---|
456 | $tmptext =~ s/^\s+//;
|
---|
457 | $tmptext =~ s/\s+$//;
|
---|
458 | $tmptext =~ s/\s+/ /gs;
|
---|
459 | $tmptext = substr ($tmptext, 0, 100);
|
---|
460 | $tmptext =~ s/\s\S*$/.../;
|
---|
461 | $doc_obj->add_utf8_metadata ($section, $field, $tmptext);
|
---|
462 | print $outhandle " extracted \"$field\" metadata \"$tmptext\"\n"
|
---|
463 | if ($self->{'verbosity'} > 2);
|
---|
464 | next;
|
---|
465 | }
|
---|
466 |
|
---|
467 |
|
---|
468 | # tag: extract the text between the first <H1> and </H1> tags
|
---|
469 | if ($field =~ /^tag[a-z0-9]+$/i) {
|
---|
470 |
|
---|
471 | my $tag = $field;
|
---|
472 | $tag =~ s/^tag//i;
|
---|
473 | my $tmptext = $$textref;
|
---|
474 | $tmptext =~ s/\s+/ /gs;
|
---|
475 | if ($tmptext =~ /<$tag[^>]*>/i) {
|
---|
476 | foreach my $word ($tmptext =~ m/<$tag[^>]*>(.*?)<\/$tag[^>]*>/g) {
|
---|
477 | $word =~ s/ / /g;
|
---|
478 | $word =~ s/<[^>]*>/ /g;
|
---|
479 | $word =~ s/^\s+//;
|
---|
480 | $word =~ s/\s+$//;
|
---|
481 | $word =~ s/\s+/ /gs;
|
---|
482 | if ($word ne "") {
|
---|
483 | $doc_obj->add_utf8_metadata ($section, $tag, $word);
|
---|
484 | print $outhandle " extracted \"$tag\" metadata \"$word\"\n"
|
---|
485 | if ($self->{'verbosity'} > 2);
|
---|
486 | }
|
---|
487 | }
|
---|
488 | }
|
---|
489 | next;
|
---|
490 | }
|
---|
491 | }
|
---|
492 | }
|
---|
493 |
|
---|
494 |
|
---|
495 | # evaluate any "../" to next directory up
|
---|
496 | # evaluate any "./" as here
|
---|
497 | sub eval_dir_dots {
|
---|
498 | my $self = shift (@_);
|
---|
499 | my ($filename) = @_;
|
---|
500 |
|
---|
501 | my $dirsep_os = &util::get_os_dirsep();
|
---|
502 | my @dirsep = split(/$dirsep_os/,$filename);
|
---|
503 |
|
---|
504 | my @eval_dirs = ();
|
---|
505 | foreach my $d (@dirsep) {
|
---|
506 | if ($d eq "..") {
|
---|
507 | pop(@eval_dirs);
|
---|
508 |
|
---|
509 | } elsif ($d eq ".") {
|
---|
510 | # do nothing!
|
---|
511 |
|
---|
512 | } else {
|
---|
513 | push(@eval_dirs,$d);
|
---|
514 | }
|
---|
515 | }
|
---|
516 |
|
---|
517 | return &util::filename_cat(@eval_dirs);
|
---|
518 | }
|
---|
519 |
|
---|
520 | sub replace_usemap_links {
|
---|
521 | my $self = shift (@_);
|
---|
522 | my ($front, $link, $back) = @_;
|
---|
523 |
|
---|
524 | $link =~ s/^\.\///;
|
---|
525 | return $front . $link . $back;
|
---|
526 | }
|
---|
527 |
|
---|
528 | sub inc_filecount {
|
---|
529 | my $self = shift (@_);
|
---|
530 |
|
---|
531 | if ($self->{'file_num'} == 1000) {
|
---|
532 | $self->{'dir_num'} ++;
|
---|
533 | $self->{'file_num'} = 0;
|
---|
534 | } else {
|
---|
535 | $self->{'file_num'} ++;
|
---|
536 | }
|
---|
537 | }
|
---|
538 |
|
---|
539 |
|
---|
540 | # Extend the BasPlug read_file so that strings like é are
|
---|
541 | # converted to UTF8 internally.
|
---|
542 | #
|
---|
543 | # We don't convert < or > or & or " in case
|
---|
544 | # they interfere with the GML files
|
---|
545 |
|
---|
546 | sub read_file {
|
---|
547 | my ($self, $filename, $encoding, $textref) = @_;
|
---|
548 |
|
---|
549 | &BasPlug::read_file($self, $filename, $encoding, $textref);
|
---|
550 |
|
---|
551 | # Convert things like é to their UTF8 equivalents
|
---|
552 | $$textref =~ s/&(lt|gt|amp|quot);/&z$1;/go;
|
---|
553 | $$textref =~ s/&([^;]+);/&unicode::ascii2utf8(\&ghtml::getcharequiv($1,1))/gseo;
|
---|
554 | $$textref =~ s/&z(lt|gt|amp|quot);/&$1;/go;
|
---|
555 | }
|
---|
556 |
|
---|
557 | 1;
|
---|