| 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 " -file_is_url Set if input filenames make up url of original source documents\n";
|
|---|
| 67 | print STDERR " e.g. if a web mirroring tool was used to create the import\n";
|
|---|
| 68 | print STDERR " directory structure\n";
|
|---|
| 69 | print STDERR " -assoc_files Perl regular expression of file extensions to associate with\n";
|
|---|
| 70 | print STDERR " html documents. Defaults to '(?i)\.(jpe?g|gif|png|css)\$'\n";
|
|---|
| 71 | print STDERR " -rename_assoc_files Renames files associated with documents (e.g. images). Also\n";
|
|---|
| 72 | print STDERR " creates much shallower directory structure (useful when creating\n";
|
|---|
| 73 | print STDERR " collections to go on cd-rom).\n\n";
|
|---|
| 74 | print STDERR " -title_sub Substitution expression to modify string stored as Title.\n";
|
|---|
| 75 | print STDERR " Used by, for example, PDFHtml to remove Page 1 etc from text\n";
|
|---|
| 76 | print STDERR " chosen to be used as the title.\n";
|
|---|
| 77 | print STDERR " -description_tags Split document into sub-sections where <Section> tags occur.\n";
|
|---|
| 78 | print STDERR " Note that by setting this option you implicitly set -no_metadata\n";
|
|---|
| 79 | print STDERR " as all metadata should be included within the <Section> tags.\n";
|
|---|
| 80 | print STDERR " Also, -keep_head will have no effect when this option is set.\n";
|
|---|
| 81 | }
|
|---|
| 82 |
|
|---|
| 83 | sub new {
|
|---|
| 84 | my $class = shift (@_);
|
|---|
| 85 | my $self = new BasPlug ($class, @_);
|
|---|
| 86 |
|
|---|
| 87 | if (!parsargv::parse(\@_,
|
|---|
| 88 | q^nolinks^, \$self->{'nolinks'},
|
|---|
| 89 | q^keep_head^, \$self->{'keep_head'},
|
|---|
| 90 | q^no_metadata^, \$self->{'no_metadata'},
|
|---|
| 91 | q^metadata_fields/.*/Title^, \$self->{'metadata_fields'},
|
|---|
| 92 | q^hunt_creator_metadata^, \$self->{'hunt_creator_metadata'},
|
|---|
| 93 | q^w3mir^, \$self->{'w3mir'},
|
|---|
| 94 | q^file_is_url^, \$self->{'file_is_url'},
|
|---|
| 95 | q^assoc_files/.*/(?i)\.(jpe?g|gif|png|css)$^, \$self->{'assoc_files'},
|
|---|
| 96 | q^rename_assoc_files^, \$self->{'rename_assoc_files'},
|
|---|
| 97 | q^title_sub/.*/^, \$self->{'title_sub'},
|
|---|
| 98 | q^description_tags^, \$self->{'description_tags'},
|
|---|
| 99 | "allow_extra_options")) {
|
|---|
| 100 |
|
|---|
| 101 | print STDERR "\nIncorrect options passed to HTMLPlug, check your collect.cfg configuration file\n";
|
|---|
| 102 | &print_usage();
|
|---|
| 103 | die "\n";
|
|---|
| 104 | }
|
|---|
| 105 |
|
|---|
| 106 | # retain this for backward compatibility (w3mir option was replaced by
|
|---|
| 107 | # file_is_url)
|
|---|
| 108 | if ($self->{'w3mir'}) {
|
|---|
| 109 | $self->{'file_is_url'} = 1;
|
|---|
| 110 | }
|
|---|
| 111 |
|
|---|
| 112 | $self->{'aux_files'} = {};
|
|---|
| 113 | $self->{'dir_num'} = 0;
|
|---|
| 114 | $self->{'file_num'} = 0;
|
|---|
| 115 |
|
|---|
| 116 | return bless $self, $class;
|
|---|
| 117 | }
|
|---|
| 118 |
|
|---|
| 119 |
|
|---|
| 120 | sub get_default_block_exp {
|
|---|
| 121 | my $self = shift (@_);
|
|---|
| 122 |
|
|---|
| 123 | return q^(?i)\.(gif|jpe?g|png|css)$^;
|
|---|
| 124 | }
|
|---|
| 125 |
|
|---|
| 126 | sub get_default_process_exp {
|
|---|
| 127 | my $self = shift (@_);
|
|---|
| 128 |
|
|---|
| 129 | # the last option is an attempt to encode the concept of an html query ...
|
|---|
| 130 | return q^(?i)(\.html?|\.shtml|\.shm|\.asp|\.php|\.cgi|.+\?.+=.*)$^;
|
|---|
| 131 | }
|
|---|
| 132 |
|
|---|
| 133 |
|
|---|
| 134 | # do plugin specific processing of doc_obj
|
|---|
| 135 | sub process {
|
|---|
| 136 | my $self = shift (@_);
|
|---|
| 137 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
|
|---|
| 138 | my $outhandle = $self->{'outhandle'};
|
|---|
| 139 |
|
|---|
| 140 | print $outhandle "HTMLPlug: processing $file\n"
|
|---|
| 141 | if $self->{'verbosity'} > 1;
|
|---|
| 142 |
|
|---|
| 143 | my $cursection = $doc_obj->get_top_section();
|
|---|
| 144 |
|
|---|
| 145 | $self->extract_metadata ($textref, $metadata, $doc_obj, $cursection)
|
|---|
| 146 | unless $self->{'no_metadata'} || $self->{'description_tags'};
|
|---|
| 147 |
|
|---|
| 148 | # Store URL for page as metadata - this can be used for an
|
|---|
| 149 | # altavista style search interface. The URL won't be valid
|
|---|
| 150 | # unless the file structure contains the domain name (i.e.
|
|---|
| 151 | # like when w3mir is used to download a website).
|
|---|
| 152 | my $web_url = "http://$file";
|
|---|
| 153 | $web_url =~ s/\\/\//g; # for windows
|
|---|
| 154 | $doc_obj->add_utf8_metadata($cursection, "URL", $web_url);
|
|---|
| 155 |
|
|---|
| 156 | if ($self->{'description_tags'}) {
|
|---|
| 157 |
|
|---|
| 158 | my $opencom = '(?:<!--|<!(?:—|—|--))';
|
|---|
| 159 | my $closecom = '(?:-->|(?:—|—|--)>)';
|
|---|
| 160 | my $lt = '(?:<|<)';
|
|---|
| 161 | my $gt = '(?:>|>)';
|
|---|
| 162 | my $quot = '(?:"|"|”|“)';
|
|---|
| 163 |
|
|---|
| 164 | my $found_something = 0; my $top = 1;
|
|---|
| 165 | while ($$textref =~ s/^(.*?)$opencom(.*?)$closecom//s) {
|
|---|
| 166 | my $text = $1;
|
|---|
| 167 | my $comment = $2;
|
|---|
| 168 | if (defined $text) {
|
|---|
| 169 | $self->process_section(\$text, $base_dir, $file, $doc_obj, $cursection);
|
|---|
| 170 | }
|
|---|
| 171 | while ($comment =~ s/$lt(.*?)$gt//s) {
|
|---|
| 172 |
|
|---|
| 173 | my $tag = $1;
|
|---|
| 174 | if ($tag eq "Section") {
|
|---|
| 175 | $found_something = 1;
|
|---|
| 176 | $cursection = $doc_obj->insert_section($doc_obj->get_end_child($cursection)) unless $top;
|
|---|
| 177 | $top = 0;
|
|---|
| 178 | } elsif ($tag eq "/Section") {
|
|---|
| 179 | $found_something = 1;
|
|---|
| 180 | $cursection = $doc_obj->get_parent_section ($cursection);
|
|---|
| 181 | } elsif ($tag =~ /^Metadata name=$quot(.*?)$quot/s) {
|
|---|
| 182 | my $metaname = $1;
|
|---|
| 183 | $comment =~ s/^(.*?)$lt\/Metadata$gt//s;
|
|---|
| 184 | my $metavalue = $1;
|
|---|
| 185 | $metavalue =~ s/^\s+//;
|
|---|
| 186 | $metavalue =~ s/\s+$//;
|
|---|
| 187 | # assume that no metadata value intentionally includes
|
|---|
| 188 | # carriage returns or HTML tags (if they're there they
|
|---|
| 189 | # were probably introduced when converting to HTML from
|
|---|
| 190 | # some other format).
|
|---|
| 191 | $metavalue =~ s/[\cJ\cM]/ /sg;
|
|---|
| 192 | $metavalue =~ s/<[^>]+>//sg;
|
|---|
| 193 | $metavalue =~ s/\s+/ /sg;
|
|---|
| 194 | $doc_obj->set_utf8_metadata_element($cursection, $metaname, $metavalue);
|
|---|
| 195 | } elsif ($tag eq "Description" || $tag eq "/Description") {
|
|---|
| 196 | # do nothing with containing Description tags
|
|---|
| 197 | } else {
|
|---|
| 198 | # simple HTML tag (probably created by the conversion
|
|---|
| 199 | # to HTML from some other format) - we'll ignore it and
|
|---|
| 200 | # hope for the best ;-)
|
|---|
| 201 | }
|
|---|
| 202 | }
|
|---|
| 203 | }
|
|---|
| 204 | if ($cursection ne "") {
|
|---|
| 205 | print $outhandle "HTMLPlug: WARNING: $file contains unmatched <Section></Section> tags\n";
|
|---|
| 206 | }
|
|---|
| 207 |
|
|---|
| 208 | $$textref =~ s/^.*?<body[^>]*>//is;
|
|---|
| 209 | $$textref =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
|
|---|
| 210 | if ($$textref =~ /\S/) {
|
|---|
| 211 | if (!$found_something) {
|
|---|
| 212 | print $outhandle "HTMLPlug: WARNING: $file appears to contain no Section tags so\n";
|
|---|
| 213 | print $outhandle " will be processed as a single section document\n";
|
|---|
| 214 | $self->process_section($$textref, $base_dir, $file, $doc_obj, $cursection);
|
|---|
| 215 | } else {
|
|---|
| 216 | print $outhandle "HTMLPlug: WARNING: $file contains the following text outside\n";
|
|---|
| 217 | print $outhandle " of the final closing </Section> tag. This text will\n";
|
|---|
| 218 | print $outhandle " be ignored.";
|
|---|
| 219 | my ($text);
|
|---|
| 220 | if (length($$textref) > 30) {
|
|---|
| 221 | $text = substr($$textref, 0, 30) . "...";
|
|---|
| 222 | } else {
|
|---|
| 223 | $text = $$textref;
|
|---|
| 224 | }
|
|---|
| 225 | $text =~ s/\n/ /isg;
|
|---|
| 226 | print $outhandle " ($text)\n";
|
|---|
| 227 | }
|
|---|
| 228 | }
|
|---|
| 229 |
|
|---|
| 230 | } else {
|
|---|
| 231 | # single section document
|
|---|
| 232 | $self->process_section($textref, $base_dir, $file, $doc_obj, $cursection);
|
|---|
| 233 | }
|
|---|
| 234 | return 1;
|
|---|
| 235 | }
|
|---|
| 236 |
|
|---|
| 237 | # note that process_section may be called multiple times for a single
|
|---|
| 238 | # section (relying on the fact that add_utf8_text appends the text to any
|
|---|
| 239 | # that may exist already).
|
|---|
| 240 | sub process_section {
|
|---|
| 241 | my $self = shift (@_);
|
|---|
| 242 | my ($textref, $base_dir, $file, $doc_obj, $cursection) = @_;
|
|---|
| 243 |
|
|---|
| 244 | # remove header and footer
|
|---|
| 245 | if (!$self->{'keep_head'} || $self->{'description_tags'}) {
|
|---|
| 246 | $$textref =~ s/^.*?<body[^>]*>//is;
|
|---|
| 247 | $$textref =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
|
|---|
| 248 | }
|
|---|
| 249 |
|
|---|
| 250 | # trap links
|
|---|
| 251 | if (!$self->{'nolinks'}) {
|
|---|
| 252 |
|
|---|
| 253 | # usemap="./#index" not handled correctly => change to "#index"
|
|---|
| 254 | $$textref =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
|---|
| 255 | $self->replace_usemap_links($1, $2, $3)/isge;
|
|---|
| 256 |
|
|---|
| 257 | $$textref =~ s/(<(?:a|area|frame|link)\s+[^>]*?\s*(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
|---|
| 258 | $self->replace_href_links ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
|
|---|
| 259 | }
|
|---|
| 260 |
|
|---|
| 261 | # trap images
|
|---|
| 262 |
|
|---|
| 263 | # allow spaces if inside quotes - jrm21
|
|---|
| 264 | $$textref =~ s/(<img[^>]*?src\s*=\s*)(\"[^\"]+\"|[^\s>]+)([^>]*>)/
|
|---|
| 265 | $self->replace_images ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
|
|---|
| 266 |
|
|---|
| 267 | # add text to document object
|
|---|
| 268 | $doc_obj->add_utf8_text($cursection, $$textref);
|
|---|
| 269 | }
|
|---|
| 270 |
|
|---|
| 271 | sub replace_images {
|
|---|
| 272 | my $self = shift (@_);
|
|---|
| 273 | my ($front, $link, $back, $base_dir,
|
|---|
| 274 | $file, $doc_obj, $section) = @_;
|
|---|
| 275 | # remove quotes from link at start and end if necessary
|
|---|
| 276 | if ($link=~/^\"/) {
|
|---|
| 277 | $link=~s/^\"//;$link=~s/\"$//;
|
|---|
| 278 | $front.='"';
|
|---|
| 279 | $back="\"$back";
|
|---|
| 280 | }
|
|---|
| 281 |
|
|---|
| 282 | $link =~ s/\n/ /g;
|
|---|
| 283 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
|---|
| 284 | return $front . $self->add_file ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) . $back;
|
|---|
| 285 | }
|
|---|
| 286 |
|
|---|
| 287 | sub replace_href_links {
|
|---|
| 288 | my $self = shift (@_);
|
|---|
| 289 | my ($front, $link, $back, $base_dir, $file, $doc_obj, $section) = @_;
|
|---|
| 290 |
|
|---|
| 291 | # attempt to sort out targets - frames are not handled
|
|---|
| 292 | # well in this plugin and some cases will screw things
|
|---|
| 293 | # up - e.g. the _parent target (so we'll just remove
|
|---|
| 294 | # them all ;-)
|
|---|
| 295 | $front =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
|---|
| 296 | $back =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
|---|
| 297 | $front =~ s/target=\"?_parent\"?//is;
|
|---|
| 298 | $back =~ s/target=\"?_parent\"?//is;
|
|---|
| 299 |
|
|---|
| 300 | return $front . $link . $back if $link =~ /^\#/s;
|
|---|
| 301 | $link =~ s/\n/ /g;
|
|---|
| 302 |
|
|---|
| 303 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
|---|
| 304 | # href may use '\'s where '/'s should be on Windows
|
|---|
| 305 | $href =~ s/\\/\//g;
|
|---|
| 306 |
|
|---|
| 307 | my ($filename) = $href =~ /^(?:.*?):(?:\/\/)?(.*)/;
|
|---|
| 308 |
|
|---|
| 309 | ##### leave all these links alone (they won't be picked up by intermediate
|
|---|
| 310 | ##### pages). I think that's safest when dealing with frames, targets etc.
|
|---|
| 311 | ##### (at least until I think of a better way to do it). Problems occur with
|
|---|
| 312 | ##### mailto links from within small frames, the intermediate page is displayed
|
|---|
| 313 | ##### within that frame and can't be seen. There is still potential for this to
|
|---|
| 314 | ##### happen even with html pages - the solution seems to be to somehow tell
|
|---|
| 315 | ##### the browser from the server side to display the page being sent (i.e.
|
|---|
| 316 | ##### the intermediate page) in the top level window - I'm not sure if that's
|
|---|
| 317 | ##### possible - the following line should probably be deleted if that can be done
|
|---|
| 318 | return $front . $link . $back if $href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/is;
|
|---|
| 319 |
|
|---|
| 320 |
|
|---|
| 321 | if (($rl == 0) || ($filename =~ /$self->{'process_exp'}/) ||
|
|---|
| 322 | ($href =~ /\/$/) || ($href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/i)) {
|
|---|
| 323 | &ghtml::urlsafe ($href);
|
|---|
| 324 | return $front . "_httpextlink_&rl=" . $rl . "&href=" . $href . $hash_part . $back;
|
|---|
| 325 |
|
|---|
| 326 | } else {
|
|---|
| 327 | # link is to some other type of file (eg image) so we'll
|
|---|
| 328 | # need to associate that file
|
|---|
| 329 | return $front . $self->add_file ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) . $back;
|
|---|
| 330 | }
|
|---|
| 331 | }
|
|---|
| 332 |
|
|---|
| 333 | sub add_file {
|
|---|
| 334 | my $self = shift (@_);
|
|---|
| 335 | my ($href, $rl, $hash_part, $base_dir, $doc_obj, $section) = @_;
|
|---|
| 336 | my ($newname);
|
|---|
| 337 |
|
|---|
| 338 | my $filename = $href;
|
|---|
| 339 | $filename =~ s/^[^:]*:\/\///;
|
|---|
| 340 | $filename = &util::filename_cat($base_dir, $filename);
|
|---|
| 341 |
|
|---|
| 342 | my ($ext) = $filename =~ /(\.[^\.]*)$/;
|
|---|
| 343 |
|
|---|
| 344 | if ((!defined $ext) || ($ext !~ /$self->{'assoc_files'}/)) {
|
|---|
| 345 | return "_httpextlink_&rl=" . $rl . "&href=" . $href . $hash_part;
|
|---|
| 346 | }
|
|---|
| 347 |
|
|---|
| 348 | if ($self->{'rename_assoc_files'}) {
|
|---|
| 349 | if (defined $self->{'aux_files'}->{$href}) {
|
|---|
| 350 | $newname = $self->{'aux_files'}->{$href}->{'dir_num'} . "/" .
|
|---|
| 351 | $self->{'aux_files'}->{$href}->{'file_num'} . $ext;
|
|---|
| 352 | } else {
|
|---|
| 353 | $newname = $self->{'dir_num'} . "/" . $self->{'file_num'} . $ext;
|
|---|
| 354 | $self->{'aux_files'}->{$href} = {'dir_num' => $self->{'dir_num'}, 'file_num' => $self->{'file_num'}};
|
|---|
| 355 | $self->inc_filecount ();
|
|---|
| 356 | }
|
|---|
| 357 | $doc_obj->associate_file($filename, $newname, undef, $section);
|
|---|
| 358 | return "_httpcollimg_/$newname";
|
|---|
| 359 |
|
|---|
| 360 | } else {
|
|---|
| 361 | ($newname) = $filename =~ /([^\/\\]*)$/;
|
|---|
| 362 | $doc_obj->associate_file($filename, $newname, undef, $section);
|
|---|
| 363 | return "_httpdocimg_/$newname";
|
|---|
| 364 | }
|
|---|
| 365 | }
|
|---|
| 366 |
|
|---|
| 367 |
|
|---|
| 368 | sub format_link {
|
|---|
| 369 | my $self = shift (@_);
|
|---|
| 370 | my ($link, $base_dir, $file) = @_;
|
|---|
| 371 |
|
|---|
| 372 | my ($before_hash, $hash_part) = $link =~ /^([^\#]*)(\#?.*)$/;
|
|---|
| 373 |
|
|---|
| 374 | $hash_part = "" if !defined $hash_part;
|
|---|
| 375 | if (!defined $before_hash || $before_hash !~ /[\w\.\/]/) {
|
|---|
| 376 | my $outhandle = $self->{'outhandle'};
|
|---|
| 377 | print $outhandle "HTMLPlug: ERROR - badly formatted tag ignored ($link)\n"
|
|---|
| 378 | if $self->{'verbosity'};
|
|---|
| 379 | return ($link, "", 0);
|
|---|
| 380 | }
|
|---|
| 381 |
|
|---|
| 382 | if ($before_hash =~ s/^((?:http|ftp|file):\/\/)//i) {
|
|---|
| 383 | my $type = $1;
|
|---|
| 384 |
|
|---|
| 385 | if ($link =~ /^(http|ftp):/i) {
|
|---|
| 386 | # Turn url (using /) into file name (possibly using \ on windows)
|
|---|
| 387 | my @http_dir_split = split('/', $before_hash);
|
|---|
| 388 | $before_hash = &util::filename_cat(@http_dir_split);
|
|---|
| 389 | }
|
|---|
| 390 |
|
|---|
| 391 | $before_hash = $self->eval_dir_dots($before_hash);
|
|---|
| 392 |
|
|---|
| 393 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
|---|
| 394 |
|
|---|
| 395 | my $rl = 0;
|
|---|
| 396 | $rl = 1 if (-e $linkfilename);
|
|---|
| 397 |
|
|---|
| 398 | # make sure there's a slash on the end if it's a directory
|
|---|
| 399 | if ($before_hash !~ /\/$/) {
|
|---|
| 400 | $before_hash .= "/" if (-d $linkfilename);
|
|---|
| 401 | }
|
|---|
| 402 |
|
|---|
| 403 | return ($type . $before_hash, $hash_part, $rl);
|
|---|
| 404 |
|
|---|
| 405 | } elsif ($link !~ /^(mailto|news|gopher|nntp|telnet|javascript):/i) {
|
|---|
| 406 | if ($before_hash =~ s/^\/// || $before_hash =~ /\\/) {
|
|---|
| 407 |
|
|---|
| 408 | # the first directory will be the domain name if file_is_url
|
|---|
| 409 | # to generate archives, otherwise we'll assume all files are
|
|---|
| 410 | # from the same site and base_dir is the root
|
|---|
| 411 | # This is not called in Windows
|
|---|
| 412 | if ($self->{'file_is_url'}) {
|
|---|
| 413 | my @dirs = split /[\/\\]/, $file;
|
|---|
| 414 | my $domname = shift (@dirs);
|
|---|
| 415 | $before_hash = &util::filename_cat($domname, $before_hash);
|
|---|
| 416 | $before_hash =~ s/\\/\//g; # for windows
|
|---|
| 417 | }
|
|---|
| 418 | else
|
|---|
| 419 | {
|
|---|
| 420 | # see if link shares directory with source document
|
|---|
| 421 | # => turn into relative link if this is so!
|
|---|
| 422 | my $before_hash_fix = "";
|
|---|
| 423 |
|
|---|
| 424 | # filename_cat appends / or \ to the beginning of the
|
|---|
| 425 | # file pathname but in Windows we don't want to do that
|
|---|
| 426 | if ($ENV{'GSDLOS'} =~ /^windows/i) {
|
|---|
| 427 | $before_hash_fix = $before_hash;
|
|---|
| 428 | }
|
|---|
| 429 | else {
|
|---|
| 430 | $before_hash_fix = &util::filename_cat("",$before_hash);
|
|---|
| 431 | }
|
|---|
| 432 |
|
|---|
| 433 | my $base_dir_win_match = $base_dir;
|
|---|
| 434 | $base_dir_win_match =~ s/\\/\\\\/g;
|
|---|
| 435 |
|
|---|
| 436 | $before_hash_fix =~ s/^$base_dir_win_match(\\|\/)//;
|
|---|
| 437 | $before_hash = $before_hash_fix;
|
|---|
| 438 |
|
|---|
| 439 | }
|
|---|
| 440 |
|
|---|
| 441 | } else {
|
|---|
| 442 | # Turn relative file path into full path
|
|---|
| 443 | my $dirname = &File::Basename::dirname($file);
|
|---|
| 444 | $before_hash = &util::filename_cat($dirname, $before_hash);
|
|---|
| 445 | $before_hash = $self->eval_dir_dots($before_hash);
|
|---|
| 446 | }
|
|---|
| 447 |
|
|---|
| 448 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
|---|
| 449 |
|
|---|
| 450 | # make sure there's a slash on the end if it's a directory
|
|---|
| 451 | if ($before_hash !~ /\/$/) {
|
|---|
| 452 | $before_hash .= "/" if (-d $linkfilename);
|
|---|
| 453 | }
|
|---|
| 454 |
|
|---|
| 455 | return ("http://" . $before_hash, $hash_part, 1);
|
|---|
| 456 |
|
|---|
| 457 | } else {
|
|---|
| 458 | # mailto, news, nntp, telnet, javascript or gopher link
|
|---|
| 459 | return ($before_hash, "", 0);
|
|---|
| 460 | }
|
|---|
| 461 | }
|
|---|
| 462 |
|
|---|
| 463 | sub extract_first_NNNN_characters {
|
|---|
| 464 | my $self = shift (@_);
|
|---|
| 465 | my ($textref, $doc_obj, $thissection) = @_;
|
|---|
| 466 |
|
|---|
| 467 | foreach my $size (split /,/, $self->{'first'}) {
|
|---|
| 468 | my $tmptext = $$textref;
|
|---|
| 469 | $tmptext =~ s/.*<body[^>]*>//i;
|
|---|
| 470 | $tmptext =~ s/$self->{'title_sub'}// if (defined $self->{'title_sub'});
|
|---|
| 471 | $tmptext =~ s/<[^>]*>/ /g;
|
|---|
| 472 | $tmptext =~ s/ / /g;
|
|---|
| 473 | $tmptext =~ s/^\s+//;
|
|---|
| 474 | $tmptext =~ s/\s+$//;
|
|---|
| 475 | $tmptext =~ s/\s+/ /gs;
|
|---|
| 476 | $tmptext = substr ($tmptext, 0, $size);
|
|---|
| 477 | $tmptext =~ s/\s\S*$/…/;
|
|---|
| 478 | $doc_obj->add_utf8_metadata ($thissection, "First$size", $tmptext);
|
|---|
| 479 | }
|
|---|
| 480 | }
|
|---|
| 481 |
|
|---|
| 482 | sub extract_metadata {
|
|---|
| 483 | my $self = shift (@_);
|
|---|
| 484 | my ($textref, $metadata, $doc_obj, $section) = @_;
|
|---|
| 485 | my $outhandle = $self->{'outhandle'};
|
|---|
| 486 |
|
|---|
| 487 | # if we don't want metadata, we may as well not be here ...
|
|---|
| 488 | return if (!defined $self->{'metadata_fields'});
|
|---|
| 489 |
|
|---|
| 490 | # hunt for an author look in the metadata elements:
|
|---|
| 491 | if (defined $self->{'hunt_creator_metadata'}) {
|
|---|
| 492 | for my $name (split /,/, "AUTHOR,AUTHOR.EMAIL,CREATOR,DC.CREATOR,DC.CREATOR.CORPORATENAME") {
|
|---|
| 493 | if ($$textref =~ /<meta(\s*?)(?:name|http-equiv)\s*=\s*\"?$name\"?([^>]*)/is) {
|
|---|
| 494 | my $content = $1 . $2;
|
|---|
| 495 | if ($content =~ /content\s*=\s*\"?(.*)\"?/is) {
|
|---|
| 496 | if (defined $1) {
|
|---|
| 497 | my $value = $1;
|
|---|
| 498 | $value =~ s/\"$//;
|
|---|
| 499 | $value =~ s/\s+/ /gs;
|
|---|
| 500 | $doc_obj->add_utf8_metadata($section, "Creator", $value);
|
|---|
| 501 | print $outhandle " extracted Creator metadata \"$value\"\n"
|
|---|
| 502 | if ($self->{'verbosity'} > 2);
|
|---|
| 503 | next;
|
|---|
| 504 | }
|
|---|
| 505 | }
|
|---|
| 506 | }
|
|---|
| 507 | }
|
|---|
| 508 | }
|
|---|
| 509 |
|
|---|
| 510 | foreach my $field (split /,/, $self->{'metadata_fields'}) {
|
|---|
| 511 |
|
|---|
| 512 | # don't need to extract field if it was passed in from a previous
|
|---|
| 513 | # (recursive) plugin
|
|---|
| 514 | next if defined $metadata->{$field};
|
|---|
| 515 |
|
|---|
| 516 | # see if there's a <meta> tag for this field
|
|---|
| 517 | if ($$textref =~ /<meta(\s*?)(?:name|http-equiv)\s*=\s*\"?$field\"?([^>]*)/is) {
|
|---|
| 518 | my $content = $1 . $2;
|
|---|
| 519 | if ($content =~ /content\s*=\s*\"?(.*)\"?/is) {
|
|---|
| 520 | if (defined $1) {
|
|---|
| 521 | my $value = $1;
|
|---|
| 522 | $value =~ s/\"$//;
|
|---|
| 523 | $value =~ s/\s+/ /gs;
|
|---|
| 524 | $value =~ s/\".*//gs;
|
|---|
| 525 | $doc_obj->add_utf8_metadata($section, $field, $value);
|
|---|
| 526 | print $outhandle " extracted \"$field\" metadata \"$value\"\n"
|
|---|
| 527 | if ($self->{'verbosity'} > 2);
|
|---|
| 528 | next;
|
|---|
| 529 | }
|
|---|
| 530 | }
|
|---|
| 531 | }
|
|---|
| 532 |
|
|---|
| 533 | # TITLE: extract the document title
|
|---|
| 534 |
|
|---|
| 535 | if ($field =~ /^title$/i) {
|
|---|
| 536 |
|
|---|
| 537 | # see if there's a <title> tag
|
|---|
| 538 | if ($$textref =~ /<title[^>]*>([^<]*)<\/title[^>]*>/is) {
|
|---|
| 539 | if (defined $1) {
|
|---|
| 540 | my $title = $1;
|
|---|
| 541 | if ($title =~ /\w/) {
|
|---|
| 542 | $title =~ s/<[^>]*>/ /g;
|
|---|
| 543 | $title =~ s/ / /g;
|
|---|
| 544 | $title =~ s/\s+/ /gs;
|
|---|
| 545 | $title =~ s/^\s+//;
|
|---|
| 546 | $title =~ s/\s+$//;
|
|---|
| 547 | $doc_obj->add_utf8_metadata ($section, $field, $title);
|
|---|
| 548 | print $outhandle " extracted \"$field\" metadata \"$title\"\n"
|
|---|
| 549 | if ($self->{'verbosity'} > 2);
|
|---|
| 550 | next;
|
|---|
| 551 | }
|
|---|
| 552 | }
|
|---|
| 553 | }
|
|---|
| 554 |
|
|---|
| 555 | # if no title use first 100 characters
|
|---|
| 556 | my $tmptext = $$textref;
|
|---|
| 557 | $tmptext =~ s/<\/([^>]+)><\1>//g; # (eg) </b><b> - no space
|
|---|
| 558 | $tmptext =~ s/<[^>]*>/ /g;
|
|---|
| 559 | $tmptext =~ s/ / /g;
|
|---|
| 560 | $tmptext =~ s/^\s+//s;
|
|---|
| 561 | $tmptext =~ s/\s+$//;
|
|---|
| 562 | $tmptext =~ s/\s+/ /gs;
|
|---|
| 563 | $tmptext =~ s/$self->{'title_sub'}// if (defined $self->{'title_sub'});
|
|---|
| 564 | $tmptext = substr ($tmptext, 0, 100);
|
|---|
| 565 | $tmptext =~ s/\s\S*$/.../;
|
|---|
| 566 | $doc_obj->add_utf8_metadata ($section, $field, $tmptext);
|
|---|
| 567 | print $outhandle " extracted \"$field\" metadata \"$tmptext\"\n"
|
|---|
| 568 | if ($self->{'verbosity'} > 2);
|
|---|
| 569 | next;
|
|---|
| 570 | }
|
|---|
| 571 |
|
|---|
| 572 |
|
|---|
| 573 | # tag: extract the text between the first <H1> and </H1> tags
|
|---|
| 574 | if ($field =~ /^tag[a-z0-9]+$/i) {
|
|---|
| 575 |
|
|---|
| 576 | my $tag = $field;
|
|---|
| 577 | $tag =~ s/^tag//i;
|
|---|
| 578 | my $tmptext = $$textref;
|
|---|
| 579 | $tmptext =~ s/\s+/ /gs;
|
|---|
| 580 | if ($tmptext =~ /<$tag[^>]*>/i) {
|
|---|
| 581 | foreach my $word ($tmptext =~ m/<$tag[^>]*>(.*?)<\/$tag[^>]*>/g) {
|
|---|
| 582 | $word =~ s/ / /g;
|
|---|
| 583 | $word =~ s/<[^>]*>/ /g;
|
|---|
| 584 | $word =~ s/^\s+//;
|
|---|
| 585 | $word =~ s/\s+$//;
|
|---|
| 586 | $word =~ s/\s+/ /gs;
|
|---|
| 587 | if ($word ne "") {
|
|---|
| 588 | $doc_obj->add_utf8_metadata ($section, $tag, $word);
|
|---|
| 589 | print $outhandle " extracted \"$tag\" metadata \"$word\"\n"
|
|---|
| 590 | if ($self->{'verbosity'} > 2);
|
|---|
| 591 | }
|
|---|
| 592 | }
|
|---|
| 593 | }
|
|---|
| 594 | next;
|
|---|
| 595 | }
|
|---|
| 596 | }
|
|---|
| 597 | }
|
|---|
| 598 |
|
|---|
| 599 |
|
|---|
| 600 | # evaluate any "../" to next directory up
|
|---|
| 601 | # evaluate any "./" as here
|
|---|
| 602 | sub eval_dir_dots {
|
|---|
| 603 | my $self = shift (@_);
|
|---|
| 604 | my ($filename) = @_;
|
|---|
| 605 |
|
|---|
| 606 | my $dirsep_os = &util::get_os_dirsep();
|
|---|
| 607 | my @dirsep = split(/$dirsep_os/,$filename);
|
|---|
| 608 |
|
|---|
| 609 | my @eval_dirs = ();
|
|---|
| 610 | foreach my $d (@dirsep) {
|
|---|
| 611 | if ($d eq "..") {
|
|---|
| 612 | pop(@eval_dirs);
|
|---|
| 613 |
|
|---|
| 614 | } elsif ($d eq ".") {
|
|---|
| 615 | # do nothing!
|
|---|
| 616 |
|
|---|
| 617 | } else {
|
|---|
| 618 | push(@eval_dirs,$d);
|
|---|
| 619 | }
|
|---|
| 620 | }
|
|---|
| 621 |
|
|---|
| 622 | return &util::filename_cat(@eval_dirs);
|
|---|
| 623 | }
|
|---|
| 624 |
|
|---|
| 625 | sub replace_usemap_links {
|
|---|
| 626 | my $self = shift (@_);
|
|---|
| 627 | my ($front, $link, $back) = @_;
|
|---|
| 628 |
|
|---|
| 629 | $link =~ s/^\.\///;
|
|---|
| 630 | return $front . $link . $back;
|
|---|
| 631 | }
|
|---|
| 632 |
|
|---|
| 633 | sub inc_filecount {
|
|---|
| 634 | my $self = shift (@_);
|
|---|
| 635 |
|
|---|
| 636 | if ($self->{'file_num'} == 1000) {
|
|---|
| 637 | $self->{'dir_num'} ++;
|
|---|
| 638 | $self->{'file_num'} = 0;
|
|---|
| 639 | } else {
|
|---|
| 640 | $self->{'file_num'} ++;
|
|---|
| 641 | }
|
|---|
| 642 | }
|
|---|
| 643 |
|
|---|
| 644 |
|
|---|
| 645 | # Extend the BasPlug read_file so that strings like é are
|
|---|
| 646 | # converted to UTF8 internally.
|
|---|
| 647 | #
|
|---|
| 648 | # We don't convert < or > or & or " in case
|
|---|
| 649 | # they interfere with the GML files
|
|---|
| 650 |
|
|---|
| 651 | sub read_file {
|
|---|
| 652 | my ($self, $filename, $encoding, $language, $textref) = @_;
|
|---|
| 653 |
|
|---|
| 654 | &BasPlug::read_file($self, $filename, $encoding, $language, $textref);
|
|---|
| 655 |
|
|---|
| 656 | # turn \ into \\ so that the rest of greenstone doesn't think there
|
|---|
| 657 | # is an escape code following.
|
|---|
| 658 | $$textref =~ s/\\/\\\\/go;
|
|---|
| 659 |
|
|---|
| 660 | # Convert things like é to their UTF8 equivalents
|
|---|
| 661 | $$textref =~ s/&(lt|gt|amp|quot);/&z$1;/go;
|
|---|
| 662 | $$textref =~ s/&([^;]+);/&unicode::ascii2utf8(\&ghtml::getcharequiv($1,1))/gseo;
|
|---|
| 663 | $$textref =~ s/&z(lt|gt|amp|quot);/&$1;/go;
|
|---|
| 664 | }
|
|---|
| 665 |
|
|---|
| 666 | 1;
|
|---|
