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