| 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 html;
|
|---|
| 42 | use util;
|
|---|
| 43 | use parsargv;
|
|---|
| 44 |
|
|---|
| 45 | sub BEGIN {
|
|---|
| 46 | @ISA = ('BasPlug');
|
|---|
| 47 | }
|
|---|
| 48 |
|
|---|
| 49 | use strict;
|
|---|
| 50 |
|
|---|
| 51 | sub print_usage {
|
|---|
| 52 | print STDERR "\nIncorrect options passed to HTMLPlug, check your collect.cfg configuration file\n";
|
|---|
| 53 |
|
|---|
| 54 | print STDERR "\n usage: plugin HTMLPlug [options]\n\n";
|
|---|
| 55 | print STDERR " options:\n";
|
|---|
| 56 | print STDERR " -process_exp A perl regular expression to match against filenames.\n";
|
|---|
| 57 | print STDERR " Matching filenames will be processed by this plugin.\n";
|
|---|
| 58 | print STDERR " Defaults to '(?i)\.html?\$' i.e. all documents ending in\n";
|
|---|
| 59 | print STDERR " .htm or .html (case-insensitive).\n";
|
|---|
| 60 | print STDERR " -nolinks Don't make any attempt to trap links (setting this flag may\n";
|
|---|
| 61 | print STDERR " improve speed of building/importing but any relative links within\n";
|
|---|
| 62 | print STDERR " documents will be broken).\n";
|
|---|
| 63 | print STDERR " -block_exp Files matching this regular expression will be blocked from\n";
|
|---|
| 64 | print STDERR " being passed to any further plugins in the list. By default\n";
|
|---|
| 65 | print STDERR " HTMLPlug blocks any files with .gif, .jpg, .jpeg, .png, .pdf\n";
|
|---|
| 66 | print STDERR " or .rtf file extensions.\n";
|
|---|
| 67 | print STDERR " -keep_head Don't remove headers from html files.\n";
|
|---|
| 68 | print STDERR " -no_metadata Don't attempt to extract any metadata from files.\n";
|
|---|
| 69 | print STDERR " -metadata_fields Comma separated list of metadata fields to attempt to extract.\n";
|
|---|
| 70 | print STDERR " Defaults to 'Title'\n";
|
|---|
| 71 | print STDERR " -w3mir Set if w3mir was used to generate input file structure.\n\n";
|
|---|
| 72 | }
|
|---|
| 73 |
|
|---|
| 74 | sub new {
|
|---|
| 75 | my $class = shift (@_);
|
|---|
| 76 | my $self = new BasPlug ();
|
|---|
| 77 |
|
|---|
| 78 | if (!parsargv::parse(\@_,
|
|---|
| 79 | q^process_exp/.*/(?i)\.html?$^, \$self->{'process_exp'},
|
|---|
| 80 | q^nolinks^, \$self->{'nolinks'},
|
|---|
| 81 | q^block_exp/.*/(?i)\.(gif|jpe?g|png|pdf|rtf)$^, \$self->{'block_exp'},
|
|---|
| 82 | q^keep_head^, \$self->{'keep_head'},
|
|---|
| 83 | q^no_metadata^, \$self->{'no_metadata'},
|
|---|
| 84 | q^metadata_fields/.*/Title^, \$self->{'metadata_fields'},
|
|---|
| 85 | q^w3mir^, \$self->{'w3mir'})) {
|
|---|
| 86 | &print_usage();
|
|---|
| 87 | die "\n";
|
|---|
| 88 | }
|
|---|
| 89 |
|
|---|
| 90 | $self->{'aux_files'} = {};
|
|---|
| 91 | $self->{'dir_num'} = 0;
|
|---|
| 92 | $self->{'file_num'} = 0;
|
|---|
| 93 |
|
|---|
| 94 | return bless $self, $class;
|
|---|
| 95 | }
|
|---|
| 96 |
|
|---|
| 97 | sub is_recursive {
|
|---|
| 98 | my $self = shift (@_);
|
|---|
| 99 |
|
|---|
| 100 | return 0; # this is not a recursive plugin
|
|---|
| 101 | }
|
|---|
| 102 |
|
|---|
| 103 | # return number of files processed, undef if can't process
|
|---|
| 104 | # Note that $base_dir might be "" and that $file might
|
|---|
| 105 | # include directories
|
|---|
| 106 | sub read {
|
|---|
| 107 | my $self = shift (@_);
|
|---|
| 108 | my ($pluginfo, $base_dir, $file, $metadata, $processor) = @_;
|
|---|
| 109 |
|
|---|
| 110 | my $filename = &util::filename_cat($base_dir, $file);
|
|---|
| 111 | return 0 if $filename =~ /$self->{'block_exp'}/;
|
|---|
| 112 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
|---|
| 113 | return undef;
|
|---|
| 114 | }
|
|---|
| 115 | $file =~ s/^[\/\\]+//;
|
|---|
| 116 |
|
|---|
| 117 | $self->{'verbosity'} = $processor->{'verbosity'};
|
|---|
| 118 | print STDERR "HTMLPlug: processing $file\n"
|
|---|
| 119 | if $self->{'verbosity'} > 1;
|
|---|
| 120 |
|
|---|
| 121 | # create a new document
|
|---|
| 122 | my $doc_obj = new doc ($file, "indexed_doc");
|
|---|
| 123 | my $cursection = $doc_obj->get_top_section();
|
|---|
| 124 |
|
|---|
| 125 | # read in HTML file
|
|---|
| 126 | open (FILE, $filename) || die "HTMLPlug::read - can't open $filename\n";
|
|---|
| 127 | undef $/;
|
|---|
| 128 | my $text = <FILE>;
|
|---|
| 129 | $/ = "\n";
|
|---|
| 130 | close FILE;
|
|---|
| 131 | if ($text !~ /\w/) {
|
|---|
| 132 | print STDERR "HTMLPlug: ERROR: $file contains no text\n" if $self->{'verbosity'};
|
|---|
| 133 | return 0;
|
|---|
| 134 | }
|
|---|
| 135 |
|
|---|
| 136 | $self->extra_metadata ($doc_obj, $cursection, $metadata);
|
|---|
| 137 | $self->extract_metadata (\$text, $metadata, $doc_obj, $cursection)
|
|---|
| 138 | unless $self->{'no_metadata'};
|
|---|
| 139 |
|
|---|
| 140 | # Store URL for page as metadata - this can be used for an
|
|---|
| 141 | # altavista style search interface. The URL won't be valid
|
|---|
| 142 | # unless the file structure contains the domain name (i.e.
|
|---|
| 143 | # like when w3mir is used to download a website).
|
|---|
| 144 | my $web_url = "http://$file";
|
|---|
| 145 | $web_url =~ s/\\/\//g; # for windows
|
|---|
| 146 | $doc_obj->add_metadata($cursection, "URL", $web_url);
|
|---|
| 147 |
|
|---|
| 148 | # remove header and footer
|
|---|
| 149 | if (!$self->{'keep_head'}) {
|
|---|
| 150 | $text =~ s/^.*?<body[^>]*>//is;
|
|---|
| 151 | $text =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
|
|---|
| 152 | }
|
|---|
| 153 |
|
|---|
| 154 | # trap links
|
|---|
| 155 | if (!$self->{'nolinks'}) {
|
|---|
| 156 |
|
|---|
| 157 | # usemap="./#index" not handled correctly => change to "#index"
|
|---|
| 158 | $text =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
|---|
| 159 | $self->replace_usemap_links($1, $2, $3)/isge;
|
|---|
| 160 |
|
|---|
| 161 | $text =~ s/(<(?:a|area|frame)\s+[^>]*?(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
|---|
| 162 | $self->replace_href_links ($1, $2, $3, $base_dir, $file, $doc_obj)/isge;
|
|---|
| 163 | }
|
|---|
| 164 |
|
|---|
| 165 | # trap images
|
|---|
| 166 | $text =~ s/(<img[^>]*?src\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
|
|---|
| 167 | $self->replace_images ($1, $2, $3, $base_dir, $file, $doc_obj)/isge;
|
|---|
| 168 |
|
|---|
| 169 | $doc_obj->add_text ($cursection, $text);
|
|---|
| 170 |
|
|---|
| 171 | # add an OID
|
|---|
| 172 | $doc_obj->set_OID();
|
|---|
| 173 |
|
|---|
| 174 | # process the document
|
|---|
| 175 | $processor->process($doc_obj);
|
|---|
| 176 |
|
|---|
| 177 | return 1; # processed the file
|
|---|
| 178 | }
|
|---|
| 179 |
|
|---|
| 180 | sub replace_images {
|
|---|
| 181 | my $self = shift (@_);
|
|---|
| 182 | my ($front, $link, $back, $base_dir, $file, $doc_obj) = @_;
|
|---|
| 183 |
|
|---|
| 184 | $link =~ s/\n/ /g;
|
|---|
| 185 |
|
|---|
| 186 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
|---|
| 187 | return $front . $self->add_file ($href, $base_dir, $doc_obj) . $back;
|
|---|
| 188 | }
|
|---|
| 189 |
|
|---|
| 190 | sub replace_href_links {
|
|---|
| 191 | my $self = shift (@_);
|
|---|
| 192 | my ($front, $link, $back, $base_dir, $file, $doc_obj) = @_;
|
|---|
| 193 |
|
|---|
| 194 | # attempt to sort out targets - frames are not handled
|
|---|
| 195 | # well in this plugin and some cases will screw things
|
|---|
| 196 | # up - e.g. the _parent target (so we'll just remove
|
|---|
| 197 | # them all ;-)
|
|---|
| 198 | $front =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
|---|
| 199 | $back =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is;
|
|---|
| 200 | $front =~ s/target=\"?_parent\"?//is;
|
|---|
| 201 | $back =~ s/target=\"?_parent\"?//is;
|
|---|
| 202 |
|
|---|
| 203 | return $front . $link . $back if $link =~ /^\#/s;
|
|---|
| 204 | $link =~ s/\n/ /g;
|
|---|
| 205 |
|
|---|
| 206 | my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file);
|
|---|
| 207 |
|
|---|
| 208 | my ($filename) = $href =~ /^(?:.*?):(?:\/\/)?(.*)/;
|
|---|
| 209 | if (($rl == 0) || ($filename =~ /$self->{'process_exp'}/) ||
|
|---|
| 210 | ($href =~ /\/$/) || ($href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/i)) {
|
|---|
| 211 | $link = $href . $hash_part;
|
|---|
| 212 | &html::urlsafe ($link);
|
|---|
| 213 | return $front . "_httpextlink_&href=" . $link . "&rl=" . $rl . $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, $base_dir, $doc_obj) . $back;
|
|---|
| 219 | }
|
|---|
| 220 | }
|
|---|
| 221 |
|
|---|
| 222 | sub add_file {
|
|---|
| 223 | my $self = shift (@_);
|
|---|
| 224 | my ($href, $base_dir, $doc_obj) = @_;
|
|---|
| 225 | my ($newname);
|
|---|
| 226 |
|
|---|
| 227 | my $filename = $href;
|
|---|
| 228 | $filename =~ s/^[^:]*:\/\///;
|
|---|
| 229 | $filename = &util::filename_cat ($base_dir, $filename);
|
|---|
| 230 | my ($ext) = $filename =~ /(\.[^\.]*)$/;
|
|---|
| 231 | if (defined $self->{'aux_files'}->{$href}) {
|
|---|
| 232 | $newname = $self->{'aux_files'}->{$href}->{'dir_num'} . "/" .
|
|---|
| 233 | $self->{'aux_files'}->{$href}->{'file_num'} . $ext;
|
|---|
| 234 | } else {
|
|---|
| 235 | $newname = $self->{'dir_num'} . "/" . $self->{'file_num'} . $ext;
|
|---|
| 236 | $self->inc_filecount ();
|
|---|
| 237 | }
|
|---|
| 238 | $doc_obj->associate_file($filename, $newname);
|
|---|
| 239 | return "_httpcollimg_/$newname";
|
|---|
| 240 | }
|
|---|
| 241 |
|
|---|
| 242 |
|
|---|
| 243 | sub format_link {
|
|---|
| 244 | my $self = shift (@_);
|
|---|
| 245 | my ($link, $base_dir, $file) = @_;
|
|---|
| 246 |
|
|---|
| 247 | my ($before_hash, $hash_part) = $link =~ /^([^\#]*)(\#?.*)$/;
|
|---|
| 248 | $hash_part = "" if !defined $hash_part;
|
|---|
| 249 | if (!defined $before_hash || $before_hash !~ /[\w\.\/]/) {
|
|---|
| 250 | print STDERR "HTMLPlug: ERROR - badly formatted tag ignored ($link)\n"
|
|---|
| 251 | if $self->{'verbosity'};
|
|---|
| 252 | return ($link, "", 0);
|
|---|
| 253 | }
|
|---|
| 254 |
|
|---|
| 255 | if ($before_hash =~ s/^((?:http|ftp|file):\/\/)//i) {
|
|---|
| 256 | my $type = $1;
|
|---|
| 257 |
|
|---|
| 258 | if ($link =~ /^(http|ftp):/i) {
|
|---|
| 259 | # Turn url (using /) into file name (possibly using \ on windows)
|
|---|
| 260 | my @http_dir_split = split('/', $before_hash);
|
|---|
| 261 | $before_hash = &util::filename_cat(@http_dir_split);
|
|---|
| 262 | }
|
|---|
| 263 |
|
|---|
| 264 | $before_hash = $self->eval_dir_dots($before_hash);
|
|---|
| 265 |
|
|---|
| 266 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
|---|
| 267 |
|
|---|
| 268 | my $rl = 0;
|
|---|
| 269 | $rl = 1 if (-e $linkfilename);
|
|---|
| 270 |
|
|---|
| 271 | # make sure there's a slash on the end if it's a directory
|
|---|
| 272 | if ($before_hash !~ /\/$/) {
|
|---|
| 273 | $before_hash .= "/" if (-d $linkfilename);
|
|---|
| 274 | }
|
|---|
| 275 |
|
|---|
| 276 | return ($type . $before_hash, $hash_part, $rl);
|
|---|
| 277 |
|
|---|
| 278 | } elsif ($link !~ /^(mailto|news|gopher|nntp|telnet|javascript):/i) {
|
|---|
| 279 |
|
|---|
| 280 | if ($before_hash =~ s/^\///) {
|
|---|
| 281 | # the first directory will be the domain name if w3mir was used
|
|---|
| 282 | # to generate archives, otherwise we'll assume all files are
|
|---|
| 283 | # from the same site and base_dir is the root
|
|---|
| 284 | if ($self->{'w3mir'}) {
|
|---|
| 285 | my @dirs = split /[\/\\]/, $file;
|
|---|
| 286 | my $domname = shift (@dirs);
|
|---|
| 287 | $before_hash = &util::filename_cat($domname, $before_hash);
|
|---|
| 288 | $before_hash =~ s/\\/\//g; # for windows
|
|---|
| 289 | }
|
|---|
| 290 |
|
|---|
| 291 | } else {
|
|---|
| 292 | # Turn relative file path into full path
|
|---|
| 293 | my $dirname = &File::Basename::dirname($file);
|
|---|
| 294 | $before_hash = &util::filename_cat($dirname, $before_hash);
|
|---|
| 295 | $before_hash = $self->eval_dir_dots($before_hash);
|
|---|
| 296 | }
|
|---|
| 297 |
|
|---|
| 298 | # make sure there's a slash on the end if it's a directory
|
|---|
| 299 | my $linkfilename = &util::filename_cat ($base_dir, $before_hash);
|
|---|
| 300 | if ($before_hash !~ /\/$/) {
|
|---|
| 301 | $before_hash .= "/" if (-d $linkfilename);
|
|---|
| 302 | }
|
|---|
| 303 |
|
|---|
| 304 | return ("http://" . $before_hash, $hash_part, 1);
|
|---|
| 305 |
|
|---|
| 306 | } else {
|
|---|
| 307 | # mailto, news, nntp, telnet, javascript or gopher link
|
|---|
| 308 | return ($before_hash, "", 0);
|
|---|
| 309 | }
|
|---|
| 310 | }
|
|---|
| 311 |
|
|---|
| 312 | sub extract_metadata {
|
|---|
| 313 | my $self = shift (@_);
|
|---|
| 314 | my ($textref, $metadata, $doc_obj, $section) = @_;
|
|---|
| 315 |
|
|---|
| 316 | foreach my $field (split /,/, $self->{'metadata_fields'}) {
|
|---|
| 317 |
|
|---|
| 318 | # don't need to extract field if it was passed in from a previous
|
|---|
| 319 | # (recursive) plugin
|
|---|
| 320 | next if defined $metadata->{$field};
|
|---|
| 321 |
|
|---|
| 322 | # see if there's a <meta> tag for this field
|
|---|
| 323 | if ($$textref =~ /<meta(.*?)(?:name|http-equiv)\s*=\s*\"?$field\"?([^>]*)/is) {
|
|---|
| 324 | my $content = $1 . $2;
|
|---|
| 325 | if ($content =~ /content\s*=\s*\"?(.*?)\"?/is) {
|
|---|
| 326 | if (defined $1) {
|
|---|
| 327 | my $value = $1;
|
|---|
| 328 | $value =~ s/\s+/ /gs;
|
|---|
| 329 | $doc_obj->add_metadata($section, $field, $value);
|
|---|
| 330 | next;
|
|---|
| 331 | }
|
|---|
| 332 | }
|
|---|
| 333 | }
|
|---|
| 334 |
|
|---|
| 335 | # special case for Title metadata - try <title> tags
|
|---|
| 336 | # then first 100 characters of text
|
|---|
| 337 |
|
|---|
| 338 | if ($field =~ /^title$/i) {
|
|---|
| 339 |
|
|---|
| 340 | # see if there's a <title> tag
|
|---|
| 341 | if ($$textref =~ /<title[^>]*>([^<]*)<\/title[^>]*>/is) {
|
|---|
| 342 | if (defined $1) {
|
|---|
| 343 | my $title = $1;
|
|---|
| 344 | if ($title =~ /\w/) {
|
|---|
| 345 | $title =~ s/\s+/ /gs;
|
|---|
| 346 | $doc_obj->add_metadata ($section, $field, $title);
|
|---|
| 347 | next;
|
|---|
| 348 | }
|
|---|
| 349 | }
|
|---|
| 350 | }
|
|---|
| 351 |
|
|---|
| 352 | # if no title use first 100 characters
|
|---|
| 353 | my $tmptext = $$textref;
|
|---|
| 354 | $tmptext =~ s/<[^>]*>//g;
|
|---|
| 355 | my $title = substr ($tmptext, 0, 100);
|
|---|
| 356 | $title =~ s/\s+/ /gs;
|
|---|
| 357 | $doc_obj->add_metadata ($section, $field, $title);
|
|---|
| 358 | }
|
|---|
| 359 | }
|
|---|
| 360 | }
|
|---|
| 361 |
|
|---|
| 362 | # evaluate any "../" to next directory up
|
|---|
| 363 | # evaluate any "./" as here
|
|---|
| 364 | sub eval_dir_dots {
|
|---|
| 365 | my $self = shift (@_);
|
|---|
| 366 | my ($filename) = @_;
|
|---|
| 367 |
|
|---|
| 368 | my $dirsep_os = &util::get_os_dirsep();
|
|---|
| 369 | my @dirsep = split(/$dirsep_os/,$filename);
|
|---|
| 370 |
|
|---|
| 371 | my @eval_dirs = ();
|
|---|
| 372 | foreach my $d (@dirsep) {
|
|---|
| 373 | if ($d eq "..") {
|
|---|
| 374 | pop(@eval_dirs);
|
|---|
| 375 |
|
|---|
| 376 | } elsif ($d eq ".") {
|
|---|
| 377 | # do nothing!
|
|---|
| 378 |
|
|---|
| 379 | } else {
|
|---|
| 380 | push(@eval_dirs,$d);
|
|---|
| 381 | }
|
|---|
| 382 | }
|
|---|
| 383 |
|
|---|
| 384 | return &util::filename_cat(@eval_dirs);
|
|---|
| 385 | }
|
|---|
| 386 |
|
|---|
| 387 | sub replace_usemap_links {
|
|---|
| 388 | my $self = shift (@_);
|
|---|
| 389 | my ($front, $link, $back) = @_;
|
|---|
| 390 |
|
|---|
| 391 | $link =~ s/^\.\///;
|
|---|
| 392 | return $front . $link . $back;
|
|---|
| 393 | }
|
|---|
| 394 |
|
|---|
| 395 | sub inc_filecount {
|
|---|
| 396 | my $self = shift (@_);
|
|---|
| 397 |
|
|---|
| 398 | if ($self->{'file_num'} == 1000) {
|
|---|
| 399 | $self->{'dir_num'} ++;
|
|---|
| 400 | $self->{'file_num'} = 0;
|
|---|
| 401 | } else {
|
|---|
| 402 | $self->{'file_num'} ++;
|
|---|
| 403 | }
|
|---|
| 404 | }
|
|---|
| 405 |
|
|---|
| 406 | 1;
|
|---|
