Changeset 808 for trunk/gsdl/perllib/plugins/HTMLPlug.pm
- Timestamp:
- 1999-12-07T10:41:49+13:00 (24 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/gsdl/perllib/plugins/HTMLPlug.pm
r734 r808 2 2 # 3 3 # HTMLPlug.pm -- basic html plugin 4 # 4 5 # A component of the Greenstone digital library software 5 6 # from the New Zealand Digital Library Project at the … … 24 25 ########################################################################### 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 26 37 27 38 package HTMLPlug; 28 39 29 40 use BasPlug; 30 use sorttools;31 41 use html; 32 42 use util; 33 34 use File::Basename; 43 use parsargv; 35 44 36 45 sub BEGIN { … … 38 47 } 39 48 49 sub print_usage { 50 print STDERR "\nIncorrect options passed to HTMLPlug, check your collect.cfg configuration file\n"; 51 52 print STDERR "\n usage: plugin HTMLPlug [options]\n\n"; 53 print STDERR " options:\n"; 54 print STDERR " -process_exp A perl regular expression to match against filenames.\n"; 55 print STDERR " Matching filenames will be processed by this plugin.\n"; 56 print STDERR " Defaults to '(?i)\.html?\$' i.e. all documents ending in\n"; 57 print STDERR " .htm or .html (case-insensitive).\n"; 58 print STDERR " -nolinks Don't make any attempt to trap links (setting this flag may\n"; 59 print STDERR " improve speed of building/importing but any relative links within\n"; 60 print STDERR " documents will be broken).\n"; 61 print STDERR " -block_exp Files matching this regular expression will be blocked from\n"; 62 print STDERR " being passed to any further plugins in the list. By default\n"; 63 print STDERR " HTMLPlug blocks any files with .gif, .jpg, .jpeg, .png, .pdf\n"; 64 print STDERR " or .rtf file extensions.\n"; 65 print STDERR " -keep_head Don't remove headers from html files.\n"; 66 print STDERR " -no_metadata Don't attempt to extract any metadata from files.\n"; 67 print STDERR " -metadata_fields Comma separated list of metadata fields to attempt to extract.\n"; 68 print STDERR " Defaults to 'Title'\n"; 69 print STDERR " -w3mir Set if w3mir was used to generate input file structure.\n\n"; 70 } 71 40 72 sub new { 41 my ($class) = @_;73 my $class = shift (@_); 42 74 $self = new BasPlug (); 43 75 44 $self->{'f2d_mapping'} = undef; 45 $self->{'f2d_partitioned_dirs'} = []; 76 if (!parsargv::parse(\@_, 77 q^process_exp/.*/(?i)\.html?$^, \$self->{'process_exp'}, 78 q^nolinks^, \$self->{'nolinks'}, 79 q^block_exp/.*/(?i)\.(gif|jpe?g|png|pdf|rtf)$^, \$self->{'block_exp'}, 80 q^keep_head^, \$self->{'keep_head'}, 81 q^no_metadata^, \$self->{'no_metadata'}, 82 q^metadata_fields/.*/Title^, \$self->{'metadata_fields'}, 83 q^w3mir^, \$self->{'w3mir'})) { 84 &print_usage(); 85 die "\n"; 86 } 87 88 $self->{'aux_files'} = {}; 89 $self->{'dir_num'} = 0; 90 $self->{'file_num'} = 0; 46 91 47 92 return bless $self, $class; … … 53 98 return 0; # this is not a recursive plugin 54 99 } 55 56 #--57 # Convert filename to lookup filename58 #--59 sub f2d_lookup_filename60 {61 my ($self,$filename) = @_;62 63 my $coldir = $ENV{'GSDLCOLLECTDIR'};64 my $dirsep_re = &util::get_re_dirsep();65 my $supress_dirs = "(import|building)";66 67 my @part_dirs = @{$self->{'f2d_partitioned_dirs'}};68 69 if (scalar(@part_dirs)>0)70 {71 my $joined_part_dirs = join('|',@part_dirs);72 $supress_dirs .= "($dirsep_re)($joined_part_dirs)";73 }74 75 my $lookup_filename = $filename;76 $lookup_filename =~ s/^$coldir($dirsep_re)//;77 $lookup_filename =~ s/^$supress_dirs($dirsep_re)//;78 79 return $lookup_filename;80 }81 82 83 #--84 # file to document mapping85 #--86 87 sub rec_f2d_mapping88 {89 my ($self,$dirname) = @_;90 91 # read all the files in the directory92 if (!opendir(DIR, $dirname))93 {94 print STDERR "HTMLPlug: WARNING - couldn't read directory $dirname";95 print STDERR " during file to doc mapping\n";96 return;97 }98 my @dir = readdir (DIR);99 closedir (DIR);100 101 # process each file102 my $subfile;103 foreach $subfile (@dir)104 {105 if ($subfile !~ /^\.\.?$/)106 {107 my $filename = &util::filename_cat($dirname, $subfile);108 109 if (-d $filename)110 {111 my $dirname = $filename;112 $self->rec_f2d_mapping($dirname);113 }114 else115 {116 if ($subfile =~ m/\.(html?(\.gz)?)$/i && (-e $filename))117 {118 # add mapping119 120 print STDERR "HTMLPlug: Precalculating OID for $subfile\n"121 if (defined($self->{'verbosity'}));122 123 my $oid = doc::_calc_OID(undef,$filename);124 my $lookup_filename125 = $self->f2d_lookup_filename($filename);126 127 $self->{'f2d_mapping'}->{$lookup_filename} = $oid;128 129 if ($subfile =~ m/index\.(html?(\.gz)?)$/i)130 {131 # Cater for links such as "/paper/" mapping to132 # "/paper/index.html"133 my $lookup_dirname134 = $self->f2d_lookup_filename($dirname);135 136 $self->{'f2d_mapping'}->{$lookup_dirname} = $oid;137 }138 }139 elsif ($subfile =~ /\.(gif|jpg|jpeg|png)$/i)140 {141 # convert to png ?142 143 # Hard link in build directory144 #--145 my $coldir = $ENV{'GSDLCOLLECTDIR'};146 my $dirsep_re = &util::get_re_dirsep();147 my $iorb_re = "import|building";148 149 my ($dirsep,$iorb)150 = ($filename =~ m/^$coldir($dirsep_re)($iorb_re)/);151 152 my $copyname = $filename;153 $copyname154 =~ s/^${coldir}${dirsep}${iorb}155 /${coldir}${dirsep}building_images${dirsep}imgsrc/x;156 157 if (!-e $copyname)158 {159 print STDERR "HTMLPlug: Hard linking $subfile\n"160 if (defined($self->{'verbosity'}));161 &util::hard_link($filename,$copyname);162 }163 164 }165 }166 167 }168 }169 }170 171 sub build_file_to_doc_mapping172 {173 my ($self,$base_dir) = @_;174 175 $self->{'f2d_mapping'} = {};176 177 my $bimages_dir178 = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},"building_images","imgsrc");179 if (!-e $bimages_dir && !mkdir ($bimages_dir, 0775))180 {181 print STDERR "HTMLPlug:: Could not create directory $bimages_dir\n";182 return;183 }184 185 $self->rec_f2d_mapping($base_dir);186 }187 188 #--189 #--190 100 191 101 # return number of files processed, undef if can't process … … 196 106 my ($pluginfo, $base_dir, $file, $metadata, $processor) = @_; 197 107 198 my $verbosity = $processor->{'verbosity'};199 $self->{'verbosity'} = $verbosity;200 201 if (!defined($self->{'f2d_mapping'}))202 {203 print STDERR "HTMLPlug: building OID mapping\n"204 if $processor->{'verbosity'};205 206 $self->build_file_to_doc_mapping($base_dir,$verbosity);207 }208 108 my $filename = &util::filename_cat($base_dir, $file); 209 my $absdir = $filename; 210 $absdir =~ s/[^\/\\]*$//; 211 212 if ($file =~ /\.(gif|jpg|jpeg|png)$/i) 213 { 214 return 0; 215 } 216 217 return undef unless ($filename =~ /\.(html?(\.gz)?)$/i && (-e $filename)); 218 return undef if (-d $filename); 219 220 my $gz = 0; 221 if (defined $2) { 222 $gz = $2; 223 $gz = 1 if ($gz =~ /\.gz/i); 224 } 225 226 print STDERR "HTMLPlug: processing $file\n" if $processor->{'verbosity'}; 109 return 0 if $filename =~ /$self->{'block_exp'}/; 110 if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) { 111 return undef; 112 } 113 $file =~ s/^[\/\\]+//; 114 115 $self->{'verbosity'} = $processor->{'verbosity'}; 116 print STDERR "HTMLPlug: processing $file\n" 117 if $self->{'verbosity'} > 1; 227 118 228 119 # create a new document 229 120 my $doc_obj = new doc ($file, "indexed_doc"); 230 231 if ($gz) { 232 open (FILE, "zcat $filename |") || die "HTMLPlug::read - zcat can't open $filename\n"; 121 my $cursection = $doc_obj->get_top_section(); 122 123 # read in HTML file 124 open (FILE, $filename) || die "HTMLPlug::read - can't open $filename\n"; 125 undef $/; 126 my $text = <FILE>; 127 $/ = "\n"; 128 close FILE; 129 if ($text !~ /\w/) { 130 print STDERR "HTMLPlug: ERROR: $file contains no text\n" if $self->{'verbosity'}; 131 return 0; 132 } 133 134 $self->add_external_metadata ($metadata, $doc_obj, $cursection); 135 $self->extract_metadata (\$text, $metadata, $doc_obj, $cursection) 136 unless $self->{'no_metadata'}; 137 138 # Store URL for page as metadata - this can be used for an 139 # altavista style search interface. The URL won't be valid 140 # unless the file structure contains the domain name (i.e. 141 # like when w3mir is used to download a website). 142 my $web_url = "http://$file"; 143 $web_url =~ s/\\/\//g; # for windows 144 $doc_obj->add_metadata($cursection, "URL", $web_url); 145 146 # remove header and footer 147 if (!$self->{'keep_head'}) { 148 $text =~ s/^.*?<body[^>]*>//is; 149 $text =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg; 150 } 151 152 # trap links 153 if (!$self->{'nolinks'}) { 154 155 # usemap="./#index" not handled correctly => change to "#index" 156 $text =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/ 157 $self->replace_usemap_links($1, $2, $3)/isge; 158 159 $text =~ s/(<(?:a|area|frame)\s+[^>]*?(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/ 160 $self->replace_href_links ($1, $2, $3, $base_dir, $file, $doc_obj)/isge; 161 } 162 163 # trap images 164 $text =~ s/(<img[^>]*?src\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/ 165 $self->replace_images ($1, $2, $3, $base_dir, $file, $doc_obj)/isge; 166 167 $doc_obj->add_text ($cursection, $text); 168 169 # add an OID 170 $doc_obj->set_OID(); 171 172 # process the document 173 $processor->process($doc_obj); 174 175 return 1; # processed the file 176 } 177 178 sub replace_images { 179 my $self = shift (@_); 180 my ($front, $link, $back, $base_dir, $file, $doc_obj) = @_; 181 182 $link =~ s/\n/ /g; 183 184 my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file); 185 return $front . $self->add_file ($href, $base_dir, $doc_obj) . $back; 186 } 187 188 sub replace_href_links { 189 my $self = shift (@_); 190 my ($front, $link, $back, $base_dir, $file, $doc_obj) = @_; 191 192 # attempt to sort out targets - frames are not handled 193 # well in this plugin and some cases will screw things 194 # up - e.g. the _parent target (so we'll just remove 195 # them all ;-) 196 $front =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is; 197 $back =~ s/(target=\"?)_top(\"?)/$1_gsdltop_$2/is; 198 $front =~ s/target=\"?_parent\"?//is; 199 $back =~ s/target=\"?_parent\"?//is; 200 201 return $front . $link . $back if $link =~ /^\#/s; 202 $link =~ s/\n/ /g; 203 204 my ($href, $hash_part, $rl) = $self->format_link ($link, $base_dir, $file); 205 206 if (($rl == 0) || ($href =~ /$self->{'process_exp'}/) || 207 ($href =~ /\/$/) || ($href =~ /^(mailto|news|gopher|nntp|telnet|javascript):/i)) { 208 $link = $href . $hash_part; 209 &html::urlsafe ($link); 210 return $front . "_httpextlink_&href=" . $link . "&rl=" . $rl . $back; 211 233 212 } else { 234 open (FILE, $filename) || die "HTMLPlug::read - can't open $filename\n"; 235 } 236 my $cursection = $doc_obj->get_top_section(); 237 238 # look up precalculated OID 239 my $lookup_filename = $self->f2d_lookup_filename($filename); 240 my $precalc_oid = $self->{'f2d_mapping'}->{$lookup_filename}; 241 242 # read in HTML file 243 my $text = ""; 244 my $line = ""; 245 my $donehead = 0; 246 my $title = ""; 247 while (defined ($line = <FILE>)) { 248 $text .= $line; 249 } 250 251 # remove line breaks 252 $text =~ s/\s+/ /g; 253 254 my $foundtitle = 0; 255 # don't use title tag if Title has been passed 256 # in from another plugin 257 if (defined $metadata->{'Title'}) { 258 $foundtitle = 1; 259 } 260 if (!$foundtitle) { 261 # see if there's a <title> tag 262 if ($text =~ /<title[^>]*>([^<]*)<\/title[^>]*>/i) { 263 if (defined $1) { 264 my $title = $1; 265 if ($title =~ /\w/) { 266 $doc_obj->add_metadata ($cursection, "Title", $title); 267 $foundtitle = 1; 268 } 213 # link is to some other type of file (image, pdf etc.) so we'll 214 # need to associate that file 215 return $front . $self->add_file ($href, $base_dir, $doc_obj) . $back; 216 } 217 } 218 219 sub add_file { 220 my $self = shift (@_); 221 my ($href, $base_dir, $doc_obj) = @_; 222 my ($newname); 223 224 my $filename = $href; 225 $filename =~ s/^[^:]*:\/\///; 226 $filename = &util::filename_cat ($base_dir, $filename); 227 my ($ext) = $filename =~ /(\.[^\.]*)$/; 228 if (defined $self->{'aux_files'}->{$href}) { 229 $newname = $self->{'aux_files'}->{$href}->{'dir_num'} . "/" . 230 $self->{'aux_files'}->{$href}->{'file_num'} . $ext; 231 } else { 232 $newname = $self->{'dir_num'} . "/" . $self->{'file_num'} . $ext; 233 $self->inc_filecount (); 234 } 235 $doc_obj->associate_file($filename, $newname); 236 return "_httpcollimg_/$newname"; 237 } 238 239 240 sub format_link { 241 my $self = shift (@_); 242 my ($link, $base_dir, $file) = @_; 243 244 my ($before_hash, $hash_part) = $link =~ /^([^\#]*)(\#?.*)$/; 245 $hash_part = "" if !defined $hash_part; 246 if (!defined $before_hash || $before_hash !~ /[\w\.\/]/) { 247 print STDERR "HTMLPlug: ERROR - badly formatted tag ignored ($front$link$back)\n" 248 if $self->{'verbosity'}; 249 return ($link, "", 0); 250 } 251 252 if ($before_hash =~ s/^((?:http|ftp|file):\/\/)//i) { 253 my $type = $1; 254 255 if ($link =~ /^(http|ftp):/i) { 256 # Turn url (using /) into file name (possibly using \ on windows) 257 my @http_dir_split = split('/', $before_hash); 258 $before_hash = &util::filename_cat(@http_dir_split); 259 } 260 261 $before_hash = $self->eval_dir_dots($before_hash); 262 263 my $linkfilename = &util::filename_cat ($base_dir, $before_hash); 264 265 my $rl = 0; 266 $rl = 1 if (-e $linkfilename); 267 268 # make sure there's a slash on the end if it's a directory 269 if ($before_hash !~ /\/$/) { 270 $before_hash .= "/" if (-d $linkfilename); 271 } 272 273 return ($type . $before_hash, $hash_part, $rl); 274 275 } elsif ($link !~ /^(mailto|news|gopher|nntp|telnet|javascript):/i) { 276 277 if ($before_hash =~ s/^\///) { 278 # the first directory will be the domain name if w3mir was used 279 # to generate archives, otherwise we'll assume all files are 280 # from the same site and base_dir is the root 281 if ($self->{'w3mir'}) { 282 my @dirs = split /[\/\\]/, $file; 283 my $domname = shift (@dirs); 284 $before_hash = &util::filename_cat($domname, $before_hash); 285 $before_hash =~ s/\\/\//g; # for windows 269 286 } 270 } 271 } 272 # if no title use first 100 characters 273 if (!$foundtitle) { 274 my $tmptext = $text; 275 $tmptext =~ s/<[^>]*>//g; 276 my $title = substr ($tmptext, 0, 100); 277 $doc_obj->add_metadata ($cursection, "Title", $title); 278 } 279 280 # remove header rubbish 281 $text =~ s/^.*?<body[^>]*>//i; 282 283 # usemap="./#index" not handled correctly => change to "#index" 284 $text =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">]+)(\"?[^>]*>)/ 285 &replace_usemap_links($1, $2, $3)/ige; 286 287 # fix up the href links 288 $f2d_mapping = $self->{'f2d_mapping'}; 289 290 $text =~ s/(<(a|area)\s+[^>]*?href\s*=\s*\"?)([^\">]+)(\"?[^>]*>)/ 291 &replace_href_links($filename,$precalc_oid,$f2d_mapping,$1,$3,$4)/ige; 292 293 # add a newline at the beginning of each paragraph 294 $text =~ s/(.)\s*<p\b/$1\n\n<p/gi; 295 296 # add a newline every 80 characters at a word boundary 297 # Note: this regular expression puts a line feed before 298 # the last word in each section, even when it is not 299 # needed. 300 $text =~ s/(.{1,80})\s/$1\n/g; 301 302 303 # Store URL for page as metadata 304 my $web_url = "http://$lookup_filename"; 305 $doc_obj->add_metadata($cursection, "URL", $web_url); 306 307 my $import_url = $filename; 308 $import_url =~ s/^$ENV{'GSDLCOLLECTDIR'}/_httpcollection_/; 309 310 # Add base tag so images can find correct location 311 my $index_url = $import_url; 312 $index_url =~ s/_httpcollection_\/import/_httpcollection_\/index\/imgsrc/; 313 314 my $dirsep_re = &util::get_re_dirsep(); 315 my @base_url_split = split(/$dirsep_re/,$index_url); 316 pop(@base_url_split); 317 my $base_url = join('/',@base_url_split,""); # force / at end 318 319 $text = "<base href=\"$base_url\">\n$text"; 320 $doc_obj->add_text ($cursection, $text); 321 322 # Add metadata that has been provided externally 323 #-- 287 288 } else { 289 # Turn relative file path into full path 290 my $dirname = &File::Basename::dirname($file); 291 $before_hash = &util::filename_cat($dirname, $before_hash); 292 $before_hash = $self->eval_dir_dots($before_hash); 293 } 294 295 # make sure there's a slash on the end if it's a directory 296 my $linkfilename = &util::filename_cat ($base_dir, $before_hash); 297 if ($before_hash !~ /\/$/) { 298 $before_hash .= "/" if (-d $linkfilename); 299 } 300 301 return ("http://" . $before_hash, $hash_part, 1); 302 303 } else { 304 # mailto, news, nntp, telnet, javascript or gopher link 305 return ($before_hash, "", 0); 306 } 307 } 308 309 # Add metadata that has been provided externally (i.e. by 310 # a previous (recursive) plugin 311 sub add_external_metadata { 312 my $self = shift (@_); 313 my ($metadata, $doc_obj, $section) = @_; 314 324 315 foreach $field (keys(%$metadata)) { 325 316 # $metadata->{$field} may be an array reference 326 317 if (ref ($metadata->{$field}) eq "ARRAY") { 327 318 map { 328 $doc_obj->add_metadata ($ cursection, $field, $_);319 $doc_obj->add_metadata ($section, $field, $_); 329 320 } @{$metadata->{$field}}; 330 321 } else { 331 $doc_obj->add_metadata ($cursection, $field, $metadata->{$field}); 332 } 333 } 334 335 # Fix OID so it is the same as the pre-calculated OID 336 $doc_obj->set_OID($precalc_oid); 337 338 my $set_oid = $doc_obj->get_OID(); 339 if ($precalc_oid ne $set_oid) # check (for super safety!) 340 { 341 print STDERR "Warning: pre-calculated OID and current OID differnt:"; 342 print STDERR " $filename\n"; 343 } 344 345 346 # process the document 347 $processor->process($doc_obj); 348 349 return 1; # processed the file 350 } 351 352 # support for fixing up links to work as a GSDL collection 353 #-- 354 355 sub eval_dir_dots 356 { 357 # evaluate any "../" to next directory up 358 # evaluate any "./" as here 359 #-- 360 my ($self,$filename) = @_; 322 $doc_obj->add_metadata ($section, $field, $metadata->{$field}); 323 } 324 } 325 } 326 327 sub extract_metadata { 328 my $self = shift (@_); 329 my ($textref, $metadata, $doc_obj, $section) = @_; 330 331 foreach $field (split /,/, $self->{'metadata_fields'}) { 332 333 # don't need to extract field if it was passed in from a previous 334 # (recursive) plugin 335 next if defined $metadata->{$field}; 336 337 # see if there's a <meta> tag for this field 338 if ($$textref =~ /<meta(.*?)(?:name|http-equiv)\s*=\s*\"?$field\"?([^>]*)/is) { 339 my $content = $1 . $2; 340 if ($content =~ /content\s*=\s*\"?(.*?)\"?/is) { 341 if (defined $1) { 342 my $value = $1; 343 $value =~ s/\s+/ /gs; 344 $doc_obj->add_metadata($section, $field, $value); 345 next; 346 } 347 } 348 } 349 350 # special case for Title metadata - try <title> tags 351 # then first 100 characters of text 352 353 if ($field =~ /^title$/i) { 354 355 # see if there's a <title> tag 356 if ($$textref =~ /<title[^>]*>([^<]*)<\/title[^>]*>/is) { 357 if (defined $1) { 358 my $title = $1; 359 if ($title =~ /\w/) { 360 $title =~ s/\s+/ /gs; 361 $doc_obj->add_metadata ($section, $field, $title); 362 next; 363 } 364 } 365 } 366 367 # if no title use first 100 characters 368 my $tmptext = $$textref; 369 $tmptext =~ s/<[^>]*>//g; 370 my $title = substr ($tmptext, 0, 100); 371 $title =~ s/\s+/ /gs; 372 $doc_obj->add_metadata ($section, $field, $title); 373 } 374 } 375 } 376 377 # evaluate any "../" to next directory up 378 # evaluate any "./" as here 379 sub eval_dir_dots { 380 my $self = shift (@_); 381 my ($filename) = @_; 361 382 362 383 my $dirsep_os = &util::get_os_dirsep(); … … 364 385 365 386 my @eval_dirs = (); 366 foreach $d (@dirsep) 367 { 368 if ($d eq "..") 369 { 387 foreach $d (@dirsep) { 388 if ($d eq "..") { 370 389 pop(@eval_dirs); 371 } 372 elsif ($d eq ".") 373 { 390 391 } elsif ($d eq ".") { 374 392 # do nothing! 375 } 376 else 377 { 393 394 } else { 378 395 push(@eval_dirs,$d); 379 396 } … … 383 400 } 384 401 385 sub replace_href_links 386 { 387 my ($this_filename,$this_oid, $f2d_mapping, $front,$link,$back) = @_; 388 389 return $front.$link.$back if ($link =~ m/\.(gif|jpg|jpeg|png)$/i); 390 391 if ($link =~ m/^(http|ftp|file):/i) 392 { 393 # this should really check that the link ends in .htm* 394 #**** 395 my $http_as_filename = $link; 396 $http_as_filename =~ s/^(http|ftp|file):\/\///i; 397 398 my ($before_hash,$after_hash) 399 = ($http_as_filename =~ m/^([^\#]*)\#?(.*)$/); 400 401 if ($link =~ m/^(http|ftp):/i) 402 { 403 # Turn url (using /) into file name (possibly using \ on windows) 404 my @http_dir_split = split('/',$before_hash); 405 $http_as_filename = &util::filename_cat(@http_dir_split); 406 } 407 408 $http_as_filename = $self->eval_dir_dots($http_as_filename); 409 410 if (defined($f2d_mapping->{$http_as_filename})) 411 { 412 # transform link into a local link (and then let it be 413 # processed by later code) 414 $link = $http_as_filename; 415 416 $link = "_httpdocument_&cl=_cgiargcl_&d=$oid"; 417 $link .= "#$after_hash" if ($after_hash ne ""); 418 } 419 else 420 { 421 # external link => set it up to pass through off-site page 422 423 my $link_safe = $link; 424 &html::urlsafe($link_safe); 425 426 $link = "_httpextlink_&href=$link_safe&d=$this_oid"; 427 428 } 429 } 430 elsif ($link !~ m/^(mailto|news):/i) 431 { 432 my ($before_hash,$after_hash) = ($link =~ m/^([^\#]*)\#?(.*)$/); 433 my $link_filename; 434 435 if ($before_hash =~ m/^\//) 436 { 437 my $dirsep_re = &util::get_re_dirsep(); 438 my $lookup = $self->f2d_lookup_filename($this_filename); 439 my @lookup_split = split(/$dirsep_re/,$lookup); 440 my $domname = shift(@lookup_split); 441 $link_filename = &util::filename_cat($domname,$before_hash); 442 } 443 else 444 { 445 # Turn relative file path into full path 446 if ($before_hash eq "") # handle links such as <a href="#x"> 447 { 448 $link_filename = $this_filename; 449 } 450 else 451 { 452 my $dirname = &File::Basename::dirname($this_filename); 453 $link_filename = &util::filename_cat($dirname,$before_hash); 454 } 455 } 456 457 $link_filename = $self->eval_dir_dots($link_filename); 458 my $lookup_filename = $self->f2d_lookup_filename($link_filename); 459 460 my $oid = $f2d_mapping->{$lookup_filename}; 461 if (defined($oid)) 462 { 463 $link = "_httpdocument_&cl=_cgiargcl_&d=$oid"; 464 $link .= "#$after_hash" if ($after_hash ne ""); 465 } 466 else 467 { 468 print STDERR "HTMLPlug WARNING:"; 469 print STDERR " Could not find link: \"$lookup_filename\". "; 470 print STDERR " Deactivating link\n"; 471 472 $link = "_httpextlink_&d=$this_oid"; 473 } 474 } 475 else 476 { 477 if ($link !~ m/^(mailto|news):/i) 478 { 479 print STDERR "HTMLPlug WARNING: Unhandled type of link, \"$link\"\n"; 480 } 481 else 482 { 483 my $link_safe = $link; 484 &html::urlsafe($link_safe); 485 486 $link = "_httpextlink_&href=$link_safe&d=$this_oid"; 487 } 488 } 489 490 my $fixed_tag = "${front}${link}${back}"; 491 492 return $fixed_tag; 493 } 494 495 496 sub replace_usemap_links 497 { 402 sub replace_usemap_links { 403 my $self = shift (@_); 498 404 my ($front, $link, $back) = @_; 499 405 500 406 $link =~ s/^\.\///; 501 502 return "${front}${link}${back}"; 503 } 504 505 407 return $front . $link . $back; 408 } 409 410 sub inc_filecount { 411 my $self = shift (@_); 412 413 if ($self->{'file_num'} == 1000) { 414 $self->{'dir_num'} ++; 415 $self->{'file_num'} = 0; 416 } else { 417 $self->{'file_num'} ++; 418 } 419 } 506 420 507 421 1; 508 509 510 511 512 513 514 515 516 517 518
Note:
See TracChangeset
for help on using the changeset viewer.