| 1 | ###########################################################################
|
|---|
| 2 | #
|
|---|
| 3 | # ConvertToPlug.pm -- plugin that inherits from HTML or TEXT Plug, depending
|
|---|
| 4 | # on plugin argument convert_to
|
|---|
| 5 | #
|
|---|
| 6 | # A component of the Greenstone digital library software
|
|---|
| 7 | # from the New Zealand Digital Library Project at the
|
|---|
| 8 | # University of Waikato, New Zealand.
|
|---|
| 9 | #
|
|---|
| 10 | # Copyright (C) 1999 New Zealand Digital Library Project
|
|---|
| 11 | #
|
|---|
| 12 | # This program is free software; you can redistribute it and/or modify
|
|---|
| 13 | # it under the terms of the GNU General Public License as published by
|
|---|
| 14 | # the Free Software Foundation; either version 2 of the License, or
|
|---|
| 15 | # (at your option) any later version.
|
|---|
| 16 | #
|
|---|
| 17 | # This program is distributed in the hope that it will be useful,
|
|---|
| 18 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|---|
| 19 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|---|
| 20 | # GNU General Public License for more details.
|
|---|
| 21 | #
|
|---|
| 22 | # You should have received a copy of the GNU General Public License
|
|---|
| 23 | # along with this program; if not, write to the Free Software
|
|---|
| 24 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
|---|
| 25 | #
|
|---|
| 26 | ###########################################################################
|
|---|
| 27 |
|
|---|
| 28 | # The plugin is inherited by such plugins as WordPlug and PDFPlug.
|
|---|
| 29 | # It facilitates the conversion of these document types to either HTML
|
|---|
| 30 | # or TEXT by setting up variable that instruct ConvertToBasPlug
|
|---|
| 31 | # how to work.
|
|---|
| 32 |
|
|---|
| 33 | # It works by dynamically inheriting HTMLPlug or TEXTPlug based on
|
|---|
| 34 | # the plugin argument 'convert_to'. If the argument is not present,
|
|---|
| 35 | # the default is to inherit HTMLPlug.
|
|---|
| 36 |
|
|---|
| 37 |
|
|---|
| 38 | package ConvertToPlug;
|
|---|
| 39 |
|
|---|
| 40 | use BasPlug;
|
|---|
| 41 | use HTMLPlug;
|
|---|
| 42 | use TEXTPlug;
|
|---|
| 43 | use ghtml;
|
|---|
| 44 |
|
|---|
| 45 | sub BEGIN {
|
|---|
| 46 | @ISA = ('HTMLPlug');
|
|---|
| 47 | # @ISA = ('HTMLPlug', 'TEXTPlug');
|
|---|
| 48 | # @ISA = ('BasPlug'); #, 'HTMLPlug', 'TEXTPlug');
|
|---|
| 49 | }
|
|---|
| 50 |
|
|---|
| 51 | sub print_usage {
|
|---|
| 52 | my ($plugin_name) = @_;
|
|---|
| 53 |
|
|---|
| 54 | # for when this function is called directly by pluginfo.pl
|
|---|
| 55 | if (ref ($plugin_name)) {
|
|---|
| 56 | $plugin_name = ref ($plugin_name);
|
|---|
| 57 | }
|
|---|
| 58 |
|
|---|
| 59 | print STDERR "\n usage: plugin $plugin_name [options]\n\n";
|
|---|
| 60 | print STDERR " options:\n";
|
|---|
| 61 | print STDERR " -convert_to (html|text) plugin converts to TEXT or HTML\n";
|
|---|
| 62 | print STDERR " (default html)\n";
|
|---|
| 63 | print STDERR " -use_strings if set a simple strings function\n";
|
|---|
| 64 | print STDERR " will be called to extract text\n";
|
|---|
| 65 | print STDERR " if the conversion utility fails\n";
|
|---|
| 66 | }
|
|---|
| 67 |
|
|---|
| 68 | sub parse_args
|
|---|
| 69 | {
|
|---|
| 70 | my $class = shift (@_);
|
|---|
| 71 | my ($args) = @_;
|
|---|
| 72 |
|
|---|
| 73 | my $plugin_name = $class;
|
|---|
| 74 | $plugin_name =~ s/\.pm$//;
|
|---|
| 75 |
|
|---|
| 76 | my $newargs = {};
|
|---|
| 77 |
|
|---|
| 78 | if (!parsargv::parse($args,
|
|---|
| 79 | q^extract_keyphrases^, \$newargs->{'kea'}, #with extra options
|
|---|
| 80 | q^extract_keyphrase_options/.*/^, \$newargs->{'kea_options'}, #no extra options
|
|---|
| 81 | q^convert_to/(html|text)/html^, \$newargs->{'generate_format'},
|
|---|
| 82 | q^use_strings^, \$newargs->{'use_strings'},
|
|---|
| 83 | "allow_extra_options")) {
|
|---|
| 84 |
|
|---|
| 85 | print STDERR "\nIncorrect options passed to $plugin_name, ";
|
|---|
| 86 | print STDERR "check your collect.cfg configuration file\n";
|
|---|
| 87 | &print_usage($plugin_name);
|
|---|
| 88 | die "\n";
|
|---|
| 89 | }
|
|---|
| 90 |
|
|---|
| 91 | return ($plugin_name, $newargs);
|
|---|
| 92 | }
|
|---|
| 93 |
|
|---|
| 94 | sub new {
|
|---|
| 95 | my $class = shift (@_);
|
|---|
| 96 | if ($class eq "ConvertToPlug") {$class = shift (@_);}
|
|---|
| 97 | my $self;
|
|---|
| 98 | # parsargv::parse might modify the list, so we do this by creating a copy
|
|---|
| 99 | # of the argument list.
|
|---|
| 100 | my @arglist = @_;
|
|---|
| 101 | my ($plugin_name, $args) = $class->parse_args(\@_);
|
|---|
| 102 |
|
|---|
| 103 | if ($class eq "PDFPlug" && $args->{'generate_format'} eq "text" &&
|
|---|
| 104 | $ENV{'GSDLOS'} =~ /^windows$/i) {
|
|---|
| 105 | print STDERR "Windows does not support pdf to text. PDFs will be converted to HTML instead\n";
|
|---|
| 106 | $args->{'generate_format'} = "html";
|
|---|
| 107 | }
|
|---|
| 108 |
|
|---|
| 109 | if ($args->{'generate_format'} eq "text")
|
|---|
| 110 | {
|
|---|
| 111 | $self = new TEXTPlug ($class, @arglist);
|
|---|
| 112 | $self->{'convert_to'} = "TEXT";
|
|---|
| 113 | $self->{'convert_to_ext'} = "txt";
|
|---|
| 114 | }
|
|---|
| 115 | else
|
|---|
| 116 | {
|
|---|
| 117 | $self = new HTMLPlug ($class, @arglist);
|
|---|
| 118 | $self->{'convert_to'} = "HTML";
|
|---|
| 119 | $self->{'convert_to_ext'} = "html";
|
|---|
| 120 |
|
|---|
| 121 | $self->{'rename_assoc_files'} = 1;
|
|---|
| 122 | $self->{'metadata_fields'} .= ",GENERATOR";
|
|---|
| 123 | }
|
|---|
| 124 |
|
|---|
| 125 | foreach my $key (keys %$args) {
|
|---|
| 126 | $self->{$key} = $args->{$key};
|
|---|
| 127 | }
|
|---|
| 128 |
|
|---|
| 129 | return bless $self, $class;
|
|---|
| 130 | }
|
|---|
| 131 |
|
|---|
| 132 |
|
|---|
| 133 |
|
|---|
| 134 | # Run conversion utility on the input file.
|
|---|
| 135 | #
|
|---|
| 136 | # The conversion takes place in a collection specific 'tmp' directory so
|
|---|
| 137 | # that we don't accidentally damage the input.
|
|---|
| 138 | #
|
|---|
| 139 | # The desired output type is indicated by $output_ext. This is usually
|
|---|
| 140 | # something like "html" or "word", but can be "best" (or the empty string)
|
|---|
| 141 | # to indicate that the conversion utility should do the best it can.
|
|---|
| 142 |
|
|---|
| 143 | sub tmp_area_convert_file {
|
|---|
| 144 | my $self = shift (@_);
|
|---|
| 145 | my ($output_ext, $input_filename, $textref) = @_;
|
|---|
| 146 |
|
|---|
| 147 | my $outhandle = $self->{'outhandle'};
|
|---|
| 148 | my $convert_to = $self->{'convert_to'};
|
|---|
| 149 | my $failhandle = $self->{'failhandle'};
|
|---|
| 150 |
|
|---|
| 151 | # softlink to collection tmp dir
|
|---|
| 152 | my $tmp_dirname
|
|---|
| 153 | = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
|---|
| 154 | &util::mk_dir($tmp_dirname) if (!-e $tmp_dirname);
|
|---|
| 155 |
|
|---|
| 156 | # derive tmp filename from input filename
|
|---|
| 157 | my ($tailname, $dirname, $suffix)
|
|---|
| 158 | = &File::Basename::fileparse($input_filename, "\\.[^\\.]+\$");
|
|---|
| 159 |
|
|---|
| 160 | # Remove any white space from filename -- no risk of name collision, and
|
|---|
| 161 | # makes later conversion by utils simpler. Leave spaces in path...
|
|---|
| 162 | $tailname =~ s/\s+//g;
|
|---|
| 163 |
|
|---|
| 164 | my $tmp_filename = &util::filename_cat($tmp_dirname, "$tailname$suffix");
|
|---|
| 165 |
|
|---|
| 166 | &util::soft_link($input_filename, $tmp_filename);
|
|---|
| 167 |
|
|---|
| 168 | my $verbosity = $self->{'verbosity'};
|
|---|
| 169 | if ($verbosity > 0) {
|
|---|
| 170 | print $outhandle "Converting $tailname$suffix to $convert_to format\n";
|
|---|
| 171 | }
|
|---|
| 172 |
|
|---|
| 173 | my $errlog = &util::filename_cat($tmp_dirname, "err.log");
|
|---|
| 174 |
|
|---|
| 175 | # Execute the conversion command and get the type of the result,
|
|---|
| 176 | # making sure the converter gives us the appropriate output type
|
|---|
| 177 | my $output_type = lc($convert_to);
|
|---|
| 178 | my $cmd = "perl -S gsConvert.pl -verbose $verbosity ";
|
|---|
| 179 | if ($self->{'use_strings'}) {
|
|---|
| 180 | $cmd .= "-use_strings ";
|
|---|
| 181 | }
|
|---|
| 182 | $cmd .= "-errlog \"$errlog\" -output $output_type \"$tmp_filename\"";
|
|---|
| 183 | $output_type = `$cmd`;
|
|---|
| 184 |
|
|---|
| 185 | # remove symbolic link to original file
|
|---|
| 186 | &util::rm($tmp_filename);
|
|---|
| 187 |
|
|---|
| 188 | # Check STDERR here
|
|---|
| 189 | chomp $output_type;
|
|---|
| 190 | if ($output_type eq "fail") {
|
|---|
| 191 | print $outhandle "Could not convert $tailname$suffix to $convert_to format\n";
|
|---|
| 192 | print $failhandle "$tailname$suffix: " . ref($self) . " failed to convert to $convert_to\n";
|
|---|
| 193 | $self->{'num_not_processed'} ++;
|
|---|
| 194 | if (-s "$errlog") {
|
|---|
| 195 | open(ERRLOG, "$errlog");
|
|---|
| 196 | while (<ERRLOG>) {
|
|---|
| 197 | print $outhandle "$_";
|
|---|
| 198 | }
|
|---|
| 199 | print $outhandle "\n";
|
|---|
| 200 | close ERRLOG;
|
|---|
| 201 | }
|
|---|
| 202 | &util::rm("$errlog") if (-e "$errlog");
|
|---|
| 203 | return "";
|
|---|
| 204 | }
|
|---|
| 205 |
|
|---|
| 206 | # store the *actual* output type and return the output filename
|
|---|
| 207 | # it's possible we requested conversion to html, but only to text succeeded
|
|---|
| 208 |
|
|---|
| 209 | $self->{'convert_to_ext'} = $output_type;
|
|---|
| 210 | if ($output_type =~ /html/i) {
|
|---|
| 211 | $self->{'converted_to'} = "HTML";
|
|---|
| 212 | } elsif ($output_type =~ /te?xt/i) {
|
|---|
| 213 | $self->{'converted_to'} = "TEXT";
|
|---|
| 214 | }
|
|---|
| 215 | my $output_filename = $tmp_filename;
|
|---|
| 216 |
|
|---|
| 217 | $output_filename =~ s/$suffix$/.$output_type/;
|
|---|
| 218 |
|
|---|
| 219 | return $output_filename;
|
|---|
| 220 | }
|
|---|
| 221 |
|
|---|
| 222 |
|
|---|
| 223 | # Remove collection specific tmp directory and all its contents.
|
|---|
| 224 |
|
|---|
| 225 | sub cleanup_tmp_area {
|
|---|
| 226 | my $self = shift (@_);
|
|---|
| 227 |
|
|---|
| 228 | my $tmp_dirname
|
|---|
| 229 | = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
|---|
| 230 | &util::rm_r($tmp_dirname);
|
|---|
| 231 | &util::mk_dir($tmp_dirname);
|
|---|
| 232 | }
|
|---|
| 233 |
|
|---|
| 234 |
|
|---|
| 235 |
|
|---|
| 236 |
|
|---|
| 237 | # Override BasPlug read
|
|---|
| 238 | # We don't want to get language encoding stuff until after we've converted
|
|---|
| 239 | # our file to either TEXT or HTML.
|
|---|
| 240 | sub read {
|
|---|
| 241 | my $self = shift (@_);
|
|---|
| 242 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs) = @_;
|
|---|
| 243 | # if ($self->is_recursive()) {
|
|---|
| 244 | # die "BasPlug::read function must be implemented in sub-class for recursive plugins\n";
|
|---|
| 245 | # }
|
|---|
| 246 |
|
|---|
| 247 | my $outhandle = $self->{'outhandle'};
|
|---|
| 248 |
|
|---|
| 249 | my $filename = $file;
|
|---|
| 250 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
|---|
| 251 |
|
|---|
| 252 | if ($self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/) {
|
|---|
| 253 | $self->{'num_blocked'} ++;
|
|---|
| 254 | return 0;
|
|---|
| 255 | }
|
|---|
| 256 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
|---|
| 257 | return undef;
|
|---|
| 258 | }
|
|---|
| 259 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
|---|
| 260 |
|
|---|
| 261 | # read in file ($text will be in utf8)
|
|---|
| 262 | my $text = "";
|
|---|
| 263 |
|
|---|
| 264 | my $output_ext = $self->{'convert_to_ext'};
|
|---|
| 265 | my $conv_filename = $self->tmp_area_convert_file($output_ext, $filename);
|
|---|
| 266 |
|
|---|
| 267 | if ("$conv_filename" eq "") {return 0;} # allows continue on errors
|
|---|
| 268 | if (! -e "$conv_filename") {return 0;} # allows continue on errors
|
|---|
| 269 | $self->{'conv_filename'} = $conv_filename;
|
|---|
| 270 |
|
|---|
| 271 | # Do encoding stuff
|
|---|
| 272 | my ($language, $encoding) = $self->textcat_get_language_encoding ($conv_filename);
|
|---|
| 273 |
|
|---|
| 274 | &BasPlug::read_file($self, $conv_filename, $encoding, $language, \$text);
|
|---|
| 275 | if (!length ($text)) {
|
|---|
| 276 | my $plugin_name = ref ($self);
|
|---|
| 277 | print $outhandle "$plugin_name: ERROR: $file contains no text\n" if $self->{'verbosity'};
|
|---|
| 278 | return 0;
|
|---|
| 279 | }
|
|---|
| 280 |
|
|---|
| 281 | # if we converted to HTML, convert é and etc to utf-8.
|
|---|
| 282 | # this should really happen before language_extraction, but that means
|
|---|
| 283 | # modifying a file on disk...
|
|---|
| 284 | $text =~ s/&([^;]+);/&ghtml::getcharequiv($1,0)/ge;
|
|---|
| 285 |
|
|---|
| 286 | # create a new document
|
|---|
| 287 | my $doc_obj = new doc ($conv_filename, "indexed_doc");
|
|---|
| 288 | $doc_obj->set_OIDtype ($processor->{'OIDtype'});
|
|---|
| 289 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
|
|---|
| 290 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
|
|---|
| 291 | my ($filemeta) = $file =~ /([^\\\/]+)$/;
|
|---|
| 292 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Source", &ghtml::dmsafe($filemeta));
|
|---|
| 293 | if ($self->{'cover_image'}) {
|
|---|
| 294 | $self->associate_cover_image($doc_obj, $filename);
|
|---|
| 295 | }
|
|---|
| 296 |
|
|---|
| 297 | # include any metadata passed in from previous plugins
|
|---|
| 298 | # note that this metadata is associated with the top level section
|
|---|
| 299 | $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
|
|---|
| 300 | # do plugin specific processing of doc_obj
|
|---|
| 301 | return undef unless defined ($self->process (\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj));
|
|---|
| 302 | # do any automatic metadata extraction
|
|---|
| 303 | $self->auto_extract_metadata ($doc_obj);
|
|---|
| 304 | # add an OID
|
|---|
| 305 | $doc_obj->set_OID();
|
|---|
| 306 | # process the document
|
|---|
| 307 | $processor->process($doc_obj);
|
|---|
| 308 | $self->cleanup_tmp_area();
|
|---|
| 309 |
|
|---|
| 310 | $self->{'num_processed'} ++;
|
|---|
| 311 |
|
|---|
| 312 | return 1;
|
|---|
| 313 | }
|
|---|
| 314 |
|
|---|
| 315 |
|
|---|
| 316 | # do plugin specific processing of doc_obj for HTML type
|
|---|
| 317 | sub process_type {
|
|---|
| 318 | my $self = shift (@_);
|
|---|
| 319 | my ($doc_ext, $textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
|
|---|
| 320 |
|
|---|
| 321 | my $conv_filename = $self->{'conv_filename'};
|
|---|
| 322 | my $tmp_dirname = File::Basename::dirname($conv_filename);
|
|---|
| 323 | my $tmp_tailname = File::Basename::basename($conv_filename);
|
|---|
| 324 |
|
|---|
| 325 | my $converted_to = $self->{'converted_to'};
|
|---|
| 326 | my $ret_val;
|
|---|
| 327 |
|
|---|
| 328 | if ($converted_to eq "TEXT")
|
|---|
| 329 | {
|
|---|
| 330 |
|
|---|
| 331 | $ret_val = &TEXTPlug::process($self, $textref, $pluginfo,
|
|---|
| 332 | $tmp_dirname, $tmp_tailname,
|
|---|
| 333 | $metadata, $doc_obj);
|
|---|
| 334 | }
|
|---|
| 335 | else
|
|---|
| 336 | {
|
|---|
| 337 | $ret_val = &HTMLPlug::process($self, $textref, $pluginfo,
|
|---|
| 338 | $tmp_dirname, $tmp_tailname,
|
|---|
| 339 | $metadata, $doc_obj);
|
|---|
| 340 | }
|
|---|
| 341 |
|
|---|
| 342 | # associate original file with doc object
|
|---|
| 343 | my $cursection = $doc_obj->get_top_section();
|
|---|
| 344 | my $filename = &util::filename_cat($base_dir, $file);
|
|---|
| 345 | $doc_obj->associate_file($filename, "doc.$doc_ext", undef, $cursection);
|
|---|
| 346 |
|
|---|
| 347 | my $doclink = "<a href=\"_httpcollection_/index/assoc/[archivedir]/doc.$doc_ext\">";
|
|---|
| 348 | $doc_obj->add_utf8_metadata ($cursection, "srclink", $doclink);
|
|---|
| 349 | $doc_obj->add_utf8_metadata ($cursection, "srcicon", "_icon".$doc_ext."_");
|
|---|
| 350 | $doc_obj->add_utf8_metadata ($cursection, "/srclink", "</a>");
|
|---|
| 351 | return $ret_val;
|
|---|
| 352 | }
|
|---|
| 353 |
|
|---|
| 354 | 1;
|
|---|
