source: trunk/gsdl/perllib/plugins/ConvertToPlug.pm@ 9853

Last change on this file since 9853 was 9853, checked in by kjdon, 19 years ago

fixed up maxdocs - now pass an extra parameter to the read function
  • Property svn:keywords set to Author Date Id Revision
File size: 13.2 KB
Line 
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
38package ConvertToPlug;
39
40use BasPlug;
41use HTMLPlug;
42use TEXTPlug;
43use ghtml;
44
45sub BEGIN {
46 @ISA = ('HTMLPlug');
47# @ISA = ('HTMLPlug', 'TEXTPlug');
48# @ISA = ('BasPlug'); #, 'HTMLPlug', 'TEXTPlug');
49}
50
51my $convert_to_list =
52 [ { 'name' => "html",
53 'desc' => "{ConvertToPlug.convert_to.html}" },
54 { 'name' => "text",
55 'desc' => "{ConvertToPlug.convert_to.text}" } ];
56
57my $arguments =
58 [ { 'name' => "convert_to",
59 'desc' => "{ConvertToPlug.convert_to}",
60 'type' => "enum",
61 'reqd' => "yes",
62 'list' => $convert_to_list,
63 'deft' => "html" },
64 { 'name' => "use_strings",
65 'desc' => "{ConvertToPlug.use_strings}",
66 'type' => "flag",
67 'reqd' => "no" } ];
68
69my $options = { 'name' => "ConvertToPlug",
70 'desc' => "{ConvertToPlug.desc}",
71 'abstract' => "yes",
72 'inherits' => "yes",
73 'args' => $arguments };
74
75sub parse_args
76{
77 my $class = shift (@_);
78 my ($args) = @_;
79
80 my $plugin_name = $class;
81 $plugin_name =~ s/\.pm$//;
82
83 my $newargs = {};
84
85 if (!parsargv::parse($args,
86 q^extract_keyphrases^, \$newargs->{'kea'}, #with extra options (undocumented)
87 q^extract_keyphrase_options/.*/^, \$newargs->{'kea_options'}, #no extra options (undocumented)
88 q^convert_to/(html|text)/html^, \$newargs->{'generate_format'},
89 q^use_strings^, \$newargs->{'use_strings'},
90 "allow_extra_options")) {
91
92 print STDERR "\nIncorrect options passed to $plugin_name, ";
93 print STDERR "check your collect.cfg configuration file\n";
94 $self->print_txt_usage(""); # Use default resource bundle
95 die "\n";
96 }
97
98 return ($plugin_name, $newargs);
99}
100
101sub new {
102 my $class = shift (@_);
103 if ($class eq "ConvertToPlug" && defined $_[0]) {$class = shift (@_);}
104 my $self;
105 # parsargv::parse might modify the list, so we do this by creating a copy
106 # of the argument list.
107 my @arglist = @_;
108 my ($plugin_name, $args) = $class->parse_args(\@_);
109
110 if ($class eq "PDFPlug" && $args->{'generate_format'} eq "text" &&
111 $ENV{'GSDLOS'} =~ /^windows$/i) {
112 print STDERR "Windows does not support pdf to text. PDFs will be converted to HTML instead\n";
113 $args->{'generate_format'} = "html";
114 }
115
116 if ($args->{'generate_format'} eq "text")
117 {
118 $self = new TEXTPlug ($class, @arglist);
119 $self->{'convert_to'} = "TEXT";
120 $self->{'convert_to_ext'} = "txt";
121 }
122 else
123 {
124 $self = new HTMLPlug ($class, @arglist);
125 $self->{'convert_to'} = "HTML";
126 $self->{'convert_to_ext'} = "html";
127
128 $self->{'rename_assoc_files'} = 1;
129 $self->{'metadata_fields'} .= ",GENERATOR";
130 }
131
132 # 14-05-02 To allow for proper inheritance of arguments - John Thompson
133 my $option_list = $self->{'option_list'};
134 push( @{$option_list}, $options );
135
136 foreach my $key (keys %$args) {
137 $self->{$key} = $args->{$key};
138 }
139
140 return bless $self, $class;
141}
142
143# we don't need to block anything, so override the one for HTMLPlug
144# files are converted in a temp dir and extra files not passed down the
145# plugin list
146sub get_default_block_exp {
147 my $self = shift (@_);
148
149 return "";
150}
151
152# Go straight to BasPlug and avoid the special case implemented by HTMLPlug
153sub store_block_files {
154 return BasPlug::store_block_files(@_);
155}
156
157# Run conversion utility on the input file.
158#
159# The conversion takes place in a collection specific 'tmp' directory so
160# that we don't accidentally damage the input.
161#
162# The desired output type is indicated by $output_ext. This is usually
163# something like "html" or "word", but can be "best" (or the empty string)
164# to indicate that the conversion utility should do the best it can.
165
166sub tmp_area_convert_file {
167 my $self = shift (@_);
168 my ($output_ext, $input_filename, $textref) = @_;
169
170 my $outhandle = $self->{'outhandle'};
171 my $convert_to = $self->{'convert_to'};
172 my $failhandle = $self->{'failhandle'};
173
174 # softlink to collection tmp dir
175 my $tmp_dirname
176 = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
177 &util::mk_dir($tmp_dirname) if (!-e $tmp_dirname);
178
179 # derive tmp filename from input filename
180 my ($tailname, $dirname, $suffix)
181 = &File::Basename::fileparse($input_filename, "\\.[^\\.]+\$");
182
183 # Remove any white space from filename -- no risk of name collision, and
184 # makes later conversion by utils simpler. Leave spaces in path...
185 $tailname =~ s/\s+//g;
186
187 my $tmp_filename = &util::filename_cat($tmp_dirname, "$tailname$suffix");
188
189 &util::soft_link($input_filename, $tmp_filename);
190
191 my $verbosity = $self->{'verbosity'};
192 if ($verbosity > 0) {
193 print $outhandle "Converting $tailname$suffix to $convert_to format\n";
194 }
195
196 my $errlog = &util::filename_cat($tmp_dirname, "err.log");
197
198 # Execute the conversion command and get the type of the result,
199 # making sure the converter gives us the appropriate output type
200 my $output_type = lc($convert_to);
201 my $cmd = "perl -S gsConvert.pl -verbose $verbosity ";
202 if (defined $self->{'convert_options'}) {
203 $cmd .= $self->{'convert_options'} . " ";
204 }
205 if ($self->{'use_strings'}) {
206 $cmd .= "-use_strings ";
207 }
208 $cmd .= "-errlog \"$errlog\" -output $output_type \"$tmp_filename\"";
209
210 $output_type = `$cmd`;
211
212 # remove symbolic link to original file
213 &util::rm($tmp_filename);
214
215 # Check STDERR here
216 chomp $output_type;
217 if ($output_type eq "fail") {
218 print $outhandle "Could not convert $tailname$suffix to $convert_to format\n";
219 print $failhandle "$tailname$suffix: " . ref($self) . " failed to convert to $convert_to\n";
220 $self->{'num_not_processed'} ++;
221 if (-s "$errlog") {
222 open(ERRLOG, "$errlog");
223 while (<ERRLOG>) {
224 print $outhandle "$_";
225 }
226 print $outhandle "\n";
227 close ERRLOG;
228 }
229 &util::rm("$errlog") if (-e "$errlog");
230 return "";
231 }
232
233 # store the *actual* output type and return the output filename
234 # it's possible we requested conversion to html, but only to text succeeded
235
236 $self->{'convert_to_ext'} = $output_type;
237 if ($output_type =~ /html/i) {
238 $self->{'converted_to'} = "HTML";
239 } elsif ($output_type =~ /te?xt/i) {
240 $self->{'converted_to'} = "TEXT";
241 }
242 my $output_filename = $tmp_filename;
243
244 $output_filename =~ s/$suffix$/.$output_type/;
245
246 return $output_filename;
247}
248
249
250# Remove collection specific tmp directory and all its contents.
251
252sub cleanup_tmp_area {
253 my $self = shift (@_);
254
255 my $tmp_dirname
256 = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
257 &util::rm_r($tmp_dirname);
258 &util::mk_dir($tmp_dirname);
259}
260
261
262
263
264# Override BasPlug read
265# We don't want to get language encoding stuff until after we've converted
266# our file to either TEXT or HTML.
267sub read {
268 my $self = shift (@_);
269 my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
270# if ($self->is_recursive()) {
271# die "BasPlug::read function must be implemented in sub-class for recursive plugins\n";
272# }
273
274 my $outhandle = $self->{'outhandle'};
275
276 my $filename = $file;
277 $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
278
279 if ($self->associate_with($file,$filename,$metadata)) {
280 # a form of smart block
281 $self->{'num_blocked'} ++;
282 return 0; # blocked
283 }
284
285 if ($self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/) {
286 $self->{'num_blocked'} ++;
287 return 0;
288 }
289 if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
290 return undef;
291 }
292 $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
293
294 # read in file ($text will be in utf8)
295 my $text = "";
296
297 my $output_ext = $self->{'convert_to_ext'};
298 my $conv_filename = $self->tmp_area_convert_file($output_ext, $filename);
299
300 if ("$conv_filename" eq "") {return 0;} # allows continue on errors
301 if (! -e "$conv_filename") {return 0;} # allows continue on errors
302 $self->{'conv_filename'} = $conv_filename;
303
304 # Do encoding stuff
305 my ($language, $encoding) = $self->textcat_get_language_encoding ($conv_filename);
306
307 &BasPlug::read_file($self, $conv_filename, $encoding, $language, \$text);
308 if (!length ($text)) {
309 my $plugin_name = ref ($self);
310 print $outhandle "$plugin_name: ERROR: $file contains no text\n" if $self->{'verbosity'};
311 return 0;
312 }
313
314 # if we converted to HTML, convert &eacute; and etc to utf-8.
315 # this should really happen before language_extraction, but that means
316 # modifying a file on disk...
317 $text =~ s/&([^;]+);/&ghtml::getcharequiv($1,0)/ge;
318
319 # create a new document
320 #my $doc_obj = new doc ($conv_filename, "indexed_doc");
321 # now we use the original filename here
322 my $doc_obj = new doc($filename, "indexed_doc");
323 $doc_obj->set_converted_filename($conv_filename);
324 $doc_obj->set_OIDtype ($processor->{'OIDtype'});
325 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
326 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
327 my ($filemeta) = $file =~ /([^\\\/]+)$/;
328 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Source", &ghtml::dmsafe($filemeta));
329 if ($self->{'cover_image'}) {
330 $self->associate_cover_image($doc_obj, $filename);
331 }
332 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
333 $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "FileSize", (-s $filename));
334
335 # include any metadata passed in from previous plugins
336 # note that this metadata is associated with the top level section
337 $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
338 # do plugin specific processing of doc_obj
339 unless (defined ($self->process(\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
340 print STDERR "<ProcessingError n='$file'>\n" if ($gli);
341 return -1;
342 }
343 # do any automatic metadata extraction
344 $self->auto_extract_metadata ($doc_obj);
345 # add an OID
346 $doc_obj->set_OID();
347 # process the document
348 $processor->process($doc_obj);
349 $self->cleanup_tmp_area();
350
351 $self->{'num_processed'} ++;
352
353 return 1;
354}
355
356
357# do plugin specific processing of doc_obj for HTML type
358sub process_type {
359 my $self = shift (@_);
360 my ($doc_ext, $textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
361
362 my $conv_filename = $self->{'conv_filename'};
363 my $tmp_dirname = File::Basename::dirname($conv_filename);
364 my $tmp_tailname = File::Basename::basename($conv_filename);
365
366 my $converted_to = $self->{'converted_to'};
367 my $ret_val;
368
369 if ($converted_to eq "TEXT")
370 {
371
372 $ret_val = &TEXTPlug::process($self, $textref, $pluginfo,
373 $tmp_dirname, $tmp_tailname,
374 $metadata, $doc_obj);
375 }
376 else
377 {
378 $ret_val = &HTMLPlug::process($self, $textref, $pluginfo,
379 $tmp_dirname, $tmp_tailname,
380 $metadata, $doc_obj);
381 }
382
383 # associate original file with doc object
384 my $cursection = $doc_obj->get_top_section();
385 my $filename = &util::filename_cat($base_dir, $file);
386 $doc_obj->associate_file($filename, "doc.$doc_ext", undef, $cursection);
387
388 my $file_type;
389
390 if ($doc_ext eq "doc") {
391 $file_type = "Word";
392 } elsif ($doc_ext eq "xls") {
393 $file_type = "Excel";
394 } elsif ($doc_ext eq "ppt") {
395 $file_type = "PPT";
396 } elsif ($doc_ext eq "pdf") {
397 $file_type = "PDF";
398 } elsif ($doc_ext eq "rtf") {
399 $file_type = "RTF";
400 } elsif ($doc_ext eq "ps") {
401 $file_type = "PS";
402 }
403
404 my $file_format = $file_type || "unknown";
405
406 # We use set instead of add here because we only want one value
407 $doc_obj->set_utf8_metadata_element($cursection, "FileFormat", $file_format);
408
409 my $doclink = "<a href=\"_httpcollection_/index/assoc/[archivedir]/doc.$doc_ext\">";
410 $doc_obj->add_utf8_metadata ($cursection, "srclink", $doclink);
411 $doc_obj->add_utf8_metadata ($cursection, "srcicon", "_icon".$doc_ext."_");
412 $doc_obj->add_utf8_metadata ($cursection, "/srclink", "</a>");
413
414 return $ret_val;
415}
416
4171;
Note: See TracBrowser for help on using the repository browser.