source: main/trunk/greenstone2/perllib/plugins/ConvertBinaryFile.pm@ 22597

Last change on this file since 22597 was 22597, checked in by kjdon, 10 years ago

code tidy up. rearranged how convertbinaryfile plugins set up their secondary plugins - now only set up the options for the one they are using. all subclass specific code moved out of convertbinaryfile.new into the appropriate plugin file.

  • Property svn:keywords set to Author Date Id Revision
File size: 17.5 KB
Line 
1###########################################################################
2#
3# ConvertBinaryFile.pm -- plugin that facilitates conversion of binary files
4# through gsConvert.pl
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# This plugin is inherited by such plugins as WordPlugin, PowerPointPlugin,
29# PostScriptPlugin,
30# RTFPlugin and PDFPlugin. It facilitates the conversion of these document types
31# to either HTML, Text or a series of images. It works by dynamically loading
32# an appropriate secondary plugin (HTMLPlug, StructuredHTMLPlug,
33# PagedImagePlugin or TextPlugin) based on the plugin argument 'convert_to'.
34
35package ConvertBinaryFile;
36
37use AutoExtractMetadata;
38use ghtml;
39use HTMLPlugin;
40use TextPlugin;
41use PagedImagePlugin;
42
43use strict;
44no strict 'refs'; # allow filehandles to be variables and viceversa
45no strict 'subs';
46
47sub BEGIN {
48 @ConvertBinaryFile::ISA = ('AutoExtractMetadata');
49}
50
51my $convert_to_list =
52 [ { 'name' => "auto",
53 'desc' => "{ConvertBinaryFile.convert_to.auto}" },
54 { 'name' => "html",
55 'desc' => "{ConvertBinaryFile.convert_to.html}" },
56 { 'name' => "text",
57 'desc' => "{ConvertBinaryFile.convert_to.text}" }
58 ];
59
60my $arguments =
61 [ { 'name' => "convert_to",
62 'desc' => "{ConvertBinaryFile.convert_to}",
63 'type' => "enum",
64 'reqd' => "yes",
65 'list' => $convert_to_list,
66 'deft' => "auto" },
67 { 'name' => "keep_original_filename",
68 'desc' => "{ConvertBinaryFile.keep_original_filename}",
69 'type' => "flag" },
70 { 'name' => "title_sub",
71 'desc' => "{HTMLPlugin.title_sub}",
72 'type' => "string",
73 #'type' => "regexp",
74 'deft' => "" },
75 { 'name' => "apply_fribidi",
76 'desc' => "{ConvertBinaryFile.apply_fribidi}",
77 'type' => "flag",
78 'reqd' => "no" },
79 { 'name' => "use_strings",
80 'desc' => "{ConvertBinaryFile.use_strings}",
81 'type' => "flag",
82 'reqd' => "no" },
83 ];
84
85my $options = { 'name' => "ConvertBinaryFile",
86 'desc' => "{ConvertBinaryFile.desc}",
87 'abstract' => "yes",
88 'inherits' => "yes",
89 'args' => $arguments };
90
91
92sub load_secondary_plugins
93{
94 my $self = shift (@_);
95 my ($class,$input_args,$hashArgOptLists) = @_;
96
97 my @convert_to_list = split(",",$self->{'convert_to_plugin'});
98 my $secondary_plugins = {};
99 # find the plugin
100
101 foreach my $convert_to (@convert_to_list) {
102 # load in "convert_to" plugin package
103 my $plugin_class = $convert_to;
104 my $plugin_package = $plugin_class.".pm";
105
106 my $colplugname = undef;
107 if (defined $ENV{'GSDLCOLLECTDIR'}) {
108 $colplugname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},
109 "perllib","plugins",
110 $plugin_package);
111 }
112
113 my $mainplugname = &util::filename_cat($ENV{'GSDLHOME'},
114 "perllib","plugins",
115 $plugin_package);
116
117 if ((defined $colplugname) && (-e $colplugname)) { require $colplugname;}
118 elsif (-e $mainplugname) { require $mainplugname; }
119 else {
120 &gsprintf(STDERR, "{plugin.could_not_find_plugin}\n",
121 $plugin_class);
122 die "\n";
123 }
124
125 # call its constructor with extra options that we've worked out!
126 my $arglist = $input_args->{$plugin_class};
127
128 my ($secondary_plugin);
129 eval("\$secondary_plugin = new $plugin_class([],\$arglist)");
130 die "$@" if $@;
131 $secondary_plugins->{$plugin_class} = $secondary_plugin;
132 }
133 $self->{'secondary_plugins'} = $secondary_plugins;
134}
135
136sub new {
137 my ($class) = shift (@_);
138 my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
139 push(@$pluginlist, $class);
140 my $classPluginName = (defined $pluginlist->[0]) ? $pluginlist->[0] : $class;
141 push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
142 push(@{$hashArgOptLists->{"OptList"}},$options);
143
144 my $self = new AutoExtractMetadata($pluginlist, $inputargs, $hashArgOptLists);
145
146 return bless $self, $class;
147}
148
149# should be called by subclasses after checking and setting
150# $self->{'convert_to'}
151sub set_standard_convert_settings {
152 my $self =shift (@_);
153
154 my $convert_to = $self->{'convert_to'};
155 if ($convert_to eq "auto") {
156 $convert_to = "html";
157 $self->{'convert_to'} = "html";
158 }
159
160 if ($convert_to eq "html") {
161 $self->{'convert_to_plugin'} = "HTMLPlugin";
162 $self->{'convert_to_ext'} = "html";
163 } elsif ($convert_to eq "text") {
164 $self->{'convert_to_plugin'} = "TextPlugin";
165 $self->{'convert_to_ext'} = "txt";
166 } elsif ($convert_to eq "structuredhtml") {
167 $self->{'convert_to_plugin'} = "StructuredHTMLPlugin";
168 $self->{'convert_to_ext'} = "html";
169 } elsif ($convert_to =~ /^pagedimg/) {
170 $self->{'convert_to_plugin'} = "PagedImagePlugin";
171 my ($convert_to_ext) = $convert_to =~ /pagedimg\_(jpg|gif|png)/i;
172 $convert_to_ext = 'jpg' unless defined $convert_to_ext;
173 $self->{'convert_to_ext'} = $convert_to_ext;
174 }
175
176}
177sub init {
178 my $self = shift (@_);
179 my ($verbosity, $outhandle, $failhandle) = @_;
180
181 $self->SUPER::init($verbosity,$outhandle,$failhandle);
182
183 my $secondary_plugins = $self->{'secondary_plugins'};
184
185 foreach my $plug_name (keys %$secondary_plugins) {
186 my $plugin = $secondary_plugins->{$plug_name};
187 $plugin->init($verbosity,$outhandle,$failhandle);
188 }
189}
190
191sub deinit {
192 # called only once, after all plugin passes have been done
193
194 my ($self) = @_;
195
196 my $secondary_plugins = $self->{'secondary_plugins'};
197
198 foreach my $plug_name (keys %$secondary_plugins) {
199 my $plugin = $secondary_plugins->{$plug_name};
200 $plugin->deinit();
201 }
202}
203
204sub convert_post_process
205{
206 # by default do no post processing
207 return;
208}
209
210
211# Run conversion utility on the input file.
212#
213# The conversion takes place in a collection specific 'tmp' directory so
214# that we don't accidentally damage the input.
215#
216# The desired output type is indicated by $output_ext. This is usually
217# something like "html" or "word", but can be "best" (or the empty string)
218# to indicate that the conversion utility should do the best it can.
219sub tmp_area_convert_file {
220 my $self = shift (@_);
221 my ($output_ext, $input_filename, $textref) = @_;
222
223 my $outhandle = $self->{'outhandle'};
224 my $convert_to = $self->{'convert_to'};
225 my $failhandle = $self->{'failhandle'};
226 my $convert_to_ext = $self->{'convert_to_ext'};
227
228 # derive tmp filename from input filename
229 my ($tailname, $dirname, $suffix)
230 = &File::Basename::fileparse($input_filename, "\\.[^\\.]+\$");
231
232 # softlink to collection tmp dir
233 my $tmp_dirname = $dirname;
234 if(defined $ENV{'GSDLCOLLECTDIR'}) {
235 $tmp_dirname = $ENV{'GSDLCOLLECTDIR'};
236 } elsif(defined $ENV{'GSDLHOME'}) {
237 $tmp_dirname = $ENV{'GSDLHOME'};
238 }
239
240 $tmp_dirname = &util::filename_cat($tmp_dirname, "tmp");
241 &util::mk_dir($tmp_dirname) if (!-e $tmp_dirname);
242
243 # add the timestamp into the path otherwise we can run into problems
244 # if documents have the same name
245 my $timestamp = time;
246 my $time_tmp_dirname = &util::filename_cat($tmp_dirname, $timestamp);
247 $tmp_dirname = $time_tmp_dirname;
248 my $i = 1;
249 while (-e $tmp_dirname) {
250 $tmp_dirname = "$time_tmp_dirname$i";
251 $i++;
252 }
253 &util::mk_dir($tmp_dirname);
254
255 # The following is not necessary and will cause problems with
256 # replacing_srcdoc_with_html in the GSDLremote case:
257 # Remove any white space from filename -- no risk of name collision, and
258 # makes later conversion by utils simpler. Leave spaces in path...
259 # tidy up the filename with space, dot, hyphen between
260 #$tailname =~ s/\s+//g;
261 #$tailname =~ s/\.+//g;
262 #$tailname =~ s/\-+//g;
263
264 # convert to utf-8 otherwise we have problems with the doc.xml file later on
265# print STDERR "**** filename $tailname$suffix is already UTF8\n" if &unicode::check_is_utf8($tailname);
266 $tailname = $self->SUPER::filepath_to_utf8($tailname) unless &unicode::check_is_utf8($tailname);
267
268 # URLEncode this since htmls with images where the html filename is utf8 don't seem
269 # to work on Windows (IE or Firefox), as browsers are looking for filesystem-encoded
270 # files on the filesystem.
271 $tailname = &util::rename_file($tailname, $self->{'file_rename_method'}, "without_suffix");
272
273 $suffix = lc($suffix);
274 my $tmp_filename = &util::filename_cat($tmp_dirname, "$tailname$suffix");
275
276 # If gsdl is remote, we're given relative path to input file, of the form import/tailname.suffix
277 # But we can't softlink to relative paths. Therefore, we need to ensure that
278 # the input_filename is the absolute path, see http://perldoc.perl.org/File/Spec.html
279 my $ensure_path_absolute = 1; # true
280 &util::soft_link($input_filename, $tmp_filename, $ensure_path_absolute);
281 my $verbosity = $self->{'verbosity'};
282 if ($verbosity > 0) {
283 print $outhandle "Converting $tailname$suffix to $convert_to format\n";
284 }
285
286 my $errlog = &util::filename_cat($tmp_dirname, "err.log");
287
288 # Execute the conversion command and get the type of the result,
289 # making sure the converter gives us the appropriate output type
290 my $output_type=$self->{'convert_to'};
291# if ($convert_to =~ m/PagedImage/i) {
292# $output_type = lc($convert_to)."_".lc($convert_to_ext);
293# } else {
294# $output_type = lc($convert_to);
295# }
296
297 my $cmd = "perl -S gsConvert.pl -verbose $verbosity ";
298 if (defined $self->{'convert_options'}) {
299 $cmd .= $self->{'convert_options'} . " ";
300 }
301 if ($self->{'use_strings'}) {
302 $cmd .= "-use_strings ";
303 }
304 $cmd .= "-errlog \"$errlog\" -output $output_type \"$tmp_filename\"";
305 print STDERR "calling cmd $cmd\n";
306 $output_type = `$cmd`;
307
308 # remove symbolic link to original file
309 &util::rm($tmp_filename);
310
311 # Check STDERR here
312 chomp $output_type;
313 if ($output_type eq "fail") {
314 print $outhandle "Could not convert $tailname$suffix to $convert_to format\n";
315 print $failhandle "$tailname$suffix: " . ref($self) . " failed to convert to $convert_to\n";
316 # The following meant that if a conversion failed, the document would be counted twice - do we need it for anything?
317 #$self->{'num_not_processed'} ++;
318 if (-s "$errlog") {
319 open(ERRLOG, "$errlog");
320 while (<ERRLOG>) {
321 print $outhandle "$_";
322 }
323 print $outhandle "\n";
324 close ERRLOG;
325 }
326 &util::rm("$errlog") if (-e "$errlog");
327 return "";
328 }
329
330 # store the *actual* output type and return the output filename
331 # it's possible we requested conversion to html, but only to text succeeded
332 #$self->{'convert_to_ext'} = $output_type;
333 if ($output_type =~ /html/i) {
334 $self->{'converted_to'} = "HTML";
335 } elsif ($output_type =~ /te?xt/i) {
336 $self->{'converted_to'} = "Text";
337 } elsif ($output_type =~ /item/i){
338 $self->{'converted_to'} = "PagedImage";
339 }
340
341 my $output_filename = $tmp_filename;
342 if ($output_type =~ /item/i) {
343 # running under windows
344 if ($ENV{'GSDLOS'} =~ /^windows$/i) {
345 $output_filename = $tmp_dirname . "\\$tailname\\" . $tailname . ".$output_type";
346 } else {
347 $output_filename = $tmp_dirname . "\/$tailname\/" . $tailname . ".$output_type";
348 }
349 } else {
350 $output_filename =~ s/$suffix$/.$output_type/;
351 }
352
353 return $output_filename;
354}
355
356
357# Override BasPlug read_into_doc_obj - we need to call secondary plugin stuff
358sub read_into_doc_obj {
359 my $self = shift (@_);
360 my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
361
362 my $outhandle = $self->{'outhandle'};
363
364 my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
365
366 my $output_ext = $self->{'convert_to_ext'};
367 my $conv_filename = "";
368 $conv_filename = $self->tmp_area_convert_file($output_ext, $filename_full_path);
369
370 if ("$conv_filename" eq "") {return -1;} # had an error, will be passed down pipeline
371 if (! -e "$conv_filename") {return -1;}
372 $self->{'conv_filename'} = $conv_filename;
373 $self->convert_post_process($conv_filename);
374
375 # Run the "fribidi" (http://fribidi.org) Unicode Bidirectional Algorithm program over the converted file
376 # Added for fixing up Persian PDFs after being processed by pdftohtml, but may be useful in other cases too
377 if ($self->{'apply_fribidi'} && $self->{'converted_to'} =~ /(HTML|Text)/) {
378 my $fribidi_command = "fribidi \"$conv_filename\" >\"${conv_filename}.tmp\"";
379 if (system($fribidi_command) != 0) {
380 print STDERR "ERROR: Cannot run fribidi on \"$conv_filename\".\n";
381 }
382 else {
383 &util::mv("${conv_filename}.tmp", $conv_filename);
384 }
385 }
386
387 my $secondary_plugins = $self->{'secondary_plugins'};
388 my $num_secondary_plugins = scalar(keys %$secondary_plugins);
389
390 if ($num_secondary_plugins == 0) {
391 print $outhandle "Warning: No secondary plugin to use in conversion. Skipping $file\n";
392 return 0; # effectively block it
393 }
394
395 my @plugin_names = keys %$secondary_plugins;
396 my $plugin_name = shift @plugin_names;
397
398 if ($num_secondary_plugins > 1) {
399 print $outhandle "Warning: Multiple secondary plugins not supported yet! Choosing $plugin_name\n.";
400 }
401
402 my $secondary_plugin = $secondary_plugins->{$plugin_name};
403
404 # note: metadata is not carried on to the next level
405## **** I just replaced $metadata with {} in following
406 my ($rv,$doc_obj)
407 = $secondary_plugin->read_into_doc_obj ($pluginfo,"", $conv_filename, $block_hash, {}, $processor, $maxdocs, $total_count, $gli);
408
409 if ((!defined $rv) || ($rv<1)) {
410 # wasn't processed
411 return $rv;
412 }
413
414 # Override previous gsdlsourcefilename set by secondary plugin
415 my $collect_file = &util::filename_within_collection($filename_full_path);
416 my $collect_conv_file = &util::filename_within_collection($conv_filename);
417 $doc_obj->set_source_filename ($collect_file, $self->{'file_rename_method'});
418 ## set_source_filename does not set the doc_obj source_path which is used in archives dbs for incremental
419 # build. so set it manually.
420 $doc_obj->{'source_path'} = $filename_full_path;
421 $doc_obj->set_converted_filename($collect_conv_file);
422
423 $self->set_Source_metadata($doc_obj, $filename_no_path);
424
425 $doc_obj->set_utf8_metadata_element($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
426 $doc_obj->set_utf8_metadata_element($doc_obj->get_top_section(), "FileSize", (-s $filename_full_path));
427
428 # ****
429 my ($tailname, $dirname, $suffix)
430 = &File::Basename::fileparse($filename_full_path, "\\.[^\\.]+\$");
431$doc_obj->set_utf8_metadata_element($doc_obj->get_top_section(), "FilenameRoot", $tailname);
432
433 # do plugin specific processing of doc_obj
434 unless (defined ($self->process($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
435 print STDERR "<ProcessingError n='$file'>\n" if ($gli);
436 return -1;
437 }
438
439 my $topsection = $doc_obj->get_top_section();
440 $self->add_associated_files($doc_obj, $filename_full_path);
441
442 # extra_metadata is already called by sec plugin in process??
443 $self->extra_metadata($doc_obj, $topsection, $metadata); # do we need this here??
444 # do any automatic metadata extraction
445 $self->auto_extract_metadata ($doc_obj);
446
447 # have we found a Title??
448 $self->title_fallback($doc_obj,$topsection,$filename_no_path);
449
450 $self->add_OID($doc_obj);
451
452 return (1, $doc_obj);
453
454}
455
456sub process {
457 my $self = shift (@_);
458 my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
459
460 return $self->process_type($base_dir, $file, $doc_obj);
461}
462
463# do plugin specific processing of doc_obj for doc_ext type
464sub process_type {
465 my $self = shift (@_);
466 my ($base_dir, $file, $doc_obj) = @_;
467
468 # need to check that not empty
469 my $doc_ext = $self->{'filename_extension'};
470 my $file_type = "unknown";
471 $file_type = $self->{'file_type'} if defined $self->{'file_type'};
472
473 # associate original file with doc object
474 my $cursection = $doc_obj->get_top_section();
475 my $filename = &util::filename_cat($base_dir, $file);
476 my $assocfilename = "doc.$doc_ext";
477 if ($self->{'keep_original_filename'} == 1) {
478 # this should be the same filename that was used for the Source and SourceFile metadata,
479 # as we will use [SourceFile] in the srclink
480 $assocfilename = $doc_obj->get_assocfile_from_sourcefile();
481 }
482 $doc_obj->associate_file($filename, $assocfilename, undef, $cursection);
483
484 # We use set instead of add here because we only want one value
485 $doc_obj->set_utf8_metadata_element($cursection, "FileFormat", $file_type);
486 my $srclink_filename = "doc.$doc_ext";
487 #my $doclink = "<a href=\"_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/doc.$doc_ext\">";
488 if ($self->{'keep_original_filename'} == 1) {
489 $srclink_filename = "[SourceFile]";
490 #$doclink = "<a href=\"_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/[SourceFile]\">";
491 }
492 #$doc_obj->add_utf8_metadata ($cursection, "srclink", $doclink);
493 $doc_obj->add_utf8_metadata ($cursection, "srcicon", "_icon".$doc_ext."_");
494 #$doc_obj->add_utf8_metadata ($cursection, "/srclink", "</a>");
495 $doc_obj->add_utf8_metadata ($cursection, "srclink_file", $srclink_filename);
496 return 1;
497}
498
4991;
500
501
502
503
504
505
506
Note: See TracBrowser for help on using the repository browser.