source: main/trunk/greenstone2/perllib/plugins/PagedImagePlugin.pm@ 28355

Last change on this file since 28355 was 28355, checked in by ak19, 8 years ago
  1. Now gsConvert.pl calls the new pptextract.vbs VBScript (which creates .item files and ppt slide.txt files in utf-8) instead of the older VB pptextract.exe executable which created .item and slide.txt files in windows default utf-16 LE. 2. PagedImagePlugin.pm::tidy_item_file now reads in the .item files in utf-8 mode, so that its strings are unicode aware. Substitutions are of unicode code points instead of byte sequences, since the strings in the file are now unicode aware.
  • Property svn:executable set to *
  • Property svn:keywords set to Author Date Id Revision
File size: 29.1 KB
Line 
1###########################################################################
2#
3# PagedImagePlugin.pm -- plugin for sets of images and OCR text that
4# make up a document
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# PagedImagePlugin
28# processes sequences of images, with optional OCR text
29#
30# This plugin takes *.item files, which contain metadata and lists of image
31# files, and produces a document containing sections, one for each page.
32# The files should be named something.item, then you can have more than one
33# book in a directory. You will need to create these files, one for each
34# document/book.
35#
36#There are two formats for the item files: a plain text format, and an xml
37#format. You can use either format, and can have both formats in the same
38#collection if you like. If you use the plain format, you must not start the
39#file off with <PagedDocument>
40
41#### PLAIN FORMAT
42# The format of the xxx.item file is as follows:
43# The first lines contain any metadata for the whole document
44# <metadata-name>metadata-value
45# eg.
46# <Title>Snail farming
47# <Date>19230102
48# Then comes a list of pages, one page per line, each line has the format
49#
50# pagenum:imagefile:textfile:r
51#
52# page num and imagefile are required. pagenum is used for the Title
53# of the section, and in the display is shown as page <pagenum>.
54# imagefile is the image for the page. textfile is an optional text
55# file containing the OCR (or any) text for the page - this gets added
56# as the text for the section. r is optional, and signals that the image
57# should be rotated 180deg. Eg use this if the image has been made upside down.
58# So an example item file looks like:
59# <Title>Snail farming
60# <Date>19960403
61# 1:p1.gif:p1.txt:
62# 2:p2.gif::
63# 3:p3.gif:p3.txt:
64# 3b:p3b.gif:p3b.txt:r
65# The second page has no text, the fourth page is a back page, and
66# should be rotated.
67#
68
69#### XML FORMAT
70# The xml format looks like the following
71#<PagedDocument>
72#<Metadata name="Title">The Title of the entire document</Metadata>
73#<Page pagenum="1" imgfile="xxx.jpg" txtfile="yyy.txt">
74#<Metadata name="Title">The Title of this page</Metadata>
75#</Page>
76#... more pages
77#</PagedDocument>
78#PagedDocument contains a list of Pages, Metadata and PageGroups. Any metadata
79#that is not inside another tag will belong to the document.
80#Each Page has a pagenum (not used at the moment), an imgfile and/or a txtfile.
81#These are both optional - if neither is used, the section will have no content.
82#Pages can also have metadata associated with them.
83#PageGroups can be introduced at any point - they can contain Metadata and Pages and other PageGroups. They are used to introduce hierarchical structure into the document.
84#For example
85#<PagedDocument>
86#<PageGroup>
87#<Page>
88#<Page>
89#</PageGroup>
90#<Page>
91#</PagedDocument>
92#would generate a structure like
93#X
94#--X
95# --X
96# --X
97#--X
98#PageGroup tags can also have imgfile/textfile metadata if you like - this way they get some content themselves.
99
100#Currently the XML structure doesn't work very well with the paged document type, unless you use numerical Titles for each section.
101#There is still a bit of work to do on this format:
102#* enable other text file types, eg html, pdf etc
103#* make the document paging work properly
104#* add pagenum as Title unless a Title is present?
105
106# All the supplemetary image amd text files should be in the same folder as
107# the .item file.
108#
109# To display the images instead of the document text, you can use [srcicon]
110# in the DocumentText format statement.
111# For example,
112#
113# format DocumentText "<center><table width=_pagewidth_><tr><td>[srcicon]</td></tr></table></center>"
114#
115# To have it create thumbnail size images, use the '-create_thumbnail' option.
116# To have it create medium size images for display, use the '-create_screenview'
117# option. As usual, running
118# 'perl -S pluginfo.pl PagedImagePlugin' will list all the options.
119
120# If you want the resulting documents to be presented with a table of
121# contents, use '-documenttype hierarchy', otherwise they will have
122# next and previous arrows, and a goto page X box.
123
124# If you have used -create_screenview, you can also use [screenicon] in the format
125# statement to display the smaller image. Here is an example that switches
126# between the two:
127#
128# format DocumentText "<center><table width=_pagewidth_><tr><td>{If}{_cgiargp_ eq full,<a href='_httpdocument_&d=_cgiargd_&p=small'>Switch to small version.</a>,<a href='_httpdocument_&d=_cgiargd_&p=full'>Switch to fullsize version</a>}</td></tr><tr><td>{If}{_cgiargp_ eq full,<a href='_httpdocument_&d=_cgiargd_&p=small' title='Switch to small version'>[srcicon]</a>,<a href='_httpdocument_&d=_cgiargd_&p=full' title='Switch to fullsize version'>[screenicon]</a>}</td></tr></table></center>"
129#
130# Additional metadata can be added into the .item files, alternatively you can
131# use normal metadata.xml files, with the name of the xxx.item file as the
132# FileName (only for document level metadata).
133
134package PagedImagePlugin;
135
136use Encode;
137use ReadXMLFile;
138use ReadTextFile;
139use ImageConverter;
140use MetadataRead;
141
142use strict;
143no strict 'refs'; # allow filehandles to be variables and viceversa
144
145sub BEGIN {
146 @PagedImagePlugin::ISA = ('MetadataRead', 'ReadXMLFile', 'ReadTextFile', 'ImageConverter');
147}
148
149my $gs2_type_list =
150 [ { 'name' => "auto",
151 'desc' => "{PagedImagePlugin.documenttype.auto2}" },
152 { 'name' => "paged",
153 'desc' => "{PagedImagePlugin.documenttype.paged2}" },
154 { 'name' => "hierarchy",
155 'desc' => "{PagedImagePlugin.documenttype.hierarchy}" }
156 ];
157
158my $gs3_type_list =
159 [ { 'name' => "auto",
160 'desc' => "{PagedImagePlugin.documenttype.auto3}" },
161 { 'name' => "paged",
162 'desc' => "{PagedImagePlugin.documenttype.paged3}" },
163 { 'name' => "hierarchy",
164 'desc' => "{PagedImagePlugin.documenttype.hierarchy}" },
165 { 'name' => "pagedhierarchy",
166 'desc' => "{PagedImagePlugin.documenttype.pagedhierarchy}" }
167 ];
168
169my $arguments =
170 [ { 'name' => "process_exp",
171 'desc' => "{BasePlugin.process_exp}",
172 'type' => "string",
173 'deft' => &get_default_process_exp(),
174 'reqd' => "no" },
175 { 'name' => "title_sub",
176 'desc' => "{HTMLPlugin.title_sub}",
177 'type' => "string",
178 'deft' => "" },
179 { 'name' => "headerpage",
180 'desc' => "{PagedImagePlugin.headerpage}",
181 'type' => "flag",
182 'reqd' => "no" },
183# { 'name' => "documenttype",
184# 'desc' => "{PagedImagePlugin.documenttype}",
185# 'type' => "enum",
186# 'list' => $type_list,
187# 'deft' => "auto",
188# 'reqd' => "no" },
189 {'name' => "processing_tmp_files",
190 'desc' => "{BasePlugin.processing_tmp_files}",
191 'type' => "flag",
192 'hiddengli' => "yes"}
193 ];
194
195my $doc_type_opt = { 'name' => "documenttype",
196 'desc' => "{PagedImagePlugin.documenttype}",
197 'type' => "enum",
198 'deft' => "auto",
199 'reqd' => "no" };
200
201my $options = { 'name' => "PagedImagePlugin",
202 'desc' => "{PagedImagePlugin.desc}",
203 'abstract' => "no",
204 'inherits' => "yes",
205 'args' => $arguments };
206
207sub new {
208 my ($class) = shift (@_);
209 my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
210 push(@$pluginlist, $class);
211
212 push(@{$hashArgOptLists->{"OptList"}},$options);
213
214 my $imc_self = new ImageConverter($pluginlist, $inputargs, $hashArgOptLists);
215
216 # we can use this plugin to check gs3 version
217 if ($imc_self->{'gs_version'} eq "3") {
218 $doc_type_opt->{'list'} = $gs3_type_list;
219 }
220 else {
221 $doc_type_opt->{'list'} = $gs2_type_list;
222 }
223 push(@$arguments,$doc_type_opt);
224 # now we add the args to the list for parsing
225 push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
226
227 my $rtf_self = new ReadTextFile($pluginlist, $inputargs, $hashArgOptLists, 1);
228 my $rxf_self = new ReadXMLFile($pluginlist, $inputargs, $hashArgOptLists);
229
230 my $self = BasePlugin::merge_inheritance($imc_self,$rtf_self,$rxf_self);
231
232 # Update $self used by XML::Parser so it finds callback functions
233 # such as start_document here and not in ReadXMLFile (which is what
234 # $self was when new XML::Parser was done)
235 #
236 # If the $self returned by this constructor is the same as the one
237 # used in ReadXMLFile (e.g. in the GreenstoneXMLPlugin) then this step isn't necessary
238 #
239 # Consider embedding this type of assignment into merge_inheritance
240 # to help catch all cases?
241
242 $rxf_self->{'parser'}->{'PluginObj'} = $self;
243
244 return bless $self, $class;
245}
246
247
248sub init {
249 my $self = shift (@_);
250 my ($verbosity, $outhandle, $failhandle) = @_;
251
252 $self->SUPER::init(@_);
253 $self->ImageConverter::init();
254}
255
256sub begin {
257 my $self = shift (@_);
258 my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
259
260 $self->SUPER::begin(@_);
261 $self->ImageConverter::begin(@_);
262}
263
264sub get_default_process_exp {
265 my $self = shift (@_);
266
267 return q^\.item$^;
268}
269
270sub get_doctype {
271 my $self = shift(@_);
272
273 return "PagedDocument";
274}
275
276
277# want to use BasePlugin's version of this, not ReadXMLFile's
278sub can_process_this_file {
279 my $self = shift(@_);
280 return $self->BasePlugin::can_process_this_file(@_);
281}
282
283# instead of a block exp, now we scan the file and record all text and img files mentioned there for blocking.
284sub store_block_files
285{
286 my $self = shift (@_);
287 my ($filename_full_path, $block_hash) = @_;
288
289 my $xml_version = $self->is_xml_item_file($filename_full_path);
290
291 # do we need to do this?
292 # does BOM interfere just with XML parsing? In that case don't need it here
293 # if we do it here, we are modifying the file before we have worked out if
294 # its new or not, so it will always be reimported.
295 #$self->tidy_item_file($filename_full_path);
296
297 my ($dir, $file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
298 if ($xml_version) {
299
300 # do something
301 $self->scan_xml_for_files_to_block($filename_full_path, $dir, $block_hash);
302 } else {
303
304 $self->scan_item_for_files_to_block($filename_full_path, $dir, $block_hash);
305 }
306
307}
308
309# we want to use BasePlugin's read, not ReadXMLFile's
310sub read
311{
312 my $self = shift (@_);
313
314 $self->BasePlugin::read(@_);
315}
316
317
318
319sub read_into_doc_obj {
320 my $self = shift (@_);
321 my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
322 my $outhandle = $self->{'outhandle'};
323 my $verbosity = $self->{'verbosity'};
324
325 my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
326
327 print $outhandle "PagedImagePlugin processing \"$filename_full_path\"\n"
328 if $verbosity > 1;
329 print STDERR "<Processing n='$file' p='PagedImagePlugin'>\n" if ($gli);
330
331 $self->{'MaxImageWidth'} = 0;
332 $self->{'MaxImageHeight'} = 0;
333
334 # here we need to decide if we have an old text .item file, or a new xml
335 # .item file
336 my $xml_version = $self->is_xml_item_file($filename_full_path);
337
338 $self->tidy_item_file($filename_full_path);
339
340 my $doc_obj;
341 if ($xml_version) {
342 # careful checking needed here!! are we using local xml handlers or super ones
343 $self->ReadXMLFile::read($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli);
344 $doc_obj = $self->{'doc_obj'};
345 } else {
346 my ($dir, $item_file);
347 ($dir, $item_file) = $filename_full_path =~ /^(.*?)([^\/\\]*)$/;
348
349 #process the .item file
350 $doc_obj = $self->process_item($filename_full_path, $dir, $item_file, $processor, $metadata);
351
352 }
353
354 my $section = $doc_obj->get_top_section();
355
356 $doc_obj->add_utf8_metadata($section, "Plugin", "$self->{'plugin_type'}");
357 $doc_obj->add_metadata($section, "FileFormat", "PagedImage");
358
359 # include any metadata passed in from previous plugins
360 # note that this metadata is associated with the top level section
361 $self->add_associated_files($doc_obj, $filename_full_path);
362 $self->extra_metadata ($doc_obj, $section, $metadata);
363 $self->auto_extract_metadata ($doc_obj);
364 $self->plugin_specific_process($base_dir, $file, $doc_obj, $gli);
365 # if we haven't found any Title so far, assign one
366 $self->title_fallback($doc_obj,$section,$filename_no_path);
367
368 $self->add_OID($doc_obj);
369 return (1,$doc_obj);
370}
371# override this for an inheriting plugin to add extra metadata etc
372sub plugin_specific_process {
373 my $self = shift(@_);
374 my ($base_dir, $file, $doc_obj, $gli) = @_;
375
376}
377
378# for now, the test is if the first non-empty line is <PagedDocument>, then its xml
379sub is_xml_item_file {
380 my $self = shift(@_);
381 my ($filename) = @_;
382
383 my $xml_version = 0;
384 open (ITEMFILE, $filename) || die "couldn't open $filename\n";
385
386 my $line = "";
387 my $num = 0;
388
389 $line = <ITEMFILE>;
390 while (defined ($line) && ($line !~ /\w/)) {
391 $line = <ITEMFILE>;
392 }
393
394 if (defined $line) {
395 chomp $line;
396 if ($line =~ /<PagedDocument/) {
397 $xml_version = 1;
398 }
399 }
400
401 close ITEMFILE;
402 return $xml_version;
403}
404
405sub tidy_item_file {
406 my $self = shift(@_);
407 my ($filename) = @_;
408
409 open (ITEMFILE, "<:encoding(UTF-8)", $filename) || die "couldn't open $filename\n";
410 my $backup_filename = "backup.item";
411 open (BACKUP,">$backup_filename")|| die "couldn't write to $backup_filename\n";
412 binmode(BACKUP, ":utf8");
413 my $line = "";
414 $line = <ITEMFILE>;
415 #$line =~ s/^\xEF\xBB\xBF//; # strip BOM in text file read in as a sequence of bytes (not unicode aware strings)
416 $line =~ s/^\x{FEFF}//; # strip BOM in file opened *as UTF-8*. Strings in the file just read in are now unicode-aware,
417 # this means the BOM is now a unicode codepoint instead of a byte sequence
418 # See http://en.wikipedia.org/wiki/Byte_order_mark and http://perldoc.perl.org/5.14.0/perlunicode.html
419 $line =~ s/\x{0B}+//ig; # removing \vt-vertical tabs using the unicode codepoint for \vt
420 $line =~ s/&/&amp;/g;
421 print BACKUP ($line);
422 #Tidy up the item file some metadata title contains \vt-vertical tab
423 while ($line = <ITEMFILE>) {
424 $line =~ s/\x{0B}+//ig; # removing \vt-vertical tabs using the unicode codepoint for \vt
425 $line =~ s/&/&amp;/g;
426 print BACKUP ($line);
427 }
428 close ITEMFILE;
429 close BACKUP;
430 &File::Copy::copy ($backup_filename, $filename);
431 &FileUtils::removeFiles($backup_filename);
432
433}
434
435sub rotate_image {
436 my $self = shift (@_);
437 my ($filename_full_path) = @_;
438
439 my ($this_filetype) = $filename_full_path =~ /\.([^\.]*)$/;
440 my $result = $self->convert($filename_full_path, $this_filetype, "-rotate 180", "ROTATE");
441 my ($new_filename) = ($result =~ /=>(.*\.$this_filetype)/);
442 if (-e "$new_filename") {
443 return $new_filename;
444 }
445 # somethings gone wrong
446 return $filename_full_path;
447
448}
449
450sub process_image {
451 my $self = shift(@_);
452 my ($filename_full_path, $filename_no_path, $doc_obj, $section, $rotate) = @_;
453 # check the filenames
454 return 0 if ($filename_no_path eq "" || !-f $filename_full_path);
455
456 # remember that this image file was one of our source files, but only
457 # if we are not processing a tmp file
458 if (!$self->{'processing_tmp_files'} ) {
459 $doc_obj->associate_source_file($filename_full_path);
460 }
461 # do rotation
462 if ((defined $rotate) && ($rotate eq "r")) {
463 # we get a new temporary file which is rotated
464 $filename_full_path = $self->rotate_image($filename_full_path);
465 }
466
467 # do generate images
468 my $result = 0;
469 if ($self->{'image_conversion_available'} == 1) {
470 # do we need to convert $filename_no_path to utf8/url encoded?
471 # We are already reading in from a file, what encoding is it in???
472 my $url_encoded_full_filename
473 = &unicode::raw_filename_to_url_encoded($filename_full_path);
474 $result = $self->generate_images($filename_full_path, $url_encoded_full_filename, $doc_obj, $section);
475 }
476 #overwrite one set in ImageConverter
477 $doc_obj->set_metadata_element ($section, "FileFormat", "PagedImage");
478 return $result;
479}
480
481
482sub xml_start_tag {
483 my $self = shift(@_);
484 my ($expat, $element) = @_;
485 $self->{'element'} = $element;
486
487 my $doc_obj = $self->{'doc_obj'};
488 if ($element eq "PagedDocument") {
489 $self->{'current_section'} = $doc_obj->get_top_section();
490 } elsif ($element eq "PageGroup" || $element eq "Page") {
491 if ($element eq "PageGroup") {
492 $self->{'has_internal_structure'} = 1;
493 }
494 # create a new section as a child
495 $self->{'current_section'} = $doc_obj->insert_section($doc_obj->get_end_child($self->{'current_section'}));
496 $self->{'num_pages'}++;
497 # assign pagenum as what??
498 my $pagenum = $_{'pagenum'}; #TODO!!
499 if (defined $pagenum) {
500 $doc_obj->set_utf8_metadata_element($self->{'current_section'}, 'PageNum', $pagenum);
501 }
502 my ($imgfile) = $_{'imgfile'};
503 if (defined $imgfile) {
504 # *****
505 # What about support for rotate image (e.g. old ':r' notation)?
506 $self->process_image($self->{'xml_file_dir'}.$imgfile, $imgfile, $doc_obj, $self->{'current_section'});
507 }
508 my ($txtfile) = $_{'txtfile'};
509 if (defined($txtfile)&& $txtfile ne "") {
510 $self->process_text ($self->{'xml_file_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
511 } else {
512 $self->add_dummy_text($doc_obj, $self->{'current_section'});
513 }
514 } elsif ($element eq "Metadata") {
515 $self->{'metadata_name'} = $_{'name'};
516 }
517}
518
519sub xml_end_tag {
520 my $self = shift(@_);
521 my ($expat, $element) = @_;
522
523 my $doc_obj = $self->{'doc_obj'};
524 if ($element eq "Page" || $element eq "PageGroup") {
525 # if Title hasn't been assigned, set PageNum as Title
526 if (!defined $doc_obj->get_metadata_element ($self->{'current_section'}, "Title") && defined $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" )) {
527 $doc_obj->add_utf8_metadata ($self->{'current_section'}, "Title", $doc_obj->get_metadata_element ($self->{'current_section'}, "PageNum" ));
528 }
529 # move the current section back to the parent
530 $self->{'current_section'} = $doc_obj->get_parent_section($self->{'current_section'});
531 } elsif ($element eq "Metadata") {
532
533 # text read in by XML::Parser is in Perl's binary byte value
534 # form ... need to explicitly make it UTF-8
535 my $meta_name = decode("utf-8",$self->{'metadata_name'});
536 my $metadata_value = decode("utf-8",$self->{'metadata_value'});
537
538 if ($meta_name =~ /\./) {
539 $meta_name = "ex.$meta_name";
540 }
541
542 $doc_obj->add_utf8_metadata ($self->{'current_section'}, $meta_name, $metadata_value);
543 $self->{'metadata_name'} = "";
544 $self->{'metadata_value'} = "";
545
546 }
547 # otherwise we ignore the end tag
548}
549
550
551sub xml_text {
552 my $self = shift(@_);
553 my ($expat) = @_;
554
555 if ($self->{'element'} eq "Metadata" && $self->{'metadata_name'}) {
556 $self->{'metadata_value'} .= $_;
557 }
558}
559
560sub xml_doctype {
561}
562
563sub open_document {
564 my $self = shift(@_);
565
566 # create a new document
567 $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc", $self->{'file_rename_method'});
568 # TODO is file filenmae_no_path??
569 $self->set_initial_doc_fields($self->{'doc_obj'}, $self->{'filename'}, $self->{'processor'}, $self->{'metadata'});
570
571 my ($dir, $file) = $self->{'filename'} =~ /^(.*?)([^\/\\]*)$/;
572 $self->{'xml_file_dir'} = $dir;
573 $self->{'num_pages'} = 0;
574 $self->{'has_internal_structure'} = 0;
575
576}
577
578sub close_document {
579 my $self = shift(@_);
580 my $doc_obj = $self->{'doc_obj'};
581
582 my $topsection = $doc_obj->get_top_section();
583
584 # add numpages metadata
585 $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', $self->{'num_pages'});
586
587 # set the document type
588 my $final_doc_type = "";
589 if ($self->{'documenttype'} eq "auto") {
590 if ($self->{'has_internal_structure'}) {
591 if ($self->{'gs_version'} eq "3") {
592 $final_doc_type = "pagedhierarchy";
593 }
594 else {
595 $final_doc_type = "hierarchy";
596 }
597 } else {
598 $final_doc_type = "paged";
599 }
600 } else {
601 # set to what doc type option was set to
602 $final_doc_type = $self->{'documenttype'};
603 }
604 $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $final_doc_type);
605 ### capiatalisation????
606# if ($self->{'documenttype'} eq 'paged') {
607 # set the gsdlthistype metadata to Paged - this ensures this document will
608 # be treated as a Paged doc, even if Titles are not numeric
609# $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Paged");
610# } else {
611# $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Hierarchy");
612# }
613
614 $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
615 $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
616 $self->{'MaxImageWidth'} = undef;
617 $self->{'MaxImageHeight'} = undef;
618
619}
620
621
622sub set_initial_doc_fields {
623 my $self = shift(@_);
624 my ($doc_obj, $filename_full_path, $processor, $metadata) = @_;
625
626 my $topsection = $doc_obj->get_top_section();
627
628 my $plugin_filename_encoding = $self->{'filename_encoding'};
629 my $filename_encoding = $self->deduce_filename_encoding($filename_full_path,$metadata,$plugin_filename_encoding);
630 $self->set_Source_metadata($doc_obj, $filename_full_path, $filename_encoding);
631
632 # if we want a header page, we need to add some text into the top section, otherwise this section will become invisible
633 if ($self->{'headerpage'}) {
634 $self->add_dummy_text($doc_obj, $topsection);
635 }
636}
637
638sub scan_xml_for_files_to_block
639{
640 my $self = shift (@_);
641 my ($filename_full_path, $dir, $block_hash) = @_;
642
643 open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
644 my $line = "";
645 while (defined ($line = <ITEMFILE>)) {
646 next unless $line =~ /\w/;
647
648 if ($line =~ /imgfile=\"([^\"]+)\"/) {
649 &util::block_filename($block_hash,&FileUtils::filenameConcatenate($dir,$1));
650 }
651 if ($line =~ /txtfile=\"([^\"]+)\"/) {
652 &util::block_filename($block_hash,&FileUtils::filenameConcatenate($dir,$1));
653 }
654 }
655 close ITEMFILE;
656
657}
658
659sub scan_item_for_files_to_block
660{
661 my $self = shift (@_);
662 my ($filename_full_path, $dir, $block_hash) = @_;
663
664
665 open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
666 my $line = "";
667 while (defined ($line = <ITEMFILE>)) {
668 next unless $line =~ /\w/;
669 chomp $line;
670 next if $line =~ /^#/; # ignore comment lines
671 next if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/); # ignore metadata lines
672 # line should be like page:imagefilename:textfilename:r
673 $line =~ s/^\s+//; #remove space at the front
674 $line =~ s/\s+$//; #remove space at the end
675 my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
676
677 # find the image file if there is one
678 if (defined $imgname && $imgname ne "") {
679 &util::block_filename($block_hash, &FileUtils::filenameConcatenate( $dir,$imgname));
680 }
681 # find the text file if there is one
682 if (defined $txtname && $txtname ne "") {
683 &util::block_filename($block_hash, &FileUtils::filenameConcatenate($dir,$txtname));
684 }
685 }
686 close ITEMFILE;
687
688}
689
690sub process_item {
691 my $self = shift (@_);
692 my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata) = @_;
693
694 my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
695 $self->set_initial_doc_fields($doc_obj, $filename_full_path, $processor, $metadata);
696 my $topsection = $doc_obj->get_top_section();
697 # simple item files are always paged unless user specified
698 if ($self->{'documenttype'} eq "auto") {
699 $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "paged");
700 } else {
701 $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $self->{'documenttype'});
702 }
703 open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path\n";
704 my $line = "";
705 my $num = 0;
706 while (defined ($line = <ITEMFILE>)) {
707
708 # Since process_item is called not on an XML item file, but a text item file
709 # don't decode into UTF8 the text that was read in, since it's already UTF-8
710 #$line = decode("utf-8",$line);
711
712 next unless $line =~ /\w/;
713 chomp $line;
714 next if $line =~ /^#/; # ignore comment lines
715 if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/) {
716 my $meta_name = $1;
717 my $meta_value = $2;
718 if ($meta_name =~ /\./) {
719 $meta_name = "ex.$meta_name";
720 }
721 $doc_obj->set_utf8_metadata_element ($topsection, $meta_name, $meta_value);
722 #$meta->{$1} = $2;
723 } else {
724 $num++;
725 # line should be like page:imagefilename:textfilename:r - the r is optional -> means rotate the image 180 deg
726 $line =~ s/^\s+//; #remove space at the front
727 $line =~ s/\s+$//; #remove space at the end
728 my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;
729
730 # create a new section for each image file
731 my $cursection = $doc_obj->insert_section($doc_obj->get_end_child($topsection));
732 # the page number becomes the Title
733 $doc_obj->set_utf8_metadata_element($cursection, 'Title', $pagenum);
734
735 # process the image for this page if there is one
736 if (defined $imgname && $imgname ne "") {
737 my $result1 = $self->process_image($dir.$imgname, $imgname, $doc_obj, $cursection, $rotate);
738 if (!defined $result1)
739 {
740 print "PagedImagePlugin: couldn't process image \"$dir$imgname\" for item \"$filename_full_path\"\n";
741 }
742 }
743 # process the text file if one is there
744 if (defined $txtname && $txtname ne "") {
745 my $result2 = $self->process_text ($dir.$txtname, $txtname, $doc_obj, $cursection);
746
747 if (!defined $result2) {
748 print "PagedImagePlugin: couldn't process text file \"$dir.$txtname\" for item \"$filename_full_path\"\n";
749 $self->add_dummy_text($doc_obj, $cursection);
750 }
751 } else {
752 # otherwise add in some dummy text
753 $self->add_dummy_text($doc_obj, $cursection);
754 }
755 }
756 }
757
758 close ITEMFILE;
759
760 # add numpages metadata
761 $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num");
762
763 $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
764 $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
765 $self->{'MaxImageWidth'} = undef;
766 $self->{'MaxImageHeight'} = undef;
767
768
769 return $doc_obj;
770}
771
772sub process_text {
773 my $self = shift (@_);
774 my ($filename_full_path, $file, $doc_obj, $cursection) = @_;
775
776 # check that the text file exists!!
777 if (!-f $filename_full_path) {
778 print "PagedImagePlugin: ERROR: File $filename_full_path does not exist, skipping\n";
779 return 0;
780 }
781
782 # remember that this text file was one of our source files, but only
783 # if we are not processing a tmp file
784 if (!$self->{'processing_tmp_files'} ) {
785 $doc_obj->associate_source_file($filename_full_path);
786 }
787 # Do encoding stuff
788 my ($language, $encoding) = $self->textcat_get_language_encoding ($filename_full_path);
789
790 my $text="";
791 &ReadTextFile::read_file($self, $filename_full_path, $encoding, $language, \$text); # already decoded as utf8
792 if (!length ($text)) {
793 # It's a bit unusual but not out of the question to have no text, so just give a warning
794 print "PagedImagePlugin: WARNING: $filename_full_path contains no text\n";
795 }
796
797 # we need to escape the escape character, or else mg will convert into
798 # eg literal newlines, instead of leaving the text as '\n'
799 $text =~ s/\\/\\\\/g; # macro language
800 $text =~ s/_/\\_/g; # macro language
801
802
803 if ($text =~ m/<html.*?>\s*<head.*?>.*<\/head>\s*<body.*?>(.*)<\/body>\s*<\/html>\s*$/is) {
804 # looks like HTML input
805 # no need to escape < and > or put in <pre> tags
806
807 $text = $1;
808
809 # add text to document object
810 $doc_obj->add_utf8_text($cursection, "$text");
811 }
812 else {
813 $text =~ s/</&lt;/g;
814 $text =~ s/>/&gt;/g;
815
816 # insert preformat tags and add text to document object
817 $doc_obj->add_utf8_text($cursection, "<pre>\n$text\n</pre>");
818 }
819
820
821 return 1;
822}
823
824
825sub clean_up_after_doc_obj_processing {
826 my $self = shift(@_);
827
828 $self->ImageConverter::clean_up_temporary_files();
829}
830
8311;
Note: See TracBrowser for help on using the repository browser.