source: gsdl/trunk/perllib/basebuildproc.pm@ 18469

Last change on this file since 18469 was 18469, checked in by davidb, 14 years ago

Support for reindexing a document added

  • Property svn:keywords set to Author Date Id Revision
File size: 21.8 KB
Line 
1###########################################################################
2#
3# basebuildproc.pm --
4# A component of the Greenstone digital library software
5# from the New Zealand Digital Library Project at the
6# University of Waikato, New Zealand.
7#
8# Copyright (C) 1999 New Zealand Digital Library Project
9#
10# This program is free software; you can redistribute it and/or modify
11# it under the terms of the GNU General Public License as published by
12# the Free Software Foundation; either version 2 of the License, or
13# (at your option) any later version.
14#
15# This program is distributed in the hope that it will be useful,
16# but WITHOUT ANY WARRANTY; without even the implied warranty of
17# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18# GNU General Public License for more details.
19#
20# You should have received a copy of the GNU General Public License
21# along with this program; if not, write to the Free Software
22# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
23#
24###########################################################################
25
26# This document processor outputs a document for indexing (should be
27# implemented by subclass) and storing in the database
28
29package basebuildproc;
30
31eval {require bytes};
32
33use classify;
34use dbutil;
35use doc;
36use docproc;
37use strict;
38no strict 'subs';
39no strict 'refs';
40use util;
41
42BEGIN {
43 @basebuildproc::ISA = ('docproc');
44}
45
46sub new()
47 {
48 my ($class, $collection, $source_dir, $build_dir, $keepold, $verbosity, $outhandle) = @_;
49 my $self = new docproc ();
50
51 # outhandle is where all the debugging info goes
52 # output_handle is where the output of the plugins is piped
53 # to (i.e. mg, database etc.)
54 $outhandle = STDERR unless defined $outhandle;
55
56 $self->{'collection'} = $collection;
57 $self->{'source_dir'} = $source_dir;
58 $self->{'build_dir'} = $build_dir;
59 $self->{'keepold'} = $keepold;
60 $self->{'verbosity'} = $verbosity;
61 $self->{'outhandle'} = $outhandle;
62
63 $self->{'classifiers'} = [];
64 $self->{'mode'} = "text";
65 $self->{'assocdir'} = $build_dir;
66 $self->{'dontdb'} = {};
67 $self->{'store_metadata_coverage'} = "false";
68
69 $self->{'index'} = "section:text";
70 $self->{'indexexparr'} = [];
71
72 $self->{'separate_cjk'} = 0;
73
74 my $found_num_data = 0;
75 my $buildconfigfile = undef;
76
77 if ($keepold) {
78 # For incremental building need to seed num_docs etc from values
79 # stored in build.cfg (if present)
80 $buildconfigfile = &util::filename_cat($build_dir, "build.cfg");
81 if (-e $buildconfigfile) {
82 $found_num_data = 1;
83 }
84 else {
85 # try the index dir
86 $buildconfigfile = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},
87 "index", "build.cfg");
88 if (-e $buildconfigfile) {
89 $found_num_data = 1;
90 }
91 }
92
93 }
94
95 if ($found_num_data)
96 {
97 #print STDERR "Found_Num_Data!\n";
98 my $buildcfg = &colcfg::read_build_cfg($buildconfigfile);
99 $self->{'starting_num_docs'} = $buildcfg->{'numdocs'};
100 #print STDERR "- num_docs: $self->{'starting_num_docs'}\n";
101 $self->{'starting_num_sections'} = $buildcfg->{'numsections'};
102 #print STDERR "- num_sections: $self->{'starting_num_sections'}\n";
103 $self->{'starting_num_bytes'} = $buildcfg->{'numbytes'};
104 #print STDERR "- num_bytes: $self->{'starting_num_bytes'}\n";
105 }
106 else
107 {
108 #print STDERR "NOT Found_Num_Data!\n";
109 $self->{'starting_num_docs'} = 0;
110 $self->{'starting_num_sections'} = 0;
111 $self->{'starting_num_bytes'} = 0;
112 }
113
114 $self->{'output_handle'} = "STDOUT";
115 $self->{'num_docs'} = $self->{'starting_num_docs'};
116 $self->{'num_sections'} = $self->{'starting_num_sections'};
117 $self->{'num_bytes'} = $self->{'starting_num_bytes'};
118
119 $self->{'num_processed_bytes'} = 0;
120 $self->{'store_text'} = 1;
121
122 # what level (section/document) the database - indexer intersection is
123 $self->{'db_level'} = "section";
124 #used by browse interface
125 $self->{'doclist'} = [];
126
127 $self->{'indexing_text'} = 0;
128
129 return bless $self, $class;
130
131}
132
133sub reset {
134 my $self = shift (@_);
135
136 $self->{'num_docs'} = $self->{'starting_num_docs'};
137 $self->{'num_sections'} = $self->{'starting_num_sections'};
138 $self->{'num_bytes'} = $self->{'starting_num_bytes'};
139
140 $self->{'num_processed_bytes'} = 0;
141}
142
143sub zero_reset {
144 my $self = shift (@_);
145
146 $self->{'num_docs'} = 0;
147 $self->{'num_sections'} = 0;
148 # reconstructed docs have no text, just metadata, so we need to
149 # remember how many bytes we had initially
150 $self->{'num_bytes'} = $self->{'starting_num_bytes'};
151
152 $self->{'num_processed_bytes'} = 0;
153}
154
155sub is_incremental_capable
156{
157 # By default we return 'no' as the answer
158 # Safer to assume non-incremental to start with, and then override in
159 # inherited classes that are.
160
161 return 0;
162}
163
164sub get_num_docs {
165 my $self = shift (@_);
166
167 return $self->{'num_docs'};
168}
169
170sub get_num_sections {
171 my $self = shift (@_);
172
173 return $self->{'num_sections'};
174}
175
176# num_bytes is the actual number of bytes in the collection
177# this is normally the same as what's processed during text compression
178sub get_num_bytes {
179 my $self = shift (@_);
180
181 return $self->{'num_bytes'};
182}
183
184# num_processed_bytes is the number of bytes actually passed
185# to mg for the current index
186sub get_num_processed_bytes {
187 my $self = shift (@_);
188
189 return $self->{'num_processed_bytes'};
190}
191
192sub set_output_handle {
193 my $self = shift (@_);
194 my ($handle) = @_;
195
196 $self->{'output_handle'} = $handle;
197}
198
199
200sub set_mode {
201 my $self = shift (@_);
202 my ($mode) = @_;
203
204 $self->{'mode'} = $mode;
205}
206
207sub get_mode {
208 my $self = shift (@_);
209
210 return $self->{'mode'};
211}
212
213sub set_assocdir {
214 my $self = shift (@_);
215 my ($assocdir) = @_;
216
217 $self->{'assocdir'} = $assocdir;
218}
219
220sub set_dontdb {
221 my $self = shift (@_);
222 my ($dontdb) = @_;
223
224 $self->{'dontdb'} = $dontdb;
225}
226
227sub set_infodbtype
228{
229 my $self = shift(@_);
230 my $infodbtype = shift(@_);
231 $self->{'infodbtype'} = $infodbtype;
232}
233
234sub set_index {
235 my $self = shift (@_);
236 my ($index, $indexexparr) = @_;
237
238 $self->{'index'} = $index;
239 $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
240}
241
242sub set_index_languages {
243 my $self = shift (@_);
244 my ($lang_meta, $langarr) = @_;
245 $self->{'lang_meta'} = $lang_meta;
246 $self->{'langarr'} = $langarr;
247}
248
249sub get_index {
250 my $self = shift (@_);
251
252 return $self->{'index'};
253}
254
255sub set_classifiers {
256 my $self = shift (@_);
257 my ($classifiers) = @_;
258
259 $self->{'classifiers'} = $classifiers;
260}
261
262sub set_indexing_text {
263 my $self = shift (@_);
264 my ($indexing_text) = @_;
265
266 $self->{'indexing_text'} = $indexing_text;
267}
268
269sub get_indexing_text {
270 my $self = shift (@_);
271
272 return $self->{'indexing_text'};
273}
274
275sub set_store_text {
276 my $self = shift (@_);
277 my ($store_text) = @_;
278
279 $self->{'store_text'} = $store_text;
280}
281
282sub set_store_metadata_coverage {
283 my $self = shift (@_);
284 my ($store_metadata_coverage) = @_;
285
286 $self->{'store_metadata_coverage'} = $store_metadata_coverage || "";
287}
288
289sub get_doc_list {
290 my $self = shift(@_);
291
292 return @{$self->{'doclist'}};
293}
294
295# the standard database level is section, but you may want to change it to document
296sub set_db_level {
297 my $self= shift (@_);
298 my ($db_level) = @_;
299
300 $self->{'db_level'} = $db_level;
301}
302
303sub set_sections_index_document_metadata {
304 my $self= shift (@_);
305 my ($index_type) = @_;
306
307 $self->{'sections_index_document_metadata'} = $index_type;
308}
309
310sub set_separate_cjk {
311 my $self = shift (@_);
312 my ($sep_cjk) = @_;
313
314 $self->{'separate_cjk'} = $sep_cjk;
315}
316
317sub process {
318 my $self = shift (@_);
319 my $method = $self->{'mode'};
320
321 $self->$method(@_);
322}
323
324# post process text depending on field. Currently don't do anything here
325# except cjk separation, and only for indexing
326# should only do this for indexed text (if $self->{'indexing_text'}),
327# but currently search term highlighting doesn't work if you do that.
328# once thats fixed up, then fix this.
329sub filter_text {
330 my $self = shift (@_);
331 my ($field, $text) = @_;
332
333 # lets do cjk seg here
334 my $new_text =$text;
335 if ($self->{'separate_cjk'}) {
336 $new_text = &cnseg::segment($text);
337 }
338 return $new_text;
339}
340
341
342sub infodb_metadata_stats
343{
344 my $self = shift (@_);
345 my ($field,$edit_mode) = @_;
346
347 # Keep some statistics relating to metadata sets used and
348 # frequency of particular metadata fields within each set
349
350 # Union of metadata prefixes and frequency of fields
351 # (both scoped for this document alone, and across whole collection)
352
353 if ($field =~ m/^(.+)\.(.*)$/) {
354 my $prefix = $1;
355 my $core_field = $2;
356
357 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
358 $self->{'doc_mdprefix_fields'}->{$prefix}->{$core_field}++;
359 $self->{'mdprefix_fields'}->{$prefix}->{$core_field}++;
360 }
361 else {
362 # delete
363 $self->{'doc_mdprefix_fields'}->{$prefix}->{$core_field}--;
364 $self->{'mdprefix_fields'}->{$prefix}->{$core_field}--;
365 }
366
367 }
368 elsif ($field =~ m/^[[:upper:]]/) {
369 # implicit 'ex' metadata set
370
371 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
372
373 $self->{'doc_mdprefix_fields'}->{'ex'}->{$field}++;
374 $self->{'mdprefix_fields'}->{'ex'}->{$field}++;
375 }
376 else {
377 # delete
378 $self->{'doc_mdprefix_fields'}->{'ex'}->{$field}--;
379 $self->{'mdprefix_fields'}->{'ex'}->{$field}--;
380 }
381 }
382
383}
384
385
386sub infodbedit {
387 my $self = shift (@_);
388 my ($doc_obj, $filename, $edit_mode) = @_;
389
390 # only output this document if it is a "indexed_doc" or "info_doc" (database only) document
391 my $doctype = $doc_obj->get_doc_type();
392 return if ($doctype ne "indexed_doc" && $doctype ne "info_doc");
393
394 my $archivedir = "";
395 if (defined $filename)
396 {
397 # doc_obj derived directly from file
398 my ($dir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
399 $dir = "" unless defined $dir;
400 $dir =~ s/\\/\//g;
401 $dir =~ s/^\/+//;
402 $dir =~ s/\/+$//;
403
404 $archivedir = $dir;
405
406 # resolve the final filenames of the files associated with this document
407 $self->assoc_files ($doc_obj, $archivedir);
408 }
409 else
410 {
411 # doc_obj reconstructed from database (has metadata, doc structure but no text)
412 my $top_section = $doc_obj->get_top_section();
413 $archivedir = $doc_obj->get_metadata_element($top_section,"archivedir");
414 }
415
416 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
417 #add this document to the browse structure
418 push(@{$self->{'doclist'}},$doc_obj->get_OID())
419 unless ($doctype eq "classification");
420 }
421 else {
422 # delete => remove this doc from browse structure
423 my $del_doc_oid = $doc_obj->get_OID();
424
425 my @filtered_doc_list = ();
426 foreach my $oid (@{$self->{'doclist'}}) {
427 push(@filtered_doc_list,$oid) if ($oid ne $del_doc_oid);
428 }
429 $self->{'doclist'} = \@filtered_doc_list;
430 }
431
432
433 # classify this document
434 &classify::classify_doc ($self->{'classifiers'}, $doc_obj, $edit_mode);
435
436 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
437 # this is another document
438 $self->{'num_docs'} += 1 unless ($doctype eq "classification");
439 }
440 else {
441 # delete
442 $self->{'num_docs'} -= 1 unless ($doctype eq "classification");
443 }
444
445 # is this a paged or a hierarchical document
446 my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
447
448 my $section = $doc_obj->get_top_section ();
449 my $doc_OID = $doc_obj->get_OID();
450 my $first = 1;
451 my $infodb_handle = $self->{'output_handle'};
452
453 $self->{'doc_mdprefix_fields'} = {};
454
455 while (defined $section)
456 {
457 my $section_OID = $doc_OID;
458 if ($section ne "")
459 {
460 $section_OID = $doc_OID . "." . $section;
461 }
462 my %section_infodb = ();
463
464 # update a few statistics
465 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
466
467 $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
468 $self->{'num_sections'} += 1 unless ($doctype eq "classification");
469 }
470 else {
471 # delete
472 $self->{'num_bytes'} -= $doc_obj->get_text_length ($section);
473 $self->{'num_sections'} -= 1 unless ($doctype eq "classification");
474 }
475
476 # output the fact that this document is a document (unless doctype
477 # has been set to something else from within a plugin
478 my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
479 if (!defined $dtype || $dtype !~ /\w/) {
480 $section_infodb{"doctype"} = [ "doc" ];
481 }
482
483 # Output whether this node contains text
484 #
485 # If doc_obj reconstructed from database file then no need to
486 # explicitly add <hastxt> as this is preserved as metadata when
487 # the database file is loaded in
488 if (defined $filename)
489 {
490 # doc_obj derived directly from file
491 if ($doc_obj->get_text_length($section) > 0) {
492 $section_infodb{"hastxt"} = [ "1" ];
493 } else {
494 $section_infodb{"hastxt"} = [ "0" ];
495 }
496 }
497
498 # output all the section metadata
499 my $metadata = $doc_obj->get_all_metadata ($section);
500 foreach my $pair (@$metadata) {
501 my ($field, $value) = (@$pair);
502
503 if ($field ne "Identifier" && $field !~ /^gsdl/ &&
504 defined $value && $value ne "") {
505
506 # escape problematic stuff
507 $value =~ s/\\/\\\\/g;
508 $value =~ s/\n/\\n/g;
509 $value =~ s/\r/\\r/g;
510
511 # special case for URL metadata
512 if ($field =~ /^URL$/i) {
513 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
514
515 &dbutil::write_infodb_entry($self->{'infodbtype'}, $infodb_handle, $value, { 'section' => [ $section_OID ] });
516 }
517 else {
518 # delete
519 &dbutil::delete_infodb_entry($self->{'infodbtype'}, $infodb_handle, $value);
520 }
521
522
523 }
524
525 if (!defined $self->{'dontdb'}->{$field}) {
526 push(@{$section_infodb{$field}}, $value);
527
528 if ($section eq "" && $self->{'store_metadata_coverage'} =~ /^true$/i)
529 {
530 $self->infodb_metadata_stats($field,$edit_mode);
531 }
532 }
533 }
534 }
535
536 if ($section eq "")
537 {
538 my $doc_mdprefix_fields = $self->{'doc_mdprefix_fields'};
539
540 foreach my $prefix (keys %$doc_mdprefix_fields)
541 {
542 push(@{$section_infodb{"metadataset"}}, $prefix);
543
544 foreach my $field (keys %{$doc_mdprefix_fields->{$prefix}})
545 {
546 push(@{$section_infodb{"metadatalist-$prefix"}}, $field);
547
548 my $val = $doc_mdprefix_fields->{$prefix}->{$field};
549 push(@{$section_infodb{"metadatafreq-$prefix-$field"}}, $val);
550 }
551 }
552 }
553
554 # If doc_obj reconstructed from database file then no need to
555 # explicitly add <archivedir> as this is preserved as metadata when
556 # the database file is loaded in
557 if (defined $filename)
558 {
559 # output archivedir if at top level
560 if ($section eq $doc_obj->get_top_section()) {
561 $section_infodb{"archivedir"} = [ $archivedir ];
562 }
563 }
564
565 # output document display type
566 if ($first) {
567 $section_infodb{"thistype"} = [ $thistype ];
568 }
569
570 if ($self->{'db_level'} eq "document") {
571 # doc num is num_docs not num_sections
572 # output the matching document number
573 $section_infodb{"docnum"} = [ $self->{'num_docs'} ];
574 }
575 else {
576 # output a list of children
577 my $children = $doc_obj->get_children ($section);
578 if (scalar(@$children) > 0) {
579 $section_infodb{"childtype"} = [ $childtype ];
580 my $contains = "";
581 foreach my $child (@$children)
582 {
583 $contains .= ";" unless ($contains eq "");
584 if ($child =~ /^.*?\.(\d+)$/)
585 {
586 $contains .= "\".$1";
587 }
588 else
589 {
590 $contains .= "\".$child";
591 }
592 }
593 $section_infodb{"contains"} = [ $contains ];
594 }
595 # output the matching doc number
596 $section_infodb{"docnum"} = [ $self->{'num_sections'} ];
597 }
598
599 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
600
601 &dbutil::write_infodb_entry($self->{'infodbtype'}, $infodb_handle, $section_OID, \%section_infodb);
602 }
603 else {
604 # delete
605 &dbutil::delete_infodb_entry($self->{'infodbtype'}, $infodb_handle, $section_OID);
606 }
607
608
609 # output a database entry for the document number, except for Lucene (which no longer needs this information)
610 unless (ref($self) eq "lucenebuildproc")
611 {
612 if (($edit_mode eq "add") || ($edit_mode eq "reindex")) {
613
614 if ($self->{'db_level'} eq "document") {
615 &dbutil::write_infodb_entry($self->{'infodbtype'}, $infodb_handle, $self->{'num_docs'}, { 'section' => [ $doc_OID ] });
616 }
617 else {
618 &dbutil::write_infodb_entry($self->{'infodbtype'}, $infodb_handle, $self->{'num_sections'}, { 'section' => [ $section_OID ] });
619 }
620 }
621 else {
622
623 if ($self->{'db_level'} eq "document") {
624 &dbutil::delete_infodb_entry($self->{'infodbtype'}, $infodb_handle, $self->{'num_docs'});
625 }
626 else {
627 &dbutil::delete_infodb_entry($self->{'infodbtype'}, $infodb_handle, $self->{'num_sections'});
628 }
629
630 }
631 }
632
633 $first = 0;
634 $section = $doc_obj->get_next_section($section);
635 last if ($self->{'db_level'} eq "document"); # if no sections wanted, only add the docs
636 }
637}
638
639
640
641
642sub infodb {
643 my $self = shift (@_);
644 my ($doc_obj, $filename) = @_;
645
646 $self->infodbedit($doc_obj,$filename,"add");
647}
648
649sub infodbreindex {
650 my $self = shift (@_);
651 my ($doc_obj, $filename) = @_;
652
653 $self->infodbedit($doc_obj,$filename,"reindex");
654}
655
656sub infodbdelete {
657 my $self = shift (@_);
658 my ($doc_obj, $filename) = @_;
659
660 $self->infodbedit($doc_obj,$filename,"delete");
661}
662
663
664sub text {
665 my $self = shift (@_);
666 my ($doc_obj) = @_;
667
668 my $handle = $self->{'outhandle'};
669 print $handle "basebuildproc::text function must be implemented in sub classes\n";
670 die "\n";
671}
672
673sub textreindex
674{
675 my $self = shift @_;
676
677 my $outhandle = $self->{'outhandle'};
678 print $outhandle "basebuildproc::textreindex function must be implemented in sub classes\n";
679 if (!$self->is_incremental_capable()) {
680
681 print $outhandle " This operation is only possible with indexing tools with that support\n";
682 print $outhandle " incremental building\n";
683 }
684 die "\n";
685}
686
687sub textdelete
688{
689 my $self = shift @_;
690
691 my $outhandle = $self->{'outhandle'};
692 print $outhandle "basebuildproc::textdelete function must be implemented in sub classes\n";
693 if (!$self->is_incremental_capable()) {
694
695 print $outhandle " This operation is only possible with indexing tools with that support\n";
696 print $outhandle " incremental building\n";
697 }
698 die "\n";
699}
700
701
702# should the document be indexed - according to the subcollection and language
703# specification.
704sub is_subcollection_doc {
705 my $self = shift (@_);
706 my ($doc_obj) = @_;
707
708 my $indexed_doc = 1;
709 foreach my $indexexp (@{$self->{'indexexparr'}}) {
710 $indexed_doc = 0;
711 my ($field, $exp, $options) = split /\//, $indexexp;
712 if (defined ($field) && defined ($exp)) {
713 my ($bool) = $field =~ /^(.)/;
714 $field =~ s/^.// if $bool eq '!';
715 my @metadata_values;
716 if ($field =~ /^filename$/i) {
717 push(@metadata_values, $doc_obj->get_source_filename());
718 }
719 else {
720 @metadata_values = @{$doc_obj->get_metadata($doc_obj->get_top_section(), $field)};
721 }
722 next unless @metadata_values;
723 foreach my $metadata_value (@metadata_values) {
724 if ($bool eq '!') {
725 if ($options =~ /^i$/i) {
726 if ($metadata_value !~ /$exp/i) {$indexed_doc = 1; last;}
727 } else {
728 if ($metadata_value !~ /$exp/) {$indexed_doc = 1; last;}
729 }
730 } else {
731 if ($options =~ /^i$/i) {
732 if ($metadata_value =~ /$exp/i) {$indexed_doc = 1; last;}
733 } else {
734 if ($metadata_value =~ /$exp/) {$indexed_doc = 1; last;}
735 }
736 }
737 }
738
739 last if ($indexed_doc == 1);
740 }
741 }
742
743 # if this doc is so far in the sub collection, and we have lang info,
744 # now we check the languages to see if it matches
745 if($indexed_doc && defined $self->{'lang_meta'}) {
746 $indexed_doc = 0;
747 my $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $self->{'lang_meta'});
748 if (defined $field) {
749 foreach my $lang (@{$self->{'langarr'}}) {
750 my ($bool) = $lang =~ /^(.)/;
751 if ($bool eq '!') {
752 $lang =~ s/^.//;
753 if ($field !~ /$lang/) {
754 $indexed_doc = 1; last;
755 }
756 } else {
757 if ($field =~ /$lang/) {
758 $indexed_doc = 1; last;
759 }
760 }
761 }
762 }
763 }
764 return $indexed_doc;
765
766}
767
768# use 'Paged' if document has no more than 2 levels
769# and each section at second level has a number for
770# Title metadata
771# also use Paged if gsdlthistype metadata is set to Paged
772sub get_document_type {
773 my $self = shift (@_);
774 my ($doc_obj) = @_;
775
776 my $thistype = "VList";
777 my $childtype = "VList";
778 my $title;
779 my @tmp = ();
780
781 my $section = $doc_obj->get_top_section ();
782
783 my $gsdlthistype = $doc_obj->get_metadata_element ($section, "gsdlthistype");
784 if (defined $gsdlthistype) {
785 if ($gsdlthistype eq "Paged") {
786 $childtype = "Paged";
787 if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
788 $thistype = "Paged";
789 } else {
790 $thistype = "Invisible";
791 }
792
793 return ($thistype, $childtype);
794 } elsif ($gsdlthistype eq "Hierarchy") {
795 return ($thistype, $childtype); # use VList, VList
796 }
797 }
798 my $first = 1;
799 while (defined $section) {
800 @tmp = split /\./, $section;
801 if (scalar(@tmp) > 1) {
802 return ($thistype, $childtype);
803 }
804 if (!$first) {
805 $title = $doc_obj->get_metadata_element ($section, "Title");
806 if (!defined $title || $title !~ /^\d+$/) {
807 return ($thistype, $childtype);
808 }
809 }
810 $first = 0;
811 $section = $doc_obj->get_next_section($section);
812 }
813 if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
814 $thistype = "Paged";
815 } else {
816 $thistype = "Invisible";
817 }
818 $childtype = "Paged";
819 return ($thistype, $childtype);
820}
821
822sub assoc_files
823{
824 my $self = shift (@_);
825 my ($doc_obj, $archivedir) = @_;
826 my ($afile);
827
828 foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
829 #rint STDERR "Processing associated file - copy " . $assoc_file->[0] . " to " . $assoc_file->[1] . "\n";
830 # if assoc file starts with a slash, we put it relative to the assoc
831 # dir, otherwise it is relative to the HASH... directory
832 if ($assoc_file->[1] =~ m@^[/\\]@) {
833 $afile = &util::filename_cat($self->{'assocdir'}, $assoc_file->[1]);
834 } else {
835 $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
836 }
837 &util::hard_link ($assoc_file->[0], $afile, $self->{'verbosity'});
838 }
839}
840
Note: See TracBrowser for help on using the repository browser.