source: main/trunk/greenstone2/perllib/plugouts/BasePlugout.pm

Last change on this file was 38851, checked in by davidb, 5 weeks ago

General upgrade to support exporting files suitable for inclusion in OpenAI's Assistants API
  • Property svn:keywords set to Author Date Id Revision
File size: 39.4 KB
Line 
1###########################################################################
2#
3# BasePlugout.pm -- base class for all the plugout modules
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) 2006 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
26package BasePlugout;
27
28eval {require bytes};
29
30use strict;
31no strict 'subs';
32no strict 'refs';
33
34use dbutil;
35use gsprintf 'gsprintf';
36use printusage;
37use parse2;
38use util;
39use FileUtils;
40use sorttools;
41
42# suppress the annoying "subroutine redefined" warning that various
43# gets cause under perl 5.6
44$SIG{__WARN__} = sub {warn($_[0]) unless ($_[0] =~ /Subroutine\s+\S+\sredefined/)};
45
46my $arguments = [
47 { 'name' => "site",
48 'desc' => "{BasPlugout.site}",
49 'type' => "string",
50 'reqd' => "no",
51 'hiddengli' => "yes" },
52 { 'name' => "xslt_file",
53 'desc' => "{BasPlugout.xslt_file}",
54 'type' => "string",
55 'reqd' => "no",
56 'deft' => "",
57 'hiddengli' => "no"},
58 { 'name' => "subdir_split_length",
59 'desc' => "{BasPlugout.subdir_split_length}",
60 'type' => "int",
61 'reqd' => "no",
62 'deft' => "8",
63 'hiddengli' => "no"},
64 { 'name' => "subdir_hash_prefix",
65 'desc' => "{BasPlugout.subdir_hash_prefix}",
66 'type' => "flag",
67 'reqd' => "no",
68 'deft' => "0",
69 'hiddengli' => "no"},
70 { 'name' => "gzip_output",
71 'desc' => "{BasPlugout.gzip_output}",
72 'type' => "flag",
73 'reqd' => "no",
74 'hiddengli' => "no"},
75 { 'name' => "verbosity",
76 'desc' => "{BasPlugout.verbosity}",
77 'type' => "int",
78 'deft' => "0",
79 'reqd' => "no",
80 'hiddengli' => "no"},
81 { 'name' => "output_info",
82 'desc' => "{BasPlugout.output_info}",
83 'type' => "string",
84 'reqd' => "yes",
85 'hiddengli' => "yes"},
86 { 'name' => "output_handle",
87 'desc' => "{BasPlugout.output_handle}",
88 'type' => "string",
89 'deft' => 'STDERR',
90 'reqd' => "no",
91 'hiddengli' => "yes"},
92 { 'name' => "debug",
93 'desc' => "{BasPlugout.debug}",
94 'type' => "flag",
95 'reqd' => "no",
96 'hiddengli' => "yes"},
97 { 'name' => 'no_rss',
98 'desc' => "{BasPlugout.no_rss}",
99 'type' => 'flag',
100 'reqd' => 'no',
101 'hiddengli' => 'yes'},
102 { 'name' => 'rss_title',
103 'desc' => "{BasPlugout.rss_title}",
104 'type' => 'string',
105 'deft' => 'dc.Title',
106 'reqd' => 'no',
107 'hiddengli' => 'yes'},
108 { 'name' => 'assocfile_copymode',
109 'desc' => "{import.assocfile_copymode}",
110 'type' => 'enum',
111 'list' => [ { 'name' => "copy", 'desc' => "{import.assocfile_copymode_copy}" },
112 { 'name' => "hardlink", 'desc' => "{import.assocfile_copymode_hardlink}" } ],
113 'deft' => 'copy',
114 'reqd' => 'yes',
115 'hiddengli' => 'no'},
116 { 'name' => "no_auxiliary_databases",
117 'desc' => "{BasPlugout.no_auxiliary_databases}",
118 'type' => "flag",
119 'reqd' => "no",
120 'hiddengli' => "yes"}
121
122];
123
124my $options = { 'name' => "BasePlugout",
125 'desc' => "{BasPlugout.desc}",
126 'abstract' => "yes",
127 'inherits' => "no",
128 'args' => $arguments};
129
130
131
132sub sigpipe_handler
133{
134 # first argument is the signal name
135 my($sig) = @_;
136
137 print "Unexpected SIG$sig ... exiting\n";
138 exit(1);
139}
140
141$SIG{PIPE} = \&sigpipe_handler;
142
143
144sub new
145{
146 my $class = shift (@_);
147
148 my ($plugoutlist,$args,$hashArgOptLists) = @_;
149 push(@$plugoutlist, $class);
150
151 my $plugout_name = (defined $plugoutlist->[0]) ? $plugoutlist->[0] : $class;
152
153 push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
154 push(@{$hashArgOptLists->{"OptList"}},$options);
155
156 my $self = {};
157 $self->{'plugout_type'} = $class;
158 $self->{'option_list'} = $hashArgOptLists->{"OptList"};
159 $self->{"info_only"} = 0;
160
161 # Check if gsdlinfo is in the argument list or not - if it is, don't parse
162 # the args, just return the object.
163 #print STDERR "#### " . join(",", @${args}) . "\n\n";
164 my $v=0;
165 foreach my $strArg (@{$args})
166 {
167 if(defined $strArg) {
168 if($strArg eq "-gsdlinfo")
169 {
170 $self->{"info_only"} = 1;
171 return bless $self, $class;
172 }
173 elsif ($strArg eq "-site") {
174 $v = $strArg;
175 }
176 elsif($v eq "-site") {
177 $self->{'site'} = $strArg;
178 }
179 }
180 }
181
182 delete $self->{"info_only"};
183
184 if(parse2::parse($args,$hashArgOptLists->{"ArgList"},$self) == -1)
185 {
186 my $classTempClass = bless $self, $class;
187 print STDERR "<BadPlugout d=$plugout_name>\n";
188 &gsprintf(STDERR, "\n{BasPlugout.bad_general_option}\n", $plugout_name);
189 $classTempClass->print_txt_usage(""); # Use default resource bundle
190 die "\n";
191 }
192
193
194 if(defined $self->{'xslt_file'} && $self->{'xslt_file'} ne "")
195 {
196 my $full_file_path = &util::locate_config_file($self->{'xslt_file'});
197 if (!defined $full_file_path) {
198 print STDERR "Can not find $self->{'xslt_file'}, please make sure you have supplied the correct file path or put the file into the collection's etc or greenstone's etc folder\n";
199 die "\n";
200 }
201 $self->{'xslt_file'} = $full_file_path;
202 }
203
204 # for group processing
205 $self->{'gs_count'} = 0;
206 $self->{'group_position'} = 1;
207
208 $self->{'keep_import_structure'} = 0;
209
210 $self->{'generate_databases'} = 1;
211 if ($self->{'no_auxiliary_databases'}) {
212 $self->{'generate_databases'} = 0;
213 }
214 undef $self->{'no_auxiliary_databases'};
215
216 &sorttools::setup_custom_sort();
217 return bless $self, $class;
218
219}
220
221# implement this in subclass if you want to do some initialization after
222# loading and setting parameters, and before processing the documents.
223sub begin {
224
225 my $self= shift (@_);
226
227}
228# implement in subclasses if it needs some non-group related cleanup (post-group cleanup
229# Like begin(), end() is also called by inexport.pm
230sub end {
231 my $self= shift (@_);
232
233}
234sub print_xml_usage
235{
236 my $self = shift(@_);
237 my $header = shift(@_);
238 my $high_level_information_only = shift(@_);
239
240 # XML output is always in UTF-8
241 gsprintf::output_strings_in_UTF8;
242
243 if ($header) {
244 &PrintUsage::print_xml_header("plugout");
245 }
246 $self->print_xml($high_level_information_only);
247}
248
249
250sub print_xml
251{
252 my $self = shift(@_);
253 my $high_level_information_only = shift(@_);
254
255 my $optionlistref = $self->{'option_list'};
256 my @optionlist = @$optionlistref;
257 my $plugoutoptions = shift(@$optionlistref);
258 return if (!defined($plugoutoptions));
259
260 gsprintf(STDERR, "<PlugoutInfo>\n");
261 gsprintf(STDERR, " <Name>$plugoutoptions->{'name'}</Name>\n");
262 my $desc = gsprintf::lookup_string($plugoutoptions->{'desc'});
263 $desc =~ s/</&amp;lt;/g; # doubly escaped
264 $desc =~ s/>/&amp;gt;/g;
265 gsprintf(STDERR, " <Desc>$desc</Desc>\n");
266 gsprintf(STDERR, " <Abstract>$plugoutoptions->{'abstract'}</Abstract>\n");
267 gsprintf(STDERR, " <Inherits>$plugoutoptions->{'inherits'}</Inherits>\n");
268 unless (defined($high_level_information_only)) {
269 gsprintf(STDERR, " <Arguments>\n");
270 if (defined($plugoutoptions->{'args'})) {
271 &PrintUsage::print_options_xml($plugoutoptions->{'args'});
272 }
273 gsprintf(STDERR, " </Arguments>\n");
274
275 # Recurse up the plugout hierarchy
276 $self->print_xml();
277 }
278 gsprintf(STDERR, "</PlugoutInfo>\n");
279}
280
281
282sub print_txt_usage
283{
284 my $self = shift(@_);
285
286 # Print the usage message for a plugout (recursively)
287 my $descoffset = $self->determine_description_offset(0);
288 $self->print_plugout_usage($descoffset, 1);
289}
290
291sub determine_description_offset
292{
293 my $self = shift(@_);
294 my $maxoffset = shift(@_);
295
296 my $optionlistref = $self->{'option_list'};
297 my @optionlist = @$optionlistref;
298 my $plugoutoptions = pop(@$optionlistref);
299 return $maxoffset if (!defined($plugoutoptions));
300
301 # Find the length of the longest option string of this download
302 my $plugoutargs = $plugoutoptions->{'args'};
303 if (defined($plugoutargs)) {
304 my $longest = &PrintUsage::find_longest_option_string($plugoutargs);
305 if ($longest > $maxoffset) {
306 $maxoffset = $longest;
307 }
308 }
309
310 # Recurse up the download hierarchy
311 $maxoffset = $self->determine_description_offset($maxoffset);
312 $self->{'option_list'} = \@optionlist;
313 return $maxoffset;
314}
315
316
317sub print_plugout_usage
318{
319 my $self = shift(@_);
320 my $descoffset = shift(@_);
321 my $isleafclass = shift(@_);
322
323 my $optionlistref = $self->{'option_list'};
324 my @optionlist = @$optionlistref;
325 my $plugoutoptions = shift(@$optionlistref);
326 return if (!defined($plugoutoptions));
327
328 my $plugoutname = $plugoutoptions->{'name'};
329 my $plugoutargs = $plugoutoptions->{'args'};
330 my $plugoutdesc = $plugoutoptions->{'desc'};
331
332 # Produce the usage information using the data structure above
333 if ($isleafclass) {
334 if (defined($plugoutdesc)) {
335 gsprintf(STDERR, "$plugoutdesc\n\n");
336 }
337 gsprintf(STDERR, " {common.usage}: plugout $plugoutname [{common.options}]\n\n");
338 }
339
340 # Display the download options, if there are some
341 if (defined($plugoutargs)) {
342 # Calculate the column offset of the option descriptions
343 my $optiondescoffset = $descoffset + 2; # 2 spaces between options & descriptions
344
345 if ($isleafclass) {
346 gsprintf(STDERR, " {common.specific_options}:\n");
347 }
348 else {
349 gsprintf(STDERR, " {common.general_options}:\n", $plugoutname);
350 }
351
352 # Display the download options
353 &PrintUsage::print_options_txt($plugoutargs, $optiondescoffset);
354 }
355
356 # Recurse up the download hierarchy
357 $self->print_plugout_usage($descoffset, 0);
358 $self->{'option_list'} = \@optionlist;
359}
360
361
362sub error
363{
364 my ($strFunctionName,$strError) = @_;
365 {
366 print "Error occoured in BasePlugout.pm\n".
367 "In Function: ".$strFunctionName."\n".
368 "Error Message: ".$strError."\n";
369 exit(-1);
370 }
371}
372
373
374# OIDtype may be "hash" or "hash_on_full_filename" or "incremental" or "filename" or "dirname" or "full_filename" or "assigned"
375sub set_OIDtype {
376 my $self = shift (@_);
377 my ($type, $metadata) = @_;
378
379 if ($type =~ /^(hash|hash_on_full_filename|incremental|filename|dirname|full_filename|assigned)$/) {
380 $self->{'OIDtype'} = $type;
381 } else {
382 $self->{'OIDtype'} = "hash";
383 }
384 if ($type =~ /^assigned$/) {
385 if (defined $metadata) {
386 $self->{'OIDmetadata'} = $metadata;
387 } else {
388 $self->{'OIDmetadata'} = "dc.Identifier";
389 }
390 }
391}
392
393sub set_output_dir
394{
395 my $self = shift @_;
396 my ($output_dir) = @_;
397
398 $self->{'output_dir'} = $output_dir;
399}
400
401sub setoutputdir
402{
403 my $self = shift @_;
404 my ($output_dir) = @_;
405
406 $self->{'output_dir'} = $output_dir;
407}
408
409sub get_output_dir
410{
411 my $self = shift (@_);
412
413 return $self->{'output_dir'};
414}
415
416sub getoutputdir
417{
418 my $self = shift (@_);
419
420 return $self->{'output_dir'};
421}
422
423sub getoutputinfo
424{
425 my $self = shift (@_);
426
427 return $self->{'output_info'};
428}
429
430
431sub get_output_handler
432{
433 my $self = shift (@_);
434
435 my ($output_file_name) = @_;
436
437 my $fh;
438 &FileUtils::openFileHandle($output_file_name, '>', \$fh) or die('Can not open a file handler for: ' . $output_file_name . "\n");
439
440 return $fh;
441}
442
443sub release_output_handler
444{
445 my $self = shift (@_);
446 my ($outhandler) = @_;
447
448 close($outhandler);
449
450}
451
452sub output_xml_header {
453 my $self = shift (@_);
454 my ($handle,$docroot,$nondoctype) = @_;
455
456 #For Dspace must be UTF in lower case
457 print $handle '<?xml version="1.0" encoding="utf-8" standalone="no"?>' . "\n";
458
459 if (!defined $nondoctype){
460 my $doctype = (defined $docroot) ? $docroot : "Section";
461
462 # Used to be '<!DOCTYPE Archive SYSTEM ...'
463
464 print $handle "<!DOCTYPE $doctype SYSTEM \"https://greenstone.org/dtd/Archive/1.0/Archive.dtd\">\n";
465 }
466
467 print $handle "<$docroot>\n" if defined $docroot;
468}
469
470sub output_xml_footer {
471 my $self = shift (@_);
472 my ($handle,$docroot) = @_;
473 print $handle "</$docroot>\n" if defined $docroot;
474}
475
476
477sub output_general_xml_header
478{
479 my $self = shift (@_);
480 my ($handle,$docroot,$opt_attributes,$opt_dtd, $opt_doctype) = @_;
481
482 print $handle '<?xml version="1.0" encoding="utf-8" standalone="no"?>' . "\n";
483
484 if (defined $opt_dtd) {
485 my $doctype = (defined $opt_doctype) ? $opt_doctype : $docroot;
486 print $handle "<!DOCTYPE $doctype SYSTEM \"$opt_dtd\">\n";
487 }
488
489 if (defined $docroot) {
490 my $full_docroot = $docroot;
491 if (defined $opt_attributes) {
492 $full_docroot .= " $opt_attributes";
493 }
494
495 print $handle "<$full_docroot>\n"
496 }
497}
498
499sub output_general_xml_footer
500{
501 output_xml_footer(@_);
502}
503
504# This is called by the plugins after read_into_doc_obj generates the doc_obj.
505sub process {
506 my $self = shift (@_);
507 my ($doc_obj) = @_;
508
509 my $output_info = $self->{'output_info'};
510 return if (!defined $output_info);
511
512 # for OAI purposes
513 $doc_obj->set_lastmodified();
514 $doc_obj->set_oailastmodified();
515
516 # find out which directory to save to
517 my $doc_dir = "";
518 if ($self->is_group()) {
519 $doc_dir = $self->get_group_doc_dir($doc_obj);
520 } else {
521 $doc_dir = $self->get_doc_dir($doc_obj);
522 }
523
524 ##############################
525 # call subclass' saveas method
526 ##############################
527 $self->saveas($doc_obj,$doc_dir);
528
529 # write out data to archiveinf-doc.db
530 if ($self->{'generate_databases'}) {
531 $self->archiveinf_db($doc_obj);
532 }
533 if ($self->is_group()) {
534 $self->{'gs_count'}++; # do we want this for all cases?
535 $self->{'group_position'}++;
536 }
537}
538
539sub store_output_info_reference {
540 my $self = shift (@_);
541 my ($doc_obj) = @_;
542
543 my $output_info = $self->{'output_info'};
544 my $metaname = $self->{'sortmeta'};
545
546 my $group_position;
547 if ($self->is_group()) {
548 $group_position = $self->{'group_position'};
549 }
550 if (!defined $metaname || $metaname !~ /\S/) {
551 my $OID = $doc_obj->get_OID();
552 $output_info->add_info($OID,$self->{'short_doc_file'}, undef, "", $group_position);
553 return;
554 }
555
556 if ($metaname eq "OID") { # sort by OID
557 my $OID = $doc_obj->get_OID();
558 $output_info->add_info($OID,$self->{'short_doc_file'}, undef, $OID, undef);
559 return;
560 }
561
562 my $metadata = "";
563 my $top_section = $doc_obj->get_top_section();
564
565 my @commameta_list = split(/,/, $metaname);
566 foreach my $cmn (@commameta_list) {
567 my $meta = $doc_obj->get_metadata_element($top_section, $cmn);
568 if ($meta) {
569 # do remove prefix/suffix - this will apply to all values
570 $meta =~ s/^$self->{'removeprefix'}// if defined $self->{'removeprefix'};
571 $meta =~ s/$self->{'removesuffix'}$// if defined $self->{'removesuffix'};
572 $meta = &sorttools::format_metadata_for_sorting($cmn, $meta, $doc_obj);
573 $metadata .= $meta if ($meta);
574 }
575 }
576
577 # store reference in the output_info
578 $output_info->add_info($doc_obj->get_OID(),$self->{'short_doc_file'}, undef, $metadata,undef);
579}
580
581
582
583sub saveas {
584 my $self = shift (@_);
585 my ($doc_obj, $doc_dir) = @_;
586
587 die "BasePlugout::saveas function must be implemented in sub classes\n";
588}
589
590sub get_group_doc_dir {
591 my $self = shift (@_);
592 my ($doc_obj) = @_;
593
594 my $outhandle = $self->{'output_handle'};
595 my $OID = $doc_obj->get_OID();
596 $OID = "NULL" unless defined $OID;
597
598 my $groupsize = $self->{'group_size'};
599 my $gs_count = $self->{'gs_count'};
600
601 my $open_new_file = (($gs_count % $groupsize)==0);
602
603 my $doc_dir;
604
605 if (!$open_new_file && scalar(@{$doc_obj->get_assoc_files()})>0) {
606 # if we have some assoc files, then we will need to start a new file
607 if ($self->{'verbosity'} > 2) {
608 print $outhandle " Starting an archives/export folder for $OID as it has associated files\n";
609 }
610 $open_new_file = 1;
611 }
612
613 # opening a new file
614 if (($open_new_file) || !defined($self->{'gs_doc_dir'})) {
615 # first we close off the old output
616 if ($gs_count>0)
617 {
618 return if (!$self->close_group_output());
619 }
620
621 # this will create the directory
622 $doc_dir = $self->get_doc_dir ($doc_obj);
623 $self->{'new_doc_dir'} = 1;
624 $self->{'gs_doc_dir'} = $doc_dir;
625 $self->{'group_position'} = 1;
626 }
627 else {
628 $doc_dir = $self->{'gs_doc_dir'};
629 $self->{'new_doc_dir'} = 0;
630 }
631
632 return $doc_dir;
633}
634
635sub get_doc_dir {
636
637 my $self = shift (@_);
638 my ($doc_obj) = @_;
639
640 my $OID = $doc_obj->get_OID();
641 $OID = "NULL" unless defined $OID;
642
643 my $working_dir = $self->get_output_dir();
644 my $working_info = $self->{'output_info'};
645 return if (!defined $working_info);
646
647 my $doc_info = $working_info->get_info($OID);
648 my $doc_dir = '';
649
650 if (defined $doc_info && scalar(@$doc_info) >= 1)
651 {
652 # This OID already has an archives directory, so use it again
653 $doc_dir = $doc_info->[0];
654 $doc_dir =~ s/\/?((doc(mets|sql)?)|(dublin_core))\.xml(\.gz)?$//;
655 }
656 elsif ($self->{'keep_import_structure'})
657 {
658 my $source_filename = $doc_obj->get_source_filename();
659 $source_filename = &File::Basename::dirname($source_filename);
660 $source_filename =~ s/[\\\/]+/\//g;
661 $source_filename =~ s/\/$//;
662
663 $doc_dir = substr($source_filename, length($ENV{'GSDLIMPORTDIR'}) + 1);
664 }
665
666 # We have to use a new archives directory for this document
667 if ($doc_dir eq "")
668 {
669 $doc_dir = $self->get_new_doc_dir ($working_info, $working_dir, $OID);
670 }
671
672 &FileUtils::makeAllDirectories(&FileUtils::filenameConcatenate($working_dir, $doc_dir));
673
674 return $doc_dir;
675}
676
677
678## @function get_new_doc_dir()
679#
680# Once a doc object is ready to write to disk (and hence has a nice OID),
681# generate a unique subdirectory to write the information to.
682#
683# - create the directory as part of this call, to try and avoid race conditions
684# found in parallel processing [jmt12]
685#
686# @todo figure out what the rule regarding $work_info->size() is meant to do
687#
688# @todo determine what $self->{'group'} is, and whether it should affect
689# directory creation
690#
691sub get_new_doc_dir
692{
693 my $self = shift (@_);
694 my($working_info,$working_dir,$OID) = @_;
695
696 my $doc_dir = "";
697 my $doc_dir_rest = $OID;
698
699 # remove any \ and / from the OID
700 $doc_dir_rest =~ s/[\\\/]//g;
701
702 # Remove ":" if we are on Windows OS, as otherwise they get confused with the drive letters
703 if ($ENV{'GSDLOS'} =~ /^windows$/i)
704 {
705 $doc_dir_rest =~ s/\://g;
706 }
707
708 # we generally create a unique directory by adding consequtive fragments of
709 # the document identifier (split by some predefined length - defaulting to
710 # 8) until we find a directory that doesn't yet exist. Note that directories
711 # that contain a document have a suffix ".dir" (whereas those that contain
712 # only subdirectories have no suffix).
713 my $doc_dir_num = 0; # how many directories deep we are
714 my $created_directory = 0; # have we successfully created a new directory
715 do
716 {
717 # (does this work on windows? - jmt12)
718 if ($doc_dir_num > 0)
719 {
720 $doc_dir .= '/';
721 }
722
723 # the default matching pattern grabs the next 'subdir_split_length'
724 # characters of the OID to act as the next subdirectory
725 my $pattern = '^(.{1,' . $self->{'subdir_split_length'} . '})';
726
727 # Do we count any "HASH" prefix against the split length limit?
728 if ($self->{'subdir_hash_prefix'} && $doc_dir_num == 0)
729 {
730 $pattern = '^((HASH)?.{1,' . $self->{'subdir_split_length'} . '})';
731 }
732
733 # Note the use of 's' to both capture the next chuck of OID and to remove
734 # it from OID at the same time
735 if ($doc_dir_rest =~ s/$pattern//i)
736 {
737 $doc_dir .= $1;
738 $doc_dir_num++;
739
740 my $full_doc_dir = &FileUtils::filenameConcatenate($working_dir, $doc_dir . '.dir');
741 if(!FileUtils::directoryExists($full_doc_dir))
742 {
743 &FileUtils::makeAllDirectories($full_doc_dir);
744 $created_directory = 1;
745 }
746
747 ###rint STDERR "[DEBUG] BasePlugout::get_new_doc_dir(<working_info>, $working_dir, $oid)\n";
748 ###rint STDERR " - create directory: $full_doc_dir => $created_directory\n";
749 ###rint STDERR " - rest: $doc_dir_rest\n";
750 ###rint STDERR " - working_info->size(): " . $working_info->size() . " [ < 1024 ?]\n";
751 ###rint STDERR " - doc_dir_num: " . $doc_dir_num . "\n";
752 }
753 }
754 while ($doc_dir_rest ne '' && ($created_directory == 0 || ($working_info->size() >= 1024 && $doc_dir_num < 2)));
755
756 # not unique yet? Add on an incremental suffix until we are unique
757 my $i = 1;
758 my $doc_dir_base = $doc_dir;
759 while ($created_directory == 0)
760 {
761 $doc_dir = $doc_dir_base . '-' . $i;
762 $created_directory = &FileUtils::makeAllDirectories(&FileUtils::filenameConcatenate($working_dir, $doc_dir . '.dir'));
763 $i++;
764 }
765
766 # in theory this should never happen
767 if (!$created_directory)
768 {
769 die("Error! Failed to create directory for document: " . $doc_dir_base . "\n");
770 }
771
772 return $doc_dir . '.dir';
773}
774## get_new_doc_dir()
775
776
777sub process_assoc_files {
778 my $self = shift (@_);
779 my ($doc_obj, $doc_dir, $handle) = @_;
780
781 my $outhandle = $self->{'output_handle'};
782
783 my $output_dir = $self->get_output_dir();
784 return if (!defined $output_dir);
785
786 &FileUtils::makeAllDirectories($output_dir) unless &FileUtils::directoryExists($output_dir);
787
788 my $working_dir = &FileUtils::filenameConcatenate($output_dir, $doc_dir);
789 &FileUtils::makeAllDirectories($working_dir) unless &FileUtils::directoryExists($working_dir);
790
791 my @assoc_files = ();
792 my $filename;;
793
794 my $source_filename = $doc_obj->get_source_filename();
795
796 my $collect_dir = $ENV{'GSDLCOLLECTDIR'};
797
798 if (defined $collect_dir) {
799 my $dirsep_regexp = &util::get_os_dirsep();
800
801 if ($collect_dir !~ /$dirsep_regexp$/) {
802 $collect_dir .= &util::get_dirsep(); # ensure there is a slash at the end
803 }
804
805 # This test is never going to fail on Windows -- is this a problem?
806
807 if ($source_filename !~ /^$dirsep_regexp/) {
808 $source_filename = &FileUtils::filenameConcatenate($collect_dir, $source_filename);
809 }
810 }
811
812
813 # set the assocfile path (even if we have no assoc files - need this for lucene)
814 $doc_obj->set_utf8_metadata_element ($doc_obj->get_top_section(),
815 "assocfilepath",
816 "$doc_dir");
817 foreach my $assoc_file_rec (@{$doc_obj->get_assoc_files()}) {
818 my ($dir, $afile) = $assoc_file_rec->[1] =~ /^(.*?)([^\/\\]+)$/;
819 $dir = "" unless defined $dir;
820
821 my $utf8_real_filename = $assoc_file_rec->[0];
822
823 # for some reasons the image associate file has / before the full path
824 $utf8_real_filename =~ s/^\\(.*)/$1/i;
825
826## my $real_filename = &util::utf8_to_real_filename($utf8_real_filename);
827 my $real_filename = $utf8_real_filename;
828 $real_filename = &util::downgrade_if_dos_filename($real_filename);
829
830 if (&FileUtils::fileExists($real_filename)) {
831
832 $filename = &FileUtils::filenameConcatenate($working_dir, $afile);
833
834 if ($self->{'assocfile_copymode'} eq "hardlink") {
835 &FileUtils::hardLink($real_filename, $filename, $self->{'verbosity'}); # Consider adding in 'strict' option??
836 }
837 else {
838 &FileUtils::copyFilesGeneral([$real_filename], $filename); # Consider adding in 'strict' option??
839 }
840
841 $doc_obj->add_utf8_metadata ($doc_obj->get_top_section(),
842 "gsdlassocfile",
843 "$afile:$assoc_file_rec->[2]:$dir");
844 } elsif ($self->{'verbosity'} > 1) {
845 print $outhandle "BasePlugout::process couldn't copy the associated file " .
846 "$real_filename to $afile\n";
847 }
848 }
849}
850
851
852sub process_metafiles_metadata
853{
854 my $self = shift (@_);
855 my ($doc_obj) = @_;
856
857 my $top_section = $doc_obj->get_top_section();
858 my $metafiles = $doc_obj->get_metadata($top_section,"gsdlmetafile");
859
860 foreach my $metafile_pair (@$metafiles) {
861 my ($full_metafile,$metafile) = split(/ : /,$metafile_pair);
862
863 $doc_obj->metadata_file($full_metafile,$metafile);
864 }
865
866 $doc_obj->delete_metadata($top_section,"gsdlmetafile");
867}
868
869sub archiveinf_files_to_field
870{
871 my $self = shift(@_);
872 my ($files,$field,$collect_dir,$oid_files,$reverse_lookups) = @_;
873
874 foreach my $file_rec (@$files) {
875 my $real_filename = (ref $file_rec eq "ARRAY") ? $file_rec->[0] : $file_rec;
876 my $full_file = (ref $file_rec eq "ARRAY") ? $file_rec->[1] : $file_rec;
877 # for some reasons the image associate file has / before the full path
878 $real_filename =~ s/^\\(.*)/$1/i;
879
880 my $raw_filename = &util::downgrade_if_dos_filename($real_filename);
881
882 #print STDERR "**** raw_filename = $raw_filename\n";
883 #print STDERR "**** real_filename = $real_filename\n\n";
884
885 if (&FileUtils::fileExists($raw_filename)) {
886
887# if (defined $collect_dir) {
888# my $collect_dir_re_safe = $collect_dir;
889# $collect_dir_re_safe =~ s/\\/\\\\/g; # use &util::filename_to_regex()
890# $collect_dir_re_safe =~ s/\./\\./g;##
891
892# $real_filename =~ s/^$collect_dir_re_safe//;
893# }
894
895 if (defined $reverse_lookups) {
896 $reverse_lookups->{$real_filename} = 1;
897 }
898
899 if($field =~ m@assoc-file|src-file|meta-file@) {
900 $raw_filename = &util::abspath_to_placeholders($raw_filename);
901 }
902
903### push(@{$oid_files->{$field}},$full_file);
904 push(@{$oid_files->{$field}},$raw_filename);
905 }
906 else {
907 print STDERR "Warning: archiveinf_files_to_field()\n $real_filename does not appear to be on the file system\n";
908 }
909 }
910}
911
912sub archiveinf_db
913{
914 my $self = shift (@_);
915 my ($doc_obj) = @_;
916
917 my $verbosity = $self->{'verbosity'};
918
919 my $collect_dir = $ENV{'GSDLCOLLECTDIR'};
920 if (defined $collect_dir) {
921 my $dirsep_regexp = &util::get_os_dirsep();
922
923 if ($collect_dir !~ /$dirsep_regexp$/) {
924 # ensure there is a slash at the end
925 $collect_dir .= &util::get_dirsep();
926 }
927 }
928
929 my $oid = $doc_obj->get_OID();
930 my $source_filename = $doc_obj->get_unmodified_source_filename();
931 my $working_info = $self->{'output_info'};
932 my $doc_info = $working_info->get_info($oid);
933
934 my ($doc_file,$index_status,$sortmeta, $group_position) = @$doc_info;
935 # doc_file is the path to the archive doc.xml. Make sure it has unix
936 # slashes, then if the collection is copied to linux, it can be built without reimport
937 $doc_file =~ s/\\/\//g;
938 my $oid_files = { 'doc-file' => $doc_file,
939 'index-status' => $index_status,
940 'src-file' => $source_filename,
941 'sort-meta' => $sortmeta,
942 'assoc-file' => [],
943 'meta-file' => [] };
944 if (defined $group_position) {
945 $oid_files->{'group-position'} = $group_position;
946 }
947 my $reverse_lookups = { $source_filename => "1" };
948
949
950 $self->archiveinf_files_to_field($doc_obj->get_source_assoc_files(),"assoc-file",
951 $collect_dir,$oid_files,$reverse_lookups);
952
953
954 $self->archiveinf_files_to_field($doc_obj->get_meta_files(),"meta-file",
955 $collect_dir,$oid_files);
956
957 # Get the infodbtype value for this collection from the arcinfo object
958 my $infodbtype = $self->{'output_info'}->{'infodbtype'};
959 my $output_dir = $self->{'output_dir'};
960
961 my $doc_db = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $output_dir);
962
963 ##print STDERR "*** To set in db: \n\t$doc_db\n\t$oid\n\t$doc_db_text\n";
964
965 if (!$self->{'no_rss'})
966 {
967 if (($oid_files->{'index-status'} eq "I") || ($oid_files->{'index-status'} eq "R")) {
968 my $top_section = $doc_obj->get_top_section();
969
970 # rss_title can be set in collect.cfg as follows:
971 # plugout GreenstoneXMLPlugout -rss_title "dc.Title; ex.Title"
972 # rss_title is a semi-colon or comma-separated list of the metadata field names that should
973 # be consulted in order to obtain a Title (anchor text) for the RSS document link.
974 # If not specified, rss_title will default to dc.Title, and fall back on Untitled
975 my $metafieldnames = $self->{'rss_title'};
976 my @metafieldarray = split(/[,;] ?/,$metafieldnames); # , or ; separator can be followed by an optional space
977 my $titles;
978 #@$titles=(); # at worst @$titles will be (), as get_metadata(dc.Titles) may return ()
979 foreach my $metafieldname (@metafieldarray) {
980 $metafieldname =~ s@^ex\.@@; # if ex.Title, need to get_metadata() on metafieldname=Title
981 $titles = $doc_obj->get_metadata($top_section,$metafieldname);
982
983 if(scalar(@$titles) != 0) { # found at least one title for one metafieldname
984 last; # break out of the loop
985 }
986 }
987
988 # if ex.Title was listed in the metafieldnames, then we'll surely have a value for title for this doc
989 # otherwise, if we have no titles at this point, add in a default of Untitled as this doc's title
990 if(scalar(@$titles) == 0) { #&& $metafieldnames !~ [email protected]@) {
991 push(@$titles, "Untitled");
992 }
993
994 # encode basic html entities like <>"& in the title(s), since the & char can break RSS links
995 for (my $i = 0; $i < scalar(@$titles); $i++) {
996 &ghtml::htmlsafe(@$titles[$i]);
997 }
998
999 my $dc_title = join("; ", @$titles);
1000
1001 if ($oid_files->{'index-status'} eq "R") {
1002 $dc_title .= " (Updated)";
1003 }
1004
1005 my $rss_entry = "<item>\n";
1006 $rss_entry .= " <title>$dc_title</title>\n";
1007 if(&util::is_gs3()) {
1008 $rss_entry .= " <link>_httpdomain__httpcollection_/document/$oid</link>\n";
1009 } else {
1010 $rss_entry .= " <link>_httpdomainHtmlsafe__httpcollection_/document/$oid</link>\n";
1011 }
1012 $rss_entry .= "</item>";
1013
1014 if (defined(&dbutil::supportsRSS) && &dbutil::supportsRSS($infodbtype))
1015 {
1016 my $rss_db = &dbutil::get_infodb_file_path($infodbtype, 'rss-items', $output_dir);
1017 my $rss_db_fh = &dbutil::open_infodb_write_handle($infodbtype, $rss_db, 'append');
1018 &dbutil::write_infodb_rawentry($infodbtype, $rss_db_fh, $oid, $rss_entry);
1019 &dbutil::close_infodb_write_handle($infodbtype, $rss_db_fh);
1020 }
1021 else
1022 {
1023 my $rss_filename = &FileUtils::filenameConcatenate($output_dir,"rss-items.rdf");
1024 my $rss_fh;
1025 if (&FileUtils::openFileHandle($rss_filename, '>>', \$rss_fh, "utf8"))
1026 {
1027 print $rss_fh $rss_entry . "\n";
1028 &FileUtils::closeFileHandle($rss_filename, \$rss_fh);
1029 }
1030 else
1031 {
1032 print STDERR "Error: Failed to open $rss_filename\n$!\n";
1033 }
1034 }
1035 }
1036 }
1037
1038 $oid_files->{'doc-file'} = [ $oid_files->{'doc-file'} ];
1039 $oid_files->{'index-status'} = [ $oid_files->{'index-status'} ];
1040 $oid_files->{'src-file'} = &util::abspath_to_placeholders($oid_files->{'src-file'});
1041 $oid_files->{'src-file'} = [ $oid_files->{'src-file'} ];
1042 $oid_files->{'sort-meta'} = [ $oid_files->{'sort-meta'} ];
1043 if (defined $oid_files->{'group-position'}) {
1044 $oid_files->{'group-position'} = [ $oid_files->{'group-position'} ];
1045 }
1046
1047 my $infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $doc_db, "append");
1048 &dbutil::write_infodb_entry($infodbtype, $infodb_file_handle, $oid, $oid_files);
1049 &dbutil::close_infodb_write_handle($infodbtype, $infodb_file_handle);
1050
1051 foreach my $rl (keys %$reverse_lookups) {
1052 $working_info->add_reverseinfo($rl,$oid);
1053 }
1054
1055 # meta files not set in reverse entry, but need to set the metadata flag
1056 if (defined $doc_obj->get_meta_files()) {
1057 foreach my $meta_file_rec(@{$doc_obj->get_meta_files()}) {
1058 my $full_file = (ref $meta_file_rec eq "ARRAY") ? $meta_file_rec->[0] : $meta_file_rec;
1059 $working_info->set_meta_file_flag($full_file);
1060 }
1061 }
1062}
1063
1064# This sub is called for every metadata.xml accepted for processing by by MetdataXMLPlugin
1065# and adds an entry into archiveinf-src.db for that file in the form:
1066# [@THISCOLLECTPATH@/import/metadata.xml]
1067# <meta-file>1
1068# This prevents blind reprocessing of the same old docs upon *incremental* building whenever
1069# we encounter a default empty metadata.xml that has no actual <FileSet> content defined.
1070sub add_metaxml_file_entry_to_archiveinfsrc {
1071 my $self = shift (@_);
1072 my ($full_file) = @_;
1073
1074 if ($self->{'verbosity'} > 2) {
1075 my $outhandle = $self->{'output_handle'};
1076 print $outhandle " Adding metaxml file entry for full_file: $full_file\n";
1077 }
1078
1079 my $working_info = $self->{'output_info'};
1080 $working_info->set_meta_file_flag($full_file);
1081}
1082
1083
1084sub set_sortmeta {
1085 my $self = shift (@_);
1086 my ($sortmeta, $removeprefix, $removesuffix) = @_;
1087
1088 $self->{'sortmeta'} = $sortmeta;
1089 if (defined ($removeprefix) && $removeprefix ) {
1090 $removeprefix =~ s/^\^//; # don't need a leading ^
1091 $self->{'removeprefix'} = $removeprefix;
1092 }
1093 if (defined ($removesuffix) && $removesuffix) {
1094 $removesuffix =~ s/\$$//; # don't need a trailing $
1095 $self->{'removesuffix'} = $removesuffix;
1096 }
1097}
1098
1099
1100
1101sub open_xslt_pipe
1102{
1103 my $self = shift @_;
1104 my ($output_file_name, $xslt_file)=@_;
1105
1106 return unless defined $xslt_file and $xslt_file ne "" and &FileUtils::fileExists($xslt_file);
1107
1108 my $java_class_path = &FileUtils::filenameConcatenate($ENV{'GSDLHOME'},"bin","java","ApplyXSLT.jar");
1109
1110 my $mapping_file_path = "";
1111
1112 if ($ENV{'GSDLOS'} eq "windows"){
1113 $java_class_path .=";".&FileUtils::filenameConcatenate($ENV{'GSDLHOME'},"bin","java","xalan.jar");
1114 # this file:/// bit didn't work for me on windows XP
1115 #$xslt_file = "\"file:///".$xslt_file."\"";
1116 #$mapping_file_path = "\"file:///";
1117 }
1118 else{
1119 $java_class_path .=":".&FileUtils::filenameConcatenate($ENV{'GSDLHOME'},"bin","java","xalan.jar");
1120 }
1121
1122
1123 $java_class_path = "\"".$java_class_path."\"";
1124
1125 my $cmd = "| java -cp $java_class_path org.nzdl.gsdl.ApplyXSLT -t \"$xslt_file\" ";
1126
1127 if (defined $self->{'mapping_file'} and $self->{'mapping_file'} ne ""){
1128 my $mapping_file_path = "\"".$self->{'mapping_file'}."\"";
1129 $cmd .= "-m $mapping_file_path";
1130 }
1131
1132 # The follow code is based around File GLOB
1133 # It has been commented out, and replaced with a filehandle variable approach
1134
1135# open(*XMLWRITER, $cmd)
1136# or die "can't open pipe to xslt: $!";
1137#
1138# $self->{'xslt_writer'} = *XMLWRITER;
1139#
1140# print XMLWRITER "<?DocStart?>\n";
1141# print XMLWRITER "$output_file_name\n";
1142
1143 open(my $FH_XMLWRITER, $cmd)
1144 or die "can't open pipe to xslt: $!";
1145
1146 $self->{'xslt_writer'} = $FH_XMLWRITER;
1147
1148 print $FH_XMLWRITER "<?DocStart?>\n";
1149 print $FH_XMLWRITER "$output_file_name\n";
1150 }
1151
1152
1153sub close_xslt_pipe
1154{
1155 my $self = shift @_;
1156
1157 return unless defined $self->{'xslt_writer'} ;
1158
1159 my $xsltwriter = $self->{'xslt_writer'};
1160
1161 print $xsltwriter "<?DocEnd?>\n";
1162 close($xsltwriter);
1163
1164 undef $self->{'xslt_writer'};
1165}
1166
1167
1168
1169#the subclass should implement this method if is_group method could return 1.
1170sub close_group_output{
1171 my $self = shift (@_);
1172}
1173
1174sub is_group {
1175 my $self = shift (@_);
1176 return 0;
1177}
1178
1179my $dc_set = { Title => 1,
1180 Creator => 1,
1181 Subject => 1,
1182 Description => 1,
1183 Publisher => 1,
1184 Contributor => 1,
1185 Date => 1,
1186 Type => 1,
1187 Format => 1,
1188 Identifier => 1,
1189 Source => 1,
1190 Language => 1,
1191 Relation => 1,
1192 Coverage => 1,
1193 Rights => 1};
1194
1195
1196# returns an XML representation of the dublin core metadata
1197# if dc meta is not found, try ex meta
1198# This method is not used by the DSpacePlugout, which has its
1199# own method to save its dc metadata
1200sub get_dc_metadata {
1201 my $self = shift(@_);
1202 my ($doc_obj, $section, $version) = @_;
1203
1204 # build up string of dublin core metadata
1205 $section="" unless defined $section;
1206
1207 my $section_ptr = $doc_obj->_lookup_section($section);
1208 return "" unless defined $section_ptr;
1209
1210
1211 my $explicit_dc = {};
1212 my $explicit_ex_dc = {};
1213 my $explicit_ex = {};
1214
1215 my $all_text="";
1216
1217 # We want high quality dc metadata to go in first, so we store all the
1218 # assigned dc.* values first. Then, for all those dc metadata names in
1219 # the official dc set that are as yet unassigned, we look to see whether
1220 # embedded ex.dc.* metadata has defined some values for them. If not,
1221 # then for the same missing dc metadata names, we look in ex metadata.
1222
1223 foreach my $data (@{$section_ptr->{'metadata'}}){
1224 my $escaped_value = &docprint::escape_text($data->[1]);
1225 if ($data->[0]=~ m/^dc\./) {
1226 $data->[0] =~ tr/[A-Z]/[a-z]/;
1227
1228 $data->[0] =~ m/^dc\.(.*)/;
1229 my $dc_element = $1;
1230
1231 if (!defined $explicit_dc->{$dc_element}) {
1232 $explicit_dc->{$dc_element} = [];
1233 }
1234 push(@{$explicit_dc->{$dc_element}},$escaped_value);
1235
1236 if (defined $version && ($version eq "oai_dc")) {
1237 $all_text .= " <dc:$dc_element>$escaped_value</dc:$dc_element>\n";
1238 }
1239 else {
1240 # qualifier???
1241 $all_text .= ' <dcvalue element="'. $dc_element.'">'. $escaped_value. "</dcvalue>\n";
1242 }
1243
1244 } elsif ($data->[0]=~ m/^ex\.dc\./) { # now look through ex.dc.* to fill in as yet unassigned fields in dc metaset
1245 $data->[0] =~ m/^ex\.dc\.(.*)/;
1246 my $ex_dc_element = $1;
1247 my $lc_ex_dc_element = lc($ex_dc_element);
1248
1249 # only store the ex.dc value for this dc metaname if no dc.* was assigned for it
1250 if (defined $dc_set->{$ex_dc_element}) {
1251 if (!defined $explicit_ex_dc->{$lc_ex_dc_element}) {
1252 $explicit_ex_dc->{$lc_ex_dc_element} = [];
1253 }
1254 push(@{$explicit_ex_dc->{$lc_ex_dc_element}},$escaped_value);
1255 }
1256 }
1257 elsif (($data->[0] =~ m/^ex\./) || ($data->[0] !~ m/\./)) { # look through ex. meta (incl. meta without prefix)
1258 $data->[0] =~ m/^(ex\.)?(.*)/;
1259 my $ex_element = $2;
1260 my $lc_ex_element = lc($ex_element);
1261
1262 if (defined $dc_set->{$ex_element}) {
1263 if (!defined $explicit_ex->{$lc_ex_element}) {
1264 $explicit_ex->{$lc_ex_element} = [];
1265 }
1266 push(@{$explicit_ex->{$lc_ex_element}},$escaped_value);
1267 }
1268 }
1269 }
1270
1271 # go through dc_set and for any element *not* defined in explicit_dc
1272 # that does exist in explicit_ex, add it in as metadata
1273 foreach my $k ( keys %$dc_set ) {
1274 my $lc_k = lc($k);
1275
1276 if (!defined $explicit_dc->{$lc_k}) {
1277 # try to find if ex.dc.* defines this dc.* meta,
1278 # if not, then look for whether there's an ex.* equivalent
1279
1280 if (defined $explicit_ex_dc->{$lc_k}) {
1281 foreach my $v (@{$explicit_ex_dc->{$lc_k}}) {
1282 my $dc_element = $lc_k;
1283 my $escaped_value = $v;
1284
1285 if (defined $version && ($version eq "oai_dc")) {
1286 $all_text .= " <dc:$dc_element>$escaped_value</dc:$dc_element>\n";
1287 }
1288 else {
1289 $all_text .= ' <dcvalue element="'. $dc_element.'">'. $escaped_value. "</dcvalue>\n";
1290 }
1291 }
1292 } elsif (defined $explicit_ex->{$lc_k}) {
1293 foreach my $v (@{$explicit_ex->{$lc_k}}) {
1294 my $dc_element = $lc_k;
1295 my $escaped_value = $v;
1296
1297 if (defined $version && ($version eq "oai_dc")) {
1298 $all_text .= " <dc:$dc_element>$escaped_value</dc:$dc_element>\n";
1299 }
1300 else {
1301 $all_text .= ' <dcvalue element="'. $dc_element.'">'. $escaped_value. "</dcvalue>\n";
1302 }
1303 }
1304 }
1305 }
1306 }
1307
1308 if ($all_text eq "") {
1309 $all_text .= " There is no Dublin Core metatdata in this document\n";
1310 }
1311 $all_text =~ s/[\x00-\x09\x0B\x0C\x0E-\x1F]//g;
1312
1313 return $all_text;
1314}
1315
1316# Build up dublin_core metadata. Priority given to dc.* over ex.*
1317# This method was apparently added by Jeffrey and committed by Shaoqun.
1318# But we don't know why it was added, so not using it anymore.
1319sub new_get_dc_metadata {
1320
1321 my $self = shift(@_);
1322 my ($doc_obj, $section, $version) = @_;
1323
1324 # build up string of dublin core metadata
1325 $section="" unless defined $section;
1326
1327 my $section_ptr=$doc_obj->_lookup_section($section);
1328 return "" unless defined $section_ptr;
1329
1330 my $all_text = "";
1331 foreach my $data (@{$section_ptr->{'metadata'}}){
1332 my $escaped_value = &docprint::escape_text($data->[1]);
1333 my $dc_element = $data->[0];
1334
1335 my @array = split('\.',$dc_element);
1336 my ($type,$name);
1337
1338 if(defined $array[1])
1339 {
1340 $type = $array[0];
1341 $name = $array[1];
1342 }
1343 else
1344 {
1345 $type = "ex";
1346 $name = $array[0];
1347 }
1348
1349 $all_text .= ' <Metadata Type="'. $type.'" Name="'.$name.'">'. $escaped_value. "</Metadata>\n";
1350 }
1351 return $all_text;
1352}
1353
1354
13551;
Note: See TracBrowser for help on using the repository browser.