source: gsdl/trunk/perllib/doc.pm@ 19617

Last change on this file since 19617 was 19617, checked in by kjdon, 12 years ago

use new util::tidy_up_oid instead of fiddling with it directly

  • Property svn:executable set to *
  • Property svn:keywords set to Author Date Id Revision
File size: 28.2 KB
Line 
1###########################################################################
2#
3# doc.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 redistr te 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# base class to hold documents
27
28package doc;
29eval {require bytes};
30
31BEGIN {
32 die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
33 unshift (@INC, "$ENV{'GSDLHOME'}/perllib/dynamic/lib/site_perl/5.005/i686-linux");
34}
35
36use strict;
37use unicode;
38use util;
39use ghtml;
40use File::stat;
41##use hashdoc;
42use docprint;
43
44# the document type may be indexed_doc, nonindexed_doc, or
45# classification
46
47our $OIDcount = 0;
48
49sub new {
50 my $class = shift (@_);
51 my ($source_filename, $doc_type, $rename_method) = @_;
52
53
54 my $self = bless {'associated_files'=>[],
55 'subsection_order'=>[],
56 'next_subsection'=>1,
57 'subsections'=>{},
58 'metadata'=>[],
59 'text'=>"",
60 'OIDtype'=>"hash"}, $class;
61
62 # used to set lastmodified here, but this can screw up the HASH ids, so
63 # the docsave processor now calls set_lastmodified
64
65 $self->{'source_path'} = $source_filename;
66
67 if (defined $source_filename) {
68 $source_filename = &util::filename_within_collection($source_filename);
69 print STDERR "****** doc.pm::new(): no file rename method provided\n" unless $rename_method;
70 $self->set_source_filename ($source_filename, $rename_method);
71 }
72
73 $self->set_doc_type ($doc_type) if defined $doc_type;
74
75 return $self;
76}
77# set lastmodified for OAI purposes, added by GRB, moved by kjdon
78sub set_lastmodified {
79 my $self = shift (@_);
80
81 my $source_path = $self->{'source_path'};
82
83 if (defined $source_path && (-e $source_path)) {
84
85 my $file_stat = stat($source_path);
86 my $mtime = $file_stat->mtime;
87 $self->add_utf8_metadata($self->get_top_section(), "lastmodified", $file_stat->mtime);
88 }
89}
90
91# clone the $self object
92sub duplicate {
93 my $self = shift (@_);
94
95 my $newobj = {};
96
97 foreach my $k (keys %$self) {
98 $newobj->{$k} = &clone ($self->{$k});
99 }
100
101 bless $newobj, ref($self);
102 return $newobj;
103}
104
105sub clone {
106 my ($from) = @_;
107 my $type = ref ($from);
108
109 if ($type eq "HASH") {
110 my $to = {};
111 foreach my $key (keys %$from) {
112 $to->{$key} = &clone ($from->{$key});
113 }
114 return $to;
115 } elsif ($type eq "ARRAY") {
116 my $to = [];
117 foreach my $v (@$from) {
118 push (@$to, &clone ($v));
119 }
120 return $to;
121 } else {
122 return $from;
123 }
124}
125
126sub set_OIDtype {
127 my $self = shift (@_);
128 my ($type, $metadata) = @_;
129
130 if (defined $type && $type =~ /^(hash|hash_on_file|hash_on_ga_xml|incremental|dirname|assigned)$/) {
131 $self->{'OIDtype'} = $type;
132 } else {
133 $self->{'OIDtype'} = "hash";
134 }
135
136 if ($type =~ /^assigned$/) {
137 if (defined $metadata) {
138 $self->{'OIDmetadata'} = $metadata;
139 } else {
140 $self->{'OIDmetadata'} = "dc.Identifier";
141 }
142 }
143}
144
145sub set_source_filename {
146 my $self = shift (@_);
147 my ($source_filename, $rename_method) = @_;
148
149 # Since the gsdlsourcefilename element goes into the doc.xml it has
150 # to be utf8. However, it should also *represent* the source filename
151 # (in the import directory) which may not be utf8 at all.
152 # For instance, if this meta element (gsdlsourcefilename) will be used
153 # by other applications that parse doc.xml in order to locate
154 # gsdlsourcefilename. Therefore, the solution is to URLencode or base64
155 # encode the real filename as this is a binary-to-text encoding meaning
156 # that the resulting string is ASCII (utf8). Decoding will give the original.
157
158# print STDERR "******URL/base64 encoding the gsdl_source_filename $source_filename ";
159
160 # URLencode just the gsdl_source_filename, not the directory. Then prepend dir
161 my ($srcfilename,$dirname,$suffix)
162 = &File::Basename::fileparse($source_filename, "\\.[^\\.]+\$");
163# print STDERR "-> $srcfilename -> ";
164 $srcfilename = &util::rename_file($srcfilename.$suffix, $rename_method);
165 $source_filename = &util::filename_cat($dirname, $srcfilename);
166# print STDERR "$source_filename\n";
167
168 $self->set_utf8_metadata_element ($self->get_top_section(),
169 "gsdlsourcefilename",
170 $source_filename);
171}
172
173sub set_converted_filename {
174 my $self = shift (@_);
175 my ($converted_filename) = @_;
176
177 # we know the converted filename is utf8
178 $self->set_utf8_metadata_element ($self->get_top_section(),
179 "gsdlconvertedfilename",
180 $converted_filename);
181}
182
183# returns the source_filename as it was provided
184sub get_source_filename {
185 my $self = shift (@_);
186
187 return $self->get_metadata_element ($self->get_top_section(), "gsdlsourcefilename");
188}
189
190# returns converted filename if available else returns source filename
191sub get_filename_for_hashing {
192 my $self = shift (@_);
193
194 my $filename = $self->get_metadata_element ($self->get_top_section(), "gsdlconvertedfilename");
195
196 if (!defined $filename) {
197 my $plugin_name = $self->get_metadata_element ($self->get_top_section(), "Plugin");
198 # if NULPlug processed file, then don't give a filename
199 if (defined $plugin_name && $plugin_name eq "NULPlug") {
200 $filename = undef;
201 } else { # returns the URL encoded source filename!
202 $filename = $self->get_metadata_element ($self->get_top_section(), "gsdlsourcefilename");
203 }
204 }
205 return $filename;
206}
207
208sub set_doc_type {
209 my $self = shift (@_);
210 my ($doc_type) = @_;
211
212 $self->set_metadata_element ($self->get_top_section(),
213 "gsdldoctype",
214 $doc_type);
215}
216
217# returns the gsdldoctype as it was provided
218# the default of "indexed_doc" is used if no document
219# type was provided
220sub get_doc_type {
221 my $self = shift (@_);
222
223 my $doc_type = $self->get_metadata_element ($self->get_top_section(), "gsdldoctype");
224 return $doc_type if (defined $doc_type);
225 return "indexed_doc";
226}
227
228
229# look up the reference to the a particular section
230sub _lookup_section {
231 my $self = shift (@_);
232 my ($section) = @_;
233
234 my ($num);
235 my $sectionref = $self;
236
237 while (defined $section && $section ne "") {
238
239 ($num, $section) = $section =~ /^\.?(\d+)(.*)$/;
240
241 $num =~ s/^0+(\d)/$1/ if defined $num ; # remove leading 0s
242
243 $section = "" unless defined $section;
244
245
246 if (defined $num && defined $sectionref->{'subsections'}->{$num}) {
247 $sectionref = $sectionref->{'subsections'}->{$num};
248 } else {
249 return undef;
250 }
251 }
252
253 return $sectionref;
254}
255
256# calculate OID by hashing the contents of the document
257sub _calc_OID {
258 my $self = shift (@_);
259 my ($filename) = @_;
260
261 my $osexe = &util::get_os_exe();
262
263 my $hashfile_exe = &util::filename_cat($ENV{'GSDLHOME'},"bin",
264 $ENV{'GSDLOS'},"hashfile$osexe");
265
266 my $result = "NULL";
267
268 if (-e "$hashfile_exe") {
269# $result = `\"$hashfile_exe\" \"$filename\"`;
270 $result = `hashfile$osexe \"$filename\"`;
271 ($result) = $result =~ /:\s*([0-9a-f]+)/i;
272 } else {
273 print STDERR "doc::_calc_OID $hashfile_exe could not be found\n";
274 }
275 return "HASH$result";
276}
277
278# methods dealing with OID, not groups of them.
279
280# if $OID is not provided one is calculated
281sub set_OID {
282 my $self = shift (@_);
283 my ($OID) = @_;
284
285 my $use_hash_oid = 0;
286 # if an OID wasn't provided calculate one
287 if (!defined $OID) {
288 $OID = "NULL";
289 if ($self->{'OIDtype'} =~ /^hash/) {
290 $use_hash_oid = 1;
291 } elsif ($self->{'OIDtype'} eq "incremental") {
292 $OID = "D" . $OIDcount;
293 $OIDcount ++;
294
295 } elsif ($self->{'OIDtype'} eq "dirname") {
296 $OID = 'J';
297 my $filename = $self->get_source_filename();
298 if (defined($filename)) { # && -e $filename) {
299 $OID = &File::Basename::dirname($filename);
300 if (defined $OID) {
301 $OID = 'J'.&File::Basename::basename($OID);
302 $OID = &util::tidy_up_oid($OID);
303 } else {
304 print STDERR "Failed to find base for filename ($filename)...generating hash id\n";
305 $use_hash_oid = 1;
306 }
307 } else {
308 print STDERR "Failed to find filename, generating hash id\n";
309 $use_hash_oid = 1;
310 }
311
312 } elsif ($self->{'OIDtype'} eq "assigned") {
313 my $identifier = $self->get_metadata_element ($self->get_top_section(), $self->{'OIDmetadata'});
314 if (defined $identifier && $identifier ne "") {
315 $OID = $identifier;
316 $OID = &util::tidy_up_oid($OID);
317 } else {
318 # need a hash id
319 print STDERR "no $self->{'OIDmetadata'} metadata found, generating hash id\n";
320 $use_hash_oid = 1;
321 }
322
323 } else {
324 $use_hash_oid = 1;
325 }
326
327 if ($use_hash_oid) {
328 my $hash_on_file = 1;
329 if ($self->{'OIDtype'} eq "hash_on_ga_xml") {
330 $hash_on_file = 0;
331 }
332 if ($hash_on_file) {
333 # "hash" OID - feed file to hashfile.exe
334 my $filename = $self->get_filename_for_hashing();
335 # -z: don't want to hash on the file if it is zero size
336 if (defined($filename) && -e $filename && !-z $filename) {
337 $OID = $self->_calc_OID ($filename);
338 } else {
339 $hash_on_file = 0;
340 }
341 }
342 if (!$hash_on_file) {
343 my $filename = &util::get_tmp_filename();
344 if (!open (OUTFILE, ">$filename")) {
345 print STDERR "doc::set_OID could not write to $filename\n";
346 } else {
347 my $doc_text = &docprint::get_section_xml($self, $self->get_top_section());
348 print OUTFILE $doc_text;
349 close (OUTFILE);
350 }
351 $OID = $self->_calc_OID ($filename);
352 &util::rm ($filename);
353 }
354 }
355 }
356 $self->set_metadata_element ($self->get_top_section(), "Identifier", $OID);
357}
358
359# this uses hashdoc (embedded c thingy) which is faster but still
360# needs a little work to be suffiently stable
361sub ___set_OID {
362 my $self = shift (@_);
363 my ($OID) = @_;
364
365 # if an OID wasn't provided then calculate hash value based on document
366 if (!defined $OID)
367 {
368 my $hash_text = &docprint::get_section_xml($self, $self->get_top_section());
369 my $hash_len = length($hash_text);
370
371 $OID = &hashdoc::buffer($hash_text,$hash_len);
372 }
373
374 $self->set_metadata_element ($self->get_top_section(), "Identifier", $OID);
375}
376
377# returns the OID for this document
378sub get_OID {
379 my $self = shift (@_);
380 my $OID = $self->get_metadata_element ($self->get_top_section(), "Identifier");
381 return $OID if (defined $OID);
382 return "NULL";
383}
384
385sub delete_OID {
386 my $self = shift (@_);
387
388 $self->set_metadata_element ($self->get_top_section(), "Identifier", "NULL");
389}
390
391
392# methods for manipulating section names
393
394# returns the name of the top-most section (the top
395# level of the document
396sub get_top_section {
397 my $self = shift (@_);
398
399 return "";
400}
401
402# returns a section
403sub get_parent_section {
404 my $self = shift (@_);
405 my ($section) = @_;
406
407 $section =~ s/(^|\.)\d+$//;
408
409 return $section;
410}
411
412# returns the first child section (or the end child
413# if there isn't any)
414sub get_begin_child {
415 my $self = shift (@_);
416 my ($section) = @_;
417
418 my $section_ptr = $self->_lookup_section($section);
419 return "" unless defined $section_ptr;
420
421 if (defined $section_ptr->{'subsection_order'}->[0]) {
422 return "$section.$section_ptr->{'subsection_order'}->[0]";
423 }
424
425 return $self->get_end_child ($section);
426}
427
428# returns the next child of a parent section
429sub get_next_child {
430 my $self = shift (@_);
431 my ($section) = @_;
432
433 my $parent_section = $self->get_parent_section($section);
434 my $parent_section_ptr = $self->_lookup_section($parent_section);
435 return undef unless defined $parent_section_ptr;
436
437 my ($section_num) = $section =~ /(\d+)$/;
438 return undef unless defined $section_num;
439
440 my $i = 0;
441 my $section_order = $parent_section_ptr->{'subsection_order'};
442 while ($i < scalar(@$section_order)) {
443 last if $section_order->[$i] eq $section_num;
444 $i++;
445 }
446
447 $i++; # the next child
448 if ($i < scalar(@$section_order)) {
449 return $section_order->[$i] if $parent_section eq "";
450 return "$parent_section.$section_order->[$i]";
451 }
452
453 # no more sections in this level
454 return undef;
455}
456
457# returns a reference to a list of children
458sub get_children {
459 my $self = shift (@_);
460 my ($section) = @_;
461
462 my $section_ptr = $self->_lookup_section($section);
463 return [] unless defined $section_ptr;
464
465 my @children = @{$section_ptr->{'subsection_order'}};
466
467 map {$_ = "$section.$_"; $_ =~ s/^\.+//;} @children;
468 return \@children;
469}
470
471# returns the child section one past the last one (which
472# is coded as "0")
473sub get_end_child {
474 my $self = shift (@_);
475 my ($section) = @_;
476
477 return $section . ".0" unless $section eq "";
478 return "0";
479}
480
481# returns the next section in book order
482sub get_next_section {
483 my $self = shift (@_);
484 my ($section) = @_;
485
486 return undef unless defined $section;
487
488 my $section_ptr = $self->_lookup_section($section);
489 return undef unless defined $section_ptr;
490
491 # first try to find first child
492 if (defined $section_ptr->{'subsection_order'}->[0]) {
493 return $section_ptr->{'subsection_order'}->[0] if ($section eq "");
494 return "$section.$section_ptr->{'subsection_order'}->[0]";
495 }
496
497 do {
498 # try to find sibling
499 my $next_child = $self->get_next_child ($section);
500 return $next_child if (defined $next_child);
501
502 # move up one level
503 $section = $self->get_parent_section ($section);
504 } while $section =~ /\d/;
505
506 return undef;
507}
508
509sub is_leaf_section {
510 my $self = shift (@_);
511 my ($section) = @_;
512
513 my $section_ptr = $self->_lookup_section($section);
514 return 1 unless defined $section_ptr;
515
516 return (scalar (@{$section_ptr->{'subsection_order'}}) == 0);
517}
518
519# methods for dealing with sections
520
521# returns the name of the inserted section
522sub insert_section {
523 my $self = shift (@_);
524 my ($before_section) = @_;
525
526 # get the child to insert before and its parent section
527 my $parent_section = "";
528 my $before_child = "0";
529 my @before_section = split (/\./, $before_section);
530 if (scalar(@before_section) > 0) {
531 $before_child = pop (@before_section);
532 $parent_section = join (".", @before_section);
533 }
534
535 my $parent_section_ptr = $self->_lookup_section($parent_section);
536 if (!defined $parent_section_ptr) {
537 print STDERR "doc::insert_section couldn't find parent section " .
538 "$parent_section\n";
539 return;
540 }
541
542 # get the next section number
543 my $section_num = $parent_section_ptr->{'next_subsection'}++;
544
545 my $i = 0;
546 while ($i < scalar(@{$parent_section_ptr->{'subsection_order'}}) &&
547 $parent_section_ptr->{'subsection_order'}->[$i] ne $before_child) {
548 $i++;
549 }
550
551 # insert the section number into the order list
552 splice (@{$parent_section_ptr->{'subsection_order'}}, $i, 0, $section_num);
553
554 # add this section to the parent section
555 my $section_ptr = {'subsection_order'=>[],
556 'next_subsection'=>1,
557 'subsections'=>{},
558 'metadata'=>[],
559 'text'=>""};
560 $parent_section_ptr->{'subsections'}->{$section_num} = $section_ptr;
561
562 # work out the full section number
563 my $section = $parent_section;
564 $section .= "." unless $section eq "";
565 $section .= $section_num;
566
567 return $section;
568}
569
570# creates a pre-named section
571sub create_named_section {
572 my $self = shift (@_);
573 my ($mastersection) = @_;
574
575 my ($num);
576 my $section = $mastersection;
577 my $sectionref = $self;
578
579 while ($section ne "") {
580 ($num, $section) = $section =~ /^\.?(\d+)(.*)$/;
581 $num =~ s/^0+(\d)/$1/; # remove leading 0s
582 $section = "" unless defined $section;
583
584 if (defined $num) {
585 if (!defined $sectionref->{'subsections'}->{$num}) {
586 push (@{$sectionref->{'subsection_order'}}, $num);
587 $sectionref->{'subsections'}->{$num} = {'subsection_order'=>[],
588 'next_subsection'=>1,
589 'subsections'=>{},
590 'metadata'=>[],
591 'text'=>""};
592 if ($num >= $sectionref->{'next_subsection'}) {
593 $sectionref->{'next_subsection'} = $num + 1;
594 }
595 }
596 $sectionref = $sectionref->{'subsections'}->{$num};
597
598 } else {
599 print STDERR "doc::create_named_section couldn't create section ";
600 print STDERR "$mastersection\n";
601 last;
602 }
603 }
604}
605
606# returns a reference to a list of subsections
607sub list_subsections {
608 my $self = shift (@_);
609 my ($section) = @_;
610
611 my $section_ptr = $self->_lookup_section ($section);
612 if (!defined $section_ptr) {
613 print STDERR "doc::list_subsections couldn't find section $section\n";
614 return [];
615 }
616
617 return [@{$section_ptr->{'subsection_order'}}];
618}
619
620sub delete_section {
621 my $self = shift (@_);
622 my ($section) = @_;
623
624# my $section_ptr = {'subsection_order'=>[],
625# 'next_subsection'=>1,
626# 'subsections'=>{},
627# 'metadata'=>[],
628# 'text'=>""};
629
630 # if this is the top section reset everything
631 if ($section eq "") {
632 $self->{'subsection_order'} = [];
633 $self->{'subsections'} = {};
634 $self->{'metadata'} = [];
635 $self->{'text'} = "";
636 return;
637 }
638
639 # find the parent of the section to delete
640 my $parent_section = "";
641 my $child = "0";
642 my @section = split (/\./, $section);
643 if (scalar(@section) > 0) {
644 $child = pop (@section);
645 $parent_section = join (".", @section);
646 }
647
648 my $parent_section_ptr = $self->_lookup_section($parent_section);
649 if (!defined $parent_section_ptr) {
650 print STDERR "doc::delete_section couldn't find parent section " .
651 "$parent_section\n";
652 return;
653 }
654
655 # remove this section from the subsection_order list
656 my $i = 0;
657 while ($i < scalar (@{$parent_section_ptr->{'subsection_order'}})) {
658 if ($parent_section_ptr->{'subsection_order'}->[$i] eq $child) {
659 splice (@{$parent_section_ptr->{'subsection_order'}}, $i, 1);
660 last;
661 }
662 $i++;
663 }
664
665 # remove this section from the subsection hash
666 if (defined ($parent_section_ptr->{'subsections'}->{$child})) {
667 undef $parent_section_ptr->{'subsections'}->{$child};
668 }
669}
670
671#--
672# methods for dealing with metadata
673
674# set_metadata_element and get_metadata_element are for metadata
675# which should only have one value. add_meta_data and get_metadata
676# are for metadata which can have more than one value.
677
678# returns the first metadata value which matches field
679
680# This version of get metadata element works much like the one above,
681# except it allows for the namespace portion of a metadata element to
682# be ignored, thus if you are searching for dc.Title, the first piece
683# of matching metadata ending with the name Title (once any namespace
684# is removed) would be returned.
685# 28-11-2003 John Thompson
686sub get_metadata_element {
687 my $self = shift (@_);
688 my ($section, $field, $ignore_namespace) = @_;
689 my ($data);
690
691 $ignore_namespace = 0 unless defined $ignore_namespace;
692
693 my $section_ptr = $self->_lookup_section($section);
694 if (!defined $section_ptr) {
695 print STDERR "doc::get_metadata_element couldn't find section ", $section, "\n";
696 return;
697 }
698
699 # Remove the any namespace if we are being told to ignore them
700 if($ignore_namespace) {
701 $field =~ s/^\w*\.//;
702 }
703
704 foreach $data (@{$section_ptr->{'metadata'}}) {
705
706 my $data_name = $data->[0];
707
708 # Remove the any namespace if we are being told to ignore them
709 if($ignore_namespace) {
710 $data_name =~ s/^\w*\.//;
711 }
712
713 return $data->[1] if (scalar(@$data) >= 2 && $data_name eq $field);
714 }
715
716 return undef; # was not found
717}
718
719# returns a list of the form [value1, value2, ...]
720sub get_metadata {
721 my $self = shift (@_);
722 my ($section, $field, $ignore_namespace) = @_;
723 my ($data);
724
725 $ignore_namespace = 0 unless defined $ignore_namespace;
726
727 my $section_ptr = $self->_lookup_section($section);
728 if (!defined $section_ptr) {
729 print STDERR "doc::get_metadata couldn't find section ",
730 $section, "\n";
731 return;
732 }
733
734 # Remove the any namespace if we are being told to ignore them
735 if($ignore_namespace) {
736 $field =~ s/^\w*\.//;
737 }
738
739 my @metadata = ();
740 foreach $data (@{$section_ptr->{'metadata'}}) {
741
742 my $data_name = $data->[0];
743 # Remove the any namespace if we are being told to ignore them
744 if($ignore_namespace) {
745 $data_name =~ s/^\w*\.//;
746 }
747
748 push (@metadata, $data->[1]) if ($data_name eq $field);
749 }
750
751 return \@metadata;
752}
753
754# returns a list of the form [[field,value],[field,value],...]
755sub get_all_metadata {
756 my $self = shift (@_);
757 my ($section) = @_;
758
759 my $section_ptr = $self->_lookup_section($section);
760 if (!defined $section_ptr) {
761 print STDERR "doc::get_all_metadata couldn't find section ", $section, "\n";
762 return;
763 }
764
765 return $section_ptr->{'metadata'};
766}
767
768# $value is optional
769sub delete_metadata {
770 my $self = shift (@_);
771 my ($section, $field, $value) = @_;
772
773 my $section_ptr = $self->_lookup_section($section);
774 if (!defined $section_ptr) {
775 print STDERR "doc::delete_metadata couldn't find section ", $section, "\n";
776 return;
777 }
778
779 my $i = 0;
780 while ($i < scalar (@{$section_ptr->{'metadata'}})) {
781 if (($section_ptr->{'metadata'}->[$i]->[0] eq $field) &&
782 (!defined $value || $section_ptr->{'metadata'}->[$i]->[1] eq $value)) {
783 splice (@{$section_ptr->{'metadata'}}, $i, 1);
784 } else {
785 $i++;
786 }
787 }
788}
789
790sub delete_all_metadata {
791 my $self = shift (@_);
792 my ($section) = @_;
793
794 my $section_ptr = $self->_lookup_section($section);
795 if (!defined $section_ptr) {
796 print STDERR "doc::delete_all_metadata couldn't find section ", $section, "\n";
797 return;
798 }
799
800 $section_ptr->{'metadata'} = [];
801}
802
803sub set_metadata_element {
804 my $self = shift (@_);
805 my ($section, $field, $value) = @_;
806
807 $self->set_utf8_metadata_element ($section, $field,
808 &unicode::ascii2utf8(\$value));
809}
810
811# set_utf8_metadata_element assumes the text has already been
812# converted to the UTF-8 encoding.
813sub set_utf8_metadata_element {
814 my $self = shift (@_);
815 my ($section, $field, $value) = @_;
816
817 $self->delete_metadata ($section, $field);
818 $self->add_utf8_metadata ($section, $field, $value);
819}
820
821
822# add_metadata assumes the text is in (extended) ascii form. For
823# text which has already been converted to the UTF-8 format use
824# add_utf8_metadata.
825sub add_metadata {
826 my $self = shift (@_);
827 my ($section, $field, $value) = @_;
828
829 $self->add_utf8_metadata ($section, $field,
830 &unicode::ascii2utf8(\$value));
831}
832
833sub add_utf8_metadata {
834 my $self = shift (@_);
835 my ($section, $field, $value) = @_;
836
837 my $section_ptr = $self->_lookup_section($section);
838 if (!defined $section_ptr) {
839 print STDERR "doc::add_utf8_metadata couldn't find section ", $section, "\n";
840 return;
841 }
842 if (!defined $value) {
843 print STDERR "doc::add_utf8_metadata undefined value for $field\n";
844 return;
845 }
846 if (!defined $field) {
847 print STDERR "doc::add_utf8_metadata undefined metadata type \n";
848 return;
849 }
850
851 #print STDERR "###$field=$value\n";
852 # double check that the value is utf-8
853 if (!&unicode::check_is_utf8($value)) {
854 print STDERR "doc::add_utf8_metadata - warning: '$field''s value $value wasn't utf8.";
855 &unicode::ensure_utf8(\$value);
856 print STDERR " Tried converting to utf8: $value\n";
857 }
858
859 push (@{$section_ptr->{'metadata'}}, [$field, $value]);
860}
861
862
863# methods for dealing with text
864
865# returns the text for a section
866sub get_text {
867 my $self = shift (@_);
868 my ($section) = @_;
869
870 my $section_ptr = $self->_lookup_section($section);
871 if (!defined $section_ptr) {
872 print STDERR "doc::get_text couldn't find section " .
873 "$section\n";
874 return "";
875 }
876
877 return $section_ptr->{'text'};
878}
879
880# returns the (utf-8 encoded) length of the text for a section
881sub get_text_length {
882 my $self = shift (@_);
883 my ($section) = @_;
884
885 my $section_ptr = $self->_lookup_section($section);
886 if (!defined $section_ptr) {
887 print STDERR "doc::get_text_length couldn't find section " .
888 "$section\n";
889 return 0;
890 }
891
892 return length ($section_ptr->{'text'});
893}
894
895sub delete_text {
896 my $self = shift (@_);
897 my ($section) = @_;
898
899 my $section_ptr = $self->_lookup_section($section);
900 if (!defined $section_ptr) {
901 print STDERR "doc::delete_text couldn't find section " .
902 "$section\n";
903 return;
904 }
905
906 $section_ptr->{'text'} = "";
907}
908
909# add_text assumes the text is in (extended) ascii form. For
910# text which has been already converted to the UTF-8 format
911# use add_utf8_text.
912sub add_text {
913 my $self = shift (@_);
914 my ($section, $text) = @_;
915
916 # convert the text to UTF-8 encoded unicode characters
917 # and add the text
918 $self->add_utf8_text($section, &unicode::ascii2utf8(\$text));
919}
920
921
922# add_utf8_text assumes the text to be added has already
923# been converted to the UTF-8 encoding. For ascii text use
924# add_text
925sub add_utf8_text {
926 my $self = shift (@_);
927 my ($section, $text) = @_;
928
929 my $section_ptr = $self->_lookup_section($section);
930 if (!defined $section_ptr) {
931 print STDERR "doc::add_utf8_text couldn't find section " .
932 "$section\n";
933 return;
934 }
935
936 $section_ptr->{'text'} .= $text;
937}
938
939# returns the Source meta, which is the utf8 filename generated.
940# Added a separate method here for convenience
941sub get_source {
942 my $self = shift (@_);
943 return $self->get_metadata_element ($self->get_top_section(), "Source");
944}
945
946# returns the SourceFile meta, which is the url reference to the URL-encoded
947# version of Source (the utf8 filename). Added a separate method here for convenience
948sub get_sourcefile {
949 my $self = shift (@_);
950 return $self->get_metadata_element ($self->get_top_section(), "SourceFile");
951}
952
953# Get the actual name of the assocfile, a url-encoded string derived from SourceFile.
954# The SourceFile meta is the (escaped) url reference to the url-encoded assocfile.
955sub get_assocfile_from_sourcefile {
956 my $self = shift (@_);
957
958 # get the SourceFile meta, which is a *URL* to a file on the filesystem
959 my $top_section = $self->get_top_section();
960 my $source_file = $self->get_metadata_element($top_section, "SourceFile");
961
962 # get the actual filename as it exists on the filesystem which this url refers to
963 $source_file = &unicode::url_to_filename($source_file);
964 my ($assocfilename) = $source_file =~ /([^\\\/]+)$/;
965 return $assocfilename;
966}
967
968# methods for dealing with associated files
969
970# a file is associated with a document, NOT a section.
971# if section is defined it is noted in the data structure
972# only so that files associated from a particular section
973# may be removed later (using delete_section_assoc_files)
974sub associate_file {
975 my $self = shift (@_);
976 my ($real_filename, $assoc_filename, $mime_type, $section) = @_;
977 $mime_type = &ghtml::guess_mime_type ($real_filename) unless defined $mime_type;
978
979 # remove all associated files with the same name
980 $self->delete_assoc_file ($assoc_filename);
981
982 push (@{$self->{'associated_files'}},
983 [$real_filename, $assoc_filename, $mime_type, $section]);
984}
985
986# returns a list of associated files in the form
987# [[real_filename, assoc_filename, mimetype], ...]
988sub get_assoc_files {
989 my $self = shift (@_);
990
991 return $self->{'associated_files'};
992}
993
994
995sub metadata_file {
996 my $self = shift (@_);
997 my ($real_filename, $filename) = @_;
998
999 push (@{$self->{'metadata_files'}},
1000 [$real_filename, $filename]);
1001}
1002
1003sub get_meta_files {
1004 my $self = shift (@_);
1005
1006 return $self->{'metadata_files'};
1007}
1008
1009sub delete_section_assoc_files {
1010 my $self = shift (@_);
1011 my ($section) = @_;
1012
1013 my $i=0;
1014 while ($i < scalar (@{$self->{'associated_files'}})) {
1015 if (defined $self->{'associated_files'}->[$i]->[3] &&
1016 $self->{'associated_files'}->[$i]->[3] eq $section) {
1017 splice (@{$self->{'associated_files'}}, $i, 1);
1018 } else {
1019 $i++;
1020 }
1021 }
1022}
1023
1024sub delete_assoc_file {
1025 my $self = shift (@_);
1026 my ($assoc_filename) = @_;
1027
1028 my $i=0;
1029 while ($i < scalar (@{$self->{'associated_files'}})) {
1030 if ($self->{'associated_files'}->[$i]->[1] eq $assoc_filename) {
1031 splice (@{$self->{'associated_files'}}, $i, 1);
1032 } else {
1033 $i++;
1034 }
1035 }
1036}
1037
1038sub reset_nextsection_ptr {
1039 my $self = shift (@_);
1040 my ($section) = @_;
1041
1042 my $section_ptr = $self->_lookup_section($section);
1043 $section_ptr->{'next_subsection'} = 1;
1044}
1045
10461;
Note: See TracBrowser for help on using the repository browser.