MT2116 : Addons to the CSV Export
[koha.git] / C4 / Record.pm
blob7c3fcc5ebb6e63ca5d88cdc61650e835bc51c47c
1 package C4::Record;
3 # Copyright 2006 (C) LibLime
4 # Joshua Ferraro <jmf@liblime.com>
6 # This file is part of Koha.
8 # Koha is free software; you can redistribute it and/or modify it under the
9 # terms of the GNU General Public License as published by the Free Software
10 # Foundation; either version 2 of the License, or (at your option) any later
11 # version.
13 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
14 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
15 # A PARTICULAR PURPOSE. See the GNU General Public License for more details.
17 # You should have received a copy of the GNU General Public License along with
18 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19 # Suite 330, Boston, MA 02111-1307 USA
22 use strict;# use warnings; #FIXME: turn off warnings before release
24 # please specify in which methods a given module is used
25 use MARC::Record; # marc2marcxml, marcxml2marc, html2marc, changeEncoding
26 use MARC::File::XML; # marc2marcxml, marcxml2marc, html2marcxml, changeEncoding
27 use MARC::Crosswalk::DublinCore; # marc2dcxml
28 use Biblio::EndnoteStyle;
29 use Unicode::Normalize; # _entity_encode
30 use XML::LibXSLT;
31 use XML::LibXML;
32 use C4::Biblio; #marc2bibtex
33 use C4::Csv; #marc2csv
34 use C4::Koha; #marc2csv
35 use Text::CSV::Encoded; #marc2csv
37 use vars qw($VERSION @ISA @EXPORT);
39 # set the version for version checking
40 $VERSION = 3.00;
42 @ISA = qw(Exporter);
44 # only export API methods
46 @EXPORT = qw(
47 &marc2endnote
48 &marc2marc
49 &marc2marcxml
50 &marcxml2marc
51 &marc2dcxml
52 &marc2modsxml
53 &marc2bibtex
54 &marc2csv
56 &html2marcxml
57 &html2marc
58 &changeEncoding
61 =head1 NAME
63 C4::Record - MARC, MARCXML, DC, MODS, XML, etc. Record Management Functions and API
65 =head1 SYNOPSIS
67 New in Koha 3.x. This module handles all record-related management functions.
69 =head1 API (EXPORTED FUNCTIONS)
71 =head2 marc2marc - Convert from one flavour of ISO-2709 to another
73 =over 4
75 my ($error,$newmarc) = marc2marc($marc,$to_flavour,$from_flavour,$encoding);
77 Returns an ISO-2709 scalar
79 =back
81 =cut
83 sub marc2marc {
84 my ($marc,$to_flavour,$from_flavour,$encoding) = @_;
85 my $error = "Feature not yet implemented\n";
86 return ($error,$marc);
89 =head2 marc2marcxml - Convert from ISO-2709 to MARCXML
91 =over 4
93 my ($error,$marcxml) = marc2marcxml($marc,$encoding,$flavour);
95 Returns a MARCXML scalar
97 =over 2
99 C<$marc> - an ISO-2709 scalar or MARC::Record object
101 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
103 C<$flavour> - MARC21 or UNIMARC
105 C<$dont_entity_encode> - a flag that instructs marc2marcxml not to entity encode the xml before returning (optional)
107 =back
109 =back
111 =cut
113 sub marc2marcxml {
114 my ($marc,$encoding,$flavour,$dont_entity_encode) = @_;
115 my $error; # the error string
116 my $marcxml; # the final MARCXML scalar
118 # test if it's already a MARC::Record object, if not, make it one
119 my $marc_record_obj;
120 if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
121 $marc_record_obj = $marc;
122 } else { # it's not a MARC::Record object, make it one
123 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
125 # conversion to MARC::Record object failed, populate $error
126 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
128 # only proceed if no errors so far
129 unless ($error) {
131 # check the record for warnings
132 my @warnings = $marc_record_obj->warnings();
133 if (@warnings) {
134 warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
135 foreach my $warn (@warnings) { warn "\t".$warn };
137 unless($encoding) {$encoding = "UTF-8"}; # set default encoding
138 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set default MARC flavour
140 # attempt to convert the record to MARCXML
141 eval { $marcxml = $marc_record_obj->as_xml_record($flavour) }; #handle exceptions
143 # record creation failed, populate $error
144 if ($@) {
145 $error .= "Creation of MARCXML failed:".$MARC::File::ERROR;
146 $error .= "Additional information:\n";
147 my @warnings = $@->warnings();
148 foreach my $warn (@warnings) { $error.=$warn."\n" };
150 # record creation was successful
151 } else {
153 # check the record for warning flags again (warnings() will be cleared already if there was an error, see above block
154 @warnings = $marc_record_obj->warnings();
155 if (@warnings) {
156 warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
157 foreach my $warn (@warnings) { warn "\t".$warn };
161 # only proceed if no errors so far
162 unless ($error) {
164 # entity encode the XML unless instructed not to
165 unless ($dont_entity_encode) {
166 my ($marcxml_entity_encoded) = _entity_encode($marcxml);
167 $marcxml = $marcxml_entity_encoded;
171 # return result to calling program
172 return ($error,$marcxml);
175 =head2 marcxml2marc - Convert from MARCXML to ISO-2709
177 =over 4
179 my ($error,$marc) = marcxml2marc($marcxml,$encoding,$flavour);
181 Returns an ISO-2709 scalar
183 =over 2
185 C<$marcxml> - a MARCXML record
187 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
189 C<$flavour> - MARC21 or UNIMARC
191 =back
193 =back
195 =cut
197 sub marcxml2marc {
198 my ($marcxml,$encoding,$flavour) = @_;
199 my $error; # the error string
200 my $marc; # the final ISO-2709 scalar
201 unless($encoding) {$encoding = "UTF-8"}; # set the default encoding
202 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set the default MARC flavour
204 # attempt to do the conversion
205 eval { $marc = MARC::Record->new_from_xml($marcxml,$encoding,$flavour) }; # handle exceptions
207 # record creation failed, populate $error
208 if ($@) {$error .="\nCreation of MARCXML Record failed: ".$@;
209 $error.=$MARC::File::ERROR if ($MARC::File::ERROR);
211 # return result to calling program
212 return ($error,$marc);
215 =head2 marc2dcxml - Convert from ISO-2709 to Dublin Core
217 =over 4
219 my ($error,$dcxml) = marc2dcxml($marc,$qualified);
221 Returns a DublinCore::Record object, will eventually return a Dublin Core scalar
223 FIXME: should return actual XML, not just an object
225 =over 2
227 C<$marc> - an ISO-2709 scalar or MARC::Record object
229 C<$qualified> - specify whether qualified Dublin Core should be used in the input or output [0]
231 =back
233 =back
235 =cut
237 sub marc2dcxml {
238 my ($marc,$qualified) = @_;
239 my $error;
240 # test if it's already a MARC::Record object, if not, make it one
241 my $marc_record_obj;
242 if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
243 $marc_record_obj = $marc;
244 } else { # it's not a MARC::Record object, make it one
245 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
247 # conversion to MARC::Record object failed, populate $error
248 if ($@) {
249 $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR;
252 my $crosswalk = MARC::Crosswalk::DublinCore->new;
253 if ($qualified) {
254 $crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 1 );
256 my $dcxml = $crosswalk->as_dublincore($marc_record_obj);
257 my $dcxmlfinal = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n";
258 $dcxmlfinal .= "<metadata
259 xmlns=\"http://example.org/myapp/\"
260 xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"
261 xsi:schemaLocation=\"http://example.org/myapp/ http://example.org/myapp/schema.xsd\"
262 xmlns:dc=\"http://purl.org/dc/elements/1.1/\"
263 xmlns:dcterms=\"http://purl.org/dc/terms/\">";
265 foreach my $element ( $dcxml->elements() ) {
266 $dcxmlfinal.="<"."dc:".$element->name().">".$element->content()."</"."dc:".$element->name().">\n";
268 $dcxmlfinal .= "\n</metadata>";
269 return ($error,$dcxmlfinal);
271 =head2 marc2modsxml - Convert from ISO-2709 to MODS
273 =over 4
275 my ($error,$modsxml) = marc2modsxml($marc);
277 Returns a MODS scalar
279 =back
281 =cut
283 sub marc2modsxml {
284 my ($marc) = @_;
285 # grab the XML, run it through our stylesheet, push it out to the browser
286 my $xmlrecord = marc2marcxml($marc);
287 my $xslfile = C4::Context->config('intrahtdocs')."/prog/en/xslt/MARC21slim2MODS3-1.xsl";
288 my $parser = XML::LibXML->new();
289 my $xslt = XML::LibXSLT->new();
290 my $source = $parser->parse_string($xmlrecord);
291 my $style_doc = $parser->parse_file($xslfile);
292 my $stylesheet = $xslt->parse_stylesheet($style_doc);
293 my $results = $stylesheet->transform($source);
294 my $newxmlrecord = $stylesheet->output_string($results);
295 return ($newxmlrecord);
298 sub marc2endnote {
299 my ($marc) = @_;
300 my $marc_rec_obj = MARC::Record->new_from_usmarc($marc);
301 my $f260 = $marc_rec_obj->field('260');
302 my $f260a = $f260->subfield('a') if $f260;
303 my $f710 = $marc_rec_obj->field('710');
304 my $f710a = $f710->subfield('a') if $f710;
305 my $f500 = $marc_rec_obj->field('500');
306 my $abstract = $f500->subfield('a') if $f500;
307 my $fields = {
308 DB => C4::Context->preference("LibraryName"),
309 Title => $marc_rec_obj->title(),
310 Author => $marc_rec_obj->author(),
311 Publisher => $f710a,
312 City => $f260a,
313 Year => $marc_rec_obj->publication_date,
314 Abstract => $abstract,
316 my $endnote;
317 my $style = new Biblio::EndnoteStyle();
318 my $template;
319 $template.= "DB - DB\n" if C4::Context->preference("LibraryName");
320 $template.="T1 - Title\n" if $marc_rec_obj->title();
321 $template.="A1 - Author\n" if $marc_rec_obj->author();
322 $template.="PB - Publisher\n" if $f710a;
323 $template.="CY - City\n" if $f260a;
324 $template.="Y1 - Year\n" if $marc_rec_obj->publication_date;
325 $template.="AB - Abstract\n" if $abstract;
326 my ($text, $errmsg) = $style->format($template, $fields);
327 return ($text);
331 =head2 marc2csv - Convert from UNIMARC to CSV
333 =over 4
335 my ($csv) = marc2csv($record, $csvprofileid, $header);
337 Returns a CSV scalar
339 =over 2
341 C<$biblio> - a biblionumber
343 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id and the GetCsvProfiles function in C4::Csv)
345 C<$header> - true if the headers are to be printed (typically at first pass)
347 =back
349 =back
351 =cut
354 sub marc2csv {
355 my ($biblio, $id, $header) = @_;
356 my $output;
358 # Getting the record
359 my $record = GetMarcBiblio($biblio);
361 # Getting the framework
362 my $frameworkcode = GetFrameworkCode($biblio);
364 # Getting information about the csv profile
365 my $profile = GetCsvProfile($id);
367 # Getting output encoding
368 my $encoding = $profile->{encoding} || 'utf8';
370 # Getting separators
371 my $csvseparator = $profile->{csv_separator} || ',';
372 my $fieldseparator = $profile->{field_separator} || '#';
373 my $subfieldseparator = $profile->{subfield_separator} || '|';
375 # TODO: Be more generic (in case we have to handle other protected chars or more separators)
376 if ($csvseparator eq '\t') { $csvseparator = "\t" }
377 if ($fieldseparator eq '\t') { $fieldseparator = "\t" }
378 if ($subfieldseparator eq '\t') { $subfieldseparator = "\t" }
380 # Init CSV
381 my $csv = Text::CSV::Encoded->new({ sep_char => $csvseparator });
382 $csv = $csv->encoding_out($encoding) if ($encoding ne 'utf8');
384 # Getting the marcfields
385 my $marcfieldslist = $profile->{marcfields};
387 # Getting the marcfields as an array
388 my @marcfieldsarray = split('\|', $marcfieldslist);
390 # Separating the marcfields from the the user-supplied headers
391 my @marcfields;
392 foreach (@marcfieldsarray) {
393 my @result = split('=', $_);
394 if (scalar(@result) == 2) {
395 push @marcfields, { header => $result[0], field => $result[1] };
396 } else {
397 push @marcfields, { field => $result[0] }
401 # If we have to insert the headers
402 if ($header) {
403 my @marcfieldsheaders;
404 my $dbh = C4::Context->dbh;
406 # For each field or subfield
407 foreach (@marcfields) {
409 my $field = $_->{field};
411 # If we have a user-supplied header, we use it
412 if (exists $_->{header}) {
413 push @marcfieldsheaders, $_->{header};
414 } else {
415 # If not, we get the matching tag name from koha
416 if (index($field, '$') > 0) {
417 my ($fieldtag, $subfieldtag) = split('\$', $field);
418 my $query = "SELECT liblibrarian FROM marc_subfield_structure WHERE tagfield=? AND tagsubfield=?";
419 my $sth = $dbh->prepare($query);
420 $sth->execute($fieldtag, $subfieldtag);
421 my @results = $sth->fetchrow_array();
422 push @marcfieldsheaders, $results[0];
423 } else {
424 my $query = "SELECT liblibrarian FROM marc_tag_structure WHERE tagfield=?";
425 my $sth = $dbh->prepare($query);
426 $sth->execute($field);
427 my @results = $sth->fetchrow_array();
428 push @marcfieldsheaders, $results[0];
432 $csv->combine(@marcfieldsheaders);
433 $output = $csv->string() . "\n";
436 # For each marcfield to export
437 my @fieldstab;
438 foreach (@marcfields) {
439 my $marcfield = $_->{field};
440 # If it is a subfield
441 if (index($marcfield, '$') > 0) {
442 my ($fieldtag, $subfieldtag) = split('\$', $marcfield);
443 my @fields = $record->field($fieldtag);
444 my @tmpfields;
446 # For each field
447 foreach my $field (@fields) {
449 # We take every matching subfield
450 my @subfields = $field->subfield($subfieldtag);
451 foreach my $subfield (@subfields) {
453 # Getting authorised value
454 my $authvalues = GetKohaAuthorisedValuesFromField($fieldtag, $subfieldtag, $frameworkcode, undef);
455 push @tmpfields, (defined $authvalues->{$subfield}) ? $authvalues->{$subfield} : $subfield;
458 push (@fieldstab, join($subfieldseparator, @tmpfields));
459 # Or a field
460 } else {
461 my @fields = ($record->field($marcfield));
462 my $authvalues = GetKohaAuthorisedValuesFromField($marcfield, undef, $frameworkcode, undef);
463 push (@fieldstab, join($fieldseparator, map((defined $authvalues->{$_->as_string}) ? $authvalues->{$_->as_string} : $_->as_string, @fields)));
467 $csv->combine(@fieldstab);
468 $output .= $csv->string() . "\n";
470 return $output;
475 =head2 html2marcxml
477 =over 4
479 my ($error,$marcxml) = html2marcxml($tags,$subfields,$values,$indicator,$ind_tag);
481 Returns a MARCXML scalar
483 this is used in addbiblio.pl and additem.pl to build the MARCXML record from
484 the form submission.
486 FIXME: this could use some better code documentation
488 =back
490 =cut
492 sub html2marcxml {
493 my ($tags,$subfields,$values,$indicator,$ind_tag) = @_;
494 my $error;
495 # add the header info
496 my $marcxml= MARC::File::XML::header(C4::Context->preference('TemplateEncoding'),C4::Context->preference('marcflavour'));
498 # some flags used to figure out where in the record we are
499 my $prevvalue;
500 my $prevtag=-1;
501 my $first=1;
502 my $j = -1;
504 # handle characters that would cause the parser to choke FIXME: is there a more elegant solution?
505 for (my $i=0;$i<=@$tags;$i++){
506 @$values[$i] =~ s/&/&amp;/g;
507 @$values[$i] =~ s/</&lt;/g;
508 @$values[$i] =~ s/>/&gt;/g;
509 @$values[$i] =~ s/"/&quot;/g;
510 @$values[$i] =~ s/'/&apos;/g;
512 if ((@$tags[$i] ne $prevtag)){
513 $j++ unless (@$tags[$i] eq "");
514 #warn "IND:".substr(@$indicator[$j],0,1).substr(@$indicator[$j],1,1)." ".@$tags[$i];
515 if (!$first){
516 $marcxml.="</datafield>\n";
517 if ((@$tags[$i] > 10) && (@$values[$i] ne "")){
518 my $ind1 = substr(@$indicator[$j],0,1);
519 my $ind2 = substr(@$indicator[$j],1,1);
520 $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
521 $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
522 $first=0;
523 } else {
524 $first=1;
526 } else {
527 if (@$values[$i] ne "") {
528 # handle the leader
529 if (@$tags[$i] eq "000") {
530 $marcxml.="<leader>@$values[$i]</leader>\n";
531 $first=1;
532 # rest of the fixed fields
533 } elsif (@$tags[$i] lt '010') { # don't compare numerically 010 == 8
534 $marcxml.="<controlfield tag=\"@$tags[$i]\">@$values[$i]</controlfield>\n";
535 $first=1;
536 } else {
537 my $ind1 = substr(@$indicator[$j],0,1);
538 my $ind2 = substr(@$indicator[$j],1,1);
539 $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
540 $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
541 $first=0;
545 } else { # @$tags[$i] eq $prevtag
546 if (@$values[$i] eq "") {
547 } else {
548 if ($first){
549 my $ind1 = substr(@$indicator[$j],0,1);
550 my $ind2 = substr(@$indicator[$j],1,1);
551 $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
552 $first=0;
554 $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
557 $prevtag = @$tags[$i];
559 $marcxml.= MARC::File::XML::footer();
560 #warn $marcxml;
561 return ($error,$marcxml);
564 =head2 html2marc
566 =over 4
568 Probably best to avoid using this ... it has some rather striking problems:
570 =over 2
572 * saves blank subfields
574 * subfield order is hardcoded to always start with 'a' for repeatable tags (because it is hardcoded in the addfield routine).
576 * only possible to specify one set of indicators for each set of tags (ie, one for all the 650s). (because they were stored in a hash with the tag as the key).
578 * the underlying routines didn't support subfield reordering or subfield repeatability.
580 =back
582 I've left it in here because it could be useful if someone took the time to fix it. -- kados
584 =back
586 =cut
588 sub html2marc {
589 my ($dbh,$rtags,$rsubfields,$rvalues,%indicators) = @_;
590 my $prevtag = -1;
591 my $record = MARC::Record->new();
592 # my %subfieldlist=();
593 my $prevvalue; # if tag <10
594 my $field; # if tag >=10
595 for (my $i=0; $i< @$rtags; $i++) {
596 # rebuild MARC::Record
597 # warn "0=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ";
598 if (@$rtags[$i] ne $prevtag) {
599 if ($prevtag < 10) {
600 if ($prevvalue) {
601 if (($prevtag ne '000') && ($prevvalue ne "")) {
602 $record->add_fields((sprintf "%03s",$prevtag),$prevvalue);
603 } elsif ($prevvalue ne ""){
604 $record->leader($prevvalue);
607 } else {
608 if (($field) && ($field ne "")) {
609 $record->add_fields($field);
612 $indicators{@$rtags[$i]}.=' ';
613 # skip blank tags, I hope this works
614 if (@$rtags[$i] eq ''){
615 $prevtag = @$rtags[$i];
616 undef $field;
617 next;
619 if (@$rtags[$i] <10) {
620 $prevvalue= @$rvalues[$i];
621 undef $field;
622 } else {
623 undef $prevvalue;
624 if (@$rvalues[$i] eq "") {
625 undef $field;
626 } else {
627 $field = MARC::Field->new( (sprintf "%03s",@$rtags[$i]), substr($indicators{@$rtags[$i]},0,1),substr($indicators{@$rtags[$i]},1,1), @$rsubfields[$i] => @$rvalues[$i]);
629 # warn "1=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ".$field->as_formatted;
631 $prevtag = @$rtags[$i];
632 } else {
633 if (@$rtags[$i] <10) {
634 $prevvalue=@$rvalues[$i];
635 } else {
636 if (length(@$rvalues[$i])>0) {
637 $field->add_subfields(@$rsubfields[$i] => @$rvalues[$i]);
638 # warn "2=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ".$field->as_formatted;
641 $prevtag= @$rtags[$i];
645 # the last has not been included inside the loop... do it now !
646 #use Data::Dumper;
647 #warn Dumper($field->{_subfields});
648 $record->add_fields($field) if (($field) && $field ne "");
649 #warn "HTML2MARC=".$record->as_formatted;
650 return $record;
653 =head2 changeEncoding - Change the encoding of a record
655 =over 4
657 my ($error, $newrecord) = changeEncoding($record,$format,$flavour,$to_encoding,$from_encoding);
659 Changes the encoding of a record
661 =over 2
663 C<$record> - the record itself can be in ISO-2709, a MARC::Record object, or MARCXML for now (required)
665 C<$format> - MARC or MARCXML (required)
667 C<$flavour> - MARC21 or UNIMARC, if MARC21, it will change the leader (optional) [defaults to Koha system preference]
669 C<$to_encoding> - the encoding you want the record to end up in (optional) [UTF-8]
671 C<$from_encoding> - the encoding the record is currently in (optional, it will probably be able to tell unless there's a problem with the record)
673 =back
675 FIXME: the from_encoding doesn't work yet
677 FIXME: better handling for UNIMARC, it should allow management of 100 field
679 FIXME: shouldn't have to convert to and from xml/marc just to change encoding someone needs to re-write MARC::Record's 'encoding' method to actually alter the encoding rather than just changing the leader
681 =back
683 =cut
685 sub changeEncoding {
686 my ($record,$format,$flavour,$to_encoding,$from_encoding) = @_;
687 my $newrecord;
688 my $error;
689 unless($flavour) {$flavour = C4::Context->preference("marcflavour")};
690 unless($to_encoding) {$to_encoding = "UTF-8"};
692 # ISO-2709 Record (MARC21 or UNIMARC)
693 if (lc($format) =~ /^marc$/o) {
694 # if we're converting encoding of an ISO2709 file, we need to roundtrip through XML
695 # because MARC::Record doesn't directly provide us with an encoding method
696 # It's definitely less than idea and should be fixed eventually - kados
697 my $marcxml; # temporary storage of MARCXML scalar
698 ($error,$marcxml) = marc2marcxml($record,$to_encoding,$flavour);
699 unless ($error) {
700 ($error,$newrecord) = marcxml2marc($marcxml,$to_encoding,$flavour);
703 # MARCXML Record
704 } elsif (lc($format) =~ /^marcxml$/o) { # MARCXML Record
705 my $marc;
706 ($error,$marc) = marcxml2marc($record,$to_encoding,$flavour);
707 unless ($error) {
708 ($error,$newrecord) = marc2marcxml($record,$to_encoding,$flavour);
710 } else {
711 $error.="Unsupported record format:".$format;
713 return ($error,$newrecord);
716 =head2 marc2bibtex - Convert from MARC21 and UNIMARC to BibTex
718 =over 4
720 my ($bibtex) = marc2bibtex($record, $id);
722 Returns a BibTex scalar
724 =over 2
726 C<$record> - a MARC::Record object
728 C<$id> - an id for the BibTex record (might be the biblionumber)
730 =back
732 =back
734 =cut
737 sub marc2bibtex {
738 my ($record, $id) = @_;
739 my $tex;
741 # Authors
742 my $marcauthors = GetMarcAuthors($record,C4::Context->preference("marcflavour"));
743 my $author;
744 for my $authors ( map { map { @$_ } values %$_ } @$marcauthors ) {
745 $author .= " and " if ($author && $$authors{value});
746 $author .= $$authors{value} if ($$authors{value});
749 # Defining the conversion hash according to the marcflavour
750 my %bh;
751 if (C4::Context->preference("marcflavour") eq "UNIMARC") {
753 # FIXME, TODO : handle repeatable fields
754 # TODO : handle more types of documents
756 # Unimarc to bibtex hash
757 %bh = (
759 # Mandatory
760 author => $author,
761 title => $record->subfield("200", "a") || "",
762 editor => $record->subfield("210", "g") || "",
763 publisher => $record->subfield("210", "c") || "",
764 year => $record->subfield("210", "d") || $record->subfield("210", "h") || "",
766 # Optional
767 volume => $record->subfield("200", "v") || "",
768 series => $record->subfield("225", "a") || "",
769 address => $record->subfield("210", "a") || "",
770 edition => $record->subfield("205", "a") || "",
771 note => $record->subfield("300", "a") || "",
772 url => $record->subfield("856", "u") || ""
774 } else {
776 # Marc21 to bibtex hash
777 %bh = (
779 # Mandatory
780 author => $author,
781 title => $record->subfield("245", "a") || "",
782 editor => $record->subfield("260", "f") || "",
783 publisher => $record->subfield("260", "b") || "",
784 year => $record->subfield("260", "c") || $record->subfield("260", "g") || "",
786 # Optional
787 # unimarc to marc21 specification says not to convert 200$v to marc21
788 series => $record->subfield("490", "a") || "",
789 address => $record->subfield("260", "a") || "",
790 edition => $record->subfield("250", "a") || "",
791 note => $record->subfield("500", "a") || "",
792 url => $record->subfield("856", "u") || ""
796 $tex .= "\@book{";
797 $tex .= join(",\n", $id, map { $bh{$_} ? qq(\t$_ = "$bh{$_}") : () } keys %bh);
798 $tex .= "\n}\n";
800 return $tex;
804 =head1 INTERNAL FUNCTIONS
806 =head2 _entity_encode - Entity-encode an array of strings
808 =over 4
810 my ($entity_encoded_string) = _entity_encode($string);
814 my (@entity_encoded_strings) = _entity_encode(@strings);
816 Entity-encode an array of strings
818 =back
820 =cut
822 sub _entity_encode {
823 my @strings = @_;
824 my @strings_entity_encoded;
825 foreach my $string (@strings) {
826 my $nfc_string = NFC($string);
827 $nfc_string =~ s/([\x{0080}-\x{fffd}])/sprintf('&#x%X;',ord($1))/sgoe;
828 push @strings_entity_encoded, $nfc_string;
830 return @strings_entity_encoded;
833 END { } # module clean-up code here (global destructor)
835 __END__
837 =head1 AUTHOR
839 Joshua Ferraro <jmf@liblime.com>
841 =head1 MODIFICATIONS
844 =cut