Bug 20892: Correct basketgroup link in histsearch.pl
[koha.git] / C4 / Record.pm
blobe741c4f63ee2972c6b6155968019065e667eb336
1 package C4::Record;
3 # Copyright 2006 (C) LibLime
4 # Parts copyright 2010 BibLibre
5 # Part copyright 2015 Universidad de El Salvador
7 # This file is part of Koha.
9 # Koha is free software; you can redistribute it and/or modify it
10 # under the terms of the GNU General Public License as published by
11 # the Free Software Foundation; either version 3 of the License, or
12 # (at your option) any later version.
14 # Koha is distributed in the hope that it will be useful, but
15 # WITHOUT ANY WARRANTY; without even the implied warranty of
16 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 # GNU General Public License for more details.
19 # You should have received a copy of the GNU General Public License
20 # along with Koha; if not, see <http://www.gnu.org/licenses>.
23 use strict;
24 #use warnings; FIXME - Bug 2505
26 # please specify in which methods a given module is used
27 use MARC::Record; # marc2marcxml, marcxml2marc, changeEncoding
28 use MARC::File::XML; # marc2marcxml, marcxml2marc, changeEncoding
29 use Biblio::EndnoteStyle;
30 use Unicode::Normalize; # _entity_encode
31 use C4::Biblio; #marc2bibtex
32 use C4::Koha; #marc2csv
33 use C4::XSLT ();
34 use YAML; #marcrecords2csv
35 use Template;
36 use Text::CSV::Encoded; #marc2csv
37 use Koha::Items;
38 use Koha::SimpleMARC qw(read_field);
39 use Koha::XSLT_Handler;
40 use Koha::CsvProfiles;
41 use Koha::AuthorisedValues;
42 use Carp;
44 use vars qw(@ISA @EXPORT);
47 @ISA = qw(Exporter);
49 # only export API methods
51 @EXPORT = qw(
52 &marc2endnote
53 &marc2marc
54 &marc2marcxml
55 &marcxml2marc
56 &marc2dcxml
57 &marc2modsxml
58 &marc2madsxml
59 &marc2bibtex
60 &marc2csv
61 &changeEncoding
64 =head1 NAME
66 C4::Record - MARC, MARCXML, DC, MODS, XML, etc. Record Management Functions and API
68 =head1 SYNOPSIS
70 New in Koha 3.x. This module handles all record-related management functions.
72 =head1 API (EXPORTED FUNCTIONS)
74 =head2 marc2marc - Convert from one flavour of ISO-2709 to another
76 my ($error,$newmarc) = marc2marc($marc,$to_flavour,$from_flavour,$encoding);
78 Returns an ISO-2709 scalar
80 =cut
82 sub marc2marc {
83 my ($marc,$to_flavour,$from_flavour,$encoding) = @_;
84 my $error;
85 if ($to_flavour =~ m/marcstd/) {
86 my $marc_record_obj;
87 if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
88 $marc_record_obj = $marc;
89 } else { # it's not a MARC::Record object, make it one
90 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
92 # conversion to MARC::Record object failed, populate $error
93 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
95 unless ($error) {
96 my @privatefields;
97 foreach my $field ($marc_record_obj->fields()) {
98 if ($field->tag() =~ m/9/ && ($field->tag() != '490' || C4::Context->preference("marcflavour") eq 'UNIMARC')) {
99 push @privatefields, $field;
100 } elsif (! ($field->is_control_field())) {
101 $field->delete_subfield(code => '9') if ($field->subfield('9'));
104 $marc_record_obj->delete_field($_) for @privatefields;
105 $marc = $marc_record_obj->as_usmarc();
107 } else {
108 $error = "Feature not yet implemented\n";
110 return ($error,$marc);
113 =head2 marc2marcxml - Convert from ISO-2709 to MARCXML
115 my ($error,$marcxml) = marc2marcxml($marc,$encoding,$flavour);
117 Returns a MARCXML scalar
119 C<$marc> - an ISO-2709 scalar or MARC::Record object
121 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
123 C<$flavour> - MARC21 or UNIMARC
125 C<$dont_entity_encode> - a flag that instructs marc2marcxml not to entity encode the xml before returning (optional)
127 =cut
129 sub marc2marcxml {
130 my ($marc,$encoding,$flavour,$dont_entity_encode) = @_;
131 my $error; # the error string
132 my $marcxml; # the final MARCXML scalar
134 # test if it's already a MARC::Record object, if not, make it one
135 my $marc_record_obj;
136 if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
137 $marc_record_obj = $marc;
138 } else { # it's not a MARC::Record object, make it one
139 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
141 # conversion to MARC::Record object failed, populate $error
142 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
144 # only proceed if no errors so far
145 unless ($error) {
147 # check the record for warnings
148 my @warnings = $marc_record_obj->warnings();
149 if (@warnings) {
150 warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
151 foreach my $warn (@warnings) { warn "\t".$warn };
153 unless($encoding) {$encoding = "UTF-8"}; # set default encoding
154 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set default MARC flavour
156 # attempt to convert the record to MARCXML
157 eval { $marcxml = $marc_record_obj->as_xml_record($flavour) }; #handle exceptions
159 # record creation failed, populate $error
160 if ($@) {
161 $error .= "Creation of MARCXML failed:".$MARC::File::ERROR;
162 $error .= "Additional information:\n";
163 my @warnings = $@->warnings();
164 foreach my $warn (@warnings) { $error.=$warn."\n" };
166 # record creation was successful
167 } else {
169 # check the record for warning flags again (warnings() will be cleared already if there was an error, see above block
170 @warnings = $marc_record_obj->warnings();
171 if (@warnings) {
172 warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
173 foreach my $warn (@warnings) { warn "\t".$warn };
177 # only proceed if no errors so far
178 unless ($error) {
180 # entity encode the XML unless instructed not to
181 unless ($dont_entity_encode) {
182 my ($marcxml_entity_encoded) = _entity_encode($marcxml);
183 $marcxml = $marcxml_entity_encoded;
187 # return result to calling program
188 return ($error,$marcxml);
191 =head2 marcxml2marc - Convert from MARCXML to ISO-2709
193 my ($error,$marc) = marcxml2marc($marcxml,$encoding,$flavour);
195 Returns an ISO-2709 scalar
197 C<$marcxml> - a MARCXML record
199 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
201 C<$flavour> - MARC21 or UNIMARC
203 =cut
205 sub marcxml2marc {
206 my ($marcxml,$encoding,$flavour) = @_;
207 my $error; # the error string
208 my $marc; # the final ISO-2709 scalar
209 unless($encoding) {$encoding = "UTF-8"}; # set the default encoding
210 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set the default MARC flavour
212 # attempt to do the conversion
213 eval { $marc = MARC::Record->new_from_xml($marcxml,$encoding,$flavour) }; # handle exceptions
215 # record creation failed, populate $error
216 if ($@) {$error .="\nCreation of MARCXML Record failed: ".$@;
217 $error.=$MARC::File::ERROR if ($MARC::File::ERROR);
219 # return result to calling program
220 return ($error,$marc);
223 =head2 marc2dcxml - Convert from ISO-2709 to Dublin Core
225 my dcxml = marc2dcxml ($marc, $xml, $biblionumber, $format);
227 EXAMPLE
229 my dcxml = marc2dcxml (undef, undef, 1, "oaidc");
231 Convert MARC or MARCXML to Dublin Core metadata (XSLT Transformation),
232 optionally can get an XML directly from biblio_metadata
233 without item information. This method take into consideration the syspref
234 'marcflavour' (UNIMARC, MARC21 and NORMARC).
235 Return an XML file with the format defined in C<$format>
237 C<$marc> - an ISO-2709 scalar or MARC::Record object
239 C<$xml> - a MARCXML file
241 C<$biblionumber> - biblionumber for database access
243 C<$format> - accept three type of DC formats (oaidc, srwdc, and rdfdc )
245 =cut
247 sub marc2dcxml {
248 my ( $marc, $xml, $biblionumber, $format ) = @_;
250 # global variables
251 my ( $marcxml, $record, $output );
253 # set the default path for intranet xslts
254 # differents xslts to process (OAIDC, SRWDC and RDFDC)
255 my $xsl = C4::Context->config('intrahtdocs') . '/prog/en/xslt/' .
256 C4::Context->preference('marcflavour') . 'slim2' . uc ( $format ) . '.xsl';
258 if ( defined $marc ) {
259 # no need to catch errors or warnings marc2marcxml do it instead
260 $marcxml = C4::Record::marc2marcxml( $marc );
261 } elsif ( not defined $xml and defined $biblionumber ) {
262 # get MARCXML biblio directly without item information
263 $marcxml = C4::Biblio::GetXmlBiblio( $biblionumber );
264 } else {
265 $marcxml = $xml;
268 # only proceed if MARC21 or UNIMARC; else clause is executed if marcflavour set it to NORMARC
269 # generate MARC::Record object to see if not a marcxml record
270 unless ( C4::Context->preference('marcflavour') eq 'NORMARC' ) {
271 eval { $record = MARC::Record->new_from_xml(
272 $marcxml,
273 'UTF-8',
274 C4::Context->preference('marcflavour')
277 } else {
278 eval { $record = MARC::Record->new_from_xml(
279 $marcxml,
280 'UTF-8',
281 'MARC21'
286 # conversion to MARC::Record object failed
287 if ( $@ ) {
288 croak "Creation of MARC::Record object failed.";
289 } elsif ( $record->warnings() ) {
290 carp "Warnings encountered while processing ISO-2709 record.\n";
291 my @warnings = $record->warnings();
292 foreach my $warn (@warnings) {
293 carp "\t". $warn;
295 } elsif ( $record =~ /^MARC::Record/ ) { # if OK makes xslt transformation
296 my $xslt_engine = Koha::XSLT_Handler->new;
297 if ( $format =~ /^(dc|oaidc|srwdc|rdfdc)$/i ) {
298 $output = $xslt_engine->transform( $marcxml, $xsl );
299 } else {
300 croak "The format argument ($format) not accepted.\n" .
301 "Please pass a valid format (oaidc, srwdc, or rdfdc)\n";
303 my $err = $xslt_engine->err; # error number
304 my $errstr = $xslt_engine->errstr; # error message
305 if ( $err ) {
306 croak "Error when processing $errstr Error number: $err\n";
307 } else {
308 return $output;
313 =head2 marc2modsxml - Convert from ISO-2709 to MODS
315 my $modsxml = marc2modsxml($marc);
317 Returns a MODS scalar
319 =cut
321 sub marc2modsxml {
322 my ($marc) = @_;
323 return _transformWithStylesheet($marc, "/prog/en/xslt/MARC21slim2MODS3-1.xsl");
326 =head2 marc2madsxml - Convert from ISO-2709 to MADS
328 my $madsxml = marc2madsxml($marc);
330 Returns a MADS scalar
332 =cut
334 sub marc2madsxml {
335 my ($marc) = @_;
336 return _transformWithStylesheet($marc, "/prog/en/xslt/MARC21slim2MADS.xsl");
339 =head2 _transformWithStylesheet - Transform a MARC record with a stylesheet
341 my $xml = _transformWithStylesheet($marc, $stylesheet)
343 Returns the XML scalar result of the transformation. $stylesheet should
344 contain the path to a stylesheet under intrahtdocs.
346 =cut
348 sub _transformWithStylesheet {
349 my ($marc, $stylesheet) = @_;
350 # grab the XML, run it through our stylesheet, push it out to the browser
351 my $xmlrecord = marc2marcxml($marc);
352 my $xslfile = C4::Context->config('intrahtdocs') . $stylesheet;
353 return C4::XSLT::engine->transform($xmlrecord, $xslfile);
356 sub marc2endnote {
357 my ($marc) = @_;
358 my $marc_rec_obj = MARC::Record->new_from_usmarc($marc);
359 my ( $abstract, $f260a, $f710a );
360 my $f260 = $marc_rec_obj->field('260');
361 if ($f260) {
362 $f260a = $f260->subfield('a') if $f260;
364 my $f710 = $marc_rec_obj->field('710');
365 if ($f710) {
366 $f710a = $f710->subfield('a');
368 my $f500 = $marc_rec_obj->field('500');
369 if ($f500) {
370 $abstract = $f500->subfield('a');
372 my $fields = {
373 DB => C4::Context->preference("LibraryName"),
374 Title => $marc_rec_obj->title(),
375 Author => $marc_rec_obj->author(),
376 Publisher => $f710a,
377 City => $f260a,
378 Year => $marc_rec_obj->publication_date,
379 Abstract => $abstract,
381 my $endnote;
382 my $style = new Biblio::EndnoteStyle();
383 my $template;
384 $template.= "DB - DB\n" if C4::Context->preference("LibraryName");
385 $template.="T1 - Title\n" if $marc_rec_obj->title();
386 $template.="A1 - Author\n" if $marc_rec_obj->author();
387 $template.="PB - Publisher\n" if $f710a;
388 $template.="CY - City\n" if $f260a;
389 $template.="Y1 - Year\n" if $marc_rec_obj->publication_date;
390 $template.="AB - Abstract\n" if $abstract;
391 my ($text, $errmsg) = $style->format($template, $fields);
392 return ($text);
396 =head2 marc2csv - Convert several records from UNIMARC to CSV
398 my ($csv) = marc2csv($biblios, $csvprofileid, $itemnumbers);
400 Pre and postprocessing can be done through a YAML file
402 Returns a CSV scalar
404 C<$biblio> - a list of biblionumbers
406 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id)
408 C<$itemnumbers> - a list of itemnumbers to export
410 =cut
412 sub marc2csv {
413 my ($biblios, $id, $itemnumbers) = @_;
414 $itemnumbers ||= [];
415 my $output;
416 my $csv = Text::CSV::Encoded->new();
418 # Getting yaml file
419 my $configfile = "../tools/csv-profiles/$id.yaml";
420 my ($preprocess, $postprocess, $fieldprocessing);
421 if (-e $configfile){
422 ($preprocess,$postprocess, $fieldprocessing) = YAML::LoadFile($configfile);
425 # Preprocessing
426 eval $preprocess if ($preprocess);
428 my $firstpass = 1;
429 if ( @$itemnumbers ) {
430 for my $itemnumber ( @$itemnumbers) {
431 my $item = Koha::Items->find( $itemnumber );
432 my $biblionumber = $item->biblio->biblionumber;
433 $output .= marcrecord2csv( $biblionumber, $id, $firstpass, $csv, $fieldprocessing, [$itemnumber] );
434 $firstpass = 0;
436 } else {
437 foreach my $biblio (@$biblios) {
438 $output .= marcrecord2csv( $biblio, $id, $firstpass, $csv, $fieldprocessing );
439 $firstpass = 0;
443 # Postprocessing
444 eval $postprocess if ($postprocess);
446 return $output;
449 =head2 marcrecord2csv - Convert a single record from UNIMARC to CSV
451 my ($csv) = marcrecord2csv($biblio, $csvprofileid, $header);
453 Returns a CSV scalar
455 C<$biblio> - a biblionumber
457 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id)
459 C<$header> - true if the headers are to be printed (typically at first pass)
461 C<$csv> - an already initialised Text::CSV object
463 C<$fieldprocessing>
465 C<$itemnumbers> a list of itemnumbers to export
467 =cut
469 sub marcrecord2csv {
470 my ($biblio, $id, $header, $csv, $fieldprocessing, $itemnumbers) = @_;
471 my $output;
473 # Getting the record
474 my $record = GetMarcBiblio({ biblionumber => $biblio });
475 return unless $record;
476 C4::Biblio::EmbedItemsInMarcBiblio( $record, $biblio, $itemnumbers );
477 # Getting the framework
478 my $frameworkcode = GetFrameworkCode($biblio);
480 # Getting information about the csv profile
481 my $profile = Koha::CsvProfiles->find($id);
483 # Getting output encoding
484 my $encoding = $profile->encoding || 'utf8';
485 # Getting separators
486 my $csvseparator = $profile->csv_separator || ',';
487 my $fieldseparator = $profile->field_separator || '#';
488 my $subfieldseparator = $profile->subfield_separator || '|';
490 # TODO: Be more generic (in case we have to handle other protected chars or more separators)
491 if ($csvseparator eq '\t') { $csvseparator = "\t" }
492 if ($fieldseparator eq '\t') { $fieldseparator = "\t" }
493 if ($subfieldseparator eq '\t') { $subfieldseparator = "\t" }
494 if ($csvseparator eq '\n') { $csvseparator = "\n" }
495 if ($fieldseparator eq '\n') { $fieldseparator = "\n" }
496 if ($subfieldseparator eq '\n') { $subfieldseparator = "\n" }
498 $csv = $csv->encoding_out($encoding) ;
499 $csv->sep_char($csvseparator);
501 # Getting the marcfields
502 my $marcfieldslist = $profile->content;
504 # Getting the marcfields as an array
505 my @marcfieldsarray = split('\|', $marcfieldslist);
507 # Separating the marcfields from the user-supplied headers
508 my @csv_structures;
509 foreach (@marcfieldsarray) {
510 my @result = split('=', $_, 2);
511 my $content = ( @result == 2 )
512 ? $result[1]
513 : $result[0];
514 my @fields;
515 while ( $content =~ m|(\d{3})\$?(.)?|g ) {
516 my $fieldtag = $1;
517 my $subfieldtag = $2 || undef;
518 push @fields, { fieldtag => $fieldtag, subfieldtag => $subfieldtag };
520 if ( @result == 2) {
521 push @csv_structures, { header => $result[0], content => $content, fields => \@fields };
522 } else {
523 push @csv_structures, { content => $content, fields => \@fields }
527 my ( @marcfieldsheaders, @csv_rows );
528 my $dbh = C4::Context->dbh;
530 my $field_list;
531 for my $field ( $record->fields ) {
532 my $fieldtag = $field->tag;
533 my $values;
534 if ( $field->is_control_field ) {
535 $values = $field->data();
536 } else {
537 $values->{indicator}{1} = $field->indicator(1);
538 $values->{indicator}{2} = $field->indicator(2);
539 for my $subfield ( $field->subfields ) {
540 my $subfieldtag = $subfield->[0];
541 my $value = $subfield->[1];
542 push @{ $values->{$subfieldtag} }, $value;
545 # We force the key as an integer (trick for 00X and OXX fields)
546 push @{ $field_list->{fields}{0+$fieldtag} }, $values;
549 # For each field or subfield
550 foreach my $csv_structure (@csv_structures) {
551 my @field_values;
552 my $tags = $csv_structure->{fields};
553 my $content = $csv_structure->{content};
555 if ( $header ) {
556 # If we have a user-supplied header, we use it
557 if ( exists $csv_structure->{header} ) {
558 push @marcfieldsheaders, $csv_structure->{header};
559 } else {
560 # If not, we get the matching tag name from koha
561 my $tag = $tags->[0];
562 if ( $tag->{subfieldtag} ) {
563 my $query = "SELECT liblibrarian FROM marc_subfield_structure WHERE tagfield=? AND tagsubfield=?";
564 my @results = $dbh->selectrow_array( $query, {}, $tag->{fieldtag}, $tag->{subfieldtag} );
565 push @marcfieldsheaders, $results[0];
566 } else {
567 my $query = "SELECT liblibrarian FROM marc_tag_structure WHERE tagfield=?";
568 my @results = $dbh->selectrow_array( $query, {}, $tag->{fieldtag} );
569 push @marcfieldsheaders, $results[0];
574 # TT tags exist
575 if ( $content =~ m|\[\%.*\%\]| ) {
576 my $tt = Template->new();
577 my $template = $content;
578 my $vars;
579 # Replace 00X and 0XX with X or XX
580 $content =~ s|fields.00(\d)|fields.$1|g;
581 $content =~ s|fields.0(\d{2})|fields.$1|g;
582 my $tt_output;
583 $tt->process( \$content, $field_list, \$tt_output );
584 push @csv_rows, $tt_output;
585 } else {
586 for my $tag ( @$tags ) {
587 my @fields = $record->field( $tag->{fieldtag} );
588 # If it is a subfield
589 my @loop_values;
590 if ( $tag->{subfieldtag} ) {
591 my $av = Koha::AuthorisedValues->search_by_marc_field({ frameworkcode => $frameworkcode, tagfield => $tag->{fieldtag}, tagsubfield => $tag->{subfieldtag}, });
592 $av = $av->count ? $av->unblessed : [];
593 my $av_description_mapping = { map { ( $_->{authorised_value} => $_->{lib} ) } @$av };
594 # For each field
595 foreach my $field (@fields) {
596 my @subfields = $field->subfield( $tag->{subfieldtag} );
597 foreach my $subfield (@subfields) {
598 push @loop_values, (defined $av_description_mapping->{$subfield}) ? $av_description_mapping->{$subfield} : $subfield;
602 # Or a field
603 } else {
604 my $av = Koha::AuthorisedValues->search_by_marc_field({ frameworkcode => $frameworkcode, tagfield => $tag->{fieldtag}, });
605 $av = $av->count ? $av->unblessed : [];
606 my $authvalues = { map { ( $_->{authorised_value} => $_->{lib} ) } @$av };
608 foreach my $field ( @fields ) {
609 my $value;
611 # If it is a control field
612 if ($field->is_control_field) {
613 $value = defined $authvalues->{$field->as_string} ? $authvalues->{$field->as_string} : $field->as_string;
614 } else {
615 # If it is a field, we gather all subfields, joined by the subfield separator
616 my @subvaluesarray;
617 my @subfields = $field->subfields;
618 foreach my $subfield (@subfields) {
619 push (@subvaluesarray, defined $authvalues->{$subfield->[1]} ? $authvalues->{$subfield->[1]} : $subfield->[1]);
621 $value = join ($subfieldseparator, @subvaluesarray);
624 # Field processing
625 my $marcfield = $tag->{fieldtag}; # This line fixes a retrocompatibility concern
626 # The "processing" could be based on the $marcfield variable.
627 eval $fieldprocessing if ($fieldprocessing);
629 push @loop_values, $value;
633 push @field_values, {
634 fieldtag => $tag->{fieldtag},
635 subfieldtag => $tag->{subfieldtag},
636 values => \@loop_values,
639 for my $field_value ( @field_values ) {
640 if ( $field_value->{subfieldtag} ) {
641 push @csv_rows, join( $subfieldseparator, @{ $field_value->{values} } );
642 } else {
643 push @csv_rows, join( $fieldseparator, @{ $field_value->{values} } );
650 if ( $header ) {
651 $csv->combine(@marcfieldsheaders);
652 $output = $csv->string() . "\n";
654 $csv->combine(@csv_rows);
655 $output .= $csv->string() . "\n";
657 return $output;
662 =head2 changeEncoding - Change the encoding of a record
664 my ($error, $newrecord) = changeEncoding($record,$format,$flavour,$to_encoding,$from_encoding);
666 Changes the encoding of a record
668 C<$record> - the record itself can be in ISO-2709, a MARC::Record object, or MARCXML for now (required)
670 C<$format> - MARC or MARCXML (required)
672 C<$flavour> - MARC21 or UNIMARC, if MARC21, it will change the leader (optional) [defaults to Koha system preference]
674 C<$to_encoding> - the encoding you want the record to end up in (optional) [UTF-8]
676 C<$from_encoding> - the encoding the record is currently in (optional, it will probably be able to tell unless there's a problem with the record)
678 FIXME: the from_encoding doesn't work yet
680 FIXME: better handling for UNIMARC, it should allow management of 100 field
682 FIXME: shouldn't have to convert to and from xml/marc just to change encoding someone needs to re-write MARC::Record's 'encoding' method to actually alter the encoding rather than just changing the leader
684 =cut
686 sub changeEncoding {
687 my ($record,$format,$flavour,$to_encoding,$from_encoding) = @_;
688 my $newrecord;
689 my $error;
690 unless($flavour) {$flavour = C4::Context->preference("marcflavour")};
691 unless($to_encoding) {$to_encoding = "UTF-8"};
693 # ISO-2709 Record (MARC21 or UNIMARC)
694 if (lc($format) =~ /^marc$/o) {
695 # if we're converting encoding of an ISO2709 file, we need to roundtrip through XML
696 # because MARC::Record doesn't directly provide us with an encoding method
697 # It's definitely less than idea and should be fixed eventually - kados
698 my $marcxml; # temporary storage of MARCXML scalar
699 ($error,$marcxml) = marc2marcxml($record,$to_encoding,$flavour);
700 unless ($error) {
701 ($error,$newrecord) = marcxml2marc($marcxml,$to_encoding,$flavour);
704 # MARCXML Record
705 } elsif (lc($format) =~ /^marcxml$/o) { # MARCXML Record
706 my $marc;
707 ($error,$marc) = marcxml2marc($record,$to_encoding,$flavour);
708 unless ($error) {
709 ($error,$newrecord) = marc2marcxml($record,$to_encoding,$flavour);
711 } else {
712 $error.="Unsupported record format:".$format;
714 return ($error,$newrecord);
717 =head2 marc2bibtex - Convert from MARC21 and UNIMARC to BibTex
719 my ($bibtex) = marc2bibtex($record, $id);
721 Returns a BibTex scalar
723 C<$record> - a MARC::Record object
725 C<$id> - an id for the BibTex record (might be the biblionumber)
727 =cut
730 sub marc2bibtex {
731 my ($record, $id) = @_;
732 my $tex;
733 my $marcflavour = C4::Context->preference("marcflavour");
735 # Authors
736 my $author;
737 my @texauthors;
738 my @authorFields = ('100','110','111','700','710','711');
739 @authorFields = ('700','701','702','710','711','721') if ( $marcflavour eq "UNIMARC" );
741 foreach my $field ( @authorFields ) {
742 # author formatted surname, firstname
743 my $texauthor = '';
744 if ( $marcflavour eq "UNIMARC" ) {
745 $texauthor = join ', ',
746 ( $record->subfield($field,"a"), $record->subfield($field,"b") );
747 } else {
748 $texauthor = $record->subfield($field,"a");
750 push @texauthors, $texauthor if $texauthor;
752 $author = join ' and ', @texauthors;
754 # Defining the conversion array according to the marcflavour
755 my @bh;
756 if ( $marcflavour eq "UNIMARC" ) {
758 # FIXME, TODO : handle repeatable fields
759 # TODO : handle more types of documents
761 # Unimarc to bibtex array
762 @bh = (
764 # Mandatory
765 author => $author,
766 title => $record->subfield("200", "a") || "",
767 editor => $record->subfield("210", "g") || "",
768 publisher => $record->subfield("210", "c") || "",
769 year => $record->subfield("210", "d") || $record->subfield("210", "h") || "",
771 # Optional
772 volume => $record->subfield("200", "v") || "",
773 series => $record->subfield("225", "a") || "",
774 address => $record->subfield("210", "a") || "",
775 edition => $record->subfield("205", "a") || "",
776 note => $record->subfield("300", "a") || "",
777 url => $record->subfield("856", "u") || ""
779 } else {
781 # Marc21 to bibtex array
782 @bh = (
784 # Mandatory
785 author => $author,
786 title => $record->subfield("245", "a") || "",
787 editor => $record->subfield("260", "f") || "",
788 publisher => $record->subfield("264", "b") || $record->subfield("260", "b") || "",
789 year => $record->subfield("264", "c") || $record->subfield("260", "c") || $record->subfield("260", "g") || "",
791 # Optional
792 # unimarc to marc21 specification says not to convert 200$v to marc21
793 series => $record->subfield("490", "a") || "",
794 address => $record->subfield("264", "a") || $record->subfield("260", "a") || "",
795 edition => $record->subfield("250", "a") || "",
796 note => $record->subfield("500", "a") || "",
797 url => $record->subfield("856", "u") || ""
801 my $BibtexExportAdditionalFields = C4::Context->preference('BibtexExportAdditionalFields');
802 my $additional_fields;
803 if ($BibtexExportAdditionalFields) {
804 $BibtexExportAdditionalFields = "$BibtexExportAdditionalFields\n\n";
805 $additional_fields = eval { YAML::Load($BibtexExportAdditionalFields); };
806 if ($@) {
807 warn "Unable to parse BibtexExportAdditionalFields : $@";
808 $additional_fields = undef;
812 if ( $additional_fields && $additional_fields->{'@'} ) {
813 my ( $f, $sf ) = split( /\$/, $additional_fields->{'@'} );
814 my ( $type ) = read_field( { record => $record, field => $f, subfield => $sf, field_numbers => [1] } );
816 if ($type) {
817 $tex .= '@' . $type . '{';
819 else {
820 $tex .= "\@book{";
823 else {
824 $tex .= "\@book{";
827 my @elt;
828 for ( my $i = 0 ; $i < scalar( @bh ) ; $i = $i + 2 ) {
829 next unless $bh[$i+1];
830 push @elt, qq|\t$bh[$i] = {$bh[$i+1]}|;
832 $tex .= join(",\n", $id, @elt);
834 if ($additional_fields) {
835 $tex .= ",\n";
836 foreach my $bibtex_tag ( keys %$additional_fields ) {
837 next if $bibtex_tag eq '@';
839 my @fields =
840 ref( $additional_fields->{$bibtex_tag} ) eq 'ARRAY'
841 ? @{ $additional_fields->{$bibtex_tag} }
842 : $additional_fields->{$bibtex_tag};
844 for my $tag (@fields) {
845 my ( $f, $sf ) = split( /\$/, $tag );
846 my @values = read_field( { record => $record, field => $f, subfield => $sf } );
847 foreach my $v (@values) {
848 $tex .= qq(\t$bibtex_tag = {$v}\n);
853 else {
854 $tex .= "\n";
857 $tex .= "}\n";
859 return $tex;
863 =head1 INTERNAL FUNCTIONS
865 =head2 _entity_encode - Entity-encode an array of strings
867 my ($entity_encoded_string) = _entity_encode($string);
871 my (@entity_encoded_strings) = _entity_encode(@strings);
873 Entity-encode an array of strings
875 =cut
877 sub _entity_encode {
878 my @strings = @_;
879 my @strings_entity_encoded;
880 foreach my $string (@strings) {
881 my $nfc_string = NFC($string);
882 $nfc_string =~ s/([\x{0080}-\x{fffd}])/sprintf('&#x%X;',ord($1))/sgoe;
883 push @strings_entity_encoded, $nfc_string;
885 return @strings_entity_encoded;
888 END { } # module clean-up code here (global destructor)
890 __END__
892 =head1 AUTHOR
894 Joshua Ferraro <jmf@liblime.com>
896 =head1 MODIFICATIONS
899 =cut