C4/Record.pm

   1 package C4::Record;
   2 #
   3 # Copyright 2006 (C) LibLime
   4 # Parts copyright 2010 BibLibre
   5 #
   6 # This file is part of Koha.
   7 #
   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.
  12 #
  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.
  16 #
  17 # You should have received a copy of the GNU General Public License along
  18 # with Koha; if not, write to the Free Software Foundation, Inc.,
  19 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  20 #
  21 #
  22 use strict;
  23 #use warnings; FIXME - Bug 2505
  24
  25 # please specify in which methods a given module is used
  26 use MARC::Record; # marc2marcxml, marcxml2marc, changeEncoding
  27 use MARC::File::XML; # marc2marcxml, marcxml2marc, changeEncoding
  28 use MARC::Crosswalk::DublinCore; # marc2dcxml
  29 use Biblio::EndnoteStyle;
  30 use Unicode::Normalize; # _entity_encode
  31 use C4::Biblio; #marc2bibtex
  32 use C4::Csv; #marc2csv
  33 use C4::Koha; #marc2csv
  34 use C4::XSLT ();
  35 use YAML; #marcrecords2csv
  36 use Text::CSV::Encoded; #marc2csv
  37
  38 use vars qw($VERSION @ISA @EXPORT);
  39
  40 # set the version for version checking
  41 $VERSION = 3.07.00.049;
  42
  43 @ISA = qw(Exporter);
  44
  45 # only export API methods
  46
  47 @EXPORT = qw(
  48   &marc2endnote
  49   &marc2marc
  50   &marc2marcxml
  51   &marcxml2marc
  52   &marc2dcxml
  53   &marc2modsxml
  54   &marc2madsxml
  55   &marc2bibtex
  56   &marc2csv
  57   &changeEncoding
  58 );
  59
  60 =head1 NAME
  61
  62 C4::Record - MARC, MARCXML, DC, MODS, XML, etc. Record Management Functions and API
  63
  64 =head1 SYNOPSIS
  65
  66 New in Koha 3.x. This module handles all record-related management functions.
  67
  68 =head1 API (EXPORTED FUNCTIONS)
  69
  70 =head2 marc2marc - Convert from one flavour of ISO-2709 to another
  71
  72   my ($error,$newmarc) = marc2marc($marc,$to_flavour,$from_flavour,$encoding);
  73
  74 Returns an ISO-2709 scalar
  75
  76 =cut
  77
  78 sub marc2marc {
  79         my ($marc,$to_flavour,$from_flavour,$encoding) = @_;
  80         my $error;
  81     if ($to_flavour =~ m/marcstd/) {
  82         my $marc_record_obj;
  83         if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
  84             $marc_record_obj = $marc;
  85         } else { # it's not a MARC::Record object, make it one
  86             eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
  87
  88 # conversion to MARC::Record object failed, populate $error
  89                 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
  90         }
  91         unless ($error) {
  92             my @privatefields;
  93             foreach my $field ($marc_record_obj->fields()) {
  94                 if ($field->tag() =~ m/9/ && ($field->tag() != '490' || C4::Context->preference("marcflavour") eq 'UNIMARC')) {
  95                     push @privatefields, $field;
  96                 } elsif (! ($field->is_control_field())) {
  97                     $field->delete_subfield(code => '9') if ($field->subfield('9'));
  98                 }
  99             }
 100             $marc_record_obj->delete_field($_) for @privatefields;
 101             $marc = $marc_record_obj->as_usmarc();
 102         }
 103     } else {
 104         $error = "Feature not yet implemented\n";
 105     }
 106         return ($error,$marc);
 107 }
 108
 109 =head2 marc2marcxml - Convert from ISO-2709 to MARCXML
 110
 111   my ($error,$marcxml) = marc2marcxml($marc,$encoding,$flavour);
 112
 113 Returns a MARCXML scalar
 114
 115 C<$marc> - an ISO-2709 scalar or MARC::Record object
 116
 117 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
 118
 119 C<$flavour> - MARC21 or UNIMARC
 120
 121 C<$dont_entity_encode> - a flag that instructs marc2marcxml not to entity encode the xml before returning (optional)
 122
 123 =cut
 124
 125 sub marc2marcxml {
 126         my ($marc,$encoding,$flavour,$dont_entity_encode) = @_;
 127         my $error; # the error string
 128         my $marcxml; # the final MARCXML scalar
 129
 130         # test if it's already a MARC::Record object, if not, make it one
 131         my $marc_record_obj;
 132         if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
 133                 $marc_record_obj = $marc;
 134         } else { # it's not a MARC::Record object, make it one
 135                 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
 136
 137                 # conversion to MARC::Record object failed, populate $error
 138                 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
 139         }
 140         # only proceed if no errors so far
 141         unless ($error) {
 142
 143                 # check the record for warnings
 144                 my @warnings = $marc_record_obj->warnings();
 145                 if (@warnings) {
 146                         warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
 147                         foreach my $warn (@warnings) { warn "\t".$warn };
 148                 }
 149                 unless($encoding) {$encoding = "UTF-8"}; # set default encoding
 150                 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set default MARC flavour
 151
 152                 # attempt to convert the record to MARCXML
 153                 eval { $marcxml = $marc_record_obj->as_xml_record($flavour) }; #handle exceptions
 154
 155                 # record creation failed, populate $error
 156                 if ($@) {
 157                         $error .= "Creation of MARCXML failed:".$MARC::File::ERROR;
 158                         $error .= "Additional information:\n";
 159                         my @warnings = $@->warnings();
 160                         foreach my $warn (@warnings) { $error.=$warn."\n" };
 161
 162                 # record creation was successful
 163         } else {
 164
 165                         # check the record for warning flags again (warnings() will be cleared already if there was an error, see above block
 166                         @warnings = $marc_record_obj->warnings();
 167                         if (@warnings) {
 168                                 warn "\nWarnings encountered while processing ISO-2709 record with title \"".$marc_record_obj->title()."\":\n";
 169                                 foreach my $warn (@warnings) { warn "\t".$warn };
 170                         }
 171                 }
 172
 173                 # only proceed if no errors so far
 174                 unless ($error) {
 175
 176                         # entity encode the XML unless instructed not to
 177                 unless ($dont_entity_encode) {
 178                         my ($marcxml_entity_encoded) = _entity_encode($marcxml);
 179                         $marcxml = $marcxml_entity_encoded;
 180                 }
 181                 }
 182         }
 183         # return result to calling program
 184         return ($error,$marcxml);
 185 }
 186
 187 =head2 marcxml2marc - Convert from MARCXML to ISO-2709
 188
 189   my ($error,$marc) = marcxml2marc($marcxml,$encoding,$flavour);
 190
 191 Returns an ISO-2709 scalar
 192
 193 C<$marcxml> - a MARCXML record
 194
 195 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
 196
 197 C<$flavour> - MARC21 or UNIMARC
 198
 199 =cut
 200
 201 sub marcxml2marc {
 202     my ($marcxml,$encoding,$flavour) = @_;
 203         my $error; # the error string
 204         my $marc; # the final ISO-2709 scalar
 205         unless($encoding) {$encoding = "UTF-8"}; # set the default encoding
 206         unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set the default MARC flavour
 207
 208         # attempt to do the conversion
 209         eval { $marc = MARC::Record->new_from_xml($marcxml,$encoding,$flavour) }; # handle exceptions
 210
 211         # record creation failed, populate $error
 212         if ($@) {$error .="\nCreation of MARCXML Record failed: ".$@;
 213                 $error.=$MARC::File::ERROR if ($MARC::File::ERROR);
 214                 };
 215         # return result to calling program
 216         return ($error,$marc);
 217 }
 218
 219 =head2 marc2dcxml - Convert from ISO-2709 to Dublin Core
 220
 221   my ($error,$dcxml) = marc2dcxml($marc,$qualified);
 222
 223 Returns a DublinCore::Record object, will eventually return a Dublin Core scalar
 224
 225 FIXME: should return actual XML, not just an object
 226
 227 C<$marc> - an ISO-2709 scalar or MARC::Record object
 228
 229 C<$qualified> - specify whether qualified Dublin Core should be used in the input or output [0]
 230
 231 =cut
 232
 233 sub marc2dcxml {
 234         my ($marc,$qualified) = @_;
 235         my $error;
 236     # test if it's already a MARC::Record object, if not, make it one
 237     my $marc_record_obj;
 238     if ($marc =~ /^MARC::Record/) { # it's already a MARC::Record object
 239         $marc_record_obj = $marc;
 240     } else { # it's not a MARC::Record object, make it one
 241                 eval { $marc_record_obj = MARC::Record->new_from_usmarc($marc) }; # handle exceptions
 242
 243                 # conversion to MARC::Record object failed, populate $error
 244                 if ($@) {
 245                         $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR;
 246                 }
 247         }
 248         my $crosswalk = MARC::Crosswalk::DublinCore->new;
 249         if ($qualified) {
 250                 $crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 1 );
 251         }
 252         my $dcxml = $crosswalk->as_dublincore($marc_record_obj);
 253         my $dcxmlfinal = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n";
 254         $dcxmlfinal .= "<metadata
 255   xmlns=\"http://example.org/myapp/\"
 256   xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"
 257   xsi:schemaLocation=\"http://example.org/myapp/ http://example.org/myapp/schema.xsd\"
 258   xmlns:dc=\"http://purl.org/dc/elements/1.1/\"
 259   xmlns:dcterms=\"http://purl.org/dc/terms/\">";
 260
 261         foreach my $element ( $dcxml->elements() ) {
 262                 $dcxmlfinal.="<"."dc:".$element->name().">".$element->content()."</"."dc:".$element->name().">\n";
 263     }
 264         $dcxmlfinal .= "\n</metadata>";
 265         return ($error,$dcxmlfinal);
 266 }
 267
 268 =head2 marc2modsxml - Convert from ISO-2709 to MODS
 269
 270   my $modsxml = marc2modsxml($marc);
 271
 272 Returns a MODS scalar
 273
 274 =cut
 275
 276 sub marc2modsxml {
 277     my ($marc) = @_;
 278     return _transformWithStylesheet($marc, "/prog/en/xslt/MARC21slim2MODS3-1.xsl");
 279 }
 280
 281 =head2 marc2madsxml - Convert from ISO-2709 to MADS
 282
 283   my $madsxml = marc2madsxml($marc);
 284
 285 Returns a MADS scalar
 286
 287 =cut
 288
 289 sub marc2madsxml {
 290     my ($marc) = @_;
 291     return _transformWithStylesheet($marc, "/prog/en/xslt/MARC21slim2MADS.xsl");
 292 }
 293
 294 =head2 _transformWithStylesheet - Transform a MARC record with a stylesheet
 295
 296     my $xml = _transformWithStylesheet($marc, $stylesheet)
 297
 298 Returns the XML scalar result of the transformation. $stylesheet should
 299 contain the path to a stylesheet under intrahtdocs.
 300
 301 =cut
 302
 303 sub _transformWithStylesheet {
 304     my ($marc, $stylesheet) = @_;
 305     # grab the XML, run it through our stylesheet, push it out to the browser
 306     my $xmlrecord = marc2marcxml($marc);
 307     my $xslfile = C4::Context->config('intrahtdocs') . $stylesheet;
 308     return C4::XSLT::engine->transform($xmlrecord, $xslfile);
 309 }
 310
 311 sub marc2endnote {
 312     my ($marc) = @_;
 313         my $marc_rec_obj =  MARC::Record->new_from_usmarc($marc);
 314     my ( $abstract, $f260a, $f710a );
 315     my $f260 = $marc_rec_obj->field('260');
 316     if ($f260) {
 317         $f260a = $f260->subfield('a') if $f260;
 318     }
 319     my $f710 = $marc_rec_obj->field('710');
 320     if ($f710) {
 321         $f710a = $f710->subfield('a');
 322     }
 323     my $f500 = $marc_rec_obj->field('500');
 324     if ($f500) {
 325         $abstract = $f500->subfield('a');
 326     }
 327         my $fields = {
 328                 DB => C4::Context->preference("LibraryName"),
 329                 Title => $marc_rec_obj->title(),
 330                 Author => $marc_rec_obj->author(),
 331                 Publisher => $f710a,
 332                 City => $f260a,
 333                 Year => $marc_rec_obj->publication_date,
 334                 Abstract => $abstract,
 335         };
 336         my $endnote;
 337         my $style = new Biblio::EndnoteStyle();
 338         my $template;
 339         $template.= "DB - DB\n" if C4::Context->preference("LibraryName");
 340         $template.="T1 - Title\n" if $marc_rec_obj->title();
 341         $template.="A1 - Author\n" if $marc_rec_obj->author();
 342         $template.="PB - Publisher\n" if  $f710a;
 343         $template.="CY - City\n" if $f260a;
 344         $template.="Y1 - Year\n" if $marc_rec_obj->publication_date;
 345         $template.="AB - Abstract\n" if $abstract;
 346         my ($text, $errmsg) = $style->format($template, $fields);
 347         return ($text);
 348
 349 }
 350
 351 =head2 marc2csv - Convert several records from UNIMARC to CSV
 352
 353   my ($csv) = marc2csv($biblios, $csvprofileid, $itemnumbers);
 354
 355 Pre and postprocessing can be done through a YAML file
 356
 357 Returns a CSV scalar
 358
 359 C<$biblio> - a list of biblionumbers
 360
 361 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id and the GetCsvProfiles function in C4::Csv)
 362
 363 C<$itemnumbers> - a list of itemnumbers to export
 364
 365 =cut
 366
 367 sub marc2csv {
 368     my ($biblios, $id, $itemnumbers) = @_;
 369     my $output;
 370     my $csv = Text::CSV::Encoded->new();
 371
 372     # Getting yaml file
 373     my $configfile = "../tools/csv-profiles/$id.yaml";
 374     my ($preprocess, $postprocess, $fieldprocessing);
 375     if (-e $configfile){
 376         ($preprocess,$postprocess, $fieldprocessing) = YAML::LoadFile($configfile);
 377     }
 378
 379     # Preprocessing
 380     eval $preprocess if ($preprocess);
 381
 382     my $firstpass = 1;
 383     if ( $itemnumbers ) {
 384         for my $itemnumber ( @$itemnumbers) {
 385             my $biblionumber = GetBiblionumberFromItemnumber $itemnumber;
 386             $output .= marcrecord2csv( $biblionumber, $id, $firstpass, $csv, $fieldprocessing, [$itemnumber] );
 387             $firstpass = 0;
 388         }
 389     } else {
 390         foreach my $biblio (@$biblios) {
 391             $output .= marcrecord2csv( $biblio, $id, $firstpass, $csv, $fieldprocessing );
 392             $firstpass = 0;
 393         }
 394     }
 395
 396     # Postprocessing
 397     eval $postprocess if ($postprocess);
 398
 399     return $output;
 400 }
 401
 402 =head2 marcrecord2csv - Convert a single record from UNIMARC to CSV
 403
 404   my ($csv) = marcrecord2csv($biblio, $csvprofileid, $header);
 405
 406 Returns a CSV scalar
 407
 408 C<$biblio> - a biblionumber
 409
 410 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id and the GetCsvProfiles function in C4::Csv)
 411
 412 C<$header> - true if the headers are to be printed (typically at first pass)
 413
 414 C<$csv> - an already initialised Text::CSV object
 415
 416 C<$fieldprocessing>
 417
 418 C<$itemnumbers> a list of itemnumbers to export
 419
 420 =cut
 421
 422
 423 sub marcrecord2csv {
 424     my ($biblio, $id, $header, $csv, $fieldprocessing, $itemnumbers) = @_;
 425     my $output;
 426
 427     # Getting the record
 428     my $record = GetMarcBiblio($biblio);
 429     next unless $record;
 430     C4::Biblio::EmbedItemsInMarcBiblio( $record, $biblio, $itemnumbers );
 431     # Getting the framework
 432     my $frameworkcode = GetFrameworkCode($biblio);
 433
 434     # Getting information about the csv profile
 435     my $profile = GetCsvProfile($id);
 436
 437     # Getting output encoding
 438     my $encoding          = $profile->{encoding} || 'utf8';
 439     # Getting separators
 440     my $csvseparator      = $profile->{csv_separator}      || ',';
 441     my $fieldseparator    = $profile->{field_separator}    || '#';
 442     my $subfieldseparator = $profile->{subfield_separator} || '|';
 443
 444     # TODO: Be more generic (in case we have to handle other protected chars or more separators)
 445     if ($csvseparator eq '\t') { $csvseparator = "\t" }
 446     if ($fieldseparator eq '\t') { $fieldseparator = "\t" }
 447     if ($subfieldseparator eq '\t') { $subfieldseparator = "\t" }
 448     if ($csvseparator eq '\n') { $csvseparator = "\n" }
 449     if ($fieldseparator eq '\n') { $fieldseparator = "\n" }
 450     if ($subfieldseparator eq '\n') { $subfieldseparator = "\n" }
 451
 452     $csv = $csv->encoding_out($encoding) ;
 453     $csv->sep_char($csvseparator);
 454
 455     # Getting the marcfields
 456     my $marcfieldslist = $profile->{content};
 457
 458     # Getting the marcfields as an array
 459     my @marcfieldsarray = split('\|', $marcfieldslist);
 460
 461    # Separating the marcfields from the user-supplied headers
 462     my @marcfields;
 463     foreach (@marcfieldsarray) {
 464         my @result = split('=', $_);
 465         if (scalar(@result) == 2) {
 466            push @marcfields, { header => $result[0], field => $result[1] };
 467         } else {
 468            push @marcfields, { field => $result[0] }
 469         }
 470     }
 471
 472     # If we have to insert the headers
 473     if ($header) {
 474         my @marcfieldsheaders;
 475         my $dbh   = C4::Context->dbh;
 476
 477         # For each field or subfield
 478         foreach (@marcfields) {
 479
 480             my $field = $_->{field};
 481         # Remove any blank char that might have unintentionally insered into the tag name
 482         $field =~ s/\s+//g;
 483
 484             # If we have a user-supplied header, we use it
 485             if (exists $_->{header}) {
 486                     push @marcfieldsheaders, $_->{header};
 487             } else {
 488                 # If not, we get the matching tag name from koha
 489                 if (index($field, '$') > 0) {
 490                     my ($fieldtag, $subfieldtag) = split('\$', $field);
 491                     my $query = "SELECT liblibrarian FROM marc_subfield_structure WHERE tagfield=? AND tagsubfield=?";
 492                     my $sth = $dbh->prepare($query);
 493                     $sth->execute($fieldtag, $subfieldtag);
 494                     my @results = $sth->fetchrow_array();
 495                     push @marcfieldsheaders, $results[0];
 496                 } else {
 497                     my $query = "SELECT liblibrarian FROM marc_tag_structure WHERE tagfield=?";
 498                     my $sth = $dbh->prepare($query);
 499                     $sth->execute($field);
 500                     my @results = $sth->fetchrow_array();
 501                     push @marcfieldsheaders, $results[0];
 502                 }
 503             }
 504         }
 505         $csv->combine(@marcfieldsheaders);
 506         $output = $csv->string() . "\n";
 507     }
 508
 509     # For each marcfield to export
 510     my @fieldstab;
 511     foreach (@marcfields) {
 512         my $marcfield = $_->{field};
 513         # If it is a subfield
 514         if (index($marcfield, '$') > 0) {
 515             my ($fieldtag, $subfieldtag) = split('\$', $marcfield);
 516             my @fields = $record->field($fieldtag);
 517             my @tmpfields;
 518
 519             # For each field
 520             foreach my $field (@fields) {
 521
 522                 # We take every matching subfield
 523                 my @subfields = $field->subfield($subfieldtag);
 524                 foreach my $subfield (@subfields) {
 525
 526                     # Getting authorised value
 527                     my $authvalues = GetKohaAuthorisedValuesFromField($fieldtag, $subfieldtag, $frameworkcode, undef);
 528                     push @tmpfields, (defined $authvalues->{$subfield}) ? $authvalues->{$subfield} : $subfield;
 529                 }
 530             }
 531             push (@fieldstab, join($subfieldseparator, @tmpfields));
 532         # Or a field
 533         } else {
 534             my @fields = ($record->field($marcfield));
 535             my $authvalues = GetKohaAuthorisedValuesFromField($marcfield, undef, $frameworkcode, undef);
 536
 537         my @valuesarray;
 538         foreach (@fields) {
 539             my $value;
 540
 541             # If it is a control field
 542             if ($_->is_control_field) {
 543                 $value = defined $authvalues->{$_->as_string} ? $authvalues->{$_->as_string} : $_->as_string;
 544             } else {
 545                 # If it is a field, we gather all subfields, joined by the subfield separator
 546                 my @subvaluesarray;
 547                 my @subfields = $_->subfields;
 548                 foreach my $subfield (@subfields) {
 549                     push (@subvaluesarray, defined $authvalues->{$subfield->[1]} ? $authvalues->{$subfield->[1]} : $subfield->[1]);
 550                 }
 551                 $value = join ($subfieldseparator, @subvaluesarray);
 552             }
 553
 554             # Field processing
 555             eval $fieldprocessing if ($fieldprocessing);
 556
 557             push @valuesarray, $value;
 558         }
 559             push (@fieldstab, join($fieldseparator, @valuesarray));
 560          }
 561     };
 562
 563     $csv->combine(@fieldstab);
 564     $output .= $csv->string() . "\n";
 565
 566     return $output;
 567
 568 }
 569
 570
 571 =head2 changeEncoding - Change the encoding of a record
 572
 573   my ($error, $newrecord) = changeEncoding($record,$format,$flavour,$to_encoding,$from_encoding);
 574
 575 Changes the encoding of a record
 576
 577 C<$record> - the record itself can be in ISO-2709, a MARC::Record object, or MARCXML for now (required)
 578
 579 C<$format> - MARC or MARCXML (required)
 580
 581 C<$flavour> - MARC21 or UNIMARC, if MARC21, it will change the leader (optional) [defaults to Koha system preference]
 582
 583 C<$to_encoding> - the encoding you want the record to end up in (optional) [UTF-8]
 584
 585 C<$from_encoding> - the encoding the record is currently in (optional, it will probably be able to tell unless there's a problem with the record)
 586
 587 FIXME: the from_encoding doesn't work yet
 588
 589 FIXME: better handling for UNIMARC, it should allow management of 100 field
 590
 591 FIXME: shouldn't have to convert to and from xml/marc just to change encoding someone needs to re-write MARC::Record's 'encoding' method to actually alter the encoding rather than just changing the leader
 592
 593 =cut
 594
 595 sub changeEncoding {
 596         my ($record,$format,$flavour,$to_encoding,$from_encoding) = @_;
 597         my $newrecord;
 598         my $error;
 599         unless($flavour) {$flavour = C4::Context->preference("marcflavour")};
 600         unless($to_encoding) {$to_encoding = "UTF-8"};
 601
 602         # ISO-2709 Record (MARC21 or UNIMARC)
 603         if (lc($format) =~ /^marc$/o) {
 604                 # if we're converting encoding of an ISO2709 file, we need to roundtrip through XML
 605                 #       because MARC::Record doesn't directly provide us with an encoding method
 606                 #       It's definitely less than idea and should be fixed eventually - kados
 607                 my $marcxml; # temporary storage of MARCXML scalar
 608                 ($error,$marcxml) = marc2marcxml($record,$to_encoding,$flavour);
 609                 unless ($error) {
 610                         ($error,$newrecord) = marcxml2marc($marcxml,$to_encoding,$flavour);
 611                 }
 612
 613         # MARCXML Record
 614         } elsif (lc($format) =~ /^marcxml$/o) { # MARCXML Record
 615                 my $marc;
 616                 ($error,$marc) = marcxml2marc($record,$to_encoding,$flavour);
 617                 unless ($error) {
 618                         ($error,$newrecord) = marc2marcxml($record,$to_encoding,$flavour);
 619                 }
 620         } else {
 621                 $error.="Unsupported record format:".$format;
 622         }
 623         return ($error,$newrecord);
 624 }
 625
 626 =head2 marc2bibtex - Convert from MARC21 and UNIMARC to BibTex
 627
 628   my ($bibtex) = marc2bibtex($record, $id);
 629
 630 Returns a BibTex scalar
 631
 632 C<$record> - a MARC::Record object
 633
 634 C<$id> - an id for the BibTex record (might be the biblionumber)
 635
 636 =cut
 637
 638
 639 sub marc2bibtex {
 640     my ($record, $id) = @_;
 641     my $tex;
 642     my $marcflavour = C4::Context->preference("marcflavour");
 643
 644     # Authors
 645     my $author;
 646     my @texauthors;
 647     my @authorFields = ('100','110','111','700','710','711');
 648     @authorFields = ('700','701','702','710','711','721') if ( $marcflavour eq "UNIMARC" );
 649
 650     foreach my $field ( @authorFields ) {
 651         # author formatted surname, firstname
 652         my $texauthor = '';
 653         if ( $marcflavour eq "UNIMARC" ) {
 654            $texauthor = join ', ',
 655            ( $record->subfield($field,"a"), $record->subfield($field,"b") );
 656        } else {
 657            $texauthor = $record->subfield($field,"a");
 658        }
 659        push @texauthors, $texauthor if $texauthor;
 660     }
 661     $author = join ' and ', @texauthors;
 662
 663     # Defining the conversion array according to the marcflavour
 664     my @bh;
 665     if ( $marcflavour eq "UNIMARC" ) {
 666
 667         # FIXME, TODO : handle repeatable fields
 668         # TODO : handle more types of documents
 669
 670         # Unimarc to bibtex array
 671         @bh = (
 672
 673             # Mandatory
 674             author    => $author,
 675             title     => $record->subfield("200", "a") || "",
 676             editor    => $record->subfield("210", "g") || "",
 677             publisher => $record->subfield("210", "c") || "",
 678             year      => $record->subfield("210", "d") || $record->subfield("210", "h") || "",
 679
 680             # Optional
 681             volume  =>  $record->subfield("200", "v") || "",
 682             series  =>  $record->subfield("225", "a") || "",
 683             address =>  $record->subfield("210", "a") || "",
 684             edition =>  $record->subfield("205", "a") || "",
 685             note    =>  $record->subfield("300", "a") || "",
 686             url     =>  $record->subfield("856", "u") || ""
 687         );
 688     } else {
 689
 690         # Marc21 to bibtex array
 691         @bh = (
 692
 693             # Mandatory
 694             author    => $author,
 695             title     => $record->subfield("245", "a") || "",
 696             editor    => $record->subfield("260", "f") || "",
 697             publisher => $record->subfield("264", "b") || $record->subfield("260", "b") || "",
 698             year      => $record->subfield("264", "c") || $record->subfield("260", "c") || $record->subfield("260", "g") || "",
 699
 700             # Optional
 701             # unimarc to marc21 specification says not to convert 200$v to marc21
 702             series  =>  $record->subfield("490", "a") || "",
 703             address =>  $record->subfield("264", "a") || $record->subfield("260", "a") || "",
 704             edition =>  $record->subfield("250", "a") || "",
 705             note    =>  $record->subfield("500", "a") || "",
 706             url     =>  $record->subfield("856", "u") || ""
 707         );
 708     }
 709
 710     $tex .= "\@book{";
 711     my @elt;
 712     for ( my $i = 0 ; $i < scalar( @bh ) ; $i = $i + 2 ) {
 713         next unless $bh[$i+1];
 714         push @elt, qq|\t$bh[$i] = {$bh[$i+1]}|;
 715     }
 716     $tex .= join(",\n", $id, @elt);
 717     $tex .= "\n}\n";
 718
 719     return $tex;
 720 }
 721
 722
 723 =head1 INTERNAL FUNCTIONS
 724
 725 =head2 _entity_encode - Entity-encode an array of strings
 726
 727   my ($entity_encoded_string) = _entity_encode($string);
 728
 729 or
 730
 731   my (@entity_encoded_strings) = _entity_encode(@strings);
 732
 733 Entity-encode an array of strings
 734
 735 =cut
 736
 737 sub _entity_encode {
 738         my @strings = @_;
 739         my @strings_entity_encoded;
 740         foreach my $string (@strings) {
 741                 my $nfc_string = NFC($string);
 742                 $nfc_string =~ s/([\x{0080}-\x{fffd}])/sprintf('&#x%X;',ord($1))/sgoe;
 743                 push @strings_entity_encoded, $nfc_string;
 744         }
 745         return @strings_entity_encoded;
 746 }
 747
 748 END { }       # module clean-up code here (global destructor)
 749 1;
 750 __END__
 751
 752 =head1 AUTHOR
 753
 754 Joshua Ferraro <jmf@liblime.com>
 755
 756 =head1 MODIFICATIONS
 757
 758
 759 =cut