C4/Record.pm

   1 package C4::Record;
   2 #
   3 # Copyright 2006 (C) LibLime
   4 # Joshua Ferraro <jmf@liblime.com>
   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 with
  18 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
  19 # Suite 330, Boston, MA  02111-1307 USA
  20 #
  21 #
  22 use strict;# use warnings; #FIXME: turn off warnings before release
  23
  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 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 $VERSION = 3.00;
  42
  43 @ISA = qw(Exporter);
  44
  45 # only export API methods
  46
  47 @EXPORT = qw(
  48   &marc2endnote
  49   &marc2marc
  50   &marc2marcxml
  51   &marcxml2marc
  52   &marc2dcxml
  53   &marc2modsxml
  54   &marc2bibtex
  55   &marc2csv
  56   &html2marcxml
  57   &html2marc
  58   &changeEncoding
  59 );
  60
  61 =head1 NAME
  62
  63 C4::Record - MARC, MARCXML, DC, MODS, XML, etc. Record Management Functions and API
  64
  65 =head1 SYNOPSIS
  66
  67 New in Koha 3.x. This module handles all record-related management functions.
  68
  69 =head1 API (EXPORTED FUNCTIONS)
  70
  71 =head2 marc2marc - Convert from one flavour of ISO-2709 to another
  72
  73 =over 4
  74
  75 my ($error,$newmarc) = marc2marc($marc,$to_flavour,$from_flavour,$encoding);
  76
  77 Returns an ISO-2709 scalar
  78
  79 =back
  80
  81 =cut
  82
  83 sub marc2marc {
  84         my ($marc,$to_flavour,$from_flavour,$encoding) = @_;
  85         my $error = "Feature not yet implemented\n";
  86         return ($error,$marc);
  87 }
  88
  89 =head2 marc2marcxml - Convert from ISO-2709 to MARCXML
  90
  91 =over 4
  92
  93 my ($error,$marcxml) = marc2marcxml($marc,$encoding,$flavour);
  94
  95 Returns a MARCXML scalar
  96
  97 =over 2
  98
  99 C<$marc> - an ISO-2709 scalar or MARC::Record object
 100
 101 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
 102
 103 C<$flavour> - MARC21 or UNIMARC
 104
 105 C<$dont_entity_encode> - a flag that instructs marc2marcxml not to entity encode the xml before returning (optional)
 106
 107 =back
 108
 109 =back
 110
 111 =cut
 112
 113 sub marc2marcxml {
 114         my ($marc,$encoding,$flavour,$dont_entity_encode) = @_;
 115         my $error; # the error string
 116         my $marcxml; # the final MARCXML scalar
 117
 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
 124
 125                 # conversion to MARC::Record object failed, populate $error
 126                 if ($@) { $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR };
 127         }
 128         # only proceed if no errors so far
 129         unless ($error) {
 130
 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 };
 136                 }
 137                 unless($encoding) {$encoding = "UTF-8"}; # set default encoding
 138                 unless($flavour) {$flavour = C4::Context->preference("marcflavour")}; # set default MARC flavour
 139
 140                 # attempt to convert the record to MARCXML
 141                 eval { $marcxml = $marc_record_obj->as_xml_record($flavour) }; #handle exceptions
 142
 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" };
 149
 150                 # record creation was successful
 151         } else {
 152
 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 };
 158                         }
 159                 }
 160
 161                 # only proceed if no errors so far
 162                 unless ($error) {
 163
 164                         # entity encode the XML unless instructed not to
 165                 unless ($dont_entity_encode) {
 166                         my ($marcxml_entity_encoded) = _entity_encode($marcxml);
 167                         $marcxml = $marcxml_entity_encoded;
 168                 }
 169                 }
 170         }
 171         # return result to calling program
 172         return ($error,$marcxml);
 173 }
 174
 175 =head2 marcxml2marc - Convert from MARCXML to ISO-2709
 176
 177 =over 4
 178
 179 my ($error,$marc) = marcxml2marc($marcxml,$encoding,$flavour);
 180
 181 Returns an ISO-2709 scalar
 182
 183 =over 2
 184
 185 C<$marcxml> - a MARCXML record
 186
 187 C<$encoding> - UTF-8 or MARC-8 [UTF-8]
 188
 189 C<$flavour> - MARC21 or UNIMARC
 190
 191 =back
 192
 193 =back
 194
 195 =cut
 196
 197 sub marcxml2marc {
 198     my ($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
 203
 204         # attempt to do the conversion
 205         eval { $marc = MARC::Record->new_from_xml($marcxml,$encoding,$flavour) }; # handle exceptions
 206
 207         # record creation failed, populate $error
 208         if ($@) {$error .="\nCreation of MARCXML Record failed: ".$@;
 209                 $error.=$MARC::File::ERROR if ($MARC::File::ERROR);
 210                 };
 211         # return result to calling program
 212         return ($error,$marc);
 213 }
 214
 215 =head2 marc2dcxml - Convert from ISO-2709 to Dublin Core
 216
 217 =over 4
 218
 219 my ($error,$dcxml) = marc2dcxml($marc,$qualified);
 220
 221 Returns a DublinCore::Record object, will eventually return a Dublin Core scalar
 222
 223 FIXME: should return actual XML, not just an object
 224
 225 =over 2
 226
 227 C<$marc> - an ISO-2709 scalar or MARC::Record object
 228
 229 C<$qualified> - specify whether qualified Dublin Core should be used in the input or output [0]
 230
 231 =back
 232
 233 =back
 234
 235 =cut
 236
 237 sub marc2dcxml {
 238         my ($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
 246
 247                 # conversion to MARC::Record object failed, populate $error
 248                 if ($@) {
 249                         $error .="\nCreation of MARC::Record object failed: ".$MARC::File::ERROR;
 250                 }
 251         }
 252         my $crosswalk = MARC::Crosswalk::DublinCore->new;
 253         if ($qualified) {
 254                 $crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 1 );
 255         }
 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/\">";
 264
 265         foreach my $element ( $dcxml->elements() ) {
 266                 $dcxmlfinal.="<"."dc:".$element->name().">".$element->content()."</"."dc:".$element->name().">\n";
 267     }
 268         $dcxmlfinal .= "\n</metadata>";
 269         return ($error,$dcxmlfinal);
 270 }
 271 =head2 marc2modsxml - Convert from ISO-2709 to MODS
 272
 273 =over 4
 274
 275 my ($error,$modsxml) = marc2modsxml($marc);
 276
 277 Returns a MODS scalar
 278
 279 =back
 280
 281 =cut
 282
 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 ($newxmlrecord);
 296 }
 297
 298 sub marc2endnote {
 299     my ($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,
 315         };
 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);
 328
 329 }
 330
 331 =head2 marc2csv - Convert several records from UNIMARC to CSV
 332 Pre and postprocessing can be done through a YAML file
 333
 334 =over 4
 335
 336 my ($csv) = marc2csv($biblios, $csvprofileid);
 337
 338 Returns a CSV scalar
 339
 340 =over 2
 341
 342 C<$biblio> - a list of biblionumbers
 343
 344 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id and the GetCsvProfiles function in C4::Csv)
 345
 346 =back
 347
 348 =back
 349
 350 =cut
 351 sub marc2csv {
 352     my ($biblios, $id) = @_;
 353     my $output;
 354     my $csv = Text::CSV::Encoded->new();
 355
 356     # Getting yaml file
 357     my $configfile = "../tools/csv-profiles/$id.yaml";
 358     my ($preprocess, $postprocess, $fieldprocessing);
 359     if (-e $configfile){
 360         ($preprocess,$postprocess, $fieldprocessing) = YAML::LoadFile($configfile);
 361     }
 362
 363     warn $fieldprocessing;
 364     # Preprocessing
 365     eval $preprocess if ($preprocess);
 366
 367     my $firstpass = 1;
 368     foreach my $biblio (@$biblios) {
 369         $output .= marcrecord2csv($biblio, $id, $firstpass, $csv, $fieldprocessing) ;
 370         $firstpass = 0;
 371     }
 372
 373     # Postprocessing
 374     eval $postprocess if ($postprocess);
 375
 376     return $output;
 377 }
 378
 379 =head2 marcrecord2csv - Convert a single record from UNIMARC to CSV
 380
 381 =over 4
 382
 383 my ($csv) = marcrecord2csv($biblio, $csvprofileid, $header);
 384
 385 Returns a CSV scalar
 386
 387 =over 2
 388
 389 C<$biblio> - a biblionumber
 390
 391 C<$csvprofileid> - the id of the CSV profile to use for the export (see export_format.export_format_id and the GetCsvProfiles function in C4::Csv)
 392
 393 C<$header> - true if the headers are to be printed (typically at first pass)
 394
 395 C<$csv> - an already initialised Text::CSV object
 396
 397 =back
 398
 399 =back
 400
 401 =cut
 402
 403
 404 sub marcrecord2csv {
 405     my ($biblio, $id, $header, $csv, $fieldprocessing) = @_;
 406     my $output;
 407
 408     # Getting the record
 409     my $record = GetMarcBiblio($biblio);
 410
 411     # Getting the framework
 412     my $frameworkcode = GetFrameworkCode($biblio);
 413
 414     # Getting information about the csv profile
 415     my $profile = GetCsvProfile($id);
 416
 417     # Getting output encoding
 418     my $encoding          = $profile->{encoding} || 'utf8';
 419     # Getting separators
 420     my $csvseparator      = $profile->{csv_separator}      || ',';
 421     my $fieldseparator    = $profile->{field_separator}    || '#';
 422     my $subfieldseparator = $profile->{subfield_separator} || '|';
 423
 424     # TODO: Be more generic (in case we have to handle other protected chars or more separators)
 425     if ($csvseparator eq '\t') { $csvseparator = "\t" }
 426     if ($fieldseparator eq '\t') { $fieldseparator = "\t" }
 427     if ($subfieldseparator eq '\t') { $subfieldseparator = "\t" }
 428
 429
 430     $csv = $csv->encoding_out($encoding) ;
 431     $csv->sep_char($csvseparator);
 432
 433     # Getting the marcfields
 434     my $marcfieldslist = $profile->{marcfields};
 435
 436     # Getting the marcfields as an array
 437     my @marcfieldsarray = split('\|', $marcfieldslist);
 438
 439    # Separating the marcfields from the the user-supplied headers
 440     my @marcfields;
 441     foreach (@marcfieldsarray) {
 442         my @result = split('=', $_);
 443         if (scalar(@result) == 2) {
 444            push @marcfields, { header => $result[0], field => $result[1] };
 445         } else {
 446            push @marcfields, { field => $result[0] }
 447         }
 448     }
 449
 450     # If we have to insert the headers
 451     if ($header) {
 452         my @marcfieldsheaders;
 453         my $dbh   = C4::Context->dbh;
 454
 455         # For each field or subfield
 456         foreach (@marcfields) {
 457
 458             my $field = $_->{field};
 459
 460             # If we have a user-supplied header, we use it
 461             if (exists $_->{header}) {
 462                     push @marcfieldsheaders, $_->{header};
 463             } else {
 464                 # If not, we get the matching tag name from koha
 465                 if (index($field, '$') > 0) {
 466                     my ($fieldtag, $subfieldtag) = split('\$', $field);
 467                     my $query = "SELECT liblibrarian FROM marc_subfield_structure WHERE tagfield=? AND tagsubfield=?";
 468                     my $sth = $dbh->prepare($query);
 469                     $sth->execute($fieldtag, $subfieldtag);
 470                     my @results = $sth->fetchrow_array();
 471                     push @marcfieldsheaders, $results[0];
 472                 } else {
 473                     my $query = "SELECT liblibrarian FROM marc_tag_structure WHERE tagfield=?";
 474                     my $sth = $dbh->prepare($query);
 475                     $sth->execute($field);
 476                     my @results = $sth->fetchrow_array();
 477                     push @marcfieldsheaders, $results[0];
 478                 }
 479             }
 480         }
 481         $csv->combine(@marcfieldsheaders);
 482         $output = $csv->string() . "\n";
 483     }
 484
 485     # For each marcfield to export
 486     my @fieldstab;
 487     foreach (@marcfields) {
 488         my $marcfield = $_->{field};
 489         # If it is a subfield
 490         if (index($marcfield, '$') > 0) {
 491             my ($fieldtag, $subfieldtag) = split('\$', $marcfield);
 492             my @fields = $record->field($fieldtag);
 493             my @tmpfields;
 494
 495             # For each field
 496             foreach my $field (@fields) {
 497
 498                 # We take every matching subfield
 499                 my @subfields = $field->subfield($subfieldtag);
 500                 foreach my $subfield (@subfields) {
 501
 502                     # Getting authorised value
 503                     my $authvalues = GetKohaAuthorisedValuesFromField($fieldtag, $subfieldtag, $frameworkcode, undef);
 504                     push @tmpfields, (defined $authvalues->{$subfield}) ? $authvalues->{$subfield} : $subfield;
 505                 }
 506             }
 507             push (@fieldstab, join($subfieldseparator, @tmpfields));
 508         # Or a field
 509         } else {
 510             my @fields = ($record->field($marcfield));
 511             my $authvalues = GetKohaAuthorisedValuesFromField($marcfield, undef, $frameworkcode, undef);
 512
 513             my @valuesarray;
 514             foreach (@fields) {
 515                 my $value;
 516
 517                 # Getting authorised value
 518                 $value = defined $authvalues->{$_->as_string} ? $authvalues->{$_->as_string} : $_->as_string;
 519
 520                 # Field processing
 521                 eval $fieldprocessing if ($fieldprocessing);
 522
 523                 push @valuesarray, $value;
 524             }
 525             push (@fieldstab, join($fieldseparator, @valuesarray));
 526          }
 527     };
 528
 529     $csv->combine(@fieldstab);
 530     $output .= $csv->string() . "\n";
 531
 532     return $output;
 533
 534 }
 535
 536
 537 =head2 html2marcxml
 538
 539 =over 4
 540
 541 my ($error,$marcxml) = html2marcxml($tags,$subfields,$values,$indicator,$ind_tag);
 542
 543 Returns a MARCXML scalar
 544
 545 this is used in addbiblio.pl and additem.pl to build the MARCXML record from
 546 the form submission.
 547
 548 FIXME: this could use some better code documentation
 549
 550 =back
 551
 552 =cut
 553
 554 sub html2marcxml {
 555     my ($tags,$subfields,$values,$indicator,$ind_tag) = @_;
 556         my $error;
 557         # add the header info
 558     my $marcxml= MARC::File::XML::header(C4::Context->preference('TemplateEncoding'),C4::Context->preference('marcflavour'));
 559
 560         # some flags used to figure out where in the record we are
 561     my $prevvalue;
 562     my $prevtag=-1;
 563     my $first=1;
 564     my $j = -1;
 565
 566         # handle characters that would cause the parser to choke FIXME: is there a more elegant solution?
 567     for (my $i=0;$i<=@$tags;$i++){
 568                 @$values[$i] =~ s/&/&amp;/g;
 569                 @$values[$i] =~ s/</&lt;/g;
 570                 @$values[$i] =~ s/>/&gt;/g;
 571                 @$values[$i] =~ s/"/&quot;/g;
 572                 @$values[$i] =~ s/'/&apos;/g;
 573
 574                 if ((@$tags[$i] ne $prevtag)){
 575                         $j++ unless (@$tags[$i] eq "");
 576                         #warn "IND:".substr(@$indicator[$j],0,1).substr(@$indicator[$j],1,1)." ".@$tags[$i];
 577                         if (!$first){
 578                                 $marcxml.="</datafield>\n";
 579                                 if ((@$tags[$i] > 10) && (@$values[$i] ne "")){
 580                         my $ind1 = substr(@$indicator[$j],0,1);
 581                                         my $ind2 = substr(@$indicator[$j],1,1);
 582                                         $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
 583                                         $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
 584                                         $first=0;
 585                                 } else {
 586                                         $first=1;
 587                                 }
 588                         } else {
 589                                 if (@$values[$i] ne "") {
 590                                         # handle the leader
 591                                         if (@$tags[$i] eq "000") {
 592                                                 $marcxml.="<leader>@$values[$i]</leader>\n";
 593                                                 $first=1;
 594                                         # rest of the fixed fields
 595                                         } elsif (@$tags[$i] lt '010') { # don't compare numerically 010 == 8
 596                                                 $marcxml.="<controlfield tag=\"@$tags[$i]\">@$values[$i]</controlfield>\n";
 597                                                 $first=1;
 598                                         } else {
 599                                                 my $ind1 = substr(@$indicator[$j],0,1);
 600                                                 my $ind2 = substr(@$indicator[$j],1,1);
 601                                                 $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
 602                                                 $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
 603                                                 $first=0;
 604                                         }
 605                                 }
 606                         }
 607                 } else { # @$tags[$i] eq $prevtag
 608                         if (@$values[$i] eq "") {
 609                         } else {
 610                                 if ($first){
 611                                         my $ind1 = substr(@$indicator[$j],0,1);
 612                                         my $ind2 = substr(@$indicator[$j],1,1);
 613                                         $marcxml.="<datafield tag=\"@$tags[$i]\" ind1=\"$ind1\" ind2=\"$ind2\">\n";
 614                                         $first=0;
 615                                 }
 616                                 $marcxml.="<subfield code=\"@$subfields[$i]\">@$values[$i]</subfield>\n";
 617                         }
 618                 }
 619                 $prevtag = @$tags[$i];
 620         }
 621         $marcxml.= MARC::File::XML::footer();
 622         #warn $marcxml;
 623         return ($error,$marcxml);
 624 }
 625
 626 =head2 html2marc
 627
 628 =over 4
 629
 630 Probably best to avoid using this ... it has some rather striking problems:
 631
 632 =over 2
 633
 634 * saves blank subfields
 635
 636 * subfield order is hardcoded to always start with 'a' for repeatable tags (because it is hardcoded in the addfield routine).
 637
 638 * only possible to specify one set of indicators for each set of tags (ie, one for all the 650s). (because they were stored in a hash with the tag as the key).
 639
 640 * the underlying routines didn't support subfield reordering or subfield repeatability.
 641
 642 =back
 643
 644 I've left it in here because it could be useful if someone took the time to fix it. -- kados
 645
 646 =back
 647
 648 =cut
 649
 650 sub html2marc {
 651     my ($dbh,$rtags,$rsubfields,$rvalues,%indicators) = @_;
 652     my $prevtag = -1;
 653     my $record = MARC::Record->new();
 654 #   my %subfieldlist=();
 655     my $prevvalue; # if tag <10
 656     my $field; # if tag >=10
 657     for (my $i=0; $i< @$rtags; $i++) {
 658         # rebuild MARC::Record
 659 #           warn "0=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ";
 660         if (@$rtags[$i] ne $prevtag) {
 661             if ($prevtag < 10) {
 662                 if ($prevvalue) {
 663                     if (($prevtag ne '000') && ($prevvalue ne "")) {
 664                         $record->add_fields((sprintf "%03s",$prevtag),$prevvalue);
 665                     } elsif ($prevvalue ne ""){
 666                         $record->leader($prevvalue);
 667                     }
 668                 }
 669             } else {
 670                 if (($field) && ($field ne "")) {
 671                     $record->add_fields($field);
 672                 }
 673             }
 674             $indicators{@$rtags[$i]}.='  ';
 675                 # skip blank tags, I hope this works
 676                 if (@$rtags[$i] eq ''){
 677                 $prevtag = @$rtags[$i];
 678                 undef $field;
 679                 next;
 680             }
 681             if (@$rtags[$i] <10) {
 682                 $prevvalue= @$rvalues[$i];
 683                 undef $field;
 684             } else {
 685                 undef $prevvalue;
 686                 if (@$rvalues[$i] eq "") {
 687                 undef $field;
 688                 } else {
 689                 $field = MARC::Field->new( (sprintf "%03s",@$rtags[$i]), substr($indicators{@$rtags[$i]},0,1),substr($indicators{@$rtags[$i]},1,1), @$rsubfields[$i] => @$rvalues[$i]);
 690                 }
 691 #           warn "1=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ".$field->as_formatted;
 692             }
 693             $prevtag = @$rtags[$i];
 694         } else {
 695             if (@$rtags[$i] <10) {
 696                 $prevvalue=@$rvalues[$i];
 697             } else {
 698                 if (length(@$rvalues[$i])>0) {
 699                     $field->add_subfields(@$rsubfields[$i] => @$rvalues[$i]);
 700 #           warn "2=>".@$rtags[$i].@$rsubfields[$i]." = ".@$rvalues[$i].": ".$field->as_formatted;
 701                 }
 702             }
 703             $prevtag= @$rtags[$i];
 704         }
 705     }
 706     #}
 707     # the last has not been included inside the loop... do it now !
 708     #use Data::Dumper;
 709     #warn Dumper($field->{_subfields});
 710     $record->add_fields($field) if (($field) && $field ne "");
 711     #warn "HTML2MARC=".$record->as_formatted;
 712     return $record;
 713 }
 714
 715 =head2 changeEncoding - Change the encoding of a record
 716
 717 =over 4
 718
 719 my ($error, $newrecord) = changeEncoding($record,$format,$flavour,$to_encoding,$from_encoding);
 720
 721 Changes the encoding of a record
 722
 723 =over 2
 724
 725 C<$record> - the record itself can be in ISO-2709, a MARC::Record object, or MARCXML for now (required)
 726
 727 C<$format> - MARC or MARCXML (required)
 728
 729 C<$flavour> - MARC21 or UNIMARC, if MARC21, it will change the leader (optional) [defaults to Koha system preference]
 730
 731 C<$to_encoding> - the encoding you want the record to end up in (optional) [UTF-8]
 732
 733 C<$from_encoding> - the encoding the record is currently in (optional, it will probably be able to tell unless there's a problem with the record)
 734
 735 =back
 736
 737 FIXME: the from_encoding doesn't work yet
 738
 739 FIXME: better handling for UNIMARC, it should allow management of 100 field
 740
 741 FIXME: shouldn't have to convert to and from xml/marc just to change encoding someone needs to re-write MARC::Record's 'encoding' method to actually alter the encoding rather than just changing the leader
 742
 743 =back
 744
 745 =cut
 746
 747 sub changeEncoding {
 748         my ($record,$format,$flavour,$to_encoding,$from_encoding) = @_;
 749         my $newrecord;
 750         my $error;
 751         unless($flavour) {$flavour = C4::Context->preference("marcflavour")};
 752         unless($to_encoding) {$to_encoding = "UTF-8"};
 753
 754         # ISO-2709 Record (MARC21 or UNIMARC)
 755         if (lc($format) =~ /^marc$/o) {
 756                 # if we're converting encoding of an ISO2709 file, we need to roundtrip through XML
 757                 #       because MARC::Record doesn't directly provide us with an encoding method
 758                 #       It's definitely less than idea and should be fixed eventually - kados
 759                 my $marcxml; # temporary storage of MARCXML scalar
 760                 ($error,$marcxml) = marc2marcxml($record,$to_encoding,$flavour);
 761                 unless ($error) {
 762                         ($error,$newrecord) = marcxml2marc($marcxml,$to_encoding,$flavour);
 763                 }
 764
 765         # MARCXML Record
 766         } elsif (lc($format) =~ /^marcxml$/o) { # MARCXML Record
 767                 my $marc;
 768                 ($error,$marc) = marcxml2marc($record,$to_encoding,$flavour);
 769                 unless ($error) {
 770                         ($error,$newrecord) = marc2marcxml($record,$to_encoding,$flavour);
 771                 }
 772         } else {
 773                 $error.="Unsupported record format:".$format;
 774         }
 775         return ($error,$newrecord);
 776 }
 777
 778 =head2 marc2bibtex - Convert from MARC21 and UNIMARC to BibTex
 779
 780 =over 4
 781
 782 my ($bibtex) = marc2bibtex($record, $id);
 783
 784 Returns a BibTex scalar
 785
 786 =over 2
 787
 788 C<$record> - a MARC::Record object
 789
 790 C<$id> - an id for the BibTex record (might be the biblionumber)
 791
 792 =back
 793
 794 =back
 795
 796 =cut
 797
 798
 799 sub marc2bibtex {
 800     my ($record, $id) = @_;
 801     my $tex;
 802
 803     # Authors
 804     my $marcauthors = GetMarcAuthors($record,C4::Context->preference("marcflavour"));
 805     my $author;
 806     for my $authors ( map { map { @$_ } values %$_  } @$marcauthors  ) {
 807         $author .= " and " if ($author && $$authors{value});
 808         $author .= $$authors{value} if ($$authors{value});
 809     }
 810
 811     # Defining the conversion hash according to the marcflavour
 812     my %bh;
 813     if (C4::Context->preference("marcflavour") eq "UNIMARC") {
 814
 815         # FIXME, TODO : handle repeatable fields
 816         # TODO : handle more types of documents
 817
 818         # Unimarc to bibtex hash
 819         %bh = (
 820
 821             # Mandatory
 822             author    => $author,
 823             title     => $record->subfield("200", "a") || "",
 824             editor    => $record->subfield("210", "g") || "",
 825             publisher => $record->subfield("210", "c") || "",
 826             year      => $record->subfield("210", "d") || $record->subfield("210", "h") || "",
 827
 828             # Optional
 829             volume  =>  $record->subfield("200", "v") || "",
 830             series  =>  $record->subfield("225", "a") || "",
 831             address =>  $record->subfield("210", "a") || "",
 832             edition =>  $record->subfield("205", "a") || "",
 833             note    =>  $record->subfield("300", "a") || "",
 834             url     =>  $record->subfield("856", "u") || ""
 835         );
 836     } else {
 837
 838         # Marc21 to bibtex hash
 839         %bh = (
 840
 841             # Mandatory
 842             author    => $author,
 843             title     => $record->subfield("245", "a") || "",
 844             editor    => $record->subfield("260", "f") || "",
 845             publisher => $record->subfield("260", "b") || "",
 846             year      => $record->subfield("260", "c") || $record->subfield("260", "g") || "",
 847
 848             # Optional
 849             # unimarc to marc21 specification says not to convert 200$v to marc21
 850             series  =>  $record->subfield("490", "a") || "",
 851             address =>  $record->subfield("260", "a") || "",
 852             edition =>  $record->subfield("250", "a") || "",
 853             note    =>  $record->subfield("500", "a") || "",
 854             url     =>  $record->subfield("856", "u") || ""
 855         );
 856     }
 857
 858     $tex .= "\@book{";
 859     $tex .= join(",\n", $id, map { $bh{$_} ? qq(\t$_ = "$bh{$_}") : () } keys %bh);
 860     $tex .= "\n}\n";
 861
 862     return $tex;
 863 }
 864
 865
 866 =head1 INTERNAL FUNCTIONS
 867
 868 =head2 _entity_encode - Entity-encode an array of strings
 869
 870 =over 4
 871
 872 my ($entity_encoded_string) = _entity_encode($string);
 873
 874 or
 875
 876 my (@entity_encoded_strings) = _entity_encode(@strings);
 877
 878 Entity-encode an array of strings
 879
 880 =back
 881
 882 =cut
 883
 884 sub _entity_encode {
 885         my @strings = @_;
 886         my @strings_entity_encoded;
 887         foreach my $string (@strings) {
 888                 my $nfc_string = NFC($string);
 889                 $nfc_string =~ s/([\x{0080}-\x{fffd}])/sprintf('&#x%X;',ord($1))/sgoe;
 890                 push @strings_entity_encoded, $nfc_string;
 891         }
 892         return @strings_entity_encoded;
 893 }
 894
 895 END { }       # module clean-up code here (global destructor)
 896 1;
 897 __END__
 898
 899 =head1 AUTHOR
 900
 901 Joshua Ferraro <jmf@liblime.com>
 902
 903 =head1 MODIFICATIONS
 904
 905
 906 =cut