|
Line 0
Link Here
|
| 0 |
- |
1 |
#!/usr/bin/perl |
|
|
2 |
package C4::DublinCoreTransformer; |
| 3 |
|
| 4 |
# |
| 5 |
# This file is part of Koha. |
| 6 |
# |
| 7 |
# Copyright (C) 2014 Universidad de El Salvador, |
| 8 |
# Facultad Multidisciplinaria Oriental |
| 9 |
# |
| 10 |
# Koha is free software; you can redistribute it and/or modify it |
| 11 |
# under the terms of the GNU General Public License as published by |
| 12 |
# the Free Software Foundation; either version 3 of the License, or |
| 13 |
# (at your option) any later version. |
| 14 |
# |
| 15 |
# Koha is distributed in the hope that it will be useful, but |
| 16 |
# WITHOUT ANY WARRANTY; without even the implied warranty of |
| 17 |
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 18 |
# GNU General Public License for more details. |
| 19 |
# |
| 20 |
# You should have received a copy of the GNU General Public License |
| 21 |
# along with Koha; if not, see <http://www.gnu.org/licenses>. |
| 22 |
# |
| 23 |
|
| 24 |
use Modern::Perl; |
| 25 |
use strict; |
| 26 |
use warnings::register; |
| 27 |
use version; |
| 28 |
use v5.10.1; |
| 29 |
|
| 30 |
BEGIN { |
| 31 |
require Exporter; |
| 32 |
|
| 33 |
# set the version for version checking |
| 34 |
our $VERSION = version->declare("v0.01"); |
| 35 |
|
| 36 |
# Inherit from Exporter to export functions and variables |
| 37 |
our @ISA = qw(Exporter); |
| 38 |
|
| 39 |
# Functions and variables which are exported by default |
| 40 |
our @EXPORT = qw(new get_xml_data get_rdf_data write_xml write_rdf); |
| 41 |
|
| 42 |
# Functions and variables which can be optionally exported |
| 43 |
our @EXPORT_OK = |
| 44 |
qw(type qualified get_marc_record root xsi_schemaLocation rdf_subject rdf_format entities optional_namespace namespace_url); |
| 45 |
|
| 46 |
} |
| 47 |
|
| 48 |
use Encode qw(encode decode); |
| 49 |
use MARC::File::USMARC; |
| 50 |
use MARC::Crosswalk::DublinCore; |
| 51 |
use XML::LibXML; |
| 52 |
use HTML::Entities; |
| 53 |
use XML::Entities; |
| 54 |
use RDF::Helper; |
| 55 |
use base qw(Class::Accessor); |
| 56 |
__PACKAGE__->mk_accessors( |
| 57 |
qw(type qualified get_marc_record root xsi_schemaLocation rdf_subject rdf_format entities optional_namespace namespace_url) |
| 58 |
); |
| 59 |
|
| 60 |
# Defining some constants |
| 61 |
my $xsi_schema = qq(http://www.w3.org/2001/XMLSchema-instance); |
| 62 |
my $unqualified = qq(http://purl.org/dc/elements/1.1/); |
| 63 |
my $qualified = qq(http://purl.org/dc/terms/); |
| 64 |
my $dcam = qq(http://purl.org/dc/dcam/); |
| 65 |
my $rdfNS = qq(http://www.w3.org/1999/02/22-rdf-syntax-ns#); |
| 66 |
my @elementupcase = |
| 67 |
qw/tableOfContents dateAccepted dateCopyrighted dateSubmitted isVersionOf hasVersion isReplacedBy isRequiredBy isPartOf hasPart isReferencedBy isFormatOf hasFormat conformsTo accrualMethod accrualPeriodicity accrualPolicy instructionalMethod rightsHolder educationLevel accessRights bibliographicCitation/; |
| 68 |
|
| 69 |
#USAGE: C4::DublinCoreTransformer->new({ |
| 70 |
# type => 'xml|rdf', |
| 71 |
# qualified => 0|1, |
| 72 |
# get_marc_record => $record |
| 73 |
#}); |
| 74 |
|
| 75 |
sub new { |
| 76 |
my ( $class, $args ) = @_; |
| 77 |
$args = {} unless defined $args; |
| 78 |
return bless( $args, $class ); |
| 79 |
} |
| 80 |
|
| 81 |
sub get_xml_data { |
| 82 |
|
| 83 |
#USAGE: oop->get_xml_data({ |
| 84 |
# root => 'root', |
| 85 |
# xsi_schemaLocation => 'http://example.org/myapp/ http://example.org/myapp/schema.xsd', |
| 86 |
# entities => 0|1, |
| 87 |
# optional_namespace => 'oai_dc' |
| 88 |
# namespace_url => 'http://www.openarchives.org/OAI/2.0/oai_dc/' |
| 89 |
#}); |
| 90 |
my $self = shift; |
| 91 |
my ($args) = @_; |
| 92 |
|
| 93 |
if (@_) { |
| 94 |
$self->root( $args->{root} ); |
| 95 |
$self->xsi_schemaLocation( $args->{xsi_schemaLocation} ); |
| 96 |
$self->entities( $args->{entities} ); |
| 97 |
$self->optional_namespace( $args->{optional_namespace} ); |
| 98 |
$self->namespace_url( $args->{namespace_url} ); |
| 99 |
} |
| 100 |
|
| 101 |
#if data not filled |
| 102 |
$self->{root} = "metadata" unless ( defined $self->{root} ); |
| 103 |
$self->{entities} = 0 unless ( defined $self->{entities} ); |
| 104 |
my %dataxml = ( |
| 105 |
xmlroot => $self->{root}, |
| 106 |
entities => $self->{entities}, |
| 107 |
optional_namespace => $self->{optional_namespace}, |
| 108 |
namespace_url => $self->{namespace_url}, |
| 109 |
); |
| 110 |
|
| 111 |
if ( exists $self->{xsi_schemaLocation} ) { |
| 112 |
$dataxml{'xsi_schemaLocation'} = $self->{xsi_schemaLocation}; |
| 113 |
} |
| 114 |
if ( exists $self->{xsi_schemaLocation} ) { |
| 115 |
$dataxml{'optional_namespace'} = $self->{optional_namespace}; |
| 116 |
} |
| 117 |
if ( exists $self->{xsi_schemaLocation} ) { |
| 118 |
$dataxml{'namespace_url'} = $self->{namespace_url}; |
| 119 |
} |
| 120 |
return \%dataxml; |
| 121 |
} |
| 122 |
|
| 123 |
sub get_rdf_data { |
| 124 |
|
| 125 |
#USAGE: oop->get_rdf_data({ |
| 126 |
# rdf_subject => 'http://opac.mydomain.edu/cgi-bin/koha/opac-detail.pl?biblionumber=17589', |
| 127 |
# rdf_format => 'ntriples | nquads| rdfxml | rdfjson | ntriples-canonical | turtle', |
| 128 |
# entities => 0|1, |
| 129 |
#}) |
| 130 |
|
| 131 |
my $self = shift; |
| 132 |
my ($args) = @_; |
| 133 |
|
| 134 |
if (@_) { |
| 135 |
$self->rdf_subject( $args->{rdf_subject} ); |
| 136 |
$self->rdf_format( $args->{rdf_format} ); |
| 137 |
$self->entities( $args->{entities} ); |
| 138 |
} |
| 139 |
|
| 140 |
#if data not filled |
| 141 |
$self->{rdf_subject} = "http://koha-community.org/" |
| 142 |
unless ( defined $self->{rdf_subject} ); |
| 143 |
$self->{rdf_format} = "rdfxml" unless ( defined $self->{rdf_format} ); |
| 144 |
$self->{entities} = 0 unless ( defined $self->{entities} ); |
| 145 |
my %rdfdescription = ( |
| 146 |
rdf_subject => $self->{rdf_subject}, |
| 147 |
rdf_format => $self->{rdf_format}, |
| 148 |
entities => $self->{entities}, |
| 149 |
); |
| 150 |
|
| 151 |
return \%rdfdescription; |
| 152 |
} |
| 153 |
|
| 154 |
sub conversion_to_dc { |
| 155 |
|
| 156 |
#Get the MARC record and return a Dublin Core elements, passed as argument in dc_order_execution method. |
| 157 |
#USAGE: |
| 158 |
#my dc = $self->conversion_to_dc(); |
| 159 |
#$self->dc_order_execution ($dc, $dcxml, $root); |
| 160 |
my $self = shift; |
| 161 |
my $marc_record_obj = $self->get_marc_record(); |
| 162 |
my $crosswalk; |
| 163 |
my $dc; |
| 164 |
|
| 165 |
#if data not filled |
| 166 |
$self->qualified(0) unless ( exists $self->{qualified} ); |
| 167 |
if ( ( $self->qualified() != 0 and $self->qualified() != 1 ) |
| 168 |
&& warnings::enabled($self) ) |
| 169 |
{ |
| 170 |
warnings::warn( $self, "Only the 0 or 1 values are accepted" ); |
| 171 |
die "Set qualified => " . $self->qualified() . " is no good, stopped"; |
| 172 |
} |
| 173 |
if ( $self->qualified() == 1 ) { |
| 174 |
$crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 1 ); |
| 175 |
} |
| 176 |
elsif ( $self->qualified() == 0 ) { |
| 177 |
$crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 0 ); |
| 178 |
} |
| 179 |
if ( $marc_record_obj =~ /^MARC::Record/ ) { |
| 180 |
$dc = $crosswalk->as_dublincore($marc_record_obj); |
| 181 |
} |
| 182 |
else { |
| 183 |
warnings::warn( $self, |
| 184 |
"MARC::Record object failed or not a MARC::Record" ); |
| 185 |
die "Stopped due get_marc_record => is not a MARC::Record"; |
| 186 |
} |
| 187 |
|
| 188 |
return $dc; |
| 189 |
|
| 190 |
} |
| 191 |
|
| 192 |
sub write_xml { |
| 193 |
my $self = shift; |
| 194 |
my $dataxml = $self->get_xml_data(); |
| 195 |
my $dc = $self->conversion_to_dc(); |
| 196 |
|
| 197 |
#Looking for bad data |
| 198 |
if ( ( $self->type() ne 'xml' and $self->type() ne 'rdf' ) |
| 199 |
&& warnings::enabled($self) ) |
| 200 |
{ |
| 201 |
warnings::warn( $self, "Only the xml or rdf values are accepted" ); |
| 202 |
die "Set type => " . $self->type() . " is no good, stopped"; |
| 203 |
} |
| 204 |
elsif ( $self->type() eq 'rdf' && warnings::enabled($self) ) { |
| 205 |
warnings::warn( $self, "The constructor specified rdf" ); |
| 206 |
die "Cannot use this method with rdf. This is no good, stopped"; |
| 207 |
} |
| 208 |
|
| 209 |
if ( ( $self->entities() != 0 and $self->entities() != 1 ) |
| 210 |
&& warnings::enabled($self) ) |
| 211 |
{ |
| 212 |
warnings::warn( $self, "Only the 0 or 1 values are accepted" ); |
| 213 |
die "Set entities => " . $self->entities() . " is no good, stopped"; |
| 214 |
} |
| 215 |
|
| 216 |
#Creating XML object |
| 217 |
my $doc = XML::LibXML::Document->new( '1.0', 'utf-8' ); |
| 218 |
my $root = $doc->createElement( $dataxml->{'xmlroot'} ); |
| 219 |
$root->setNamespace( $xsi_schema, 'xsi', 0 ); |
| 220 |
|
| 221 |
if (defined $dataxml->{'optional_namespace'}){ |
| 222 |
$root->setNamespace( $dataxml->{'namespace_url'}, |
| 223 |
$dataxml->{'optional_namespace'}, 0 ); |
| 224 |
} |
| 225 |
|
| 226 |
if ( $self->type() eq 'xml' ) { |
| 227 |
if ( $self->qualified() == 0 ) { |
| 228 |
if ( defined $dataxml->{'xsi_schemaLocation'} ) { |
| 229 |
$root->setAttribute( |
| 230 |
'xsi:schemaLocation' => $dataxml->{'xsi_schemaLocation'} ); |
| 231 |
$root->setNamespace( $unqualified, 'dc', 0 ); |
| 232 |
$self->dc_order_execution( $dc, $doc, $root ); |
| 233 |
} |
| 234 |
else { |
| 235 |
$root->setNamespace( $unqualified, 'dc', 0 ); |
| 236 |
$self->dc_order_execution( $dc, $doc, $root ); |
| 237 |
} |
| 238 |
} |
| 239 |
elsif ( $self->qualified() == 1 ) { |
| 240 |
if ( defined $dataxml->{'xsi_schemaLocation'} ) { |
| 241 |
$root->setAttribute( |
| 242 |
'xsi:schemaLocation' => $dataxml->{'xsi_schemaLocation'} ); |
| 243 |
$root->setNamespace( $unqualified, 'dc', 0 ); |
| 244 |
$root->setNamespace( $qualified, 'dcterms', 0 ); |
| 245 |
$self->dc_order_execution( $dc, $doc, $root ); |
| 246 |
} |
| 247 |
else { |
| 248 |
$root->setNamespace( $unqualified, 'dc', 0 ); |
| 249 |
$root->setNamespace( $qualified, 'dcterms', 0 ); |
| 250 |
$self->dc_order_execution( $dc, $doc, $root ); |
| 251 |
} |
| 252 |
} |
| 253 |
} |
| 254 |
|
| 255 |
$doc->setDocumentElement($root); |
| 256 |
|
| 257 |
#Enabling or desabling entities and print the xml |
| 258 |
if ( $dataxml->{'entities'} == 1 ) { |
| 259 |
my @xml = $doc->toString(1); |
| 260 |
foreach (@xml) { |
| 261 |
my ($dcxml_entity_encoded) = |
| 262 |
XML_entity_encode( Encode::decode( 'UTF-8', $_ ) ); |
| 263 |
$xml[-1] = $dcxml_entity_encoded; |
| 264 |
} |
| 265 |
my $xmljoiner = join "", @xml; |
| 266 |
return $xmljoiner; |
| 267 |
} |
| 268 |
else { |
| 269 |
return $doc->toString(1); |
| 270 |
} |
| 271 |
} |
| 272 |
|
| 273 |
sub write_rdf { |
| 274 |
my $self = shift; |
| 275 |
my $rdfdescription = $self->get_rdf_data(); |
| 276 |
my $dc = $self->conversion_to_dc(); |
| 277 |
my $rdf; |
| 278 |
my $subject = $rdfdescription->{'rdf_subject'}; |
| 279 |
|
| 280 |
#Looking for bad data |
| 281 |
if ( ( $self->type() ne 'xml' and $self->type() ne 'rdf' ) |
| 282 |
&& warnings::enabled($self) ) |
| 283 |
{ |
| 284 |
warnings::warn( $self, "Only the xml or rdf values are accepted" ); |
| 285 |
die "Set type => " . $self->type() . " is no good, stopped"; |
| 286 |
} |
| 287 |
elsif ( $self->type() eq 'xml' && warnings::enabled($self) ) { |
| 288 |
warnings::warn( $self, "The constructor specified xml" ); |
| 289 |
die "Cannot use this method with xml. This is no good, stopped"; |
| 290 |
} |
| 291 |
|
| 292 |
if ( ( $self->entities() != 0 and $self->entities() != 1 ) |
| 293 |
&& warnings::enabled($self) ) |
| 294 |
{ |
| 295 |
warnings::warn( $self, "Only the 0 or 1 values are accepted" ); |
| 296 |
die "Set entities => " . $self->entities() . " is no good, stopped"; |
| 297 |
} |
| 298 |
|
| 299 |
if ( $self->qualified() == 0 ) { |
| 300 |
|
| 301 |
#Creating RDF::Helper object |
| 302 |
$rdf = RDF::Helper->new( |
| 303 |
BaseInterface => 'RDF::Trine', |
| 304 |
namespaces => { |
| 305 |
rdf => $rdfNS, |
| 306 |
dc => $unqualified, |
| 307 |
}, |
| 308 |
ExpandQNames => 1, |
| 309 |
base_uri => '', |
| 310 |
); |
| 311 |
} |
| 312 |
else { |
| 313 |
$rdf = RDF::Helper->new( |
| 314 |
BaseInterface => 'RDF::Trine', |
| 315 |
namespaces => { |
| 316 |
rdf => $rdfNS, |
| 317 |
dcterms => $qualified, |
| 318 |
dcam => $dcam, |
| 319 |
}, |
| 320 |
ExpandQNames => 1, |
| 321 |
base_uri => '', |
| 322 |
); |
| 323 |
} |
| 324 |
|
| 325 |
if ( $self->type() eq 'rdf' ) { |
| 326 |
$self->dc_order_execution( $dc, $rdf, $subject ); |
| 327 |
} |
| 328 |
|
| 329 |
#Enabling or desabling entities and print the xml, fixing encode with rdfjson |
| 330 |
if ( $rdfdescription->{'entities'} == 1 |
| 331 |
&& $rdfdescription->{'rdf_format'} eq 'rdfxml' ) |
| 332 |
{ |
| 333 |
my @xml = $rdf->serialize( format => $rdfdescription->{'rdf_format'} ); |
| 334 |
foreach (@xml) { |
| 335 |
my ($dcxml_entity_encoded) = |
| 336 |
XML_entity_encode( Encode::decode( 'UTF-8', $_ ) ); |
| 337 |
$xml[-1] = $dcxml_entity_encoded; |
| 338 |
} |
| 339 |
my $xmljoiner = join "", @xml; |
| 340 |
return $xmljoiner; |
| 341 |
} |
| 342 |
elsif ( $rdfdescription->{'rdf_format'} eq 'rdfjson' ) { |
| 343 |
my $encoding = |
| 344 |
$rdf->serialize( format => $rdfdescription->{'rdf_format'} ); |
| 345 |
$encoding = Encode::encode( 'UTF-8', $encoding ); |
| 346 |
return $encoding; |
| 347 |
} |
| 348 |
else { |
| 349 |
return $rdf->serialize( format => $rdfdescription->{'rdf_format'} ); |
| 350 |
} |
| 351 |
} |
| 352 |
|
| 353 |
sub upper_or_lowercase { |
| 354 |
|
| 355 |
#This subroutine define if is necessary to convert lowercase some qualifiers or let them as crosswalk method basis |
| 356 |
#USAGE: upper_or_lowercase (element->qualifier()) |
| 357 |
my $dcelement = $_[0]; |
| 358 |
my $count = 0; |
| 359 |
my $result; |
| 360 |
while ( $count <= $#elementupcase ) { |
| 361 |
if ( $elementupcase[$count] eq $dcelement ) { |
| 362 |
$result = $dcelement; |
| 363 |
last; |
| 364 |
} |
| 365 |
elsif ( $dcelement ne $elementupcase[$count] ) { |
| 366 |
$result = lc($dcelement); |
| 367 |
$count++; |
| 368 |
next; |
| 369 |
} |
| 370 |
} |
| 371 |
|
| 372 |
return $result; |
| 373 |
|
| 374 |
} |
| 375 |
|
| 376 |
sub XML_entity_encode { |
| 377 |
|
| 378 |
#This subroutine change characteres to XML entities. Since XML::LibXML only encode the five basics |
| 379 |
#USAGE: This subroutine is called when get_xml_data is populated with entities => 1 |
| 380 |
my @strings = @_; |
| 381 |
my @strings_entity_encoded; |
| 382 |
foreach my $string (@strings) { |
| 383 |
my $html_entities = |
| 384 |
HTML::Entities::encode_entities( $string, '^\n\t\x20-\x26\x28-\x7e' ); |
| 385 |
my $xml_entities = XML::Entities::numify( 'all', $html_entities ); |
| 386 |
push @strings_entity_encoded, $xml_entities; |
| 387 |
} |
| 388 |
return @strings_entity_encoded; |
| 389 |
} |
| 390 |
|
| 391 |
sub xml_constructor { |
| 392 |
|
| 393 |
#xml_constructor ( doc, root, namespace, attr, element_name, element_content, element_qualifier, element_scheme ) |
| 394 |
#USAGE: define all posible elements (name, content, qualifier, scheme) as neccesary |
| 395 |
#doc: the XML object |
| 396 |
#root: append root to doc |
| 397 |
#namespace: defined as "dc" or "dcterms" |
| 398 |
#attr: a boolean value to determinate xsi:type xml namespace for dcterms |
| 399 |
#examples: |
| 400 |
#xml_constructor ( $dcxml, $root, "dc", 0, element->name(), element->content() ); |
| 401 |
#xml_constructor ( $dcxml, $root, "dcterms", 1, element->name(), element->content(), $element->qualifier() ); |
| 402 |
|
| 403 |
my ( $doc, $root, $namespace, $attr, $element_name, $element_content, |
| 404 |
$element_qualifier, $element_scheme ) |
| 405 |
= @_; |
| 406 |
|
| 407 |
# This scalar is used for deleting spaces in scheme elements |
| 408 |
my $deletespaces; |
| 409 |
|
| 410 |
# This scalar is used for making upper or lowercase the qualifier elements |
| 411 |
my $lowercase; |
| 412 |
my $tagxml; |
| 413 |
|
| 414 |
given ($namespace) { |
| 415 |
when ( $namespace eq "dc" ) { |
| 416 |
$lowercase = lc($element_name); |
| 417 |
} |
| 418 |
when ( $namespace eq "dcterms" ) { |
| 419 |
$lowercase = upper_or_lowercase($element_qualifier); |
| 420 |
|
| 421 |
#replacing spacial to spatial, bug from dublin or crosswalk. |
| 422 |
$lowercase =~ s/spacial/spatial/; |
| 423 |
} |
| 424 |
} |
| 425 |
|
| 426 |
if ( ( $namespace eq "dc" or "dcterms" ) and $attr == 0 ) { |
| 427 |
|
| 428 |
# This will do something like: <dc:title>Title</dc:title> |
| 429 |
$tagxml = XML::LibXML::Element->new($lowercase); |
| 430 |
$tagxml->setNamespace( undef, $namespace, 1 ); |
| 431 |
$tagxml->appendTextNode($element_content); |
| 432 |
$root->appendChild($tagxml); |
| 433 |
} |
| 434 |
elsif ( ( $namespace eq "dcterms" ) and $attr == 1 ) { |
| 435 |
|
| 436 |
# This will do something like: <dcterms:spatial xsi:type="dcterms:ISO3166">us</dcterms:spatial> |
| 437 |
$deletespaces = $element_scheme; |
| 438 |
$deletespaces =~ s/\s//; |
| 439 |
$tagxml = XML::LibXML::Element->new($lowercase); |
| 440 |
$tagxml->setNamespace( undef, $namespace, 1 ); |
| 441 |
my $attribute = $namespace . ':' . $deletespaces; |
| 442 |
$tagxml->setAttribute( 'xsi:type', $attribute ); |
| 443 |
$tagxml->appendTextNode($element_content); |
| 444 |
$root->appendChild($tagxml); |
| 445 |
} |
| 446 |
elsif ( ( $namespace eq "dc" ) and $attr == 1 ) { |
| 447 |
given ($element_scheme) { |
| 448 |
|
| 449 |
# This will do something like: <dc:subject xsi:type="dcterms:LCSH">Water supply.</dc:subject> |
| 450 |
when ( $element_scheme ne "DCMI Type Vocabulary" ) |
| 451 |
{ #Dublin core returns DCMI Type Vocabulary |
| 452 |
$deletespaces = $element_scheme; |
| 453 |
$deletespaces =~ s/\s//; |
| 454 |
$tagxml = XML::LibXML::Element->new($lowercase); |
| 455 |
$tagxml->setNamespace( undef, $namespace, 1 ); |
| 456 |
my $attribute = 'dcterms:' . $deletespaces; |
| 457 |
$tagxml->setAttribute( 'xsi:type', $attribute ); |
| 458 |
$tagxml->appendTextNode($element_content); |
| 459 |
$root->appendChild($tagxml); |
| 460 |
} |
| 461 |
when ( $element_scheme eq "DCMI Type Vocabulary" ) |
| 462 |
{ #Changing for DCMIType |
| 463 |
# This will do something like: <dc:type xsi:type="dcterms:DCMIType">Text</dc:type> |
| 464 |
$tagxml = XML::LibXML::Element->new($lowercase); |
| 465 |
$tagxml->setNamespace( undef, $namespace, 1 ); |
| 466 |
$tagxml->setAttribute( 'xsi:type', 'dcterms:DCMIType' ); |
| 467 |
$tagxml->appendTextNode($element_content); |
| 468 |
$root->appendChild($tagxml); |
| 469 |
} |
| 470 |
} |
| 471 |
} #if...elsif |
| 472 |
|
| 473 |
return $tagxml; |
| 474 |
|
| 475 |
} #subroutine |
| 476 |
|
| 477 |
sub rdf_constructor { |
| 478 |
|
| 479 |
#xml_constructor ( rdf, subject, namespace, attr, element_name, element_content, element_qualifier, element_scheme ) |
| 480 |
#USAGE: define all posible elements (name, content, qualifier, scheme) as neccesary |
| 481 |
#rdf: the RDF object |
| 482 |
#subject: the subject for the triples |
| 483 |
#namespace: defined as "dc" or "dcterms". This is for Simple DC or DC-RDF recommendations. |
| 484 |
#attr: a boolean value to determine the nodeID attribute for dcterms |
| 485 |
|
| 486 |
my ( $rdf, $subject, $namespace, $attr, $element_name, $element_content, |
| 487 |
$element_qualifier, $element_scheme ) |
| 488 |
= @_; |
| 489 |
|
| 490 |
#This scalar is used for deleting spaces in scheme elements |
| 491 |
my $deletespaces; |
| 492 |
|
| 493 |
#This scalar is used for making upper or lowercase the qualifier elements |
| 494 |
my $lowercase; |
| 495 |
|
| 496 |
given ($namespace) { |
| 497 |
when ( $namespace eq "dc" ) { |
| 498 |
$lowercase = lc($element_name); |
| 499 |
} |
| 500 |
when ( $namespace eq "dcterms" ) { |
| 501 |
$lowercase = upper_or_lowercase($element_qualifier); |
| 502 |
|
| 503 |
#replacing spacial to spatial, bug from dublin or crosswalk. |
| 504 |
$lowercase =~ s/spacial/spatial/; |
| 505 |
} |
| 506 |
} |
| 507 |
|
| 508 |
if ( ( $namespace eq "dc" or "dcterms" ) and $attr == 0 ) { |
| 509 |
|
| 510 |
# This will do something like: <dc:title>Title</dc:title> |
| 511 |
my $predicate = $namespace . ':' . $lowercase; |
| 512 |
my $object = $element_content; |
| 513 |
$rdf->assert_literal( $subject, $predicate, $object ); |
| 514 |
} |
| 515 |
elsif ( ( $namespace eq "dc" ) and $attr == 2 ) { |
| 516 |
|
| 517 |
# This will do something like: <dcterms:title>Title</dcterms:title> |
| 518 |
my $predicate = 'dcterms:' . $lowercase; |
| 519 |
my $object = $element_content; |
| 520 |
$rdf->assert_literal( $subject, $predicate, $object ); |
| 521 |
} |
| 522 |
elsif ( ( $namespace eq "dc" ) and $attr == 1 ) { |
| 523 |
$deletespaces = $element_scheme; |
| 524 |
$deletespaces =~ s/\s//; |
| 525 |
my $predicate = 'dcterms:' . $lowercase; |
| 526 |
my $object = $element_content; |
| 527 |
given ($element_scheme) { |
| 528 |
when ( $element_scheme ne "DCMI Type Vocabulary" ) { |
| 529 |
|
| 530 |
# Dublin core returns DCMI Type Vocabulary |
| 531 |
# This will do something like: |
| 532 |
# <rdf:Description rdf:nodeID="ISO3166"> |
| 533 |
# <dcam:memberOf rdf:resource="http://purl.org/dc/terms/ISO3166"/> |
| 534 |
# <rdf:value>us</rdf:value> |
| 535 |
# </rdf:Description> |
| 536 |
# <dcterms:spatial rdf:nodeID="ISO3166"/> |
| 537 |
my $bnode = $rdf->new_bnode($deletespaces); |
| 538 |
my $uri_resource = join "", $qualified, $deletespaces; |
| 539 |
my $resource = $rdf->new_resource($uri_resource); |
| 540 |
$rdf->assert_literal( $subject, $predicate, $bnode ); |
| 541 |
$rdf->assert_literal( $bnode, 'dcam:memberOf', $resource ); |
| 542 |
$rdf->assert_literal( $bnode, 'rdf:value', $object ); |
| 543 |
} |
| 544 |
when ( $element_scheme eq "DCMI Type Vocabulary" ) |
| 545 |
{ #Changing for DCMIType |
| 546 |
my $bnode = $rdf->new_bnode('DCMIType'); |
| 547 |
my $uri_resource = $qualified . 'DCMIType'; |
| 548 |
my $resource = $rdf->new_resource($uri_resource); |
| 549 |
$rdf->assert_literal( $subject, $predicate, $bnode ); |
| 550 |
$rdf->assert_literal( $bnode, 'dcam:memberOf', $resource ); |
| 551 |
$rdf->assert_literal( $bnode, 'rdf:value', $object ); |
| 552 |
} |
| 553 |
} |
| 554 |
} #if-elsif |
| 555 |
return $rdf; |
| 556 |
} #subroutine |
| 557 |
|
| 558 |
sub dc_order_execution { |
| 559 |
|
| 560 |
#USAGE: USAGE: Get the parameters for xml or rdf and execute in order. |
| 561 |
my $self = shift; |
| 562 |
|
| 563 |
if ( $self->type() eq 'xml' ) { #Order for xml |
| 564 |
my ( $dc, $dcxml, $root ) = @_; |
| 565 |
foreach my $element ( $dc->elements() ) { |
| 566 |
SWITCH: { |
| 567 |
if ( $element->name() |
| 568 |
and not defined $element->scheme() |
| 569 |
and not defined $element->qualifier() ) |
| 570 |
{ |
| 571 |
$dcxml = |
| 572 |
xml_constructor( $dcxml, $root, "dc", 0, $element->name(), |
| 573 |
$element->content() ); |
| 574 |
last SWITCH; |
| 575 |
} |
| 576 |
|
| 577 |
if ( $element->name() |
| 578 |
and not defined $element->scheme() |
| 579 |
and not defined $element->qualifier() ) |
| 580 |
{ |
| 581 |
$dcxml = |
| 582 |
xml_constructor( $dcxml, $root, "dcterms", 0, |
| 583 |
$element->name(), $element->content() ); |
| 584 |
last SWITCH; |
| 585 |
} |
| 586 |
|
| 587 |
if ( $element->name() |
| 588 |
and $element->scheme() |
| 589 |
and $element->qualifier() ) |
| 590 |
{ |
| 591 |
$dcxml = xml_constructor( |
| 592 |
$dcxml, $root, |
| 593 |
"dcterms", 1, |
| 594 |
$element->name(), $element->content(), |
| 595 |
$element->qualifier(), $element->scheme() |
| 596 |
); |
| 597 |
last SWITCH; |
| 598 |
} |
| 599 |
|
| 600 |
if ( $element->name() |
| 601 |
and $element->scheme() |
| 602 |
and ( $element->scheme ne "DCMI Type Vocabulary" ) ) |
| 603 |
{ |
| 604 |
$dcxml = |
| 605 |
xml_constructor( $dcxml, $root, "dc", 1, $element->name(), |
| 606 |
$element->content(), $element->qualifier(), |
| 607 |
$element->scheme() ); |
| 608 |
last SWITCH; |
| 609 |
} |
| 610 |
|
| 611 |
if ( $element->name() |
| 612 |
and $element->scheme() |
| 613 |
and ( $element->scheme() eq "DCMI Type Vocabulary" ) ) |
| 614 |
{ |
| 615 |
$dcxml = |
| 616 |
xml_constructor( $dcxml, $root, "dc", 1, $element->name(), |
| 617 |
$element->content(), $element->qualifier(), |
| 618 |
$element->scheme() ); |
| 619 |
last SWITCH; |
| 620 |
} |
| 621 |
|
| 622 |
if ( $element->name() and $element->qualifier() ) { |
| 623 |
$dcxml = xml_constructor( |
| 624 |
$dcxml, $root, |
| 625 |
"dcterms", 0, |
| 626 |
$element->name(), $element->content(), |
| 627 |
$element->qualifier(), $element->scheme() |
| 628 |
); |
| 629 |
last SWITCH; |
| 630 |
} #if |
| 631 |
} #SWITCH |
| 632 |
} #foreach 1 |
| 633 |
} |
| 634 |
elsif ( $self->type() eq 'rdf' ) { #Order for rdf/xml |
| 635 |
my ( $dc, $rdf, $subject ) = @_; |
| 636 |
foreach my $element ( $dc->elements() ) { |
| 637 |
SWITCH: { |
| 638 |
if ( $element->name() |
| 639 |
and not defined $element->scheme() |
| 640 |
and not defined $element->qualifier() ) |
| 641 |
{ |
| 642 |
rdf_constructor( $rdf, $subject, "dc", 0, $element->name(), |
| 643 |
$element->content() ) |
| 644 |
unless $self->qualified() == 1; |
| 645 |
rdf_constructor( $rdf, $subject, "dc", 2, $element->name(), |
| 646 |
$element->content() ) |
| 647 |
unless $self->qualified() == 0; |
| 648 |
last SWITCH; |
| 649 |
} |
| 650 |
|
| 651 |
if ( $element->name() |
| 652 |
and $element->scheme() |
| 653 |
and ( $element->scheme ne "DCMI Type Vocabulary" ) ) |
| 654 |
{ |
| 655 |
rdf_constructor( $rdf, $subject, "dc", 1, $element->name(), |
| 656 |
$element->content(), $element->qualifier(), |
| 657 |
$element->scheme() ); |
| 658 |
last SWITCH; |
| 659 |
} |
| 660 |
|
| 661 |
if ( $element->name() |
| 662 |
and $element->scheme() |
| 663 |
and ( $element->scheme() eq "DCMI Type Vocabulary" ) ) |
| 664 |
{ |
| 665 |
rdf_constructor( $rdf, $subject, "dc", 1, $element->name(), |
| 666 |
$element->content(), $element->qualifier(), |
| 667 |
$element->scheme() ); |
| 668 |
last SWITCH; |
| 669 |
} |
| 670 |
|
| 671 |
if ( $element->name() and $element->qualifier() ) { |
| 672 |
rdf_constructor( |
| 673 |
$rdf, $subject, |
| 674 |
"dcterms", 0, |
| 675 |
$element->name(), $element->content(), |
| 676 |
$element->qualifier(), $element->scheme() |
| 677 |
); |
| 678 |
last SWITCH; |
| 679 |
} #if |
| 680 |
} #SWITCH |
| 681 |
} #foreach 2 |
| 682 |
} |
| 683 |
return; |
| 684 |
} |
| 685 |
|
| 686 |
END { |
| 687 |
|
| 688 |
#When opened with MARC::File::USMARC, close marc record if not closed |
| 689 |
sub DESTROY { |
| 690 |
my $self = shift; |
| 691 |
if ( $self->{get_marc_record} =~ /^MARC:File::USMARC/ ) { |
| 692 |
$self->{get_marc_record}->close(); |
| 693 |
undef $self->{get_marc_record}; |
| 694 |
} |
| 695 |
} |
| 696 |
} |
| 697 |
1; |
| 698 |
__END__ |
| 699 |
|
| 700 |
=pod |
| 701 |
|
| 702 |
=encoding UTF-8 |
| 703 |
|
| 704 |
=head1 NAME |
| 705 |
|
| 706 |
C4::DublinCoreTransformer - transform the values of MARC::Crosswalk::DublinCore |
| 707 |
into metadata in XML or RDF format. |
| 708 |
|
| 709 |
=head1 SYNOPSIS |
| 710 |
|
| 711 |
=head2 XML-DC |
| 712 |
|
| 713 |
use C4::DublinCoreTransformer; |
| 714 |
use MARC::File::USMARC; |
| 715 |
my $batch = MARC::File::USMARC->in('in.mrc'); |
| 716 |
|
| 717 |
## the $record will be a MARC::Record object. |
| 718 |
my $record = $batch->next(); |
| 719 |
|
| 720 |
#For Simple DC-XML |
| 721 |
my $objectDC = DublinCoreTransformer->new({ |
| 722 |
type => 'xml', |
| 723 |
qualified => 0, |
| 724 |
get_marc_record => $record}); |
| 725 |
|
| 726 |
#For Qualified DC-XML |
| 727 |
my $objectDC = DublinCoreTransformer->new({ |
| 728 |
type => 'xml', |
| 729 |
qualified => 1, |
| 730 |
get_marc_record => $record}); |
| 731 |
|
| 732 |
#After call constructor you need to pass the root element and |
| 733 |
#the schema in .xsd format. If you prefer or you don't know nothing |
| 734 |
#about it pass the method in blank, i.e., without arguments. |
| 735 |
#Also you can determine if you want to encode with xml entities or |
| 736 |
#just with the five basics entities. |
| 737 |
#Set entities to 0 for the five basics or 1 to encode with the |
| 738 |
#complete xml entities. |
| 739 |
$objectDC->get_xml_data({ |
| 740 |
root => 'metadata', |
| 741 |
xsi_schemaLocation => 'http://example.org/myapp/ http://example.org/myapp/schema.xsd', |
| 742 |
}); |
| 743 |
#This print your xml file in STDOUT |
| 744 |
my $XML-DC = $objectDC->write_xml(); |
| 745 |
print $XML-DC; |
| 746 |
|
| 747 |
#Close the file according to MARC::File::USMARC |
| 748 |
$batch->close(); |
| 749 |
undef $batch; |
| 750 |
|
| 751 |
#Or if you want to generate a file |
| 752 |
open (OUTPUT, '> dc2xml.xml') or die $!; |
| 753 |
print OUTPUT print $XML-DC; |
| 754 |
close(OUTPUT); |
| 755 |
|
| 756 |
=head2 DC-RDF |
| 757 |
|
| 758 |
use C4::DublinCoreTransformer; |
| 759 |
use MARC::File::USMARC; |
| 760 |
my $marc = MARC::File::USMARC->in('in.mrc'); |
| 761 |
my $record = $marc->next(); |
| 762 |
|
| 763 |
#Simple DC-RDF is not used but not deprecated, DublinCoreTransformer |
| 764 |
#can generate it as follow: |
| 765 |
my $objectDC = DublinCoreTransformer->new({ |
| 766 |
type => 'rdf', |
| 767 |
qualified => 0, |
| 768 |
get_marc_record => $record}); |
| 769 |
|
| 770 |
#DC-RDF |
| 771 |
my $objectDC = DublinCoreTransformer->new({ |
| 772 |
type => 'rdf', |
| 773 |
qualified => 1, |
| 774 |
get_marc_record => $record}); |
| 775 |
|
| 776 |
#The get_rdf_data method for RDF does not need to use the root element |
| 777 |
#and the schema, but you need to declare the resource being described, |
| 778 |
#i.e., the record. |
| 779 |
#Note: For Koha is the IRI identifier of the record |
| 780 |
$objectDC->get_rdf_data({ |
| 781 |
rdf_subject => 'http://opac.mydomain.edu/cgi-bin/koha/opac-detail.pl?biblionumber=17589', |
| 782 |
}); |
| 783 |
|
| 784 |
my $XML-RDF = $objectDC->write_rdf(); |
| 785 |
print $XML-RDF; |
| 786 |
|
| 787 |
#Close the file according to MARC::File::USMARC |
| 788 |
$batch->close(); |
| 789 |
undef $marc; |
| 790 |
|
| 791 |
=head1 DESCRIPTION |
| 792 |
|
| 793 |
The C4::DublinCoreTransformer defines the interface to transform |
| 794 |
and manipulate MARC records into DC-XML or DC-RDF recommendation. |
| 795 |
This package transform MARC records to XML or RDF/XML Dublin Core |
| 796 |
and follows the recommendations listed below: |
| 797 |
|
| 798 |
L<DC-XML-GUIDELINES [DCMI Recommendation]|http://dublincore.org/documents/dc-xml-guidelines/> |
| 799 |
|
| 800 |
L<Expressing Simple Dublin Core in RDF/XML|http://dublincore.org/documents/dcmes-xml/> This recommendation is |
| 801 |
superseded by DC-RDFrecommendation but not obsoleted at all. |
| 802 |
|
| 803 |
L<DC-RDF [DCMI Recommendation]|http://dublincore.org/documents/dc-rdf/> |
| 804 |
|
| 805 |
=head2 Methods |
| 806 |
|
| 807 |
=over 12 |
| 808 |
|
| 809 |
=item C<new> |
| 810 |
|
| 811 |
Return a new C4::DublinCoreTransformer object. |
| 812 |
This constructor sets the output type (RDF or XML), the level of the standard (Simple DC or Qualified DC) and the MARC record to be transformed to DC. |
| 813 |
|
| 814 |
my $objectDC = DublinCoreTransformer->new({ |
| 815 |
type => 'rdf', |
| 816 |
qualified => 0, |
| 817 |
get_marc_record => $record}); |
| 818 |
|
| 819 |
=item C<get_xml_data> |
| 820 |
|
| 821 |
This method is just for DC-XML: Fetch the xml root element and the schema in .xsd format. |
| 822 |
|
| 823 |
$objectDC->get_xml_data({ |
| 824 |
root => 'metadata', |
| 825 |
xsi_schemaLocation => 'http://example.org/myapp/ http://example.org/myapp/schema.xsd', |
| 826 |
entities => 1, |
| 827 |
}); |
| 828 |
|
| 829 |
The entities argument is for extended xml entities, i.e., placing entities to zero will codify only the five basic entities e.g., <, >, ", &, and &apos. Conversely if it is encoded to one, all entities will coded into decimal. |
| 830 |
|
| 831 |
Additionally, has been included the optional_namespace and namespace_url arguments to include another XML name space. |
| 832 |
|
| 833 |
NOTE: This method is only for use of the DC-XML recommendation, i.e., type => 'xml'. |
| 834 |
See C<new> method or the section SYNOPSIS. |
| 835 |
|
| 836 |
=item C<get_rdf_data> |
| 837 |
|
| 838 |
This method is only for DC-RDF: Fetch the IRI identifier of the resource being described. As well, you can pass the format you want to generate with the rdf, by default the DC-RDF specify RDF/XML but RDF can represent in other formats listed below: |
| 839 |
|
| 840 |
. ntriples |
| 841 |
|
| 842 |
. nquads |
| 843 |
|
| 844 |
. rdfxml |
| 845 |
|
| 846 |
. rdfjson |
| 847 |
|
| 848 |
. ntriples-canonical |
| 849 |
|
| 850 |
. turtle |
| 851 |
|
| 852 |
$objectDC->get_rdf_data({ |
| 853 |
rdf_subject => 'http://opac.fmoues.edu.sv/cgi-bin/koha/opac-detail.pl?biblionumber=17589', |
| 854 |
rdf_format => 'rdfxml', |
| 855 |
entities => 0|1, |
| 856 |
}); |
| 857 |
|
| 858 |
NOTE: The entities is needed only for rdfxml value and is used like C<get_xml_data> method. |
| 859 |
|
| 860 |
=item C<write_xml> |
| 861 |
|
| 862 |
Return the xml object as string. |
| 863 |
|
| 864 |
NOTE: This method is only for use of the DC-XML recommendation, i.e., type => 'xml'. |
| 865 |
See C<new> and C<get_xml_data> methods or the section SYNOPSIS. |
| 866 |
|
| 867 |
=item C<write_rdf> |
| 868 |
|
| 869 |
Return the rdf object as string. |
| 870 |
|
| 871 |
NOTE: This method is only for use of the DC-RDF recommendation, i.e., type => 'rdf'. |
| 872 |
See C<new> and C<get_rdf_data> methods or the section SYNOPSIS. |
| 873 |
|
| 874 |
=back |
| 875 |
|
| 876 |
=head1 VERSION |
| 877 |
|
| 878 |
version 0.01 |
| 879 |
|
| 880 |
=head1 RELATES MODULES |
| 881 |
|
| 882 |
This package requires the following modules: |
| 883 |
|
| 884 |
DublinCore::Record |
| 885 |
|
| 886 |
MARC::Record |
| 887 |
|
| 888 |
MARC::File::USMARC |
| 889 |
|
| 890 |
MARC::Crosswalk::DublinCore |
| 891 |
|
| 892 |
HTML::Entities |
| 893 |
|
| 894 |
XML::Entities |
| 895 |
|
| 896 |
XML::LibXML |
| 897 |
|
| 898 |
RDF::Helper |
| 899 |
|
| 900 |
=head2 SEE ALSO |
| 901 |
|
| 902 |
Dublin Core Web Page L<http://dublincore.org>. |
| 903 |
|
| 904 |
=head1 LICENSE |
| 905 |
|
| 906 |
Copyright (c) 2014 Koha Community. All rights reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. |
| 907 |
|
| 908 |
=head1 AUTHOR |
| 909 |
|
| 910 |
Hector Eduardo Catro Avalos E<lt> hector.hecaxmmx at gmail dot com E<gt> |
| 911 |
and the Koha Development Team. |
| 912 |
|
| 913 |
=cut |