class SCHEMAOrganization(RDFClass):
type = RDFField(predicate=RDF.type, object=ns.SCHEMA.Organization)
title_pl = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='en'),
allow_null=False)
class GeonamesDCTLocation(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.Location)
geonames_url = RDFField(predicate=ns.DCT.identifier,
object_type=URIRef,
required=False)
centroid = RDFField(predicate=ns.DCAT.centroid,
object_type=partial(Literal, datatype=ns.GSP.asWKT),
required=False)
class SKOSConceptScheme(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.SKOS.ConceptScheme)
title_pl = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='pl'),
required=False)
title_en = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='en'),
required=False)
class SKOSConcept(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.SKOS.Concept)
title_pl = RDFField(predicate=ns.SKOS.prefLabel,
object_type=partial(Literal, lang='pl'),
required=False)
title_en = RDFField(predicate=ns.SKOS.prefLabel,
object_type=partial(Literal, lang='en'),
required=False)
scheme = RDFNestedField('SKOSConceptScheme', predicate=ns.SKOS.inScheme)
class VCARDKind(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.VCARD.Kind)
type = RDFField(predicate=RDF.type, object=ns.VCARD.Kind)
name = RDFField(predicate=ns.VCARD.fn,
object_type=partial(Literal, datatype=XSD.string),
allow_null=False)
email = RDFField(predicate=ns.VCARD.hasEmail,
object_type=URIRef,
required=False)
class BaseDCATDeserializer(RDFClass):
ext_ident = RDFField(predicate=ns.DCT.identifier)
title_pl = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='pl'),
try_non_lang=True)
title_en = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='en'))
description_pl = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='pl'),
try_non_lang=True)
description_en = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='en'))
created = RDFField(predicate=ns.DCT.issued,
object_type=partial(Literal, datatype=XSD.dateTime))
modified = RDFField(predicate=ns.DCT.modified,
object_type=partial(Literal, datatype=XSD.dateTime))
def get_ext_ident_object(self, triple_store, subject, field):
ident = [
value.split('/')[-1] if _is_valid_uri(value) else value
for value in triple_store.objects(subject=subject,
predicate=field.predicate)
]
if not ident and _is_valid_uri(subject):
ident = [subject.split('/')[-1]]
return ident
class FOAFAgent(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.FOAF.Agent)
title_pl = RDFField(predicate=ns.FOAF.name,
object_type=partial(Literal, lang='pl'),
allow_null=False)
title_en = RDFField(predicate=ns.FOAF.name,
object_type=partial(Literal, lang='en'),
allow_null=False)
email = RDFField(predicate=ns.FOAF.mbox,
object_type=Literal,
required=False)
homepage = RDFField(predicate=ns.FOAF.homepage,
object_type=URIRef,
required=False)
agent_type = RDFNestedField('SKOSConcept',
predicate=ns.DCT.type,
required=False)
class DCATDatasetDeserializer(BaseDCATDeserializer):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCAT.Dataset)
categories = DCATVocabularyField(predicate=ns.DCAT.theme,
vocabulary_name='theme',
many=True)
resources = RDFNestedField('DCATDistributionDeserializer')
update_frequency = DCATVocabularyField(predicate=ns.DCT.AccrualPeriodicity,
vocabulary_name='frequency')
tags = RDFField(predicate=ns.DCAT.keyword, many=True)
def get_tags_object(self, triple_store, subject, field):
return [{
'name': value.value,
'lang': value.language
} if value.language in ('pl', 'en') else {
'name': value.value
} if not value.language else {}
for value in triple_store.objects(subject=subject,
predicate=field.predicate)]
class HYDRAPagedCollection(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.HYDRA.PagedCollection)
count = RDFField(predicate=ns.HYDRA.totalItems)
per_page = RDFField(predicate=ns.HYDRA.itemsPerPage)
next_page = RDFField(predicate=ns.HYDRA.nextPage, required=False)
last_page = RDFField(predicate=ns.HYDRA.lastPage, required=False)
prev_page = RDFField(predicate=ns.HYDRA.previousPage, required=False)
first_page = RDFField(predicate=ns.HYDRA.firstPage, required=False)
def get_subject(self, data):
return BNode('PagedCollection')
class SCHEMACatalog(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.SCHEMA.DataCatalog)
language_pl = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='pl')
language_en = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='en')
title_pl = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='en'),
allow_null=False)
description_pl = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
description_en = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
modified = RDFField(predicate=ns.SCHEMA.dateModified)
homepage = RDFField(predicate=ns.SCHEMA.url)
dataset = RDFField(predicate=ns.SCHEMA.dataset,
object_type=URIRef,
many=True)
def get_subject(self, data):
return URIRef(CATALOG_URL)
def get_data(self, data):
return {
'title_pl': config.CATALOG__TITLE_PL,
'title_en': config.CATALOG__TITLE_EN,
'description_pl': config.CATALOG__DESCRIPTION_PL,
'description_en': config.CATALOG__DESCRIPTION_EN,
'modified': data.get('catalog_modified'),
'homepage': settings.BASE_URL,
'dataset': data.get('dataset_refs'),
}
class DCATDistributionDeserializer(BaseDCATDeserializer):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCAT.Distribution)
format = DCATVocabularyField(predicate=ns.DCT['format'],
vocabulary_name='file-type')
file_mimetype = DCATVocabularyField(predicate=ns.DCAT.mediaType,
vocabulary_name='media-type')
description_pl = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='pl'))
description_en = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='en'))
created = RDFField(predicate=ns.DCT.issued,
object_type=partial(Literal, datatype=XSD.dateTime))
modified = RDFField(predicate=ns.DCT.modified,
object_type=partial(Literal, datatype=XSD.dateTime))
link = RDFField(predicate=ns.DCAT.accessURL, object_type=URIRef)
license = LicenseDCATVocabularyField()
class SCHEMADataset(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.SCHEMA.Dataset)
resources = RDFNestedField('SCHEMADistribution',
predicate=ns.SCHEMA.distribution,
many=True)
organization = RDFNestedField('SCHEMAOrganization',
predicate=ns.SCHEMA.creator)
language_pl = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='pl')
language_en = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='en')
identifier = RDFField(predicate=ns.SCHEMA.url)
id = RDFField(predicate=ns.SCHEMA.identifier)
title_pl = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='en'),
allow_null=False)
notes_pl = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
notes_en = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
status = RDFField(predicate=ns.SCHEMA.creativeWorkStatus)
created = RDFField(predicate=ns.SCHEMA.dateCreated)
modified = RDFField(predicate=ns.SCHEMA.dateModified)
version = RDFField(predicate=ns.SCHEMA.version)
tags = RDFField(predicate=ns.SCHEMA.keywords)
license = RDFField(predicate=ns.SCHEMA.license, required=False)
def get_subject(self, data):
return URIRef(data['frontend_absolute_url'])
def get_data(self, data):
data['tags'] = ', '.join(data['tags'])
data['catalog'] = CATALOG_URL
return data
def get_organization_subject(self, data):
return URIRef(data['access_url'])
def get_resources_subject(self, data):
return URIRef(data['access_url'])
class SCHEMADistribution(RDFClass):
type = RDFField(predicate=RDF.type, object=ns.SCHEMA.DataDownload)
language_pl = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='pl')
language_en = RDFField(predicate=ns.SCHEMA.inLanguage, object_value='en')
title_pl = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.SCHEMA.name,
object_type=partial(Literal, lang='en'),
allow_null=False)
description_pl = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
description_en = RDFField(predicate=ns.SCHEMA.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
status = RDFField(predicate=ns.SCHEMA.creativeWorkStatus)
created = RDFField(predicate=ns.SCHEMA.dateCreated)
modified = RDFField(predicate=ns.SCHEMA.dateModified)
access_url = RDFField(predicate=ns.SCHEMA.url)
download_url = RDFField(predicate=ns.SCHEMA.contentUrl)
file_mimetype = RDFField(predicate=ns.SCHEMA.encodingFormat)
license = RDFField(predicate=ns.SCHEMA.license)
def get_subject(self, data):
return URIRef(data['access_url'])
class DCATDataset(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCAT.Dataset)
resources = RDFNestedField('DCATDistribution',
predicate=ns.DCAT.distribution,
many=True)
organization = RDFNestedField('FOAFAgent', predicate=ns.DCT.publisher)
contact_point = RDFNestedField('VCARDKind', predicate=ns.DCAT.contactPoint)
categories = RDFNestedField('SKOSConcept',
predicate=ns.DCAT.theme,
many=True)
landing_page = RDFNestedField('FOAFDocument',
predicate=ns.DCAT.landingPage)
update_frequency = RDFNestedField('DCTFrequency',
predicate=ns.DCT.accrualPeriodicity)
language_pl = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
language_en = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
if is_enabled('S38_dcat_spatial_data.be'):
spatial = RDFNestedField('GeonamesDCTLocation',
predicate=ns.DCT.spatial,
many=True)
identifier = RDFField(predicate=ns.DCT.identifier)
title_pl = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='en'),
allow_null=False)
notes_pl = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
notes_en = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
created = RDFField(predicate=ns.DCT.issued)
modified = RDFField(predicate=ns.DCT.modified)
version = RDFField(predicate=ns.OWL.versionInfo)
tags = RDFField(predicate=ns.DCAT.keyword, many=True)
def get_subject(self, data):
return URIRef(data['frontend_absolute_url'])
def get_organization_subject(self, data):
return URIRef(data['access_url'])
def get_resources_subject(self, data):
return URIRef(data['access_url'])
def get_landing_page_subject(self, data):
return URIRef(data)
def get_update_frequency_data(self, data):
update_frequency = data['update_frequency']
if update_frequency is None:
update_frequency = f"{VOCABULARIES['frequency']}UNKNOWN"
return {
'subject': URIRef(update_frequency),
'scheme': URIRef(VOCABULARIES['frequency'].rstrip('/'))
}
def get_language_pl_data(self, data):
return {
'subject': URIRef(VOCABULARIES['language'] + 'POL'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
}
def get_language_en_data(self, data):
return {
'subject': URIRef(VOCABULARIES['language'] + 'ENG'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
}
def get_categories_subject(self, data):
return URIRef(VOCABULARIES['theme'] + data['code'])
def get_categories_data(self, data):
for category in data['categories']:
category['scheme'] = {
'subject': URIRef(VOCABULARIES['theme'].rstrip('/')),
'title_pl': 'Kategoria danych',
'title_en': 'Data theme',
}
return data['categories']
def get_contact_point_subject(self, data):
return BNode('VcardKind')
def get_contact_point_data(self, data):
result = {
'name': config.DATASET__CONTACT_POINT__FN,
}
if config.DATASET__CONTACT_POINT__HAS_EMAIL:
result['email'] = config.DATASET__CONTACT_POINT__HAS_EMAIL
return result
def get_spatial_data(self, data):
spatial_details = [{
'centroid': d['centroid']
} if d['geonames_url'] is None else {
'geonames_url': d['geonames_url']
} for d in data['spatial']]
return spatial_details
def get_spatial_subject(self, data):
data_str = data.get('geonames_url', data.get('centroid', ''))
return BNode('DCTLocation' +
hashlib.sha256(data_str.encode('utf-8')).hexdigest())
class DCTMediaType(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.MediaType)
class DCATCatalog(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCAT.Catalog)
publisher = RDFNestedField('FOAFAgent', predicate=ns.DCT.publisher
) # TODO warning, brak słownika kontrolowanego
language_pl = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
language_en = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
homepage = RDFNestedField('FOAFDocument', predicate=ns.FOAF.homepage)
spatial = RDFNestedField('DCTLocation', predicate=ns.DCT.spatial)
theme_taxonomy = RDFNestedField('SKOSConceptScheme',
predicate=ns.DCAT.themeTaxonomy)
dataset = RDFField(predicate=ns.DCAT.dataset,
object_type=URIRef,
many=True)
title_pl = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='pl'),
allow_null=False)
title_en = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='en'),
allow_null=False)
description_pl = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
description_en = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
issued = RDFField(predicate=ns.DCT.issued,
object_type=partial(Literal, datatype=XSD.date))
modified = RDFField(predicate=ns.DCT.modified,
object_type=partial(Literal, datatype=XSD.dateTime))
def get_subject(self, data):
return URIRef(CATALOG_URL)
def get_publisher_subject(self, data):
return BNode('CatalogPublisher')
def get_data(self, data):
result = {
'title_pl': config.CATALOG__TITLE_PL,
'title_en': config.CATALOG__TITLE_EN,
'description_pl': config.CATALOG__DESCRIPTION_PL,
'description_en': config.CATALOG__DESCRIPTION_EN,
'issued': config.CATALOG__ISSUED,
'spatial': {
'subject': URIRef(VOCABULARIES['country'] + 'POL'),
'scheme': URIRef(VOCABULARIES['country'].rstrip('/'))
},
'language_pl': {
'subject': URIRef(VOCABULARIES['language'] + 'POL'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
},
'language_en': {
'subject': URIRef(VOCABULARIES['language'] + 'ENG'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
},
'modified': data.get('catalog_modified'),
'homepage': settings.BASE_URL,
'publisher': {
'title_pl': config.CATALOG__PUBLISHER__NAME_PL,
'title_en': config.CATALOG__PUBLISHER__NAME_EN,
'agent_type': {
'subject':
URIRef(VOCABULARIES['publishertype-prefix'] +
'NationalAuthority'),
'scheme': {
'subject':
URIRef(VOCABULARIES['publishertype'].rstrip('/')),
'title_en':
'Publisher Type',
},
'title_pl':
'Organ krajowy',
'title_en':
'National authority',
}
},
'theme_taxonomy': {
'subject': URIRef(VOCABULARIES['theme'].rstrip('/')),
'title_pl': 'Kategoria danych',
'title_en': 'Data theme',
},
'dataset': data.get('dataset_refs'),
}
if config.CATALOG__PUBLISHER__EMAIL:
result['publisher']['email'] = config.CATALOG__PUBLISHER__EMAIL
if config.CATALOG__PUBLISHER__HOMEPAGE:
result['publisher'][
'homepage'] = config.CATALOG__PUBLISHER__HOMEPAGE
return result
def get_homepage_subject(self, data):
return URIRef(data)
class DCTMediaTypeOrExtent(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.MediaTypeOrExtent)
scheme = RDFField(predicate=ns.SKOS.inScheme, object_type=URIRef)
class DCTFrequency(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.Frequency)
scheme = RDFField(predicate=ns.SKOS.inScheme, object_type=URIRef)
class FOAFDocument(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.FOAF.Document)
class DCATDistribution(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCAT.Distribution)
format = RDFNestedField('DCTMediaTypeOrExtent',
predicate=ns.DCT['format'],
required=False)
file_mimetype = RDFNestedField('DCTMediaType', predicate=ns.DCAT.mediaType)
language_pl = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
language_en = RDFNestedField('DCTLinguisticSystem',
predicate=ns.DCT.language)
status = RDFNestedField('SKOSConcept', predicate=ns.ADMS.status)
title_pl = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='pl'))
title_en = RDFField(predicate=ns.DCT.title,
object_type=partial(Literal, lang='en'),
allow_null=False)
description_pl = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='pl'),
allow_null=False)
description_en = RDFField(predicate=ns.DCT.description,
object_type=partial(Literal, lang='en'),
allow_null=False)
created = RDFField(predicate=ns.DCT.issued,
object_type=partial(Literal, datatype=XSD.dateTime))
modified = RDFField(predicate=ns.DCT.modified,
object_type=partial(Literal, datatype=XSD.dateTime))
access_url = RDFField(predicate=ns.DCAT.accessURL, object_type=URIRef)
download_url = RDFField(predicate=ns.DCAT.downloadURL,
object_type=URIRef,
allow_null=False)
file_size = RDFField(predicate=ns.DCAT.byteSize,
object_type=partial(Literal, datatype=XSD.decimal))
license = RDFField(predicate=ns.DCAT.license, object_type=URIRef)
def get_subject(self, data):
return URIRef(data['access_url'])
def get_format_data(self, data):
if data['format'] is not None:
return {
'subject':
URIRef(VOCABULARIES['file-type'] + data['format'].upper()),
'scheme':
URIRef(VOCABULARIES['file-type'].rstrip('/'))
}
def get_file_mimetype_subject(self, data):
if data is not None:
return URIRef(VOCABULARIES['media-type'] + data)
def get_status_data(self, data):
return {
'subject': URIRef(VOCABULARIES['status-prefix'] + 'Completed'),
'scheme': {
'subject': URIRef(VOCABULARIES['status']),
'title_en': 'Status',
},
'title_en': 'Completed',
}
def get_language_pl_data(self, data):
return {
'subject': URIRef(VOCABULARIES['language'] + 'POL'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
}
def get_language_en_data(self, data):
return {
'subject': URIRef(VOCABULARIES['language'] + 'ENG'),
'scheme': URIRef(VOCABULARIES['language'].rstrip('/'))
}
class DCTLocation(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.Location)
scheme = RDFField(predicate=ns.SKOS.inScheme, object_type=URIRef)
class DCTLinguisticSystem(RDFClass):
rdf_type = RDFField(predicate=RDF.type, object=ns.DCT.LinguisticSystem)
scheme = RDFField(predicate=ns.SKOS.inScheme, object_type=URIRef)