def map_oai_pmh_record(self, collection_id, record_etree):
record_identifier = record_etree.find('header').find('identifier').text
object_id = ObjectId.parse(str(collection_id) + '/' + urllib.quote(record_identifier, ''))
metadata_etree = record_etree.find('metadata')
object_builder = \
self._ObjectBuilder(
collection_id=collection_id,
institution_id=collection_id.getInstitutionId(),
record_identifier=record_identifier
)
for etree in metadata_etree.find(self._UNTL_NS + 'metadata'):
assert etree.tag.startswith(self._UNTL_NS)
tag = etree.tag[len(self._UNTL_NS):]
method_name = '_parse_record_metadata_' + tag + '_element'
try:
method = getattr(self, method_name)
except AttributeError:
self._logger.warn("no such method '%s' for record %s with text '%s'", method_name, record_identifier, etree.text)
continue
method(
element=etree,
object_builder=object_builder
)
return ObjectEntry(object_id, object_builder.build())
def map_omeka_item(self, collection_id, endpoint_url, omeka_item, omeka_item_files, square_thumbnail_height_px, square_thumbnail_width_px):
object_id = ObjectId.parse(str(collection_id) + '/' + str(omeka_item.id))
object_builder = \
self._ObjectBuilder(
endpoint_url=endpoint_url,
logger=self._logger,
object_id=object_id,
omeka_item=omeka_item
)
for element_text in omeka_item.element_texts:
text = element_text.text.strip()
if len(text) == 0:
continue
self._map_omeka_item_element(
element_name=element_text.element.name,
element_set_name=element_text.element_set.name,
object_builder=object_builder,
text=text
)
self._map_omeka_item_files(
object_builder=object_builder,
omeka_item=omeka_item,
omeka_item_files=omeka_item_files,
square_thumbnail_height_px=square_thumbnail_height_px,
square_thumbnail_width_px=square_thumbnail_width_px
)
if len(object_builder.work_types) == 0 and omeka_item.item_type is not None:
self._map_omeka_item_type(
object_builder=object_builder,
omeka_item_type=omeka_item.item_type
)
tag_names = [tag.name for tag in omeka_item.tags]
if len(tag_names) > 0:
self._map_omeka_item_tags(
object_builder=object_builder,
tag_names=tuple(tag_names)
)
return ObjectEntry(object_id, object_builder.build())
def map_omeka_item(self, collection_id, endpoint_url, omeka_item, omeka_item_files, square_thumbnail_height_px, square_thumbnail_width_px):
object_id = ObjectId.parse(str(collection_id) + '/' + str(omeka_item.id))
vocab_ref = VocabRef.Builder().setVocab(Vocab.COSTUME_CORE).build()
feature_name = None
omeka_collection_id = int(collection_id.getUnqualifiedCollectionId())
for item in self.OMEKA_COLLECTIONS.iteritems():
if item[1] == omeka_collection_id:
feature_name = item[0]
break
assert feature_name is not None
feature_value = None
item_image_credit_line = item_image_license = None
for element_text in omeka_item.element_texts:
if len(element_text.text) == 0:
continue
if element_text.element_set.name == 'Dublin Core':
if element_text.element.name == 'Title':
if feature_value is None:
feature_value = element_text.text
elif element_text.element_set.name == 'Item Type Metadata':
if element_text.element.name == 'Image Creator':
item_image_credit_line = element_text.text
elif element_text.element.name == 'Image License':
item_image_license = element_text.text
else:
self._logger.warn("Omeka item %d has unknown element set name '%s'", omeka_item.id, element_text.element_set.name)
object_builder = \
Object.Builder()\
.setCollectionId(collection_id)\
.setHidden(True)\
.setInstitutionId(collection_id.getInstitutionId())\
.setStructures(\
StructureSet.Builder().setElements(ImmutableList.of(
Structure.Builder()
.setText(feature_value)
.setType(
StructureType.Builder()
.setText(feature_name)
.setVocabRef(vocab_ref)
.build()
)
.build()
))
.build()
)\
.setTitles(
TitleSet.Builder().setElements(ImmutableList.of(
Title.Builder()
.setText("%(feature_value)s" % locals())
.setType(TitleType.DESCRIPTIVE)
.build()
))
.build()
)\
.setViewType(ViewType.DETAIL)
images = []
for file_ in omeka_item_files:
if not file_.mime_type.startswith('image/'):
continue
image_credit_line = item_image_credit_line
image_license = item_image_license
for element_text in file_.element_texts:
if element_text.element_set.name == 'Dublin Core':
if element_text.element.name == 'License':
image_license = element_text.text
elif element_text.element.name == 'Provenance':
image_credit_line = element_text.text
if image_credit_line is None or len(image_credit_line) == 0:
self._logger.warn("Omeka item %d has a file %d missing a Provenance", omeka_item.id, file_.id)
continue
if image_license is None or len(image_license) == 0:
self._logger.warn("Omeka item %d has a file %d missing a License", omeka_item.id, file_.id)
continue
license_vocab_ref = None
if image_license.lower() == 'public domain':
rights_type = RightsType.PUBLIC_DOMAIN
elif image_license == 'CC0':
rights_type = RightsType.LICENSED
license_vocab_ref = \
VocabRef.Builder()\
.setVocab(Vocab.CREATIVE_COMMONS)\
.setUri(Uri.parse('https://creativecommons.org/publicdomain/zero/1.0/'))\
.build()
elif image_license.startswith('CC BY-SA '):
rights_type = RightsType.LICENSED
version = image_license[len('CC BY-SA '):]
float(version)
license_vocab_ref = \
VocabRef.Builder()\
.setVocab(Vocab.CREATIVE_COMMONS)\
.setUri(Uri.parse("https://creativecommons.org/licenses/by-sa/%s/" % version))\
.build()
else:
rights_type = RightsType.LICENSED
image_builder = Image.Builder() # @UndefinedVariable
file_urls = file_.file_urls
image_builder.setOriginal(ImageVersion.Builder().setUrl(Url.parse(file_urls.original)).build())
image_builder.setRights(
RightsSet.Builder().setElements(ImmutableList.of(
Rights.Builder()
.setLicenseVocabRef(Optional.fromNullable(license_vocab_ref))
.setRightsHolder(image_credit_line)
.setText(image_license)
.setType(rights_type)
.build()
))
.build()
)
if file_urls.square_thumbnail is None:
self._logger.warn("Omeka item %d has a file %d missing a square thumbnail", omeka_item.id, file_.id)
continue
image_builder.setSquareThumbnail(
ImageVersion.Builder()
.setHeightPx(UnsignedInteger.valueOf(square_thumbnail_height_px))
.setUrl(Url.parse(file_urls.square_thumbnail))
.setWidthPx(UnsignedInteger.valueOf(square_thumbnail_width_px))
.build()
)
images.append(image_builder.build())
if len(images) > 0:
object_builder.setImages(ImmutableList.copyOf(images))
else:
self._logger.warn("Omeka item %d has no valid images", omeka_item.id)
object_ = object_builder.build()
object_id = ObjectId.parse(str(collection_id) + '/' + urllib.quote(feature_value, ''))
return ObjectEntry(object_id, object_)
