X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/38cc2bc1fec598a380371358d0a9d0b6cf4d9f05..49c55f5af6d008f4c8039eebd648e6654194e3e4:/apps/oai/handlers.py?ds=sidebyside diff --git a/apps/oai/handlers.py b/apps/oai/handlers.py index d098a64ad..4ad5c138f 100644 --- a/apps/oai/handlers.py +++ b/apps/oai/handlers.py @@ -10,6 +10,8 @@ from datetime import datetime from lxml import etree from lxml.etree import ElementTree from django.db.models import Q +from django.conf import settings +from django.contrib.sites.models import Site wl_dc_reader = metadata.MetadataReader( @@ -26,6 +28,8 @@ wl_dc_reader = metadata.MetadataReader( 'identifier': ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'), 'source': ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'), 'language': ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'), + #'isPartOf': ('textList', 'rdf:RDF/rdf:Description/dc:relation.isPartOf/text()'), + 'hasPart': ('textList', 'rdf:RDF/rdf:Description/dc:relation.hasPart/text()'), # 'relation': ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'), # 'coverage': ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'), 'rights': ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()') @@ -36,11 +40,21 @@ wl_dc_reader = metadata.MetadataReader( ) +NS_DCTERMS = "http://purl.org/dc/terms/" + + +def nsdcterms(name): + return '{%s}%s' % (NS_DCTERMS, name) + + class Catalogue(common.ResumptionOAIPMH): TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre'] - - def __init__(self): + + def __init__(self, metadata_registry): super(Catalogue, self).__init__() + self.metadata_registry = metadata_registry + + self.oai_id = "oai:" + Site.objects.get_current().domain + ":%s" # earliest change year_zero = datetime(1990, 1, 1, 0, 0, 0) @@ -52,23 +66,26 @@ class Catalogue(common.ResumptionOAIPMH): try: earliest_delete = \ - Deleted.objects.ordery_by('deleted_at')[0].deleted_at + Deleted.objects.exclude(slug__exact=u'').ordery_by('deleted_at')[0].deleted_at except: earliest_delete = year_zero self.earliest_datestamp = earliest_change <= earliest_delete and \ earliest_change or earliest_delete - # admins - self.admin_emails = [u.email for u in User.objects.filter(is_superuser=True)] - def metadata(self, book): - xml = etree.parse(book.xml_file) + try: + xml = etree.parse(book.xml_file) + finally: + book.xml_file.close() md = wl_dc_reader(xml) - return md.getMap() + m = md.getMap() + if book.parent: + m['isPartOf'] = [str(WLURI.from_slug(book.parent.slug))] + return m def record_for_book(self, book, headers_only=False): meta = None - identifier = str(WLURI.from_slug(book.slug)) + identifier = self.slug_to_identifier(book.slug) if isinstance(book, Book): # setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES)) header = common.Header(identifier, book.changed_at, [], False) @@ -89,7 +106,7 @@ class Catalogue(common.ResumptionOAIPMH): 'Wolne Lektury', # generate '%s/oaipmh' % WL_BASE, # generate '2.0', # version - self.admin_emails, # adminEmails + [m[1] for m in settings.MANAGERS], # adminEmails self.earliest_datestamp, # earliest datestamp of any change 'persistent', # deletedRecord 'YYYY-MM-DDThh:mm:ssZ', # granularity @@ -101,11 +118,11 @@ class Catalogue(common.ResumptionOAIPMH): def books(self, tag, from_, until): if tag: # we do not support sets, since they are problematic for deleted books. - raise errror.NoSetHierarchyError("Wolne Lektury does not support sets.") + raise error.NoSetHierarchyError("Wolne Lektury does not support sets.") # books = Book.tagged.with_all([tag]) else: books = Book.objects.all() - deleted = Deleted.objects.filter(slug__isnull=False) + deleted = Deleted.objects.exclude(slug__exact=u'') books = books.order_by('changed_at') deleted = deleted.order_by('deleted_at') @@ -136,7 +153,7 @@ class Catalogue(common.ResumptionOAIPMH): """ Returns (header, metadata, about) for given record. """ - slug = WLURI(kw['identifier']).slug + slug = self.identifier_to_slug(kw['identifier']) try: book = Book.objects.get(slug=slug) return self.record_for_book(book) @@ -149,7 +166,20 @@ Returns (header, metadata, about) for given record. raise error.IdDoesNotExistError("No item for this identifier") return self.record_for_book(deleted_book) + def validate_kw(self, kw): + if 'resumptionToken' in kw: + raise error.BadResumptionTokenError("No resumption token support at this point") + if 'metadataPrefix' in kw and not self.metadata_registry.hasWriter(kw['metadataPrefix']): + raise error.CannotDisseminateFormatError("This format is not supported") + + def identifier_to_slug(self, ident): + return ident.split(':')[-1] + + def slug_to_identifier(self, slug): + return self.oai_id % slug + def listIdentifiers(self, **kw): + self.validate_kw(kw) records = [self.record_for_book(book, headers_only=True) for book in self.books(None, kw.get('from_', None), @@ -161,6 +191,7 @@ Returns (header, metadata, about) for given record. can get a resumptionToken kw. returns result, token """ + self.validate_kw(kw) records = [self.record_for_book(book) for book in self.books(None, kw.get('from_', None), @@ -169,9 +200,26 @@ returns result, token return records, None def listMetadataFormats(self, **kw): - return [('oai_dc', - 'http://www.openarchives.org/OAI/2.0/oai_dc.xsd', - server.NS_OAIDC)] + formats = [ + ('oai_dc', + 'http://www.openarchives.org/OAI/2.0/oai_dc.xsd', + server.NS_OAIDC), + ('qdc', + 'http://dublincore.org/schemas/xmls/qdc/2006/01/06/dcterms.xsd', + NS_DCTERMS)] + if 'identifier' in kw: + slug = self.identifier_to_slug(kw['identifier']) + try: + b = Book.objects.get(slug=slug) + return formats + except: + try: + d = Deleted.objects.get(slug=slug) + return [] + except: + raise error.IdDoesNotExistError("This id does not exist") + else: + return formats def listSets(self, **kw): raise error.NoSetHierarchyError("Wolne Lektury does not support sets.") @@ -182,5 +230,3 @@ returns result, token # tag.name, # tag.description)) # return tags, None - -