X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/a728fedba796dcf8d969c6f96def0f630d05f2f9..d6c2be651d0cc889b1dbc5b73c7659a8a2ff7704:/apps/oai/handlers.py diff --git a/apps/oai/handlers.py b/apps/oai/handlers.py index 9e00b2164..30669ac95 100644 --- a/apps/oai/handlers.py +++ b/apps/oai/handlers.py @@ -3,32 +3,38 @@ from catalogue.models import Book, Tag from api.models import Deleted from api.handlers import WL_BASE from librarian.dcparser import BookInfo -from librarian import WLURI from django.contrib.contenttypes.models import ContentType from django.contrib.auth.models import User 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_XPATH = '(.|*)/rdf:RDF/rdf:Description/%s/text()' wl_dc_reader = metadata.MetadataReader( fields={ - 'title': ('textList', 'rdf:RDF/rdf:Description/dc:title/text()'), - 'creator': ('textList', 'rdf:RDF/rdf:Description/dc:creator/text()'), - 'subject': ('textList', 'rdf:RDF/rdf:Description/dc:subject.period/text() | rdf:RDF/rdf:Description/dc:subject.type/text() | rdf:RDF/rdf:Description/dc:subject.genre/text()'), - 'description': ('textList', 'rdf:RDF/rdf:Description/dc:description/text()'), - 'publisher': ('textList', 'rdf:RDF/rdf:Description/dc:publisher/text()'), - 'contributor': ('textList', 'rdf:RDF/rdf:Description/dc:contributor.editor/text() | rdf:RDF/rdf:Description/dc:contributor.translator/text() | rdf:RDF/rdf:Description/dc:contributor.technical_editor/text()'), - 'date': ('textList', 'rdf:RDF/rdf:Description/dc:date/text()'), - 'type': ('textList', 'rdf:RDF/rdf:Description/dc:type/text()'), - 'format': ('textList', 'rdf:RDF/rdf:Description/dc:format/text()'), - '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()'), + 'title': ('textList', WL_DC_READER_XPATH % 'dc:title'), + 'creator': ('textList', WL_DC_READER_XPATH % 'dc:creator'), + 'subject': ('textList', (WL_DC_READER_XPATH + ' | ' + WL_DC_READER_XPATH + ' | ' + WL_DC_READER_XPATH) % + ('dc:subject.period', 'dc:subject.type', 'dc:subject.genre')), + 'description': ('textList', WL_DC_READER_XPATH % 'dc:description'), + 'publisher': ('textList', WL_DC_READER_XPATH % 'dc:publisher'), + 'contributor': ('textList', (WL_DC_READER_XPATH + ' | ' + WL_DC_READER_XPATH + ' | ' + WL_DC_READER_XPATH) % + ('dc:contributor.editor', 'dc:contributor.translator', 'dc:contributor.technical_editor')), + 'date': ('textList', WL_DC_READER_XPATH % 'dc:date'), + 'type': ('textList', WL_DC_READER_XPATH % 'dc:type'), + 'format': ('textList', WL_DC_READER_XPATH % 'dc:format'), + 'identifier': ('textList', WL_DC_READER_XPATH % 'dc:identifier.url'), + 'source': ('textList', WL_DC_READER_XPATH % 'dc:source'), + 'language': ('textList', WL_DC_READER_XPATH % 'dc:language'), + #'isPartOf': ('textList', 'rdf:RDF/rdf:Description/dc:relation.isPartOf/text()'), + 'hasPart': ('textList', WL_DC_READER_XPATH % 'dc:relation.hasPart'), # '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()') + 'rights': ('textList', WL_DC_READER_XPATH % 'dc:rights') }, namespaces={ 'dc': 'http://purl.org/dc/elements/1.1/', @@ -36,11 +42,21 @@ 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) @@ -52,26 +68,26 @@ 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() 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, setSpec, False) + # setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES)) + header = common.Header(identifier, book.changed_at, [], False) if not headers_only: meta = common.Metadata(self.metadata(book)) about = None @@ -89,7 +105,7 @@ 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 @@ -100,17 +116,18 @@ class Catalogue(common.ResumptionOAIPMH): def books(self, tag, from_, until): if tag: - books = Book.tagged.with_all([tag]) + # we do not support sets, since they are problematic for deleted books. + 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') if from_: books = books.filter(changed_at__gte=from_) deleted = deleted.filter(deleted_at__gte=from_) - print "DELETED:%s" % deleted if until: books = books.filter(changed_at__lte=until) deleted = deleted.filter(deleted_at__lte=until) @@ -126,16 +143,16 @@ class Catalogue(common.ResumptionOAIPMH): cs = s.split(':') if len(cs) == 2: if not cs[0] in Catalogue.TAG_CATEGORIES: - raise error.NoSetHierarchyError() + raise error.NoSetHierarchyError("No category part in set") tag = Tag.objects.get(slug=cs[1], category=cs[0]) return tag - raise error.NoSetHierarchyError() + raise error.NoSetHierarchyError("Setspec should have two components: category:slug") def getRecord(self, **kw): """ 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) @@ -145,15 +162,25 @@ Returns (header, metadata, about) for given record. deleted_book = Deleted.objects.get(content_type=book_type, slug=slug) except: - raise error.NoRecordsMatchError() + 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): - print "list identifiers %s" % (kw, ) + self.validate_kw(kw) records = [self.record_for_book(book, headers_only=True) for - book in self.books( - self.setspec_to_tag( - kw.get('set', None)), + book in self.books(None, kw.get('from_', None), kw.get('until', None))] return records, None @@ -163,27 +190,42 @@ 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( - self.setspec_to_tag( - kw.get('set', None)), + book in self.books(None, kw.get('from_', None), kw.get('until', None))] return records, None def listMetadataFormats(self, **kw): - return [('oaidc', - '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): - tags = [] - for category in Catalogue.TAG_CATEGORIES: - for tag in Tag.objects.filter(category=category): - tags.append(("%s:%s" % (tag.category, tag.slug), - tag.name, - tag.description)) - return tags, None - - + raise error.NoSetHierarchyError("Wolne Lektury does not support sets.") + # tags = [] + # for category in Catalogue.TAG_CATEGORIES: + # for tag in Tag.objects.filter(category=category): + # tags.append(("%s:%s" % (tag.category, tag.slug), + # tag.name, + # tag.description)) + # return tags, None