X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/e702c3da468a55bb04ba5899ef26ad789a5c8cd1..0e79f07515678e3e678c3637124a27cb81ee8a94:/apps/oai/handlers.py?ds=sidebyside diff --git a/apps/oai/handlers.py b/apps/oai/handlers.py index e37fc084d..197ca7987 100644 --- a/apps/oai/handlers.py +++ b/apps/oai/handlers.py @@ -1,34 +1,44 @@ +# -*- coding: utf-8 -*- +# This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. +# from oaipmh import server, common, metadata, error 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 +from django.utils import timezone + +make_time_naive = lambda d: timezone.localtime(d).replace(tzinfo=None) + +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/', @@ -36,14 +46,24 @@ 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) + year_zero = timezone.make_aware(datetime(1990, 1, 1, 0, 0, 0), timezone.utc) try: earliest_change = \ @@ -59,21 +79,27 @@ class Catalogue(common.ResumptionOAIPMH): earliest_change or earliest_delete 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) + header = common.Header(identifier, make_time_naive(book.changed_at), [], False) if not headers_only: meta = common.Metadata(self.metadata(book)) about = None elif isinstance(book, Deleted): - header = common.Header(identifier, book.deleted_at, [], True) + header = common.Header(identifier, make_time_naive(book.deleted_at), [], True) if not headers_only: meta = common.Metadata({}) about = None @@ -87,7 +113,7 @@ class Catalogue(common.ResumptionOAIPMH): '%s/oaipmh' % WL_BASE, # generate '2.0', # version [m[1] for m in settings.MANAGERS], # adminEmails - self.earliest_datestamp, # earliest datestamp of any change + make_time_naive(self.earliest_datestamp), # earliest datestamp of any change 'persistent', # deletedRecord 'YYYY-MM-DDThh:mm:ssZ', # granularity ['identity'], # compression @@ -98,7 +124,7 @@ 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() @@ -133,7 +159,7 @@ 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) @@ -146,7 +172,20 @@ 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), @@ -158,6 +197,7 @@ 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), @@ -166,9 +206,26 @@ 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.") @@ -179,5 +236,3 @@ returns result, token # tag.name, # tag.description)) # return tags, None - -