X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/9fc17f14ed2cdf669ae931051513097c0378413a..9c2555ad140dca1c09cde628bd22aac811e05b7e:/src/oai/handlers.py diff --git a/src/oai/handlers.py b/src/oai/handlers.py index 9ad9e67b7..54a0a20bc 100644 --- a/src/oai/handlers.py +++ b/src/oai/handlers.py @@ -1,17 +1,19 @@ -# -*- 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 import WLURI from datetime import datetime from lxml import etree + from django.conf import settings +from django.contrib.contenttypes.models import ContentType from django.contrib.sites.models import Site from django.utils import timezone +from oaipmh import server, common, metadata, error + +from api.handlers import WL_BASE +from api.models import Deleted +from catalogue.models import Book +from librarian import WLURI def make_time_naive(d): @@ -53,7 +55,6 @@ def nsdcterms(name): return '{%s}%s' % (NS_DCTERMS, name) -# WTF class Catalogue(common.ResumptionOAIPMH): TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre'] @@ -67,81 +68,82 @@ class Catalogue(common.ResumptionOAIPMH): year_zero = timezone.make_aware(datetime(1990, 1, 1, 0, 0, 0), timezone.utc) try: - earliest_change = Book.objects.order_by('changed_at')[0].changed_at + earliest_change = Book.objects.filter(findable=True, preview=False).order_by('changed_at')[0].changed_at except IndexError: earliest_change = year_zero try: earliest_delete = \ - Deleted.objects.exclude(slug__exact=u'').order_by('deleted_at')[0].deleted_at + Deleted.objects.exclude(slug__exact='').order_by('deleted_at')[0].deleted_at except IndexError: earliest_delete = year_zero self.earliest_datestamp = earliest_change if earliest_change <= earliest_delete else earliest_delete - # @staticmethod - # def metadata(book): - # try: - # xml = etree.parse(book.xml_file) - # finally: - # book.xml_file.close() - # md = wl_dc_reader(xml) - # m = md.getMap() - # if book.parent: - # m['isPartOf'] = [str(WLURI.from_slug(book.parent.slug))] - # return m - - # WTF - # def record_for_book(self, book, headers_only=False): - # meta = None - # 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, 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, make_time_naive(book.deleted_at), [], True) - # if not headers_only: - # meta = common.Metadata({}) - # about = None - # if headers_only: - # return header - # return header, meta, about - - # def identify(self, **kw): - # ident = common.Identify( - # 'Wolne Lektury', # generate - # '%s/oaipmh' % unicode(WL_BASE), # generate - # '2.0', # version - # [m[1] for m in settings.MANAGERS], # adminEmails - # make_time_naive(self.earliest_datestamp), # earliest datestamp of any change - # 'persistent', # deletedRecord - # 'YYYY-MM-DDThh:mm:ssZ', # granularity - # ['identity'], # compression - # [] # descriptions - # ) - # return ident - - # def books(self, tag, from_, until): - # if 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.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_) - # if until: - # books = books.filter(changed_at__lte=until) - # deleted = deleted.filter(deleted_at__lte=until) - # return list(books) + list(deleted) + @staticmethod + def metadata(book): + try: + xml = etree.parse(book.xml_file) + finally: + book.xml_file.close() + md = wl_dc_reader(xml) + 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 = 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(None, identifier, make_time_naive(book.changed_at), [], False) + if not headers_only: + meta = common.Metadata(None, self.metadata(book)) + about = None + elif isinstance(book, Deleted): + header = common.Header(None, identifier, make_time_naive(book.deleted_at), [], True) + if not headers_only: + meta = common.Metadata(None, {}) + about = None + else: + raise TypeError('Unknown book class') + if headers_only: + return header + return header, meta, about + + def identify(self, **kw): + ident = common.Identify( + 'Wolne Lektury', # generate + '%s/oaipmh' % str(WL_BASE), # generate + '2.0', # version + [m[1] for m in settings.MANAGERS], # adminEmails + make_time_naive(self.earliest_datestamp), # earliest datestamp of any change + 'persistent', # deletedRecord + 'YYYY-MM-DDThh:mm:ssZ', # granularity + ['identity'], # compression + [] # descriptions + ) + return ident + + def books(self, tag, from_, until): + if 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.filter(findable=True, preview=False) + deleted = Deleted.objects.exclude(slug__exact='') + + 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_) + if until: + books = books.filter(changed_at__lte=until) + deleted = deleted.filter(deleted_at__lte=until) + return list(books) + list(deleted) # @staticmethod # def tag_to_setspec(tag): @@ -158,82 +160,79 @@ class Catalogue(common.ResumptionOAIPMH): # return tag # raise error.NoSetHierarchyError("Setspec should have two components: category:slug") - # def getRecord(self, **kw): - # """Returns (header, metadata, about) for given record.""" - # slug = self.identifier_to_slug(kw['identifier']) - # try: - # book = Book.objects.get(slug=slug) - # return self.record_for_book(book) - # except Book.DoesNotExist: - # book_type = ContentType.objects.get_for_model(Book) - # try: - # deleted_book = Deleted.objects.get(content_type=book_type, - # slug=slug) - # except: - # 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), - # kw.get('until', None))] - # return records, None - - # def listRecords(self, **kw): - # """ - # 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), - # kw.get('until', None))] - # - # return records, None - - # def listMetadataFormats(self, **kw): - # 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.") - # # 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 + def getRecord(self, **kw): + """Returns (header, metadata, about) for given record.""" + slug = self.identifier_to_slug(kw['identifier']) + try: + book = Book.objects.get(slug=slug) + return self.record_for_book(book) + except Book.DoesNotExist: + book_type = ContentType.objects.get_for_model(Book) + try: + deleted_book = Deleted.objects.get(content_type=book_type, slug=slug) + except: + 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_'), kw.get('until'))] + return records, None + + def listRecords(self, **kw): + """ + 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), + kw.get('until', None))] + + return records, None + + def listMetadataFormats(self, **kw): + 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: + Book.objects.get(slug=slug) + return formats + except Book.DoesNotExist: + try: + Deleted.objects.get(slug=slug) + return [] + except Deleted.DoesNotExist: + raise error.IdDoesNotExistError("This id does not exist") + else: + return formats + + def listSets(self, **kw): + 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