def collection_from_details(_db, client, collection_details):
if not (client and isinstance(client, IntegrationClient)):
return None
if collection_details:
# A DataSource may be sent for collections with the
# ExternalIntegration.OPDS_IMPORT protocol.
data_source_name = request.args.get('data_source')
if data_source_name:
data_source_name = urllib.unquote(data_source_name)
collection, ignore = Collection.from_metadata_identifier(
_db, collection_details, data_source=data_source_name)
return collection
return None
def collection_from_details(_db, client, collection_details):
if not (client and isinstance(client, IntegrationClient)):
return None
if collection_details:
# A DataSource may be sent for collections with the
# ExternalIntegration.OPDS_IMPORT protocol.
data_source_name = request.args.get('data_source')
if data_source_name:
data_source_name = urllib.unquote(data_source_name)
collection, ignore = Collection.from_metadata_identifier(
_db, collection_details, data_source=data_source_name
)
return collection
return None
def add_items(self, collection_details):
"""Adds identifiers to a Collection's catalog"""
client = authenticated_client_from_request(self._db)
if isinstance(client, ProblemDetail):
return client
collection, ignore = Collection.from_metadata_identifier(
self._db, collection_details)
urns = request.args.getlist('urn')
messages = []
for urn in urns:
message = None
identifier = None
try:
identifier, ignore = Identifier.parse_urn(self._db, urn)
except Exception as e:
identifier = None
if not identifier:
message = OPDSMessage(urn, INVALID_URN.status_code,
INVALID_URN.detail)
else:
status = HTTP_OK
description = "Already in catalog"
if identifier not in collection.catalog:
collection.catalog_identifier(self._db, identifier)
status = HTTP_CREATED
description = "Successfully added"
message = OPDSMessage(urn, status, description)
messages.append(message)
title = "%s Catalog Item Additions for %s" % (collection.protocol,
client.url)
url = cdn_url_for("add",
collection_metadata_identifier=collection.name,
urn=urns)
addition_feed = AcquisitionFeed(self._db,
title,
url, [],
VerboseAnnotator,
precomposed_entries=messages)
return feed_response(addition_feed)
def remove_items(self, collection_details):
"""Removes identifiers from a Collection's catalog"""
client = authenticated_client_from_request(self._db)
if isinstance(client, ProblemDetail):
return client
collection, ignore = Collection.from_metadata_identifier(
self._db, collection_details)
urns = request.args.getlist('urn')
messages = []
for urn in urns:
message = None
identifier = None
try:
identifier, ignore = Identifier.parse_urn(self._db, urn)
except Exception as e:
identifier = None
if not identifier:
message = OPDSMessage(urn, INVALID_URN.status_code,
INVALID_URN.detail)
else:
if identifier in collection.catalog:
collection.catalog.remove(identifier)
message = OPDSMessage(urn, HTTP_OK, "Successfully removed")
else:
message = OPDSMessage(urn, HTTP_NOT_FOUND,
"Not in catalog")
messages.append(message)
title = "%s Catalog Item Removal for %s" % (collection.protocol,
client.url)
url = cdn_url_for("remove",
collection_metadata_identifier=collection.name,
urn=urns)
removal_feed = AcquisitionFeed(self._db,
title,
url, [],
VerboseAnnotator,
precomposed_entries=messages)
return feed_response(removal_feed)
def updates_feed(self, collection_details):
client = authenticated_client_from_request(self._db)
if isinstance(client, ProblemDetail):
return client
collection, ignore = Collection.from_metadata_identifier(
self._db, collection_details)
last_update_time = request.args.get('last_update_time', None)
if last_update_time:
last_update_time = datetime.strptime(last_update_time,
"%Y-%m-%dT%H:%M:%SZ")
updated_works = collection.works_updated_since(self._db,
last_update_time)
pagination = load_pagination_from_request()
works = pagination.apply(updated_works).all()
title = "%s Collection Updates for %s" % (collection.protocol,
client.url)
def update_url(time=last_update_time, page=None):
kw = dict(_external=True,
collection_metadata_identifier=collection_details)
if time:
kw.update({'last_update_time': last_update_time})
if page:
kw.update(page.items())
return cdn_url_for("updates", **kw)
entries = []
for work in works[:]:
entry = work.verbose_opds_entry or work.simple_opds_entry
entry = etree.fromstring(entry)
if entry:
entries.append(entry)
works.remove(work)
works = [(work.identifier, work) for work in works]
update_feed = LookupAcquisitionFeed(self._db,
title,
update_url(),
works,
VerboseAnnotator,
precomposed_entries=entries)
if len(updated_works.all()) > pagination.size + pagination.offset:
update_feed.add_link_to_feed(
update_feed.feed,
rel="next",
href=update_url(page=pagination.next_page))
if pagination.offset > 0:
update_feed.add_link_to_feed(
update_feed.feed,
rel="first",
href=update_url(page=pagination.first_page))
previous_page = pagination.previous_page
if previous_page:
update_feed.add_link_to_feed(update_feed.feed,
rel="previous",
href=update_url(page=previous_page))
return feed_response(update_feed)
class URNLookupController(CoreURNLookupController):
UNRESOLVABLE_IDENTIFIER = "I can't gather information about an identifier of this type."
IDENTIFIER_REGISTERED = "You're the first one to ask about this identifier. I'll try to find out about it."
WORKING_TO_RESOLVE_IDENTIFIER = "I'm working to locate a source for this identifier."
SUCCESS_DID_NOT_RESULT_IN_PRESENTATION_READY_WORK = "Something's wrong. I have a record of covering this identifier but there's no presentation-ready work to show you."
OPERATION = CoverageRecord.RESOLVE_IDENTIFIER_OPERATION
NO_WORK_DONE_EXCEPTION = u'No work done yet'
log = logging.getLogger("URN lookup controller")
def presentation_ready_work_for(self, identifier):
"""Either return a presentation-ready work associated with the
given `identifier`, or return None.
"""
pools = identifier.licensed_through
if not pools:
return None
# All LicensePools for a given Identifier have the same Work.
work = pools[0].work
if not work or not work.presentation_ready:
return None
return work
def can_resolve_identifier(self, identifier):
"""A chance to determine whether resolution should proceed."""
# We can resolve any ISBN and any Overdrive ID.
#
# We can resolve any Gutenberg ID by looking it up in the open-access
# content server.
#
# We can attempt to resolve URIs by looking them up in the
# open-access content server, though there's no guarantee
# it will work.
if identifier.type in (Identifier.ISBN, Identifier.OVERDRIVE_ID,
Identifier.GUTENBERG_ID, Identifier.URI):
return True
# We can resolve any identifier that's associated with a
# presentation-ready work, since the resolution has already
# been done--no need to speculate about how.
work = self.presentation_ready_work_for(identifier)
if work is None:
return False
return True
def process_urn(self, urn, collection_details=None, **kwargs):
"""Turn a URN into a Work suitable for use in an OPDS feed.
"""
try:
identifier, is_new = Identifier.parse_urn(self._db, urn)
except ValueError, e:
identifier = None
if not identifier:
# Not a well-formed URN.
return self.add_message(urn, 400, INVALID_URN.detail)
if not self.can_resolve_identifier(identifier):
return self.add_message(urn, HTTP_NOT_FOUND,
self.UNRESOLVABLE_IDENTIFIER)
# We are at least willing to try to resolve this Identifier.
# If a Collection was provided by an authenticated IntegrationClient,
# this Identifier is part of the Collection's catalog.
client = authenticated_client_from_request(self._db, required=False)
if client and collection_details:
collection, ignore = Collection.from_metadata_identifier(
self._db, collection_details)
collection.catalog_identifier(self._db, identifier)
if (identifier.type == Identifier.ISBN and not identifier.work):
# There's not always enough information about an ISBN to
# create a full Work. If not, we scrape together the cover
# and description information and force the entry.
return self.make_opds_entry_from_metadata_lookups(identifier)
# All other identifiers need to be associated with a
# presentation-ready Work for the lookup to succeed. If there
# isn't one, we need to register it as unresolved.
work = self.presentation_ready_work_for(identifier)
if work:
# The work has been done.
return self.add_work(identifier, work)
# Work remains to be done.
return self.register_identifier_as_unresolved(urn, identifier)
production_session,
)
log = logging.getLogger(name="Metadata Wrangler configuration import")
try:
_db = production_session()
# Get all of the OPDS_IMPORT collections.
collections = Collection.by_protocol(_db, ExternalIntegration.OPDS_IMPORT)
for collection in collections:
opds_url = collection.external_account_id
if not opds_url:
decoded_collection, ignore = Collection.from_metadata_identifier(
_db, collection.name
)
opds_url = decoded_collection.external_account_id
if not opds_url:
# This shouldn't happen.
log.warn(
'Could not find external_account_id for %r' % collection
)
continue
if opds_url and collection == decoded_collection:
log.info(
'Added URL "%s" to collection %r',
decoded_collection.external_account_id, decoded_collection
)
else:
ExternalIntegration,
production_session,
)
log = logging.getLogger(name="Metadata Wrangler configuration import")
try:
_db = production_session()
# Get all of the OPDS_IMPORT collections.
collections = Collection.by_protocol(_db, ExternalIntegration.OPDS_IMPORT)
for collection in collections:
opds_url = collection.external_account_id
if not opds_url:
decoded_collection, ignore = Collection.from_metadata_identifier(
_db, collection.name)
opds_url = decoded_collection.external_account_id
if not opds_url:
# This shouldn't happen.
log.warn('Could not find external_account_id for %r' %
collection)
continue
if opds_url and collection == decoded_collection:
log.info('Added URL "%s" to collection %r',
decoded_collection.external_account_id,
decoded_collection)
else:
# Somehow the collection has been duplicated. This shouldn't
# happen, but if it does, we shouldn't update the collection's
# data_source on faulty information.
