def indexer_receiver(sender,
json=None,
record=None,
index=None,
**dummy_kwargs):
"""Connect to before_record_index signal to transform record for ES."""
def string_to_object(val, field_name):
return {field_name: val}
if 'keywords' in json and len(json['keywords']) > 0 and isinstance(
json['keywords'][0], str):
json['keywords'] = [
string_to_object(s, 'keyword') for s in json.get('keywords', )
]
if 'disciplines' in json and len(json['disciplines']) > 0 and isinstance(
json['disciplines'][0], str):
json['disciplines'] = [
string_to_object(s, 'discipline_name') for s in json['disciplines']
]
if 'external_pids' in json['_deposit']:
# Keep the 'external_pids' if the record is a draft (deposit) or
# if the files are public.
if (not is_deposit(record.model) and allow_public_file_metadata(json)):
json['external_pids'] = json['_deposit']['external_pids']
del json['_deposit']['external_pids']
if not index.startswith('records'):
return
try:
if '_files' in json:
if not allow_public_file_metadata(json):
for f in json['_files']:
del f['key']
del json['_deposit']
json['_created'] = pytz.utc.localize(record.created).isoformat()
json['_updated'] = pytz.utc.localize(record.updated).isoformat()
json['owners'] = record['_deposit']['owners']
json['_internal'] = dict()
# add the 'is_last_version' flag
parent_pid = b2share_parent_pid_fetcher(None, record).pid_value
pid = b2share_record_uuid_fetcher(None, record).pid_value
last_version_pid = PIDVersioning(
parent=RecordUUIDProvider.get(parent_pid).pid).last_child
json['_internal']['is_last_version'] = \
(last_version_pid.pid_value == pid)
# insert the bucket id for link generation in search results
record_buckets = RecordsBuckets.query.filter(
RecordsBuckets.record_id == record.id).all()
if record_buckets:
json['_internal']['files_bucket_id'] = \
str(record_buckets[0].bucket_id)
except Exception:
raise
def indexer_receiver(sender, json=None, record=None, index=None,
**dummy_kwargs):
"""Connect to before_record_index signal to transform record for ES."""
if not index.startswith('records'):
return
try:
if '_files' in json:
if not allow_public_file_metadata(json):
for f in json['_files']:
del f['key']
del json['_deposit']
json['_created'] = pytz.utc.localize(record.created).isoformat()
json['_updated'] = pytz.utc.localize(record.updated).isoformat()
json['owners'] = record['_deposit']['owners']
json['_internal'] = dict()
# add the 'is_last_version' flag
parent_pid = b2share_parent_pid_fetcher(None, record).pid_value
pid = b2share_record_uuid_fetcher(None, record).pid_value
last_version_pid = PIDVersioning(
parent=RecordUUIDProvider.get(parent_pid).pid
).last_child
json['_internal']['is_last_version'] = \
(last_version_pid.pid_value == pid)
# insert the bucket id for link generation in search results
record_buckets = RecordsBuckets.query.filter(
RecordsBuckets.record_id == record.id).all()
if record_buckets:
json['_internal']['files_bucket_id'] = \
str(record_buckets[0].bucket_id)
except Exception:
raise
def filter_internal(self, data):
"""Remove internal fields from the record metadata."""
if '_deposit' in data['metadata']:
data['metadata']['owners'] = data['metadata']['_deposit']['owners']
del data['metadata']['_deposit']
if '_files' in data['metadata']:
if allow_public_file_metadata(data['metadata']):
data['files'] = data['metadata']['_files']
del data['metadata']['_files']
if '_pid' in data['metadata']:
epic_pids = [
p for p in data['metadata']['_pid']
if p.get('type') == 'ePIC_PID'
]
dois = [
p for p in data['metadata']['_pid'] if p.get('type') == 'DOI'
]
if len(epic_pids) > 0:
data['metadata']['ePIC_PID'] = epic_pids[0].get('value')
if len(dois) > 0:
data['metadata']['DOI'] = DOI_URL_PREFIX + dois[0].get('value')
del data['metadata']['_pid']
if '_oai' in data['metadata']:
del data['metadata']['_oai']
if '_internal' in data['metadata']:
del data['metadata']['_internal']
return data
def indexer_receiver(sender,
json=None,
record=None,
index=None,
**dummy_kwargs):
"""Connect to before_record_index signal to transform record for ES."""
if not index.startswith('records'):
return
try:
if '_files' in json:
if not allow_public_file_metadata(json):
for f in json['_files']:
del f['key']
del json['_deposit']
json['_created'] = pytz.utc.localize(record.created).isoformat()
json['_updated'] = pytz.utc.localize(record.updated).isoformat()
json['owners'] = record['_deposit']['owners']
# insert the bucket id for link generation in search results
record_buckets = RecordsBuckets.query.filter(
RecordsBuckets.record_id == record.id).all()
if record_buckets:
json['_internal'] = {
'files_bucket_id': str(record_buckets[0].bucket_id),
}
except Exception:
raise
def filter_internal(self, data):
"""Remove internal fields from the record metadata."""
if '_deposit' in data['metadata']:
data['metadata']['owners'] = data['metadata']['_deposit']['owners']
del data['metadata']['_deposit']
if '_files' in data['metadata']:
if allow_public_file_metadata(data['metadata']):
data['files'] = data['metadata']['_files']
del data['metadata']['_files']
if '_pid' in data['metadata']:
# move PIDs to metadata top level
epic_pids = [p for p in data['metadata']['_pid']
if p.get('type') == 'ePIC_PID']
dois = [p for p in data['metadata']['_pid']
if p.get('type') == 'DOI']
if len(epic_pids) > 0:
data['metadata']['ePIC_PID'] = epic_pids[0].get('value')
if len(dois) > 0:
data['metadata']['DOI'] = DOI_URL_PREFIX + dois[0].get('value')
# add parent version pid
# data['metadata']['parent_id'] = next(
# pid['value'] for pid in data['metadata']['_pid']
# if pid['type'] == RecordUUIDProvider.parent_pid_type
# )
del data['metadata']['_pid']
if '_oai' in data['metadata']:
del data['metadata']['_oai']
if '_internal' in data['metadata']:
del data['metadata']['_internal']
return data
def indexer_receiver(sender, json=None, record=None, index=None,
**dummy_kwargs):
"""Connect to before_record_index signal to transform record for ES."""
from b2share.modules.access.policies import allow_public_file_metadata
from b2share.modules.records.fetchers import b2share_parent_pid_fetcher, b2share_record_uuid_fetcher
if 'external_pids' in json['_deposit']:
# Keep the 'external_pids' if the record is a draft (deposit) or
# if the files are public.
if (not is_deposit(record.model) and allow_public_file_metadata(json)):
json['external_pids'] = json['_deposit']['external_pids']
del json['_deposit']['external_pids']
if not index.startswith('records'):
return
try:
if '_files' in json:
if not allow_public_file_metadata(json):
for f in json['_files']:
del f['key']
del json['_deposit']
json['_created'] = pytz.utc.localize(record.created).isoformat()
json['_updated'] = pytz.utc.localize(record.updated).isoformat()
json['owners'] = record['_deposit']['owners']
json['_internal'] = dict()
# add the 'is_last_version' flag
parent_pid = b2share_parent_pid_fetcher(None, record).pid_value
pid = b2share_record_uuid_fetcher(None, record).pid_value
last_version_pid = PIDNodeVersioning(
pid=RecordUUIDProvider.get(parent_pid).pid
).last_child
json['_internal']['is_last_version'] = \
(last_version_pid.pid_value == pid)
# insert the bucket id for link generation in search results
record_buckets = RecordsBuckets.query.filter(
RecordsBuckets.record_id == record.id).all()
if record_buckets:
json['_internal']['files_bucket_id'] = \
str(record_buckets[0].bucket_id)
except Exception:
raise
def filter_internal(self, data):
"""Remove internal fields from the record metadata."""
external_pids = []
bucket = None
record = None
# differentiating between search results and
# single record requests
if hasattr(g, 'record'):
record = g.record
if record.files:
bucket = record.files.bucket
if is_deposit(record.model):
external_pids = generate_external_pids(record)
# if it is a published record don't generate external pids
# as they are immutable and stored in _deposit
else:
external_pids = record.model.json[
'_deposit'].get('external_pids')
user_has_permission = \
allow_public_file_metadata(data['metadata']) if bucket \
is None else files_permission_factory(
bucket, 'bucket-read').can()
elif hasattr(g, 'record_hit'):
user_has_permission = allow_public_file_metadata(
g.record_hit['_source'])
if '_deposit' in data['metadata']:
data['metadata']['owners'] = data['metadata']['_deposit']['owners']
# Add the external_pids only if the
# user is allowed to read the bucket
if external_pids and bucket and user_has_permission:
data['metadata']['external_pids'] = external_pids
del data['metadata']['_deposit']
if '_files' in data['metadata']:
# Also add the files field only if the user is allowed
if user_has_permission:
data['files'] = data['metadata']['_files']
if external_pids and bucket:
external_dict = {x['key']: x['ePIC_PID']
for x in external_pids}
for _file in data['files']:
if _file['key'] in external_dict:
_file['b2safe'] = True
_file['ePIC_PID'] = external_dict[_file['key']]
del data['metadata']['_files']
if '_pid' in data['metadata']:
# move PIDs to metadata top level
epic_pids = [p for p in data['metadata']['_pid']
if p.get('type') == 'ePIC_PID']
dois = [p for p in data['metadata']['_pid']
if p.get('type') == 'DOI']
if len(epic_pids) > 0:
data['metadata']['ePIC_PID'] = epic_pids[0].get('value')
if len(dois) > 0:
data['metadata']['DOI'] = DOI_URL_PREFIX + dois[0].get('value')
# add parent version pid
# data['metadata']['parent_id'] = next(
# pid['value'] for pid in data['metadata']['_pid']
# if pid['type'] == RecordUUIDProvider.parent_pid_type
# )
del data['metadata']['_pid']
if '_oai' in data['metadata']:
del data['metadata']['_oai']
if '_internal' in data['metadata']:
del data['metadata']['_internal']
return data
def filter_internal(self, data):
"""Remove internal fields from the record metadata."""
external_pids = []
bucket = None
record = None
# differentiating between search results and
# single record requests
if hasattr(g, 'record'):
record = g.record
if record.files:
bucket = record.files.bucket
if is_deposit(record.model):
external_pids = generate_external_pids(record)
# if it is a published record don't generate external pids
# as they are immutable and stored in _deposit
else:
external_pids = record.model.json[
'_deposit'].get('external_pids')
user_has_permission = \
allow_public_file_metadata(data['metadata']) if bucket \
is None else files_permission_factory(
bucket, 'bucket-read').can()
elif hasattr(g, 'record_hit'):
user_has_permission = allow_public_file_metadata(
g.record_hit['_source'])
if '_deposit' in data['metadata']:
if hasattr(g, 'record') and is_deposit(record.model) and current_app.config.get('AUTOMATICALLY_ASSIGN_DOI', False):
# add future DOI string
data['metadata'].update({'$future_doi': generate_doi(data['metadata']['_deposit']['id']) })
data['metadata']['owners'] = data['metadata']['_deposit']['owners']
# Add the external_pids only if the
# user is allowed to read the bucket
if external_pids and bucket and user_has_permission:
data['metadata']['external_pids'] = external_pids
del data['metadata']['_deposit']
if '_files' in data['metadata']:
# Also add the files field only if the user is allowed
if user_has_permission:
data['files'] = data['metadata']['_files']
if external_pids and bucket:
external_dict = {x['key']: x['ePIC_PID']
for x in external_pids}
for _file in data['files']:
if _file['key'] in external_dict:
_file['b2safe'] = True
_file['ePIC_PID'] = external_dict[_file['key']]
del data['metadata']['_files']
if '_pid' in data['metadata']:
# move PIDs to metadata top level
epic_pids = [p for p in data['metadata']['_pid']
if p.get('type') == 'ePIC_PID']
dois = [p for p in data['metadata']['_pid']
if p.get('type') == 'DOI']
if len(epic_pids) > 0:
data['metadata']['ePIC_PID'] = epic_pids[0].get('value')
if len(dois) > 0:
data['metadata']['DOI'] = DOI_URL_PREFIX + dois[0].get('value')
# add parent version pid
# data['metadata']['parent_id'] = next(
# pid['value'] for pid in data['metadata']['_pid']
# if pid['type'] == RecordUUIDProvider.parent_pid_type
# )
del data['metadata']['_pid']
if '_oai' in data['metadata']:
del data['metadata']['_oai']
if '_internal' in data['metadata']:
del data['metadata']['_internal']
return data