def drop_single_dataset_metadata(self, accession):
"""Drop all metadata entries associated with `accession` from `self.mongo_collections.metadata`"""
collection = self.mongo_collections.metadata
with collection.database.client.start_session() as session:
with session.start_transaction():
run_mongo_action(
"delete_many",
collection,
query={"id.accession": accession},
)
return "dropped", "removed from database", None
def drop_status(collection,
prefix="Status update",
accession=None,
status=None,
info=None,
warning=None,
error=None,
**kwargs):
"""Drop all references to accession from `collection`"""
query = {"accession": accession}
run_mongo_action(action="delete_many", collection=collection, query=query)
log_status(prefix, status, info, warning, error, query)
def update_metadata_value_lookup(mongo_collections, cacher_id, keys=("investigation", "study", "assay")):
"""Collect existing keys and values for lookups"""
m = "{}:\n reindexing metadata lookup records ('{}')"
GeneFabLogger.info(m.format(cacher_id, mongo_collections.metadata_aux.name))
index = deepcopy_keys(METADATA_AUX_TEMPLATE, *keys)
INPLACE_update_metadata_value_lookup_keys(index, mongo_collections)
INPLACE_update_metadata_value_lookup_values(index, mongo_collections)
collection = mongo_collections.metadata_aux
with collection.database.client.start_session() as session:
with session.start_transaction():
for isa_category in index:
for subkey in index[isa_category]:
run_mongo_action(
action="replace", collection=collection,
query={"isa_category": isa_category, "subkey": subkey},
data={"content": index[isa_category][subkey]},
)
m = "{}:\n finished reindexing metadata lookup records ('{}')"
GeneFabLogger.info(m.format(cacher_id, mongo_collections.metadata_aux.name))
def update_status(collection,
prefix="Status update",
report_type=None,
accession=None,
assay_name=None,
sample_name=None,
status=None,
info=None,
warning=None,
error=None,
**kwargs):
"""Update status of dataset (and, optionally, assay/sample) in `collection`, log with logger"""
query = {
"status":
status,
"report type":
report_type
or ("dataset status" if sample_name is None else "parser message"),
"accession":
accession,
"assay name":
assay_name,
"sample name":
sample_name,
"info":
info,
"warning":
warning,
"error":
None if (error is None) else type(error).__name__,
"args":
getattr(error, "args", []),
"kwargs":
kwargs,
}
run_mongo_action(
action="replace",
collection=collection,
query=query,
data={"report timestamp": int(datetime.now().timestamp())},
)
log_status(prefix, status, info, warning, error, query)
def __init__(self, identifier, collection, value):
"""Match existing documents by base64-encoded `value`, update if changed, report state in self.changed"""
if not isinstance(identifier, dict):
msg = "ValueCheckedRecord(): `identifier` is not a dictionary"
raise GeneFabConfigurationException(msg, identifier=identifier)
elif "base64value" in identifier:
msg = "ValueCheckedRecord(): `identifier` uses a reserved key"
_kw = dict(identifier=identifier, key="base64value")
raise GeneFabConfigurationException(msg, **_kw)
else:
self.identifier, self.value = identifier, value
try:
dumped = dumps(value, sort_keys=True, default=funcdump)
self.base64value = compress(encodebytes(dumped.encode()))
except TypeError as e:
msg, _erep = "ValueCheckedRecord(): TypeError", repr(e)
_kw = dict(identifier=identifier,
value=value,
debug_info=_erep)
raise GeneFabConfigurationException(msg, **_kw)
else:
self.changed, n_stale_entries = True, 0
for entry in collection.find(identifier):
if entry["base64value"] == self.base64value:
self.changed = False
else:
n_stale_entries += 1
if (n_stale_entries != 0) or self.changed:
msg = f"ValueCheckedRecord updated:\n {identifier}"
GeneFabLogger.info(msg)
with collection.database.client.start_session() as session:
with session.start_transaction():
run_mongo_action(
"replace",
collection,
query=identifier,
data={"base64value": self.base64value},
)