def find_missing_ids(self):
id_fields = [
self.to_mongo().get(x, None)
for x in ['doi', 'pubmed_id', 'pmcid']
]
ids_not_none = [x is not None for x in id_fields]
#We need at least one of the id fields complete in order to find the others
if not all(ids_not_none) and any(ids_not_none):
present_id = next(x for x in id_fields if x is not None)
remaining_ids = find_remaining_ids(present_id)
for k, v in remaining_ids.items():
if v is not None:
self[k] = v
def _parse_pubmed_id(self, doc):
""" Returns the PubMed ID of a document as a <class 'str'>."""
return find_remaining_ids(self._parse_doi(doc))['pubmed_id']
def _parse_pmcid(self, doc):
""" Returns the pmcid of a document as a <class 'str'>."""
return find_remaining_ids(self._parse_doi(doc))['pmcid']
def _parse_doi(self, doc):
""" Returns the DOI of a document as a <class 'str'>"""
doi_fetch = find_remaining_ids(str(doc['pmid'])).get('doi', None)
if doi_fetch != None:
return doi_fetch
return None
def _parse_pmcid(self, doc):
""" Returns the pmcid of a document as a <class 'str'>."""
try:
return doc['pmcid']
except:
return find_remaining_ids(str(doc['pmid']))['pmcid']
def _parse_pubmed_id(self, doc):
""" Returns the PubMed ID of a document as a <class 'str'>."""
if 'pmid' in doc.keys():
if doc['pmcid'] != '':
return doc['pmcid']
return find_remaining_ids(self._parse_doi(doc))['pubmed_id']
def update_PubMed_entries(mongo_db):
col_name = 'CDCN_extracted_PubMed'
col = mongo_db[col_name]
col_entries = mongo_db['entries']
col_entries_vespa = mongo_db['entries_vespa']
query = col.find({}, )
print('query.count()', query.count())
found_entries = set()
found_entries_vespa = set()
for doc in query:
PMID = doc['PMID']
ids = find_remaining_ids(PMID)
doi = ids['doi']
drugs = doc.get('Repurposed Drug Name', None)
if not drugs:
continue
# TODO: remove drug name NR
if isinstance(drugs, str):
drugs = [drugs]
if doi is not None:
print('doi', doi)
col.find_one_and_update({'_id': doc['_id']},
{'$set': {
'doi': doi
}})
# # update entries
# if doi is not None:
# entry = col_entries.find_one({'doi': doi})
# if entry:
# col_entries.find_one_and_update(
# {
# '_id': entry['_id']
# },
# {
# "$set": {
# 'drug_names': drugs
# },
# }
# )
# found_entries.add(entry['_id'])
#
# # update vespa entries
# if PMID is not None:
# entry = col_entries_vespa.find_one({'pubmed_id': PMID})
# if entry:
# col_entries_vespa.find_one_and_update(
# {
# '_id': entry['_id']
# },
# {
# "$set": {
# 'drug_names': drugs
# },
# }
# )
# found_entries_vespa.add(entry['_id'])
#
# if doi is not None:
# entry = col_entries_vespa.find_one({'doi': doi})
# if entry:
# col_entries_vespa.find_one_and_update(
# {
# '_id': entry['_id']
# },
# {
# "$set": {
# 'drug_names': drugs
# },
# }
# )
# found_entries_vespa.add(entry['_id'])
print('found_entries', len(found_entries))
print('found_entries_vespa', len(found_entries_vespa))