def main(db, infile, mdbdb, mdbcollection, esindex,
user=None, password=None, host=None, port=None, recreateindex=False):
if db == "Elasticsearch":
dbc = DBconnection(db, esindex, host=host, port=port,
recreateindex=recreateindex)
read_and_index_faers_records(infile, dbc, es_index_reports)
dbc.es.indices.refresh(index=esindex)
elif db == "MongoDB":
dbc = DBconnection(db, mdbdb, mdbcollection=mdbcollection,
host=host, port=port, user=user, password=password,
recreateindex=recreateindex)
read_and_index_faers_records(infile, dbc.mdbi[mdbcollection],
mongodb_index_reports)
mongodb_indices(dbc.mdbi[mdbcollection])
if __name__ == '__main__':
args = argparse.ArgumentParser(
description='Index FDA FAERS dataset json files with Elasticsearch,'
' or MongoDB, downloaded from ' + SOURCEURL)
args.add_argument('--infile',
required=True,
help='drug-event .json or .json.zip files or'
'folder that includes the .json.zip files')
dbargs(args)
args = args.parse_args()
main(args.dbtype, args.infile, args.mdbdb, args.mdbcollection,
args.esindex,
args.user, args.password, args.host, args.port, args.recreateindex)
recreateindex=True):
indxr = Indexer(dbtype,
mdbdb,
mdbcollection,
esindex,
host,
port,
recreateindex=recreateindex)
indxr.parse_and_index_xmlfile(infile)
pool.close()
pool.join()
pool.terminate()
print("\nCompleted reading and indexing the ClinVar entries")
if dbtype == 'Elasticsearch':
indxr.es.indices.refresh(index=esindex)
else:
mongodb_indices(indxr.mcl)
if __name__ == '__main__':
args = argparse.ArgumentParser(
description='Index ClinVar Variation Archive xml files,'
' with Elasticsearch or MongoDB')
args.add_argument('infile',
help='Input file name of ClinVar Variation Archive,'
' compressed or uncompressed xml file')
dbargs(args, mdbcollection='clinvarvariation', esindex='clinvarvariation')
args = args.parse_args()
main(args.infile, args.dbtype, args.mdbdb, args.mdbcollection,
args.esindex, args.host, args.port)