def run_query()

in source/benchmark-sample/emr-benchmark.py [0:0]

• 15 lines of code
• 5 McCabe index (conditional complexity)

def run_query(uri, scale, fs_proto, cleanup=False, gendata=True, restore=False, engine='spark'):
    if cleanup:
        cleanup_benchmark_data(uri, scale, fs_proto)
    if gendata:
        if restore:
            print('Trying to restore database')
        gen_benchmark_data(uri, scale, fs_proto, restore)

    if engine == 'spark':
        printH('init thriftserver')
        sh(INIT_THRIFT_SERVER_BIN)
    else:
        printH('stop thriftserver')
        sh('sudo /usr/lib/spark/sbin/stop-thriftserver.sh')

    printH(f'run {engine} queries {scale} {fs_proto}')
    sh(f'{RUN_QUERY_BIN} -g {engine} -d tpcds_parquet_{scale}_{fs_proto},tpcds_orc_{scale}_{fs_proto} -s {uri}')