def parse_args()

in python/dataproc_templates/gcs/gcs_to_gcs.py [0:0]
91 lines of code
3 McCabe index (conditional complexity)

    def parse_args(args: Optional[Sequence[str]] = None) -> Dict[str, Any]:

        parser: argparse.ArgumentParser = argparse.ArgumentParser()

        parser.add_argument(
            f'--{constants.GCS_TO_GCS_INPUT_LOCATION}',
            dest=constants.GCS_TO_GCS_INPUT_LOCATION,
            required=True,
            help='Cloud Storage location of the input files'
        )
        parser.add_argument(
            f'--{constants.GCS_TO_GCS_INPUT_FORMAT}',
            dest=constants.GCS_TO_GCS_INPUT_FORMAT,
            required=True,
            help='Cloud Storage input file format (one of: avro,parquet,csv,json,delta)',
            choices=[
                constants.FORMAT_AVRO,
                constants.FORMAT_PRQT,
                constants.FORMAT_CSV,
                constants.FORMAT_JSON,
                constants.FORMAT_DELTA
            ]
        )
        add_spark_options(parser, constants.get_csv_input_spark_options("gcs.gcs.input."))
        add_spark_options(parser, constants.get_csv_output_spark_options("gcs.gcs.output."))
        parser.add_argument(
            f'--{constants.GCS_TO_GCS_TEMP_VIEW_NAME}',
            dest=constants.GCS_TO_GCS_TEMP_VIEW_NAME,
            required=False,
            default="",
            help='Temp view name for creating a spark sql view on source data. This name has to match with the table name that will be used in the SQL query'
        )
        parser.add_argument(
            f'--{constants.GCS_TO_GCS_SQL_QUERY}',
            dest=constants.GCS_TO_GCS_SQL_QUERY,
            required=False,
            default="",
            help='SQL query for data transformation. This must use the temp view name as the table to query from.'
        )

        parser.add_argument(
            f'--{constants.GCS_TO_GCS_OUTPUT_PARTITION_COLUMN}',
            dest=constants.GCS_TO_GCS_OUTPUT_PARTITION_COLUMN,
            required=False,
            default="",
            help='Partition column name to partition the final output in destination bucket'
        )
        parser.add_argument(
            f'--{constants.GCS_TO_GCS_OUTPUT_FORMAT}',
            dest=constants.GCS_TO_GCS_OUTPUT_FORMAT,
            required=False,
            default=constants.FORMAT_PRQT,
            help=(
                'Output write format '
                '(one of: avro,parquet,csv,json)'
                '(Defaults to parquet)'
            ),
            choices=[
                constants.FORMAT_AVRO,
                constants.FORMAT_PRQT,
                constants.FORMAT_CSV,
                constants.FORMAT_JSON
            ]
        )

        parser.add_argument(
            f'--{constants.GCS_TO_GCS_OUTPUT_MODE}',
            dest=constants.GCS_TO_GCS_OUTPUT_MODE,
            required=False,
            default=constants.OUTPUT_MODE_APPEND,
            help=(
                'Output write mode '
                '(one of: append,overwrite,ignore,errorifexists) '
                '(Defaults to append)'
            ),
            choices=[
                constants.OUTPUT_MODE_OVERWRITE,
                constants.OUTPUT_MODE_APPEND,
                constants.OUTPUT_MODE_IGNORE,
                constants.OUTPUT_MODE_ERRORIFEXISTS
            ]
        )

        parser.add_argument(
            f'--{constants.GCS_TO_GCS_OUTPUT_LOCATION}',
            dest=constants.GCS_TO_GCS_OUTPUT_LOCATION,
            required=True,
            help=(
                'Destination Cloud Storage location'
            )
        )

        known_args: argparse.Namespace
        known_args, _ = parser.parse_known_args(args)

        if getattr(known_args, constants.GCS_TO_GCS_SQL_QUERY) and not getattr(known_args, constants.GCS_TO_GCS_TEMP_VIEW_NAME):
            sys.exit('ArgumentParser Error: Temp view name cannot be null if you want to do data transformations with query')

        return vars(known_args)