(convention): Add Arrow prefix to parquet datasource for consistency (#1724)

Author: jaidisido

awswrangler/distributed/ray/datasources/__init__.py

@@ -1,21 +1,21 @@
 """Ray Datasources Module."""
 
 from awswrangler.distributed.ray.datasources.arrow_csv_datasource import ArrowCSVDatasource
+from awswrangler.distributed.ray.datasources.arrow_parquet_datasource import ArrowParquetDatasource
 from awswrangler.distributed.ray.datasources.pandas_file_based_datasource import UserProvidedKeyBlockWritePathProvider
 from awswrangler.distributed.ray.datasources.pandas_text_datasource import (
     PandasCSVDataSource,
     PandasFWFDataSource,
     PandasJSONDatasource,
     PandasTextDatasource,
 )
-from awswrangler.distributed.ray.datasources.parquet_datasource import ParquetDatasource
 
 __all__ = [
     "ArrowCSVDatasource",
+    "ArrowParquetDatasource",
     "PandasCSVDataSource",
     "PandasFWFDataSource",
     "PandasJSONDatasource",
-    "ParquetDatasource",
     "PandasTextDatasource",
     "UserProvidedKeyBlockWritePathProvider",
 ]

awswrangler/distributed/ray/datasources/parquet_datasource.py renamed to awswrangler/distributed/ray/datasources/arrow_parquet_datasource.py

@@ -1,4 +1,4 @@
-"""Ray ParquetDatasource Module."""
+"""Ray ArrowParquetDatasource Module."""
 
 import logging
 from typing import Any, Callable, Dict, Iterator, List, Optional, Union
@@ -86,7 +86,7 @@ def _read_pieces(
 ray.data.datasource.parquet_datasource._read_pieces = _read_pieces  # pylint: disable=protected-access
 
 
-class ParquetDatasource(PandasFileBasedDatasource):  # pylint: disable=abstract-method
+class ArrowParquetDatasource(PandasFileBasedDatasource):  # pylint: disable=abstract-method
     """Parquet datasource, for reading and writing Parquet files."""
 
     _FILE_EXTENSION = "parquet"

awswrangler/distributed/ray/modin/s3/_read_parquet.py

@@ -6,7 +6,7 @@
 import pyarrow as pa
 from ray.data import read_datasource
 
-from awswrangler.distributed.ray.datasources import ParquetDatasource
+from awswrangler.distributed.ray.datasources import ArrowParquetDatasource
 from awswrangler.distributed.ray.modin._utils import _to_modin
 
 
@@ -27,7 +27,7 @@ def _read_parquet_distributed(  # pylint: disable=unused-argument
     if coerce_int96_timestamp_unit:
         dataset_kwargs["coerce_int96_timestamp_unit"] = coerce_int96_timestamp_unit
     dataset = read_datasource(
-        datasource=ParquetDatasource(),  # type: ignore
+        datasource=ArrowParquetDatasource(),  # type: ignore
         parallelism=parallelism,
         use_threads=use_threads,
         paths=paths,

awswrangler/distributed/ray/modin/s3/_write_parquet.py

@@ -10,7 +10,7 @@
 from ray.data import from_modin, from_pandas
 from ray.data.datasource.file_based_datasource import DefaultBlockWritePathProvider
 
-from awswrangler.distributed.ray.datasources import ParquetDatasource, UserProvidedKeyBlockWritePathProvider
+from awswrangler.distributed.ray.datasources import ArrowParquetDatasource, UserProvidedKeyBlockWritePathProvider
 
 _logger: logging.Logger = logging.getLogger(__name__)
 
@@ -49,7 +49,7 @@ def _to_parquet_distributed(  # pylint: disable=unused-argument
     # Repartition by max_rows_by_file
     elif max_rows_by_file and (max_rows_by_file > 0):
         ds = ds.repartition(math.ceil(ds.count() / max_rows_by_file))
-    datasource = ParquetDatasource()
+    datasource = ArrowParquetDatasource()
     ds.write_datasource(
         datasource,  # type: ignore
         path=path or path_root,