rmnskb commented on code in PR #47253:
URL: https://github.com/apache/arrow/pull/47253#discussion_r2688163647
##########
python/pyarrow/parquet/core.py:
##########
@@ -90,6 +90,69 @@ def _check_filters(filters, check_null_strings=True):
return filters
+def _map_spark_to_arrow_types(datatype: pa.DataType) -> str | None:
+ lookup = {
+ "NA": "null",
+ "BOOL": "boolean",
+ "INT8": "byte",
+ "INT16": "short",
+ **dict.fromkeys(
+ ["UINT8", "UINT16", "UINT32", "UINT64", "INT32"], "integer"),
+ "INT64": "long",
+ **dict.fromkeys(["HALF_FLOAT", "FLOAT"], "float"),
+ "DOUBLE": "double",
+ "BINARY": "binary",
+ "STRING": "string",
+ **dict.fromkeys(
+ ["DECIMAL" + str(2 ** i) for i in range(5, 9)], "decimal"),
+ **dict.fromkeys(
+ ["LIST", "LARGE_LIST", "LIST_VIEW", "LARGE_LIST_VIEW",
"FIXED_SIZE_LIST"],
+ "array",
+ ),
+ "MAP": "map",
+ **dict.fromkeys(["DATE32", "DATE64"], "date"),
+ "TIMESTAMP": "timestamp",
+ "INTERVAL_MONTH_DAY_NANO": "Calendar Interval", # TODO: Correct this
Review Comment:
So if we're continue with this approach, do you propose handling a fraction
of data types?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
