ueshin commented on code in PR #45699:
URL: https://github.com/apache/spark/pull/45699#discussion_r1552202465
##########
python/pyspark/sql/connect/session.py:
##########
@@ -418,6 +425,28 @@ def createDataFrame(
# If no schema supplied by user then get the names of columns only
if schema is None:
_cols = [str(x) if not isinstance(x, str) else x for x in
data.columns]
+ infer_pandas_dict_as_map = (
+
str(self.conf.get("spark.sql.execution.pandas.inferPandasDictAsMap")).lower()
+ == "true"
+ )
+ if infer_pandas_dict_as_map:
+ struct = StructType()
+ pa_schema = pa.Schema.from_pandas(data)
+ spark_type: Union[MapType, DataType]
+ for field in pa_schema:
+ field_type = field.type
+ if isinstance(field_type, pa.StructType):
+ if len(field_type) == 0:
+ raise PySparkValueError(
+ error_class="CANNOT_INFER_EMPTY_SCHEMA",
+ message_parameters={},
+ )
+ arrow_type = field_type.field(0).type
Review Comment:
The API `StructType.field` seems to be available since pyarrow `10.0`.
https://arrow.apache.org/docs/10.0/python/generated/pyarrow.StructType.html
We may want to bump up the minimum pyarrow version, or we should avoid this
API?
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
