Hyukjin Kwon created SPARK-61071:
------------------------------------
Summary: Type pandas columns of Decimals and other values as
PySpark does in createDataFrame of the Rust client
Key: SPARK-61071
URL: https://issues.apache.org/jira/browse/SPARK-61071
Project: Spark
Issue Type: Bug
Components: Connect
Affects Versions: 5.0.0
Reporter: Hyukjin Kwon
createDataFrame of a pandas object column that mixes Decimal values with other
values, such as pd.DataFrame({"d": [Decimal("1.5"), 2]}), raises
AttributeError: 'int' object has no attribute 'as_tuple' in the Rust client.
PySpark 4.3.0 leaves such a column to Arrow, which types it decimal(2,1) and
takes the int as 2.0, and raises PySparkTypeError or PySparkValueError for
floats, strings, bools and values that do not fit.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
