Hyukjin Kwon created SPARK-61393:
------------------------------------

             Summary: Accept the values that PySpark accepts in the list 
arguments of the Column methods, and raise RecursionError for a deep list, in 
the Python client
                 Key: SPARK-61393
                 URL: https://issues.apache.org/jira/browse/SPARK-61393
             Project: Spark
          Issue Type: Bug
          Components: Connect
    Affects Versions: connect-rust-4.3.0
            Reporter: Hyukjin Kwon


A follow-up of SPARK-61290, found in a post-merge review. Column.when, 
otherwise, getItem and isin convert a list with literal_as, which accepts only 
an exact literal type: a float list for a FloatType or DoubleType element, an 
np.float32 for DoubleType, or a datetime for DateType raises a bare 
AssertionError, where PySpark's LiteralExpression accepts them. The conversion 
also has no depth bound and grows super-linearly (52 s for a 2000-level nested 
list), where PySpark raises RecursionError.



--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to