Hyukjin Kwon created SPARK-21296:
------------------------------------
Summary: Avoid per-record type dispatch in PySpark createDataFrame
schema verification
Key: SPARK-21296
URL: https://issues.apache.org/jira/browse/SPARK-21296
Project: Spark
Issue Type: Improvement
Components: PySpark
Affects Versions: 2.2.0
Reporter: Hyukjin Kwon
Here, we do the type dispatch per record -
https://github.com/apache/spark/blob/d935e0a9d9bb3d3c74e9529e161648caa50696b7/python/pyspark/sql/types.py#L1252-L1366
and
https://github.com/apache/spark/blob/d935e0a9d9bb3d3c74e9529e161648caa50696b7/python/pyspark/sql/session.py#L517-L537
This requires per-record operation cost. We should get rid of this.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
