Zamil Majdy created SPARK-45604:
-----------------------------------
Summary: Converting timestamp_ntz to array<timestamp_ntz> can
cause NPE or SEGFAULT on parquet vectorized reader
Key: SPARK-45604
URL: https://issues.apache.org/jira/browse/SPARK-45604
Project: Spark
Issue Type: Bug
Components: Spark Core
Affects Versions: 3.5.0
Reporter: Zamil Majdy
Repro:
{{{}```{}}}{{{}{}}}
spark.conf.set("spark.databricks.photon.enabled", "false")
{{}}
val path = "/tmp/somepath"
val df = sql("SELECT MAP('key', CAST('2019-01-01 00:00:00' AS TIMESTAMP_NTZ))
AS field")
{{}}
df.write.mode("overwrite").parquet(path)
spark.read.schema("field map<string,
array<timestamp_ntz>>").parquet(path).collect()
{{{}{}}}{{{}```{}}}
Depending on the memory mode is used, it will produced NPE on on-heap mode, and
segfault on off-heap
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
