[
https://issues.apache.org/jira/browse/SPARK-40468?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dongjoon Hyun updated SPARK-40468:
----------------------------------
Labels: correctness (was: )
> Column pruning is not handled correctly in CSV when _corrupt_record is used
> ---------------------------------------------------------------------------
>
> Key: SPARK-40468
> URL: https://issues.apache.org/jira/browse/SPARK-40468
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 3.3.0, 3.2.2, 3.4.0
> Reporter: Ivan Sadikov
> Assignee: Ivan Sadikov
> Priority: Major
> Labels: correctness
> Fix For: 3.4.0, 3.3.2
>
>
> I have found that depending on the name of the corrupt record in CSV, the
> field is populated incorrectly. Here is an example:
> {code:java}
> 1,a > /tmp/file.csv
> ===
> val df = spark.read
> .schema("c1 int, c2 string, x string, _corrupt_record string")
> .csv("file:/tmp/file.csv")
> .withColumn("x", lit("A"))
> Result:
> +---+---+---+---------------+
> |c1 |c2 |x |_corrupt_record|
> +---+---+---+---------------+
> |1 |a |A |1,a |
> +---+---+---+---------------+{code}
>
> However, if you rename the {{_corrupt_record}} column to something else, the
> result is different:
> {code:java}
> val df = spark.read
> .option("columnNameCorruptRecord", "corrupt_record")
> .schema("c1 int, c2 string, x string, corrupt_record string")
> .csv("file:/tmp/file.csv") .withColumn("x", lit("A"))
> Result:
> +---+---+---+--------------+
> |c1 |c2 |x |corrupt_record|
> +---+---+---+--------------+
> |1 |a |A |null |
> +---+---+---+--------------+{code}
>
> This is due to inconsistency in CSVFileFormat, when enabling columnPruning,
> we check SQLConf option for corrupt records but CSV reader relies on
> {{columnNameCorruptRecord}} option instead.
> Also, this disables column pruning which used to work in Spark version prior
> to
> https://github.com/apache/spark/commit/959694271e30879c944d7fd5de2740571012460a.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
