You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Dongjoon Hyun (JIRA)" <ji...@apache.org> on 2019/03/26 05:30:00 UTC
[jira] [Commented] (SPARK-27269) File source v2 should validate
data schema only
[ https://issues.apache.org/jira/browse/SPARK-27269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16801393#comment-16801393 ]
Dongjoon Hyun commented on SPARK-27269:
---------------------------------------
Hi, [~Gengliang.Wang].
- If this is a regression, this should have a type `BUG`.
{quote}This is actually a regression.
{quote}
- If you want to put this under your umbrella JIRA, please put it there. Then, it will have `subtask` as you want.
> File source v2 should validate data schema only
> -----------------------------------------------
>
> Key: SPARK-27269
> URL: https://issues.apache.org/jira/browse/SPARK-27269
> Project: Spark
> Issue Type: Task
> Components: SQL
> Affects Versions: 3.0.0
> Reporter: Gengliang Wang
> Priority: Major
>
> Currently, File source v2 allows each data source to specify the supported data types by implementing the method `supportsDataType` in `FileScan` and `FileWriteBuilder`.
> However, in the read path, the validation checks all the data types in `readSchema`, which might contain partition columns. This is actually a regression. E.g. Text data source only supports String data type, while the partition columns can still contain Integer type since partition columns are processed by Spark.
> This PR is to:
> 1. Refactor schema validation and check data schema only
> 2. Filter the partition columns in data schema if user specified schema provided.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org