[
https://issues.apache.org/jira/browse/HUDI-5807?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17763201#comment-17763201
]
Ethan Guo commented on HUDI-5807:
---------------------------------
[~linliu] You tested the default behavior of merging on master where the
`HoodieAvroRecordMerger` and record payload logic is used. You need to set
`hoodie.datasource.write.record.merger.impls` to
`org.apache.hudi.HoodieSparkRecordMerger` so that the Spark record type and
Spark record merger are used, which uses `HoodieSparkParquetReader`. Try
testing that again with "hoodie.datasource.write.drop.partition.columns" as
"true".
> HoodieSparkParquetReader is not appending partition-path values
> ---------------------------------------------------------------
>
> Key: HUDI-5807
> URL: https://issues.apache.org/jira/browse/HUDI-5807
> Project: Apache Hudi
> Issue Type: Bug
> Components: spark
> Affects Versions: 0.13.0
> Reporter: Alexey Kudinkin
> Assignee: Lin Liu
> Priority: Blocker
> Fix For: 1.0.0
>
>
> Current implementation of HoodieSparkParquetReader isn't supporting the case
> when "hoodie.datasource.write.drop.partition.columns" is set to true.
> In that case partition-path values are expected to be parsed from
> partition-path and be injected w/in the File Reader (this is behavior of
> Spark's own readers)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
