[
https://issues.apache.org/jira/browse/SPARK-46633?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon reassigned SPARK-46633:
------------------------------------
Assignee: Ivan Sadikov
> Reading a non-empty Avro file with empty blocks returns 0 records
> -----------------------------------------------------------------
>
> Key: SPARK-46633
> URL: https://issues.apache.org/jira/browse/SPARK-46633
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 3.5.0, 4.0.0
> Reporter: Ivan Sadikov
> Assignee: Ivan Sadikov
> Priority: Major
> Labels: pull-request-available
>
> When an Avro file contains empty blocks, Spark returns 0 records while
> "fastavro" and "avro-python-3" both read the file correctly and return
> records.
>
> This is due to the way Spark handles empty blocks (or does not handle). Call
> to `hasNext` loads the next block and if that block is empty, it returns
> false. But instead of exiting the loop, we need to probe the next block until
> sync point is reached.
--
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
