You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Yuepeng Pan (Jira)" <ji...@apache.org> on 2026/05/19 13:36:00 UTC

[jira] [Updated] (FLINK-39710) Release Testing: Verify FLIP-565: Improve ProcessTableFunctions for late data handling and state access

     [ https://issues.apache.org/jira/browse/FLINK-39710?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yuepeng Pan updated FLINK-39710:
--------------------------------
    Description: 
The state access part of FLIP-565 has not been implemented yet, so the release testing focuses around late data handling.

Testing instructions:
 * Create some append-only source (e.g. Kafka) with some out of order data and a watermark strategy that produces late data
 * Create a ProcessTableFunction ([from example|https://nightlies.apache.org/flink/flink-docs-master/docs/dev/table/functions/ptfs/#ordering-guarantees-and-late-events]) and submit a Flink job with it
 * Case 1: Check whether that only ordered data enters the PTF and late data is dropped
 * Case 2: Check that late data is not dropped when the ORDER BY clause is omitted
 * Stop and restore both functions from a savepoint and check whether they are checkpoint correctly

> Release Testing: Verify FLIP-565: Improve ProcessTableFunctions for late data handling and state access
> -------------------------------------------------------------------------------------------------------
>
>                 Key: FLINK-39710
>                 URL: https://issues.apache.org/jira/browse/FLINK-39710
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Tests
>            Reporter: Yuepeng Pan
>            Priority: Blocker
>             Fix For: 2.3.0
>
>
> The state access part of FLIP-565 has not been implemented yet, so the release testing focuses around late data handling.
> Testing instructions:
>  * Create some append-only source (e.g. Kafka) with some out of order data and a watermark strategy that produces late data
>  * Create a ProcessTableFunction ([from example|https://nightlies.apache.org/flink/flink-docs-master/docs/dev/table/functions/ptfs/#ordering-guarantees-and-late-events]) and submit a Flink job with it
>  * Case 1: Check whether that only ordered data enters the PTF and late data is dropped
>  * Case 2: Check that late data is not dropped when the ORDER BY clause is omitted
>  * Stop and restore both functions from a savepoint and check whether they are checkpoint correctly



--
This message was sent by Atlassian Jira
(v8.20.10#820010)