[
https://issues.apache.org/jira/browse/FLINK-8118?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16262211#comment-16262211
]
Fabian Hueske commented on FLINK-8118:
--------------------------------------
Hi [~xccui],
1. I think offsets should be configured via the builder as an optional
parameters. IMO, we should not split the configuration into a builder part and
a non-builder part.
2. For testing, you can use Mockito to replace the {{FlinkConsumer}} which is
returned from {{getKafkaConsumer}} by a mock and check if the correct method to
configure the offsets is called. I did something similar to test the
configuration of the Orc readers in the
[OrcRowInputFormatTest|https://github.com/fhueske/flink/blob/d80506e3785268f541457a69ade3118c634cf7e7/flink-connectors/flink-orc/src/test/java/org/apache/flink/orc/OrcRowInputFormatTest.java].
Best, Fabian
> Allow to specify the offsets of KafkaTableSources
> -------------------------------------------------
>
> Key: FLINK-8118
> URL: https://issues.apache.org/jira/browse/FLINK-8118
> Project: Flink
> Issue Type: New Feature
> Components: Table API & SQL
> Affects Versions: 1.4.0
> Reporter: Timo Walther
> Assignee: Xingcan Cui
>
> Right now the Kafka TableSources can only read from the current group offset.
> We should expose the possibilities of the Kafka Consumer:
> https://ci.apache.org/projects/flink/flink-docs-release-1.4/dev/connectors/kafka.html#kafka-consumers-start-position-configuration
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
