You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@flink.apache.org by "jinshuangxian (Jira)" <ji...@apache.org> on 2022/05/27 10:08:00 UTC

[jira] [Created] (FLINK-27816) file source reader has Some requirements

jinshuangxian created FLINK-27816:
-------------------------------------

             Summary: file source reader has Some requirements
                 Key: FLINK-27816
                 URL: https://issues.apache.org/jira/browse/FLINK-27816
             Project: Flink
          Issue Type: Improvement
    Affects Versions: 1.15.0
            Reporter: jinshuangxian


I use the flink sql file-system connector to consume data, write it to object storage, and use the file-system consumption to process the data in the object storage. The whole process works fine. I have 2 new requirements:
1. can I specify a timestamp to consume files within a specified time range
2. It is hoped that the data written to the object storage can be ordered in the partition (for example, partitioned according to the device id), and the file source reader can consume the files in an orderly manner similar to kafka when consuming files.
Can some enhancements be made to the file source reader?



--
This message was sent by Atlassian Jira
(v8.20.7#820007)