You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by rtudoran <gi...@git.apache.org> on 2017/08/01 07:09:37 UTC

[GitHub] flink issue #4380: Time sort with offset/fetch without retraction

Github user rtudoran commented on the issue:

    https://github.com/apache/flink/pull/4380
  
    @fhueske
    
    Thanks for the remarks/clarification. 
    
    I am fine with taking an approach to tackle both scenarios (also the simple addons you suggest but also having the actual useful code).
    Hence i would propose that the code developed with retraction support for ORDER BY rowtime ASC FETCH/OFFSET  we use it for  "ORDER BY rowtime DESC offset/FETCH.  Do you agree with this?
    
    I can add then in another PR/update also the code you intend to use for the ASC case with the simple modifications.
    
    Putting this aside there is another issue with LIMIT x. The intended behavior of LIMIT is to limit the number of results considered at a point. The way it is translated by Calcite is similar with FETCH. Basically at the moment of conversion you do not see any difference between 
    
    ORDER BY rowtime ASC FETCH 10 ROWS ONLY and ORDER BY rowtime LIMIT 10.
    How should we handle this situation?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---