You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pinot.apache.org by "gortiz (via GitHub)" <gi...@apache.org> on 2023/05/03 14:25:30 UTC

[GitHub] [pinot] gortiz commented on issue #10713: Adding sorted / columns with inv index to noDictionaryColumns followed by a reload causes query failures

gortiz commented on issue #10713:
URL: https://github.com/apache/pinot/issues/10713#issuecomment-1533123219

   Here we have a case where the user explicitly say they don't want to have a dictionary but some other index requires the dictionary. Historically, in this situation we decided to ignore user request and create the dictionary anyway, but the new isn't consistent: It doesn't fail at validation time and it fails at runtime as the code follows a path we don't expect.
   
   It is clear that we need to either go back to the previous behavior or break compatibility. My suggestion is to introduce these special cases in `DictionaryIndex.getConfig` (or more specifically in `createDeserializer`). Other alternatives are focused on changing the behavior at reading time, but that would break the key idea introduced in `index-spi` where getConfig returns the actual configuration the index should use and therefore the invariant that FieldIndexConfigs can be seen as the source of trust. 


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For additional commands, e-mail: commits-help@pinot.apache.org