You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@drill.apache.org by GitBox <gi...@apache.org> on 2018/08/26 01:55:45 UTC

[GitHub] sachouche opened a new pull request #1445: DRILL-6706: fixed null pointer exception in HashJoin

sachouche opened a new pull request #1445: DRILL-6706: fixed null pointer exception in HashJoin
URL: https://github.com/apache/drill/pull/1445
 
 
   **Findings after Code Inspection -**
   Looking at the code, having back-ticks within SchemaPath is not necessary (this is my understanding)
   - Back-ticks are useful in the context of a compound name such as T.`column.with.a.dot`.another_column
   - As soon as individual parts are parsed then the back-ticks can be omitted
   
   Then why Aggregator was able to handle columns having back-ticks (and whatever I could throw at it )
   - I found logic to strip the extra back-ticks within the aggregator code
   
   **Suggested Fix -**
   - Stripping the back-tick from the MaterializedField names within Parquet seems to fix the HashJoin issue
   - The Aggregator didn't seem bothered by this change either
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services