cleanup for group by in semanticanalyzer
----------------------------------------
Key: HIVE-651
URL: https://issues.apache.org/jira/browse/HIVE-651
Project: Hadoop Hive
Issue Type: Bug
Components: Query Processor
Reporter: Namit Jain
The code in semantic analyzer is getting pretty complex. There are many
functions which do similar stuff and it becoming painful to debug these
scenarios.
It would be useful to break up that code into smaller modules, which will make
it more readable.
Some common pieces which can be factored out are as follows:
1. Resolution of the aggregation function - this can happen before any group
by, and is independent of whether distinct or non-distinct aggregation is being
performed.
2. After the resolution, a mode can be passed which indicates whether partial
aggregation is required or not.
Some more detailed inspection of the code is needed, but the amount of common
code can be refactored and reduced.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
