rdblue commented on issue #28026: [SPARK-31257][SQL] Unify create table syntax
URL: https://github.com/apache/spark/pull/28026#issuecomment-611035487
 
 
   @gatorsmile, this isn't a new feature. This is fixing a long-standing 
problem that there are two different parser rules for create table that 
generate different plans. Now, there is one rule and additional metadata passed 
to conversions that already take place. Updating the existing paths for those 
properties is a reasonable part of the fix, instead of throwing exceptions.
   
   What you're asking for is to purposely disable passing only some of that 
metadata through one code path, but not the other. I've heard no rationale for 
that or discussion of risks introduced by passing it.
   
   What is the risk that justifies doing more work to remove passing a few 
metadata properties?
   
   If the only justification is that this is a "new feature" then should we 
also detect when Hive's `PARTITION BY` syntax is used for DataSource tables and 
throw an exception there? I think we all agree that would be silly. So what is 
the distinction between that and what you're asking?

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to