rdblue commented on issue #25453: [SPARK-28730][SQL] Configurable type coercion 
policy for table insertion
URL: https://github.com/apache/spark/pull/25453#issuecomment-523540307
 
 
   > For Spark users, I think it makes more sense to choose the table insertion 
policy with one flag.
   
   I don't think that v2 should allow the legacy mode. ANSI mode and strict 
modes make sense, but carrying the legacy mode forward just to avoid an extra 
config property doesn't seem worth it. This mode is user-facing, while the 
configuration properties are primarily administrator settings. Reducing the 
administrator settings by one doesn't seem worth corrupting user data in the v2 
code path.
   
   And, the v1 code path will eventually be removed along with a v1 setting for 
this mode. If v2 doesn't support legacy mode, then we can remove it as well.
   
   > I don't think it's possible to make strict mode the default.
   
   I think it is fine to make ANSI mode the default -- assuming that's what the 
community votes for. But I don't think that carrying legacy mode forward is not 
the right way to avoid making strict mode the default.
   
   If legacy mode is available then we will always need to support it even when 
v1 is gone. If it is the default, then we can't actually make progress in this 
area by combining this with breaking changes for v2.

----------------------------------------------------------------
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:
us...@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

Reply via email to