rdblue commented on issue #3700:
URL: https://github.com/apache/iceberg/issues/3700#issuecomment-991252584


   @jackye1995, a way to "fix" the problem is to move `spec_id` so that the 
original column ordinals match. The problem with doing that is that we leave a 
known bug in the codebase. And it's very unlikely that we'll get lucky and have 
someone catch it before release next time. There was no test that caught this.
   
   Another reason not to is that we state that Iceberg metadata files follow 
Iceberg schema evolution rules. If we violate that here and start avoiding 
certain valid changes because of fear something might break, then we're back to 
Hadoop usability: you have to know what not to do and you're artificially 
limited. We don't want Iceberg developers to be afraid of using tools that are 
perfectly valid according to the spec.


-- 
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: [email protected]

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



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

Reply via email to