[GitHub] [hudi] ivorzhou edited a comment on pull request #2091: HUDI-1283 Fill missing columns with default value when spark dataframe save to hudi table

[GitBox <gi...@apache.org>]

ivorzhou edited a comment on pull request #2091:
URL: https://github.com/apache/hudi/pull/2091#issuecomment-697058418


   > Thank you for creating this PR. At this point, I am not fully convinced if we really need this logic. A missing column in the DataFrame could also mean that column has been dropped, although Hudi schema evolution does not really support dropping of fields at this point of time. But if in future, if we are planning to support something like that then this would contradict with it.
   > 
   > While the logic LGTM, lets get a second opinion. @vinothchandar @bvaradar thoughts on this use-case ?
   
   Thank you for your reviewing, I think sometimes a missing column does not means dropped ,but ignored when update specified columns. Like update SQL, sometimes I only want to update one column , keep others unchanged. 


----------------------------------------------------------------
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:
users@infra.apache.org