HyukjinKwon edited a comment on issue #27725: [SPARK-30234][SQL][FOLLOWUP] 
Rename `spark.sql.legacy.addDirectory.recursive.enabled` to 
`spark.sql.legacy.addSingleFileInAddFile`
URL: https://github.com/apache/spark/pull/27725#issuecomment-593029589
 
 
   >  To me, legacy just sounds like the previous behavior is not good, 
especially there's some problem with it.
   
   It doesn't necessarily have to have a problem with it. We have a legacy 
configuration based on case-by-case, for example, when the previous behaviour 
has been widely used, newer behaviour supersedes the old behaviour, etc. Note 
that bugs have to be problem without legacy config as a general guideline!
   
   `spark.sql.legacy.addSingleFileInAddFile` supersedes old behaviour, and we 
will remove this configuration away in the future because disabling 
`spark.sql.legacy.addSingleFileInAddFile` will support **both a directory and 
single file**.
   
   Also, I don't actually completely buy this argument 
https://github.com/apache/spark/pull/26863#discussion_r378715586 since we 
already have different default values for legacy configurations, e.g., 
`spark.sql.legacy.replaceDatabricksSparkAvro.enabled` vs 
`spark.sql.legacy.setopsPrecedence.enabled`. I only buy the argument about the 
namespace. Doesn't quite matter if it's called directory or file.
   

----------------------------------------------------------------
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