[GitHub] [spark] srowen commented on pull request #31682: [SPARK-34545] Fix row comparisons

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

srowen commented on pull request #31682:
URL: https://github.com/apache/spark/pull/31682#issuecomment-787509798


   The error is:
   `Incorrect evaluation (codegen off): [true,-0.0] IN ([false,-0.0],[true,3.1807339850248914E35],null,null,[null,NaN],[true,-Infinity],[false,-1.7976931348623157E308],[true,0.0],[null,Infinity]), actual: true, expected: null`
   
   Looks like the issue is comparison of 0.0 to -0.0. Eh, I've forgotten if 0.0 should equal -0.0 in SQL? It looks like it's matching now, which is weird to me, because for example `new java.lang.Double(0.0).equals(new java.lang.Double(-0.0))` is false but `0.0 == -0.0` is true. We've moved to .equals() for comparison in this case (right?)
   
   Anyway it looks like it's expecting 'null' for this IN predicate, so maybe it's something slightly different. I don't quite see why. Maybe you know better than I do how to analyze this?


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



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