[GitHub] [spark] turboFei commented on issue #25701: [SPARK-29000][SQL] Decimal precision overflow when don't allow precision loss

[GitBox]

turboFei commented on issue #25701: [SPARK-29000][SQL] Decimal precision 
overflow when don't allow precision loss
URL: https://github.com/apache/spark/pull/25701#issuecomment-528739147
 
 
   > @turboFei thanks for this PR. I don't understand very well the difference 
before and after your patch. Indeed, it seems to be that the case when operand 
always becomes a `decimal(34, 24)` as well as the `1` or `1.0`. May you please 
explain what changes before and after?
   
   We can check the code  below:
   
https://github.com/apache/spark/blob/4664a082c2c7ac989e818958c465c72833d3ccfe/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/DecimalPrecision.scala#L123-L128
   
   If  a decimal(34,24) * 1.0, p1=34, s1=24, p2=2,s2=1,
   And the result type  is Decimal(38,25)
   <img width="1076" alt="image-20190905145151701" 
src="https://user-images.githubusercontent.com/6757692/64407902-0f412c00-d0b8-11e9-9682-ef0b5477d5df.png";>
   
   If a decimal(34,24) * 1,  p1 =34,s1=24,p2=34,s2=24.
   And the result type is Decimal(38,38).
   <img width="1167" alt="image-20190905145246902" 
src="https://user-images.githubusercontent.com/6757692/64407919-1c5e1b00-d0b8-11e9-8247-1e780401c10b.png";>
   
   The result will overflow.
   
   
   
   
   

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