>From [design 
>document](https://docs.google.com/document/d/1kfWUZ_OsEmLOJu_tKZO0-ZUJwc52drrm15fyif7BmzQ/edit#)
> there is a note which says Drill computes scale/precision for mod function 
>differently and formula is as below. Not sure what the reason was. Also the 
>comment above that line says precision should be atleast equal or greater than 
>input precision. So if max of input precision is > 38 then that will be chosen 
>as precision, if < 38 then a greater number 38 will be chosen. Do you still 
>think this is a bug?
```
For mod function resulting precision also differs:
p = Max(range(Max(s1,s2)+p1-s1),Max(p1,p2))
```

[ Full content available at: https://github.com/apache/drill/pull/1451 ]
This message was relayed via gitbox.apache.org for [email protected]

Reply via email to