It was a comment which described older behavior. After deprecating old decimal 
types, there was no need in determining the range, so it was replaced with 
`DRILL_REL_DATATYPE_SYSTEM.getMaxNumericPrecision()`. But we cannot return 
precision greater than maxNumericPrecision. Or we can try to use formula from 
SQL Server implementation:
```
min(p1-s1, p2 -s2) + max( s1,s2 )
```
But again, add `Math.min(DRILL_REL_DATATYPE_SYSTEM.getMaxNumericPrecision(), 
outputPrecision)` to be sure that the value is not greater than 
maxNumericPrecision.

[ 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