Github user mgaido91 commented on the issue:
https://github.com/apache/spark/pull/20023
@gatorsmile I answered to your comments about DB2 in the e-mail.
@cloud-fan that would help, but not solve the problem. It would just make
the problem being generated by bigger numbers.
As you can see from the e-mail, DB2 behavior is actually in accordance to
SQL standards and the other DBs, it just have a smaller maximum precision. And
the case of throwing an exception is point 3 of my e-mails and it is out of
scope of this PR, because I think we best discuss before which is the right
approach in that case and then I can eventually create a PR.
Therefore, also DB2 behavior is aligned to the other SQL engines, the SQL
standard and Spark is the only one which is currently behaving differently.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
