coderfender commented on PR #1996:
URL:
https://github.com/apache/datafusion-comet/pull/1996#issuecomment-3122030778
@andygrove , @parthchandra
The casting operation to DecimalType ( in case the operands are arent
already decimal types) is failing the test `"test integral divide"` . I think
we should update the test to reflect the fact that there will surely be a spark
operator (cast / project) in case the operands arent already decimal .
Also, I was wondering if we should stick to max precision and scale while
casting operands to decimal or should we perhaps cap it at precison of
Long.MAX_VALUE (10 digits) ?
Finally, below is the rust code which helped me understand the inconsistency
/ proof of concept of why we couldnt cast the operands to DoubleType (like
native Spark operation does)
`
fn main() {
let a:f64 = i64::MIN as f64;
assert!(((a / -1.0) as i64) == i64::MAX);
}`
https://play.rust-lang.org/?version=stable&mode=debug&edition=2024&gist=68515908b492b787be6973861504924c
--
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.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
