alamb commented on code in PR #26164:
URL: https://github.com/apache/datafusion/pull/26164#discussion_r4234123283
##########
datafusion/expr/src/expr_schema.rs:
##########
@@ -435,6 +435,24 @@ impl ExprSchemable for Expr {
Expr::ScalarSubquery(subquery) =>
Ok(scalar_subquery_nullable(subquery)),
Expr::BinaryExpr(BinaryExpr { left, right, op }) => match op {
Operator::IsDistinctFrom | Operator::IsNotDistinctFrom =>
Ok(false),
+ Operator::And => {
+ if contains_is_not_null(left.as_ref(), right.as_ref()) {
Review Comment:
Maybe we can add some more comments explaining why we need to explicitly
check IS NOT NULL
ALso, the example you gave above is only valid when the argument (`x`) is on
both sides
`x IS NOT NULL AND x`
However it doesn't hold for:
`y IS NOT NULL and x` (different arguments).
This seems very similar to something @pepijnve hit a while ago with Case
nullability -- specifically that the after simplification the nullability can
sometimes be smarter than before the simplification.
I am still missing why we need to make the nullability check smarter (why
isn't leaving it as `nullable` ok, even if overly conservative)?
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
