alamb commented on code in PR #5342:
URL: https://github.com/apache/arrow-datafusion/pull/5342#discussion_r1120839022
##########
datafusion/expr/src/type_coercion/binary.rs:
##########
@@ -238,13 +238,32 @@ fn comparison_binary_numeric_coercion(
(_, Decimal128(_, _)) => get_comparison_common_decimal_type(rhs_type,
lhs_type),
(Float64, _) | (_, Float64) => Some(Float64),
(_, Float32) | (Float32, _) => Some(Float32),
- (Int64, _) | (_, Int64) => Some(Int64),
- (Int32, _) | (_, Int32) => Some(Int32),
- (Int16, _) | (_, Int16) => Some(Int16),
- (Int8, _) | (_, Int8) => Some(Int8),
+ // start checking from Int64, that is the most inclusive integer type
Review Comment:
I think I am just confused
For example this PR adds this rule:
```
(Int8, UInt16) => Some(Int32),
```
Which isn't just merging the schema (e.g. UInt16) in my mind, but is
extending it in some way. In this case, so that the entire range of `Int8` and
`UInt16` can be represented -- which now makes sense
I think what would help me would be:
1. Add some comments explaining the rationale for these rules
2. (maybe) add test coverage (ideally looking like your beautiful table from
the PR description) that helps to illustrate the point
However, I don't think this is needed prior to merging this PR
--
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]
