Added documentation on the remainder (Rem) operator for floating points. #59529
Conversation
|
r? @Kimundi (rust_highfive has picked a reviewer for you, use r? to override) |
rust-highfive
added
the
S-waiting-on-review
|
I'm not sure this is enough to clarify the operation. LLVM
We could demonstrate that in Rust code like: |
|
@cuviper I think you're right. I will adjust it tomorrow and write a doc test as an example? Or do you think that would be overkill? |
|
As long as it's kept simple, an example sounds good to me. |
…int values is computed internally.
|
@cuviper What do you think about this? |
| /// ``` | ||
| /// let x: f32 = 4.0; | ||
| /// let y: f32 = 2.5; | ||
| /// let remainder = x - (x / y).trunc() * y; |
There was a problem hiding this comment.
I'd comment the expected value (1.5) for demonstration purposes. Might be nicer to choose inputs with a greater quotient too, so this doesn't look like just subtraction.
There was a problem hiding this comment.
Alright! So what about 15.2 and 2.3? (Just some random numbers). Then I would also add a comment that the result is: 1.4
There was a problem hiding this comment.
Random is fine, but maybe stick to base-2 numbers that float can represent exactly.
There was a problem hiding this comment.
@cuviper I updated the example! What about this?
jonas-schievink
added
the
A-docs
…floats and added a comment with the solution.
| /// ``` | ||
| /// let x: f32 = 4.0; | ||
| /// let y: f32 = 2.5; | ||
| /// let remainder = x - (x / y).trunc() * y; |
|
r? @cuviper |
|
@bors r+ |
|
📌 Commit a1c7905 has been approved by |
bors
added
S-waiting-on-bors
Added documentation on the remainder (Rem) operator for floating points. # Description As has been explained in rust-lang#57738 the remainder operator on floating points is not clear. This PR requests adds some information on how the `Rem` / remainder operator on floating points works. Note also that this description is for both `Rem<f32> for f32` and `Rem<f64> for f64` implementations. Ps. I wasn't really sure on how to formulate things. So please suggest changes if you have better idea's! closes rust-lang#57738
Rollup of 4 pull requests Successful merges: - #59166 (resolve: collect trait aliases along with traits) - #59341 (Fix custom relative libdir) - #59446 (Fix stack overflow when generating debuginfo for 'recursive' type) - #59529 (Added documentation on the remainder (Rem) operator for floating points.) Failed merges: r? @ghost
Description
As has been explained in #57738 the remainder operator on floating points is not clear.
This PR requests adds some information on how the
Rem/ remainder operator on floating points works.Note also that this description is for both
Rem<f32> for f32andRem<f64> for f64implementations.Ps. I wasn't really sure on how to formulate things. So please suggest changes if you have better idea's!
closes #57738