r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Is there actually any usage of this div_rem? A search locally only finds usage in coretests.

Maybe we can just delete it?

$ rg -w div_rem library
library/core/src/num/bignum.rs
341:            pub fn div_rem(&self, d: &$name, q: &mut $name, r: &mut $name) {

library/coretests/tests/num/bignum.rs
172:    fn div_rem(n: u64, d: u64) -> (Big, Big) {
175:        Big::from_u64(n).div_rem(&Big::from_u64(d), &mut q, &mut r);
178:    assert_eq!(div_rem(1, 1), (Big::from_small(1), Big::from_small(0)));
179:    assert_eq!(div_rem(4, 3), (Big::from_small(1), Big::from_small(1)));
180:    assert_eq!(div_rem(1, 7), (Big::from_small(0), Big::from_small(1)));
181:    assert_eq!(div_rem(45, 9), (Big::from_small(5), Big::from_small(0)));
182:    assert_eq!(div_rem(103, 9), (Big::from_small(11), Big::from_small(4)));
183:    assert_eq!(div_rem(123456, 77), (Big::from_u64(1603), Big::from_small(25)));
184:    assert_eq!(div_rem(0xffff, 1), (Big::from_u64(0xffff), Big::from_small(0)));
185:    assert_eq!(div_rem(0xeeee, 0xffff), (Big::from_small(0), Big::from_u64(0xeeee)));
186:    assert_eq!(div_rem(2_000_000, 2), (Big::from_u64(1_000_000), Big::from_u64(0)));

Would it be reasonable to check if other methods were ever used and delete them as well?

But, looking at this from the other side, what if someone were to write something in dec2flt or flt2dec and needed div_rem, but there was nothing?

I'm not particularly worried about someone needing it and needing to write it (or dig it up in git history). I definitely don't want to spend reviewer time reviewing optimizations for dead code though.

Fair enough, then first question about other unused methods is still open, and I'm fine to delete them as well if found any other

I don't have any strong opinion on whether to proactively audit for other unused code. It's probably relatively harmless to keep it and just not touch it.

Oh, well

It's probably relatively harmless to keep it and just not touch it.

Do you mean it's better not to touch this function or any others? If first so feel free to close this one I guess

No strong opinion either way, changes that are just deleting (dead code) lines I'm generally happy to rubber stamp. They're easy to add back later.

@bors r+ rollup

📌 Commit a2d66db has been approved by Mark-Simulacrum

It is now in the queue for this repository.