lint Iterator::folds that can be one of the special methods

[oli-obk]

I just found .fold(false, |acc, x| acc || x.starts_with("--target")); in rustc. That should obviously be .any(|x| x.starts_with("--target"))

[clarfonthey]

Also .fold(1, |acc, x| acc * x) and .fold(0, |acc, x| acc + x) should be replaced with .sum() and .product() respectively.

[clarfonthey]

Bonus points: .rfold(1, |acc, x| acc * x) and .rfold(0, |acc, x| acc * x) could be .rev().product() and .rev().sum().

[theotherphil]

I'd like to give this a go. Any pointers to get started?

[oli-obk]

@theotherphil

There are already many iterator method lints in

https://github.com/rust-lang-nursery/rust-clippy/blob/master/clippy_lints/src/methods.rs#L718

You should be able to add a new branch for fold. You'll need to extract some info from the closure (if there is one). Have a look at

https://github.com/rust-lang-nursery/rust-clippy/blob/f9682ca563ad9ec6bc2da520ad92970d30d5c1e0/clippy_lints/src/map_clone.rs#L34

which does this for the |x| *x closure

[theotherphil]

Great, thanks!

[ghost]

@oli-obk
@theotherphil
.any and .all are not always safe replacements of .fold: those two are short-circuiting, while .fold evaluates every element in the iterator. If the iterator incurs side effects the two methods will have different behaviour.
Imho clippy should at least warn the user for the possible difference instead of suggesting .all and .any indifferently.

[camsteffen]

Fixed by #2350. We also have #2401 for more fold cases.