Results 475 comments of Jacob Lifshay

I think also having a `CompareEq` trait would be useful where you don't have ordering, but just `eq`/`ne`, like `PartialEq`/`Eq` but with state.

> [@programmerjake](https://github.com/programmerjake) Yeah. Then if we follow the Hash/Hasher/BuildeHasher analogy, we may have: > > * Compare/BuildCompare > * PartialCompare/BuildPartialCompare > * CompareEq/BuildCompareEq > * ~PartialEq~ (UPDATE: I noticed that...

> > they have significant semantic differences, so I think we need both ComparePartialEq and CompareEq. > > Well. Then maybe for naming: > > * Compare > * ComparePartial...

since hashes don't easily combine and it's usually better to pass all input parts into a hashing algorithm and extract a hash at the end, maybe have an API like...

well, if only the completely external to the container approach were in std for all the other things that use comparison, e.g. `BTreeMap/Set`, `BinaryHeap`, etc.

Any time your depending on `Drop` being run on some guard for safety, it's trivial to show that your API design is not safe: just `mem::forget` the guard, which is...

that could work, see [`rayon::scope`](https://docs.rs/rayon/1.4.0/rayon/fn.scope.html) for a similar scoped API.

I don't think you even need `Rc` or `Weak`. Assuming I didn't mess up somewhere, the following should work: ```rust #[pyclass] pub struct MyMutRef { v: Option, } impl MyMutRef...

I encountered this issue too last time I tried (like a month ago), so this isn't stale.

can you at least convert control characters to the [unicode control pictures block](https://en.wikipedia.org/wiki/Control_Pictures)? that way you can see what is actually there instead of seeing a bunch of `.` and...