Abdullah Atta

Hey! Thank you for the suggestion. Yes, that is the plan; to implement Async Iterator API. It will require a bit of redesign I think but I will have a...

After some "alpha" testing of the sync iterator api, there is a little bit of a performance degradation. However, I don't think Iterator APIs should be used to get a...

> I would like the ability to log if a file was skipped because it was a symlink and symlink was disabled. This doesn't "sound" good for performance though but...

> I read an article though about Async Iterator being bad for perf The testing I have done implies the same. In a very crude sense, you could say that...

I considered going native before but the benefit of a 0 dependency purely Javascript library is far greater — I don't have to deal with platform differences or worry about...

Alright, curiosity got the better of me and I benchmarked `jwalk` with `fdir` using `napi-rs`. ## Code _I am no expert in Rust so this code is probably sub-par._ ```rs...

@Brooooooklyn Unfortunately getting this: ```rs error[E0599]: the method `filter_map` exists for struct `WalkDirGeneric`, but its trait bounds were not satisfied --> src/lib.rs:25:10 | 25 | .filter_map(|dir_entry| { | ^^^^^^^^^^ method...

Ah, had to do `into_iter`.

@vjpr Can you share your benchmark code? I'd like to take a look and see what is causing that huge bottleneck.

After a couple "optimizations", here are the results: ``` jwalk: 2.6862s ignore: 5.747s fdir: 3.2936s ``` These results are average of 5 iterations on the root directory containing 1.6million files...