A checkout of my repository [0] with many pdf and audio files (20GB) is slow with -u. These data files are normally ignored because 1) they are in .gitignore and 2) they are binary.

The repository contains CI files in .woodpecker. These are scripts that I'd normally expect to be searching in. Until a week ago I used -uu to do so, but that made rg take over 4 seconds for a search. Using -. brings the search time down to 24ms.

    git ls-files -z | time xargs -0 rg -w e23
    40ms

    rg -w. e23
    24ms

    rgg -w e23
    16ms

    rg -wuu e23
    2754ms

To reproduce this with the given repository, fill it with 20GB of binary files.

The -. flag makes this point moot though.

[0] https://codeberg.org/vandenoever/rehorse

Oh I see now. I now understand that you thought you couldn't convince ripgrep to search hidden files without also searching files typically ignored by gitignore. Thus, `git ls-files`.

Yes, now it makes sense. And yes, `-./--hidden` makes it moot. Thanks for following up!

I don't think this is the same thing as using gitignore.

It will only search tracked files. For that it can just use the index. I would expect the index to be faster than looking at the fs for listings.

I was extremely careful with my wording. Re-quoting, with added emphasis:

> ripgrep should generally already be approximating `git ls-files` by respecting gitignore.

See also: https://news.ycombinator.com/item?id=45629515

I'm just trying to be helpful, not call you out.

I've implemented gitignore aware file scanning before, and it was slower than git native operations when you only care about tracked files.

It's the speed that is the part I was speaking too, not the semantics.

Yes, I agree, `git ls-files` can indeed be faster than just `rg --files`. On my Chromium checkout:

    $ hyperfine --output pipe 'rg --files' 'git ls-files'
    Benchmark 1: rg --files
      Time (mean ± σ):     141.2 ms ±   7.1 ms    [User: 1134.5 ms, System: 376.3 ms]
      Range (min … max):   128.7 ms … 154.8 ms    20 runs

    Benchmark 2: git ls-files
      Time (mean ± σ):      54.9 ms ±   1.7 ms    [User: 41.7 ms, System: 13.1 ms]
      Range (min … max):    52.2 ms …  62.0 ms    54 runs

    Summary
      git ls-files ran
        2.57 ± 0.15 times faster than rg --files

But the semantics here are important, because ripgrep doesn't only try to approximate `git ls-files`. It also needs to deal with `.rgignore` and `.ignore`. And no git repository state will help there. ripgrep also supports respecting `.gitignore` even when it isn't inside a git repository.