Use config previewer

[yujinyuz]

One problem I encounter so far in staged_grep is when I'm typing too, neovim hangs when using buffer previews.

Using the default cat/bat previewer doesn't. Also, the builtin live_grep doesn't hang as well

[Conni2461]

LGTM Thanks. About the slowdown its a known issue and it happens because of sync ts parsing. Everything else about the new buffer previewers is async besides that but i have an idea how to make it a little bit more responsive. When the selected entries change a lot. e.g. when scrolling. If you want i can ping you :] Also should i just merge that?

[yujinyuz]

@Conni2461 Sure! I've already tested that so far and works for me.

Off topic:

Would it be possible for this to implement the tags as well? LIke using fzf/rg instead?

[Conni2461]

I have no idea how this works 😆 I just have the power to merge. But i guess not because i can't see a way how we could read the tags file with fzf The problem with tags right now is that the reading is not happening async. There is a PR that tries to achieve that https://github.com/nvim-telescope/telescope.nvim/pull/288 but he is currently stuck and i don't know why its not working

[yujinyuz]

Thanks @Conni2461

I just wanted to use fzf as the default sorter/file selector since it fits my use case better and returns my expected results (via fzf_writer.files) compared to https://github.com/nvim-telescope/telescope-fzy-native.nvim

[Conni2461]

Sure thing but fzf-writer works differently then fzy-native because here we are starting a fzf termjob and collecting data and let fzf do everything. With fzy we have a compiled shared lib and wrote some ffi so we can call those c functions and that let them do the sorting.

For us to achieve something similar with fzf would be create a c fork of fzf sorting algorithm and write ffi bindings. Then we could have something similar to fzy-native.

Edit: Basically we have to fork that file in c: https://github.com/junegunn/fzf/blob/master/src/algo/algo.go