Using rg inside mounted dir via sshfs.

[ApplicationDeveloper]

I use ripgrep in fzf. It works awesome.
Is it expected that rg --files is slow when used in a mounted dir using sshfs?
Or is it just because of ssh?
Or is there an option that I have to use that I overlooked?

[BurntSushi]

In absolute terms, absolutely, it is expected. Consider what sshfs is doing: instead of reading file metadata from your local disk, it has to read that metadata over a network. Depending on your network latency, this could be significantly slower.

If you run rg --files twice in a row on an sshfs mounted directory, the second run may be faster due to the file metadata being cached in memory on your local system. For this reason, it's hard to do proper speed comparisons. If you're on Linux, then you can clear your local system's file cache with the following command:

$ sudo sh -c 'echo 3 > /proc/sys/vm/drop_caches'

That should then cause rg --files to run more slowly again.

The more relevant comparison here is whether rg --files is similar in speed to, say, something like find ./ -type f. They aren't doing strictly the same thing (since ripgrep respects your gitignore rules), but they should be similarish. You could use rg -uuu --files for a more fair comparison. Just remember to clear your file cache, as suggested above, between runs.

Or is there an option that I have to use that I overlooked?

There is no option. ripgrep doesn't care about sshfs. It just uses (and minimizes the number of) normal file stat calls to get metadata and read directory entries.