Josh McKinney

Rebased and re-ran benchmark: Before: ``` ❯ cargo bench buffer/diff Compiling ratatui v0.28.1 (/Users/joshka/local/ratatui) Finished `bench` profile [optimized] target(s) in 20.36s Running benches/main.rs (target/release/deps/main-104385649c38cd56) buffer/diff time: [129.09 µs 129.51 µs...

I'd recommend seeing if you can find a more minimal repro for this, perhaps with your own code if possible. If it's only happening on a single terminal, then it's...

> This doesn't seem like a ratatui problem and should be taken up with foot right? If foot is drawing zero-width characters as width 1 and all other terminals don't,...

Noted some concerns about foots response in the thread there. From a Ratatui perspective, I'd suggest pre-filtering your output strings. This doesn't seem like it's a useful character to render....

I ran the following to get a feel for how different unicode-width is from the reality of terminals in a few different apps: ```toml [package] name = "unicode-width-term" version =...

This one is going to be OS and terminal emulator specific. I'd recommend that you take a step back and look at how your terminal choice(es) handle this with any...

This is mainly an experiment. In theory Widget possibly should be Implemented for FnOnce, but that runs into a conflict with the blanket implementation of Widget for &W: WidgetRef. (Can't...

> I think it would more natural to add a method to `Frame` that accepts a `FnOnce(&mut Buffer)`. That would be what I'd look for and expect to find if...

> I don't want to derail this PR with a broad discussion of issues, so after this reply I'll attempt to politely step back and let a more focused discussion...

More likely Line than span. Possibly even Text.