[BUG]: Byte literal detected as possible spelling mistake

[jusexton]

Issue Description

Currently, a byte literal is detected as a possible spelling mistake. I would expect this value to not be seen as a possible mistake.

Operating System

macOS

Editor

Zed

Codebook Version

No response

Configuration

N/A

Steps to Reproduce

1. Instantiate byte literal however you prefer
2. Notice the possible spelling error

Expected Behavior

No possible spelling mistake displayed

Actual Behavior

Possible spelling mistake displayed

Code Sample

let value = b'a';

Log Output

Additional Context

No response

[jusexton]

I'm no tree sitter expert, but this seems like an inconsistency with tree sitter.

Tree sitter is capturing "b'a'" instead of just "a" for the char_literal node type, which is kinda strange considering char_literal can now capture both "a" and "b'a'". This pattern doesn't seem to emerge anywhere else when matching on literals or raw strings ie r#"a"# br"a" b"a".

[blopker]

Hey! Thanks for the report. Yeah, this is happening in Python with f strings as well. Ideally the tree sitter package would allow us to just select the string part, but, yeah, I won't know if it's possible until I can take a closer look.

If you find a way, I'd love to know!

[blopker]

Now that I've slept on it, it doesn't really make sense to capture char_literal at all for spell checking since it's only ever 1 character, right?

[jusexton]

Haha, so easy to get wrapped up thinking about why something works a certain way instead of asking if it should even exist at all. Genius.

But yeah, I agree, I cant think of an edge case.

Sadly this doesn't solve the issue on the python side though so maybe a little investigation is still needed there. Is there already an issue open for the Python issue? I'm not able tor reproduce it with something like name = f"correct spelling"

[blopker]

Hm, I guess I fixed the Python issue at some point too... Anyway, I've removed char_literal from Rust, will be fixed on next release. Thank you!