Choosing the parsing backend

[pintariching]

Some options are:

• nom
• chumsky
• combine
• pest

nom and pest seem to be the most popular right now.
Any other suggestions? Tell me what we should use

[linrium]

I found a benchmark between chumsky and nom: zesterer/chumsky#94
And some related topics between nom and pest: https://unhandledexpression.com/general/2018/10/04/no-pest-is-not-faster-than-nom.html

[pintariching]

Chumsky seems well maintained and apparently very fast. I'm leaning towards it.

[qwertydelle]

From looking into Svelte Parser HTML is parsed using its own parser while javascript an CSS are both parsed by 3rd party libraries. Rustle could use a similar approach by parsing javascript with swc_ecma_parser or another parser and CSS with another library.

[pintariching]

Yeah after looking into it, I think I'm going to use the swc crates for parsing the js and css and try and rewrite their html compiler

[qwertydelle]

Does swc have a css parser?

[pintariching]

I looked around in the swc crates and I found it