feat(rome_css_parser): CSS parser #4682

[denbezrukov]

Summary

Initial implementation of CssTokenSource and CssParser for the CSS parser.
Added CssParserOptions which has an allow_single_line_comments option. This allows the writing of single line comments in CSS syntax.

Test Plan

cargo test

[netlify]

✅ Deploy Preview for docs-rometools canceled.

Name	Link
🔨 Latest commit	190ed1c
🔍 Latest deploy log	https://app.netlify.com/sites/docs-rometools/deploys/64bedeb1dfda2a0008514efd

[github-actions]

Parser conformance results on ubuntu-latest

js/262

Test result	main count	This PR count	Difference
Total	48863	48863	0
Passed	47810	47810	0
Failed	1053	1053	0
Panics	0	0	0
Coverage	97.84%	97.84%	0.00%

jsx/babel

Test result	main count	This PR count	Difference
Total	40	40	0
Passed	37	37	0
Failed	3	3	0
Panics	0	0	0
Coverage	92.50%	92.50%	0.00%

symbols/microsoft

Test result	main count	This PR count	Difference
Total	6212	6212	0
Passed	1764	1764	0
Failed	4448	4448	0
Panics	0	0	0
Coverage	28.40%	28.40%	0.00%

ts/babel

Test result	main count	This PR count	Difference
Total	639	639	0
Passed	573	573	0
Failed	66	66	0
Panics	0	0	0
Coverage	89.67%	89.67%	0.00%

ts/microsoft

Test result	main count	This PR count	Difference
Total	17224	17224	0
Passed	13121	13121	0
Failed	4103	4103	0
Panics	0	0	0
Coverage	76.18%	76.18%	0.00%

[denbezrukov]

It's the entry point for syntax parsing.

[denbezrukov]

It's a copy of the JSON test spec.

I'd like to write tests the same way we use in the JS parser. But I'm unsure what is the best way. I think that we can combine both approaches.

[ematipico]

Personally, I don't like the test suite of the JS parser:

• we have to run two commands in order to run the tests
• we always need to tweak the test generator parser to accommodate new things - like parser options, which is becoming quite difficult

The best way, personally, would be just to write a file to test and then check the snapshot with insta; of course the testing suite will do much more, but that's the gist of it.

[denbezrukov]

It seems that we don't need any special behavior for comments here. When allow_comments is true, it implies that COMMENT and MULTILINE_COMMENT kinds are considered as trivia.

tools/crates/rome_json_parser/src/token_source.rs

Lines 62 to 71 in 1dcd62b

	Ok(trivia_kind) => {
	if trivia_kind.is_newline() {
	trailing = false;
	self.preceding_line_break = true;
	}
	self.trivia
	.push(Trivia::new(trivia_kind, token.range(), trailing));
	}
	}