Andrey Sitnik

@chriseppstein thanks, This is very useful feedback. > Sometimes comments are their own node, sometimes comments are part of raws, etc. Let’s make the rule for new selector/value API. Comments...

> PostCSS also [does not follow this rule][declaration_comment_ast]. Yeap. It is impossible if you don’t have value and selectors parser. But now we will have them ;) > I think...

I think we should focus API on how people will use it. This is why I am not sure that putting whitespace as a tokens is a great idea. User...

Are you sure, that all cases could be solved by walkers? 😏 For instance, in Autoprefixer I need to find function and replace it (yeap, it is easy for walker)....

I am right now working on my new project. Will not start a new feature in PostCSS at least for a few months.

Tailwind contains many different tools inside. It is hard to debug it. Can you try to remove Tailwind tools to find a few plugins to reproduce the issue?

Very strange. We didn’t change anything with plugin API between 8.1.13 and 8.1.14. Here is only commit related to JS sources: https://github.com/postcss/postcss/commit/a187464a8e8647caa0a75183bab91f50ab9d6f99

1. What are CSS escaping rules? Ignore the next symbol after `\`? 2. What is the current output?

Do we really need to change it in the tokenizer if it affects performance? We can change parser and avoid any performance investigations.

I recommend copying `indexOf`/RegExps jumps first since they are bringing the main performance improvements https://github.com/postcss/postcss/blob/master/lib/tokenize.es6#L253-L254 https://github.com/postcss/postcss/blob/master/lib/tokenize.es6#L161-L175 https://github.com/postcss/postcss/blob/master/lib/tokenize.es6#L116 https://github.com/postcss/postcss/blob/master/lib/tokenize.es6#L284-L285 JS is very slow when you are tokenizing symbol by symbol.