David Zearing

Responding to both above: > Peers make sense for any ecosystem with a core - eslint, babel, webpack, typescript, react, etc. It declares compatibility, and does already require something to...

I've summarized my thoughts on how to address peer dependencies here: https://hackmd.io/@dzearing/BJifNnpsq The important parts are at the bottom but the TLDR: * app hosts can define their own policy...

FWIW just to add to this issue; we've been evaluating this plugin and we're blocked on using this as well. For us it creates large bundle regressions. Most css has...

I believe @LFFATE means if we use the previously suggested custom prefix to strip the dirs, then that will create the dupe rule situation, but that could be worked around...

If a user renders styles but does not provide the `dir` attribute, styles still render with this approach. Currently if the attribute is omitted, your directionally sensitive properties won't even...

@manucorporat Created a small repro repo here: https://github.com/dzearing/qwik-library-styling-repro README explains everything; look for the "red background Counter". Works in isolation within the library, but styling won't come along to the...

@manucorporat did you get anywhere on this one?

I have some advice on things to validate for this rule: * For each file path, validate the file exists * For each path that has an "import" conditional, validate...

I think it's fine to do that for performance and the css variable nondeterministic reasons, but I think there needs to be feedback to train developers to do the right...

Another thought... are there APIs browsers could add here to simplify some of this? (E.g. shorthand parser api or something.) We should consider if there are asks to browser vendors...