ci(lint): validate name metadata

[uncenter]

No description provided.

[isabelroses]

This should probably also validate other @'s in the userstyles meta block too. For example using @license outside will also create an error.

[uncenter]

I think you misunderstand. This lints the actual value of @name. I think what you are suggesting is to make sure there aren't other @name/@<usercss-metadata-variable> outside of the meta block?

[uncenter]

Other than the first two (different issue - fixed in #399), this is working as expected. Verifies that the name provided in the userstyles.yml file matches the actual userstyle's name. WhatsApp Web has an extra space, others have capitalization issues or differ in some way. Might want to make it clear that one of the names isn't necessarily better but just that they don't match?

[isabelroses]

I think you misunderstand. This lints the actual value of @name. I think what you are suggesting is to make sure there aren't other @name/@<usercss-metadata-variable> outside of the meta block?

ohhh that makes sense, thanks for informing me

[uncenter]

I think you misunderstand. This lints the actual value of @name. I think what you are suggesting is to make sure there aren't other @name/@<usercss-metadata-variable> outside of the meta block?

ohhh that makes sense, thanks for informing me

Np. Not a bad idea though, but this PR (and the linter in general) will in a way let us know if @name (or other vars) is/are used elsewhere since that will be flagged when parsed incorrectly.

[uncenter]

Semi-dependent on #399 btw.

[nekowinston]

Semi-dependent on #399 btw.

in the sense of it throwing warnings until that PR is merged? Or are there other blockers?

[uncenter]

Yep, just that there is an issue with the usercss-meta package where if you have @name (or any @<variable> from the meta block) anywhere outside of the meta block (lengthy conversation about it), it messes up the parsing. We fixed it in #399 since it was relevant there.

[nekowinston]

LGTM then, don't really care about the warnings showing up for the next few commits.

[uncenter]

There are some legitimate issues raised by the linter with the addition of this PR though. Should I fix them here or in a separate PR after this merges?

[nekowinston]

Sure, let's go for the mass ping again 😅

[uncenter]

I think we should make the name uppercase in userstyles.yml (and here), see #346 (comment) for the conversation from a while back

[uncenter]

Actually we need to change userstyles.yml here as well:

[uncenter]

Bruh-

[uncenter]

@nekowinston is there a reason we are using esm.sh for type-fest but the npm: modifier for other NPM packages?

[nekowinston]

Yeah, type-fest is already in cache since our deno-lib uses ESM imports.

[uncenter]

Ohhh right neat.