Codecov Report

Merging#425 intomaster willdecrease coverage by0.73%.
The diff coverage is68.88%.

@@            Coverage Diff            @@##           master    #425      +/-   ##=========================================- Coverage   96.23%   95.5%   -0.74%=========================================  Files          83      88       +5       Lines        4067    4093      +26       Branches     1149    1150       +1     =========================================- Hits         3914    3909       -5- Misses         52      82      +30- Partials      101     102       +1

@typescript-eslint/experimental-utils

Sounds like a good idea!
I'll make the change now :)

Hello! 👋

This should make it easier for people to consume our libraries when writing rules they don't want to submit here (because they're internal rules, etc), orwhen they're writing an eslint plugin in typescript.

I tried to do this, and it broke horribly since the utils depend ontypescript. Is that on purpose? I'd like to author the plugin in TS, but that shouldn't be visible to consumers of the plugin.

Sorta related, I got a type error in the test unless I specified the TS parser - are the plugins compatible with the standard parser, or will the type safety be wrong if another parser is used?

Refjest-community/eslint-plugin-jest#269

@SimenB - there's no direct dependency on typescript, but there is a peer dependency:
https://github.com/typescript-eslint/typescript-eslint/blob/master/packages/experimental-utils/package.json#L42-L45

The package re-exports some things fromtypescript-estree via the default entry point
https://github.com/typescript-eslint/typescript-eslint/blob/master/packages/experimental-utils/src/index.ts#L9-L14

Which causes the package to init, and it has a dependency of typescript (it's marked as a peer dep to help out thecreate-react-app folks that do some dual install stuff (#443).

I'll put up a fix so that this doesn't happen.

Ah, that's awesome! Thanks.

Thoughts on not specifyingparser inRuleTester?

I'd like to run tests with the default eslint parser

feel free to open a PR if you want!
I thought about that but was deferring on it till someone complained :)
Happy to have it as a generic so that it is easy to enforce a specific parser string at compile time.

Hehe, fair enough! I'll see what I can do

Happy to have it as a generic so that it is easy to enforce a specific parser string at compile time.

How so? I thought to just make itparser?: string