Feature: Add support to exhaustive-deps rule for any hook ending with Effect

[airjp73]

Summary

Following up on @gaearon 's comment here. My original motivation was to document the additionalHooks option in the exhaustive-deps lint rule. He suggested an alternate solution by linting any hook ending with Effect. This PR adds support for that.

Test Plan

I find that eslint rule tests are pretty thorough so all my testing is just through those test cases. I copied all the tests I could find for the additionalHooks option and reworked them to use a useXEffect naming convention instead of the additionalHooks option. I also tweaked some of the additionalHooks tests that were testing for useCustomEffect and changed it to useCustomHook because useCustomEffect would get caught anyway after these changes.

[codesandbox-ci]

This pull request is automatically built and testable in CodeSandbox.

To see build info of the built libraries, click here or the icon next to each commit SHA.

Latest deployment of this branch, based on commit 47994e7:

Sandbox	Source
billowing-bash-qn7n4	Configuration

[sizebot]

No significant bundle size changes to report.

Size changes (experimental)

Generated by 🚫 dangerJS against 47994e7

[sizebot]

No significant bundle size changes to report.

Size changes (stable)

Generated by 🚫 dangerJS against 47994e7

[karlvr]

Oh my code is so much nicer. Thank you thank you.

[gaearon]

Thanks.

[jquense]

This heuristic seems very prone to false positives, we have a bunch of custom "Effect" hooks that don't take dep arguments and now require pragma's to avoid warnings about potential infinite updates

[Glinkis]

I am using hooks.macro, which adds a few hook utilities that doesn't take a dependency array. This now incorrectly shows up as a warning.

[gaearon]

One thing we could do is to only run checks if you have a function followed by an array, in addition to ending with Effect. But I'm leaning towards reverting this altogether. There's indeed been too many false positives being reported.

[gaearon]

I'm going to have to revert this because the heuristic has too many reasonable false positives.
Thanks for trying!
#19004

[gaearon]

This is removed in eslint-plugin-react-hooks@4.0.3. Thanks all for the input.