[no-duplicate-hooks]: allow subsequent describe.each?

[kwonoj]

Is there way to allow no-duplicate-hooks allow to work with describe.each, or is there recommended hook setup conforms those rules?

Interestingly, this doesn't error out for all describe.each but only if top-level describe includes subsequent nested describe.each.

repo: https://github.com/kwonoj/eslint-jest-duplicated-hooks

code:

describe('top', () => {
  describe.each(['approve', 'decline'])('inner1', () => {
		beforeEach(() => {
      console.log('set');
		});

		it('renders correctly', () => {
      expect(true).toBeTruthy();
		});
	});

	describe.each(['approve', 'decline'])('inner2', () => {
		beforeEach(() => {
			console.log('set2');
		});

		it('renders correctly 2', () => {
			expect(true).toBeTruthy
		});
	});
});

config:

module.exports = {
    "env": {
        "browser": true,
        "commonjs": true,
        "es2020": true,
        "jest/globals": true
    },
    "extends": "eslint:recommended",
    "parserOptions": {
        "ecmaVersion": 11
    },
    "plugins": [
        "jest"
    ],
    "rules": {
        "jest/no-duplicate-hooks": "error"
    }
};

[github-actions]

🎉 This issue has been resolved in version 24.3.3 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀

[brewster1134]

I use the context alias for describe (as mentioned here), but i get a false negative for jest/no-duplicate-hooks if i have multiple setup/tear-down functions defined

describe('X, () => {
  context('when foo', () => {
    beforeEach(() => {});
  });

  context('when bar', () => {
    beforeEach(() => {});
  });
});

the tests run fine, just eslint complains. is there a way to configure it to treat context as it's own scope?

[G-Rath]

@brewster1134 as you know, support for general context as an alias of describe is tracked by #83

However, what you're describing looks like a bug to me - ultimately, I think no-duplicate-hooks should just look for multiple uses of the same hook within the same scope (so regardless of if it's in a describe, context, it, etc)

[brewster1134]

that makes sense. i had assumed the same that it would work with any function name VS a white/black list of approved scopes, but when i only had the issue of beforeEach incorrectly flagged as being defined multiple times when using context, i thought maybe it needed explicit support.

let me know if i can help out by filing a bug