New Rule: `no-duplicate-selectors`

[przemyslaw-zan]

Rule details

Disallow selectors that duplicate an earlier rule's selector, and repeated selectors within a single selector list.

What type of rule is this?

Warns about a potential problem

Example code

.card {
    color: black;
}

/* could be far apart - a copy-paste or merge-conflict artifact */

.card {
    color: gray; /* silently wins; editing the first block appears to do nothing */
}

.button,
.button {
    padding: 8px;
}

Prior Art

• no-duplicate-selectors, enabled in stylelint's error-catching preset: https://github.com/stylelint/stylelint-config-recommended/blob/02f3006b4bb99c33b9002dc94477946f84c225d6/index.js#L35

Participation

• I am willing to submit a pull request to implement this rule.

AI acknowledgment

• I did not use AI to generate this issue report.
• (If the above is not checked) I have reviewed the AI-generated content before submitting.

Additional comments

• The failure mode this catches is the same one as no-duplicate-imports: the duplicate is valid CSS handled deterministically by the browser, but it usually enters the file unintentionally (copy-paste, merge conflict) and far away from the original, so it survives review. The result is confusing - declarations in the later block silently win ties, so edits to the earlier block appear to have no effect.
• Only selectors in the same context would be compared: an identical selector under a different @media/@supports/@layer condition or a different nesting parent is not a duplicate. Comparison would be done on normalized selector text (whitespace-insensitive).
• Optional for first implementation: detecting equivalent resolved selectors written differently (.a .b vs .a { & .b }).

[DMartens]

I agree that this would be a helpful rule.
I think the we should clearly define how duplicates would work with selector nesting:

a {}
a b {} /* Should nesting always be preferred? Maybe an option */

a {
	b {}
}
a b {} /* This should be detected as a duplicate */

As this is the 5th issue asking to port a stylelint rule in a short timeframe, I ask you to refrain from opening more before the existing issues are closed.

[przemyslaw-zan]

As this is the 5th issue asking to port a stylelint rule in a short timeframe, I ask you to refrain from opening more before the existing issues are closed.

Sure, I believe I exhausted my requests for now anyway :)