**:not()** with a selector list doesn't expand & symbol

[WORMSS]

:not(&--a):not(&--b) does expand the & symbol into the parent's selector as expected.
:not(&--a, &--b) doesn't expand the & symbol into the parent's selector.

To reproduce:

.works {
  &:not(&--a):not(&--b):hover {
    background-color: green;
  }
}

.broken {
  &:not(&--a, &--b):hover {
    background-color: green;
  }
}

Current behavior:

.works:not(.works--a):not(.works--b):hover {
  background-color: green;
}

.broken:not(&--a, &--b):hover {
  background-color: green;
}

Expected behavior:

.works:not(.works--a):not(.works--b):hover {
  background-color: green;
}

.broken:not(.broken--a, .broken--b):hover {
  background-color: green;
}

Environment information:

• less version: 4.2.2
• nodejs version: any
• operating system: any

We are currently in the middle of a battle of linters. We are told :not(a):not(b) is being classed as complex, and should be changed to :not(a, b)but it appears less forgets to expand when this feature is being used.4.2.2at least compiles the syntax (though incorrectly) our older version4.1.3` just keels over completely.. So I am guessing this is a relatively new feature and hasn't had much usage yet.

[dosubot]

The issue you're encountering with the & symbol not expanding within a selector list inside :not() is a known limitation in Less. In version 4.2.2, there was a fix related to nested pseudo-selector parsing, which might have improved some aspects of handling :not(), but it seems that the specific case of selector lists within :not() still doesn't expand the & symbol as expected ^[1].

This behavior indicates that Less currently does not support expanding the & symbol within a selector list inside :not(). The workaround would be to avoid using selector lists in :not() when you need the & symbol to expand, and instead use separate :not() pseudo-classes as you did in the .works example.

If this is causing issues with linters, you might need to adjust the linter rules or continue using the workaround until a future update potentially addresses this limitation.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[WORMSS]

I am guessing @puckowski would be the best person for this type of issue? Due to his work on the latest fix for listed selectors?

[puckowski]

Thank you for the issue report. I'll spend some time this weekend on this issue and I will try to come up with a fix PR.

[puckowski]

I think I have a fix. Will work on adding more tests and try to get a PR submitted this weekend.

[puckowski]

PR #4314 should resolve this issue, but it looks like we may be having new CI problems.

[puckowski]

FYI figured out the new CI issue so PR #4314 can be merged after review.