Fix declaration-block-no-redundant-longhand-properties with basic keywords

[mattxwang]

Which issue, if any, is this issue related to?

Closes #6738.

Is there anything in the PR that needs further explanation?

The scope of this PR is broader than autofixing, since we also shouldn't flag instances which use a basic keyword. Let me know if additional test cases / documentation are desired!

Note that we already had this feature for inherit; this PR just expands it to use basicKeywords.

If we'd like, we could also create a more "aggressive" approach that converts

margin-top: inset;
margin-bottom: inset;
margin-right: inset;
margin-left: inset;

to

margin: inset;

which is valid (to my understanding). The rule doesn't currently behave this way with inherit; but, happy to add this if we think it's helpful!

---

Spec ref for shorthand properties not supporting CSS-wide keywords:

If a shorthand is specified as one of the CSS-wide keywords [css-values-3], it sets all of its sub-properties to that keyword, including any that are reset-only sub-properties. (Note that these keywords cannot be combined with other values in a single declaration, not even in a shorthand.)

[changeset-bot]

🦋 Changeset detected

Latest commit: ceafc35

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
stylelint	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[ybiquitous]

@mattxwang Thanks for the pull request!

If we'd like, we could also create a more "aggressive" approach that converts

How do you estimate the work? We could address it with another pull request if it is a bit hard.
Of course, if you prefer, you can do it with this pull request. I'll leave it to you.

[mattxwang]

How do you estimate the work? We could address it with another pull request if it is a bit hard.

I think leaving it to another PR is a good idea; in particular, this PR reduces false positives, while the other change would reduce false negatives. I can open an issue and allow for some discussion after merging this PR!

[ybiquitous]

Thanks, LGTM 👍🏼

[pbojinov]

After upgrading to 15.4.0 I started noticing declaration-block-no-redundant-longhand-properties come up for styles that have a top, right, bottom, left set to 0 for position absolutes. What would be the best way to handle this inset besides disabling the next line?

.example {
  position: absolute;
  top: 0;
  right: 0;
  bottom: 0;
  /* stylelint-disable-next-line declaration-block-no-redundant-longhand-properties */
  left: 0;
}

[ybiquitous]

@pbojinov You mean you don't want to use the inset shorthand, right? If so, using a disabled comment is a way as you commented.

As another way, you can either turn off the rule or ignore only inset via a config like this:

{
  "rules": {
    "declaration-block-no-redundant-longhand-properties": null
  }
}

{
  "rules": {
    "declaration-block-no-redundant-longhand-properties": [true, {"ignoreShorthands": ["inset"]}]
  }
}

[Mouvedia]

related: #7214 (comment)

[mattxwang]

related: #7214 (comment)

Sorry, I think I'm missing something - how is this related to the linked comment? I can submit a fix if so (since I probably introduced the bug 😓 )

[Mouvedia]

It's just a reminder so that, if we tackle "CSS-wide keywords" for this rule, we take into account what has been learned from this PR. You don't have to do anything :)