Selectors - and indexed pseudo classes #2168
Conversation
github-actions
bot
added
the
feature definition
features/nth-child.yml
Outdated
| @@ -1,6 +1,19 @@ | |||
| name: ":nth-child() (initial support)" | |||
| description: "The `:nth-child()` and `:nth-last-child()` CSS functional pseudo-classes match elements by the relative position of elements (first, second, third, fourth, and so on), counted from the first or last sibling." | |||
| description: "The `:nth-child()` and `:nth-last-child()` CSS functional pseudo-classes match elements by their relative position, counted from the first or last sibling. The `:first-child`, `:last-child`, and `:only-child` pseudo-classes represent common use cases." | |||
There was a problem hiding this comment.
Position within what?
Possible suggestion:
| description: "The `:nth-child()` and `:nth-last-child()` CSS functional pseudo-classes match elements by their relative position, counted from the first or last sibling. The `:first-child`, `:last-child`, and `:only-child` pseudo-classes represent common use cases." | |
| description: "The `:nth-child()` and `:nth-last-child()` CSS functional pseudo-classes match elements based on their relative position within a list of elements, counted from the first or last sibling. The `:first-child`, `:last-child`, and `:only-child` pseudo-classes represent common use cases." |
Also, "represent common use cases" feels improvable. This doesn't describe what these pseudos do.
| - css.selectors.first-child.no_parent_required | ||
| - css.selectors.last-child.no_parent_required | ||
| - css.selectors.nth-child.no_parent_required | ||
| - css.selectors.nth-last-child.no_parent_required | ||
| - css.selectors.only-child.no_parent_required |
There was a problem hiding this comment.
I'm unclear as to what these subfeatures cover. Can you help me understand?
Would they make sense as a separate feature?
There was a problem hiding this comment.
Selectors level 3 had text referring to the children's parent, but selectors level 4 adds the note-
As there was no reason to exclude them from matching elements without parents, or with non-element parents, they have been rephrased to refer to an element’s relative index amongst its siblings.
For instance, the document root matches :first-child despite having no parent.
This doesn't seem like a feature to me. It also looks like Safari data might be incorrect, as document.documentElement.matches(":first-child") returns true in Safari.
There was a problem hiding this comment.
Thanks! Keeping them here, ignored, seems right then.
features/nth-of-type.yml
Outdated
| @@ -0,0 +1,10 @@ | |||
| name: ":nth-of-type() pseudo-classes" | |||
| description: "The `:nth-of-type()` CSS functional pseudo-classes match elements based on their position among siblings of the same type, counted from the first or last sibling. The `:only-of-type` pseudo-class matches any element that is the only child of its type, of its parent." | |||
There was a problem hiding this comment.
The first sentence only mentions :nth-of-type(), but uses the plural "pseudo-classes", which feels odd.
We might want to list all the pseudo-classes, like we do in the nth-child feature above.
There was a problem hiding this comment.
I updated this- it's a lot to cram into 300 characters, and I'd welcome other verbage.
| - css.selectors.first-child.no_parent_required | ||
| - css.selectors.last-child.no_parent_required | ||
| - css.selectors.nth-child.no_parent_required | ||
| - css.selectors.nth-last-child.no_parent_required | ||
| - css.selectors.only-child.no_parent_required |
There was a problem hiding this comment.
Thanks! Keeping them here, ignored, seems right then.
Co-authored-by: Patrick Brosset <patrickbrosset@gmail.com>
See #2170 for more info