Add HTMLHyperlinkElementUtils. #144
Conversation
| @@ -0,0 +1,41 @@ | |||
| -- https://html.spec.whatwg.org/multipage/links.html#htmlhyperlinkelementutils | |||
There was a problem hiding this comment.
Can we extend this comment a bit futher - the specs used in this library are generally from the w3c not the whatwg... however, in this case, after doing some research it turns out this might be the only place the IDL is actually stated for this API.
It's called URLUtils in the HTML5 spec, which points to a URL standard document, which now points to a whatwg document, which doesn't have the interface listed. As far as I can tell from the non-IDL descriptions in the w3c spec though, this HTMLHyperlinkElementUtils IDL has all the same elements as the now-missing URLUtils API.
So I'd suggest something like:
The `URLUtils` interface referenced in https://www.w3.org/TR/html50/infrastructure.html#urlutils, now documented at https://html.spec.whatwg.org/multipage/links.html#htmlhyperlinkelementutils
There was a problem hiding this comment.
I wonder which name we should use for this module and type, between "URLUtils" and "HTMLHyperlinkElementUtils". I feel like the latter term actually has a specification, which is good enough reason to choose this name rather than "URLUtils", and also it better fits the names of other types in this lib, like "HTMLAnchorElement".
Yeah, I was curious the relationship between w3 and whatwg's specs.
BTW, I chose the whatwg spec because that's the one MDN references:
Specifications
HTML Living Standard
The definition of 'HTMLHyperlinkElementUtils.href' in that specification.
Source: https://developer.mozilla.org/en-US/docs/Web/API/HTMLHyperlinkElementUtils/href
There was a problem hiding this comment.
Thanks for the review!
There was a problem hiding this comment.
As far as I understand it, the whatwg/living standard stuff is basically "encode what is commonly available in a spec" 😄 (rather than the w3c specs that describe how things are intended to work, before an implementation exists). I said in the readme we were going off the w3c specs, but maybe the living standard(s) would be more practical.
Really I was just hoping to avoid implementing browser quirks in the API / experimental stuff that is not widely implemented and agreed upon yet. LS would probably also satisfy that.
|
Bump - this one is reviewable again, possibly good to merge. |
|
Sorry, thanks! |
#143
This API is required to get the
hrefattribute from an anchor element. I just threw this together to get a concrete proposal for this. I think it's working in my experiment project.