Remove spec_url from api.Document.execCommand

[SamB]

The only mentions of execCommand I saw in Trusted Types were in the names of tests.

According to https://respec.org/xref/?term=execCommand%28%29, this API is defined in just one spec, so use that one.

[SamB]

It is true that the spec doesn't exactly have good standing, but it is the only one we appear to have, and the api is implemented ~everywhere.

[queengooborg]

It looks like the spec URL you had tried to use is an unofficial draft, which Webref does not consider as a valid standard. Upon further research, it appears that the feature was originally defined in the First Public Working Draft of the trusted types spec, but was then removed in the following Working Draft.

As such, I think that it's actually better to mark this as non-standard and remove the spec URL.

[SamB]

Note that html appears to make quite a few normative references to execCommand. The Editing APIs section, in particular, appears to incorporate by reference all of the following:

{
  "https://w3c.github.io/editing/docs/execCommand/": {
    "anchors": [
      "execcommand%28%29",
      "querycommandenabled%28%29",
      "querycommandindeterm%28%29",
      "querycommandstate%28%29",
      "querycommandsupported%28%29",
      "querycommandvalue%28%29",
      "editable",
      "active-range",
      "editing-host-of",
      "delete-the-selection"
    ]
  }
}

(that's every anchor it references from execComman, as listed at https://github.com/w3c/webref/blob/48fcf5576be961ce5bc6cd2a134301b572c5952d/ed/links/html.json#L2807) as well as some user interface requirements. (It also apparently attempts to include a definition of text selection, which presumably should now refer to https://w3c.github.io/selection-api/#dfn-selection instead.)

[caugner]

I don't think we should link to a normative reference that doesn't specify the feature, and since there is no official spec for this (deprecated) feature, the right thing to do here is to remove the spec_url:

[caugner]

@Elchi3 While waiting for the spec state change, should we either keep the (broken) spec link as is (i.e. close this PR), or remove the spec_url? I would prefer to not keep this PR hanging around.

[Elchi3]

I think I would remove the spec_url for now. Hopefully the Web Editing Working Group will publish an active spec for this at some point and then we can add it here.

[jpc-ae]

w3c lists the spec url as https://w3c.github.io/editing/docs/execCommand/

Considering the ubiquitousness of this command, could the spec_url not be updated to point there rather than removing it completely? I get that it's an "unofficial draft", but until it starts being taking out of browsers, I don't see a reason not to include it.

[caugner]

@jpc-ae See the discussion above. As per our guidelines, we don't link to "unofficial" specs.