Inline IDLDefined #20
Comments
|
I think it would be fine to just talk about properties defined by the IDL |
|
I think that might end up being confusing. E.g., when IDL itself talks about members of of interfaces or consequential interfaces that excludes the members that will be added as part of IDL syntax, such as an iterable declaration adding "entries". |
|
I think the way to define this is by simply putting all the properties that are IDL-defined in a slot after the object is allocated. That way this and #22 can be defined in one simple sweep. |
This addresses #20 partially. We still need to know what internal methods IDL uses and make sure we don’t override those until after we allocated the object (through the standardDefinedProperties slot). This fixes #19 by defining the properties listed through internal methods rather than IDL directly after allocation. It also makes them unenumerable (a change from IDL requested by @zcorpan).
|
IDL doesn't really define the object initialization explicitly, but seems like it can all be done with |
In https://www.w3.org/Bugs/Public/show_bug.cgi?id=29376 we concluded that this would be better defined here as only
Locationneeds it. What exactly it needs to say is still a little bit unclear as noted there, I hope @bholley or @bzbarsky can weigh in on that.As @bzbarsky noted it would be best not to align this fully with
WindowProxy's handling of[[Configurable]]as that would less likely be web-compatible.The text was updated successfully, but these errors were encountered: