[Webkit-unassigned] [Bug 9181] Complete DOMUIEvent Obj-C API to reflect UIEvent

[bugzilla-daemon at opendarwin.org]

http://bugzilla.opendarwin.org/show_bug.cgi?id=9181





------- Comment #6 from darin at apple.com  2006-05-31 17:13 PDT -------
(In reply to comment #5)
> Much of what I added is actually part of the DOM Level 3 Events specs. I'm all
> for standards, so maybe the goal could be to have an API that covers the DOM
> {2,3} Events specs completely (and nothing else)?

Sounds OK.

> It is probably also unnecessary to expose (DOM)UIEventWithKeyState in the API -
> it's used in the C++ implementation, but if we want to model the API after the
> DOM specs, then it would be confusing.

The way this works is that we use .idl files to define what we want to expose
publicly. This doesn't have to exactly match our C++.

> Regarding generating the API: The C++ implementation does not closely follow
> the specs, so generating the ObjC API from that would expose unnecessary
> classes like UIEventWithKeyState. Writing the ObjC wrapper is pretty
> straightforward and it won't need updating very often, so I think it would be
> OK to do that manually.
> 

(In reply to comment #5)
> Much of what I added is actually part of the DOM Level 3 Events specs. I'm all
> for standards, so maybe the goal could be to have an API that covers the DOM
> {2,3} Events specs completely (and nothing else)?
> 
> It is probably also unnecessary to expose (DOM)UIEventWithKeyState in the API -
> it's used in the C++ implementation, but if we want to model the API after the
> DOM specs, then it would be confusing.
> 
> Regarding generating the API: The C++ implementation does not closely follow
> the specs, so generating the ObjC API from that would expose unnecessary
> classes like UIEventWithKeyState. Writing the ObjC wrapper is pretty
> straightforward and it won't need updating very often, so I think it would be
> OK to do that manually.

As I said, we do this from .idl files, not the C++ code. We're currently using
this for the JavaScript bindings.


-- 
Configure bugmail: http://bugzilla.opendarwin.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.