Re: Request for review: Outstanding bugs on canvas tests

On 04/04/2011 05:43 PM, Philip Taylor wrote:
> Ms2ger wrote:
>> http://www.w3.org/Bugs/Public/show_bug.cgi?id=12142
>> http://dvcs.w3.org/hg/html/rev/eb60791dc704
>
> How does this case interact with the overload resolution algorithm? I
> don't understand WebIDL at all, but in
> http://dev.w3.org/2006/webapi/WebIDL/#dfn-overload-resolution-algorithm
> it sounds like an argument with ECMAScript value 'undefined' would not
> add object types into the 'types' set, and so it wouldn't match any of
> the drawImage overloads (which all expect some object type as their
> first argument) and should trigger a TypeError. (It'll never reach the
> stage of converting the ES value 'undefined' into an IDL value 'null',
> since it never gets far enough to pick a function to convert the
> argument list for). (I could be completely misunderstanding everything,
> though.)

Cameron agrees with you, so I'm going to assume you're right and I'm 
wrong. I reverted my change. (He said that the spec is subject to 
change, though, but we can handle that once it happens.)

> The changes to the other tests look fine to me.

Thanks.

-- 
Ms2ger

Received on Thursday, 7 April 2011 22:18:28 UTC