[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] WebIDL cleanup



On Thu, May 31, 2012 at 8:06 AM, Boris Zbarsky <bzbarsky@mit.edu> wrote:
>
> On 5/31/12 10:42 AM, Glenn Maynard wrote:
>>
>> (Boris: does this look right to you?)
>
>
> It's right in that the IDL will lead to identical JS-perceived behavior.
>
> I don't have an opinion on which is simpler.

I agree that Glenn's suggestion is more terse, and that it would
shorten the IDL a fair amount. Any objections to making this change?

> I also don't know which is faster on the UA side; we don't have union
> support in our codegen yet, but once we do I could probably do an experiment
> and get that data.  I expect it to be pretty much a wash, honestly, with the
> union _maybe_ requiring one extra branch, which is pretty invisible on most
> of the functions we're talking about.  The uniform setters are the only ones
> where every cycle might matter.

Overloading and union support would probably have to be implemented
similarly. In WebKit's code generator (which is not currently Web
IDL), overloading is implemented with a series of nested if-tests
within a single function called by the JavaScript engine.

-Ken

-----------------------------------------------------------
You are currently subscribed to public_webgl@khronos.org.
To unsubscribe, send an email to majordomo@khronos.org with
the following command in the body of your email:
unsubscribe public_webgl
-----------------------------------------------------------