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

Re: [Public WebGL] isBuffer, isTexture etc.



OK, I see your points. Any objections to more strongly typing these
entry points? If so, I'll be happy to update the spec.

-Ken

On Sat, Apr 24, 2010 at 5:10 AM, Cedric Vivier <cedricv@neonux.com> wrote:
> On Sat, Apr 24, 2010 at 02:56, Gregg Tavares <gman@google.com> wrote:
>>
>> I disagree.  By that argument ctx.bindTexture should be ctx.bindTexture(in
>> GLenum target, in WebGLObject texture)
>
> I agree with Gregg, though the naming can be misleading (as Chris pointed
> out), I believe isTexture's rationale is to check that a texture object is
> _valid_ in the current context [1], not to check that the object _is_ a
> texture - something that would not require an API method in the first place
> (ie. to do that one would use standard instanceof).
> I do not believe as well there is real use cases that would make weak-typing
> those is* functions a useful feature, on the other hand changing the
> signature to the related object type makes API's intent clearer to
> developers and can help catch logic errors earlier (possibly by static
> analysis even).
> Regards,
>
> [1] : the object has been generated by current context or sharing is
> possible, and the object has not been created before a context loss.
> [2] : using isBuffer when isTexture was intented for instance, imho this
> case of "typo" is way more likely to happen than usage of the supposed
> "type-checking" feature allowed by weak-typed is* functions.
>

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