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

Re: [Public WebGL] WebGL conformance test updates



Great. Thanks for these updates.

boolean-argument-conversion.html seems like it's on the boundary of a
"webgl" test -- seems more like a test of Web IDL's behavior. Not sure
we should be moving the conformance tests in that direction, but
generally agree that if it's an area where implementations could
differ and affect application behavior then we should test it.

-Ken


On Mon, May 28, 2012 at 4:33 PM, Benoit Jacob <bjacob@mozilla.com> wrote:
>
> Hi,
>
> I've just made a few WebGL conformance test updates. This completes the changes that were currently known to me as needed (the webgl-test-needed flag on Mozilla bugzilla, see query link below [1]). Thanks to Gregg and Ken for having gone over several of these bugs.
>
> New file:
>
> conformance/misc/boolean-argument-conversion.html
>
> Modified files:
>
> conformance/context/context-attributes-alpha-depth-stencil-antialias.html
> conformance/renderbuffers/framebuffer-object-attachment.html
> conformance/extensions/get-extension.html
> conformance/uniforms/gl-uniform-arrays.html
>
> Log:
>
> ------------------------------------------------------------------------
> r17926 | bjacob | 2012-05-28 15:21:54 -0400 (Mon, 28 May 2012) | 5 lines
>
> Add new test page covering conversion of boolean arguments
>
> When an object is converted to a boolean, it should always evaluate as true. Any valueOf() method should not even get called. See Mozilla bug 727590 where Gecko incorrectly converted such an argument to a Number instead of a Boolean, giving the wrong behavior. See 9.2 and 9.3 in the EcmaScript specification.
>
>
> ------------------------------------------------------------------------
> r17927 | bjacob | 2012-05-28 15:25:06 -0400 (Mon, 28 May 2012) | 3 lines
>
> fixed properties for new test
>
>
> ------------------------------------------------------------------------
> r17928 | bjacob | 2012-05-28 16:36:16 -0400 (Mon, 28 May 2012) | 3 lines
>
> attempt to trigger gc in getExtension test
>
>
> ------------------------------------------------------------------------
> r17929 | bjacob | 2012-05-28 17:14:18 -0400 (Mon, 28 May 2012) | 3 lines
>
> check the values returned by getParameter(XXX_BITS) with complete framebuffer objects bound
>
>
> ------------------------------------------------------------------------
> r17930 | bjacob | 2012-05-28 17:18:12 -0400 (Mon, 28 May 2012) | 3 lines
>
> remove debugging leftovers
>
>
> ------------------------------------------------------------------------
> r17932 | bjacob | 2012-05-28 17:49:34 -0400 (Mon, 28 May 2012) | 3 lines
>
> check that the width of each component/buffer is compatible with context creation flags
>
>
> ------------------------------------------------------------------------
> r17933 | bjacob | 2012-05-28 19:18:49 -0400 (Mon, 28 May 2012) | 3 lines
>
> add tests covering valid uniform setter calls with smaller and larger arrays (multiples of array element size).
>
>
> ------------------------------------------------------------------------
>
>
> Cheers,
> Benoit
>
>
> [1] https://bugzilla.mozilla.org/buglist.cgi?classification=Components;status_whiteboard_type=allwordssubstr;query_format=advanced;status_whiteboard=webgl-test-needed;product=Core;list_id=3219606
>
> -----------------------------------------------------------
> 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
> -----------------------------------------------------------
>

-----------------------------------------------------------
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
-----------------------------------------------------------