Closed
Bug 637077
Opened 14 years ago
Closed 13 years ago
missing ImageData interface makes WebGL conformance test fail
Categories
(Core :: Graphics: Canvas2D, defect)
Core
Graphics: Canvas2D
Tracking
()
RESOLVED
DUPLICATE
of bug 550309
People
(Reporter: bjacob, Unassigned)
References
Details
getImageData() and putImageData() are implemented by duck-typing; the ImageData interface is missing. This is preventing me from passing this WebGL test,
https://cvs.khronos.org/svn/repos/registry/trunk/public/webgl/sdk/tests/conformance/more/conformance/quickCheckAPI.html
as it fails on this code (quickCheckAPI.js line 1050):
ximage instanceof ImageData
with the exception 'ImageData is not defined'.
If you think that this conformance test is wrong, I'd be interested in your opinion (I don't really know myself if the spec's interface names such as ImageData are normative).
Reporter | ||
Updated•14 years ago
|
Summary: missing ImageData interface → missing ImageData interface make WebGL conformance test fail
Reporter | ||
Comment 1•14 years ago
|
||
Also, in WebGL, certain functions such as texImage2D are overloaded and the overload resolution requires differentiating between ImageData and HTML elements such as 'img', 'canvas', 'video'.
Currently we have to resort to very clumsy code, basically trying to interprete the passed object as a HTML element first, and if that fails then try to interprete it as a duck-typed ImageData.
Having a proper ImageData type would presumably allow to do that in a cleaner way.
Reporter | ||
Updated•14 years ago
|
Summary: missing ImageData interface make WebGL conformance test fail → missing ImageData interface makes WebGL conformance test fail
Reporter | ||
Updated•14 years ago
|
OS: Linux → All
Hardware: x86_64 → All
![]() |
||
Comment 2•14 years ago
|
||
The interface thing is normative, but postdates our implementation. It's also been somewhat controversial; I'm not sure what the current thinking is on keeping it in the spec.
![]() |
||
Comment 3•14 years ago
|
||
And it's also not obvious whether it should be [NoInterfaceObject] or not, of course. It's not in the spec right now, but that could be oversight.
Comment 4•13 years ago
|
||
This has long been fixed, no?
Reporter | ||
Updated•13 years ago
|
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → DUPLICATE
You need to log in
before you can comment on or make changes to this bug.
Description
•