Test reflection of GPUTexture.textureBindingViewDimension. #4537
Conversation
In core it should always be undefined. In compat it should be what was passed in or the default.
| t.expect(texture.mipLevelCount === (descriptor.mipLevelCount || 1)); | ||
| t.expect(texture.sampleCount === (descriptor.sampleCount || 1)); | ||
| t.expect( | ||
| texture.textureBindingViewDimension === |
There was a problem hiding this comment.
aside: We should also have a separate (idl) test for 'textureBindingViewDimension' in GPUTexture.prototype. It's a really minor interoperability thing, but I think we do want all browsers to implement it
There was a problem hiding this comment.
Do we? At the moment the only thing a core implementation needs to do is add the 4 new limits. They do not need to add this textureBindingViewDimension and have it be undefined. It's undefined by the fact that it doesn't exist. I feel like that's fine. I also feel like apple would push back. From their POV it would be a wart and things would arguably work without adding it. (vs the limits which were required)
There was a problem hiding this comment.
Going to mark this as resolved of that's ok. If we decide the getter is required then we can add an idl test at that time.
2 participants
In core it should always be undefined.
In compat it should be what was passed in or the default.
This passes in a build of chromium that exposes this property
In compatibility mode
When not in compatibility mode
Also checked Firefox stable (as chrome stable has trials on)