
Ref tests fail on XB PointStream
Reported by Andor Salga | April 18th, 2011 @ 11:49 PM | in 1.0
Not sure if it's something I'm doing that's wrong, but all my cases are failing. I'm uploading the zip of just the test cases along with a screenshot.
You can download the library from github: https://github.com/asalga/XB-PointStream/tree/staged
In the library there is a tools/ directory which has an older version of sundae you can replace with your 0.9 and plug in the tests.
Comments and changes to this ticket
-
cwdesautels April 19th, 2011 @ 02:15 PM
Upon first looking at the tests.json, I notice that both Acorn test 1 & 2 call the same function for the second canvas.
Script loading is not sandboxed they are loaded on global scope, so what I think is happening is that the functions are stepping on each other. I'm not controlling when the resources scripts get loaded or in what order the tests finish loading.
Let me confirm this
-
cwdesautels April 19th, 2011 @ 04:41 PM
- no changes were found...
-
cwdesautels April 19th, 2011 @ 04:45 PM
After hacking away a bit, I updated my the tester in 'staged'. I appended "_debug" to your tests presented here.
I noticed a few things:
*In the screen shot you'll see that something is coming out inverted when I'm grabbing the pixels, the image or XBS
*What are these warnings being thrown? I feel like they could be affecting my pixel manipulation but I've never seen them before.
-
Andor Salga April 19th, 2011 @ 05:08 PM
Ah, this is the same reason why I had to write that patch to invert the pixels the first time I tried using Sundae. But why does it only show with MY tests I wonder?
If I get time later this evening I'll see if there are any errors/warnings.
-
cwdesautels April 19th, 2011 @ 05:29 PM
Warnings are in the screenshot, in both firefox 4 and minefield 6
Could it be left hand system vs right hand system when getting pixels from a 3D vs a 2D context?
I realized that my webGL tests are not complex enough to even see an inversion like that. Where as my 2D context tests are complex enough and I know my 2D context get pixels are correct.
I'll cook up a few tests to confirm
-
-
cwdesautels April 21st, 2011 @ 04:21 PM
- State changed from new to resolved
(from [504585486a43d0ec892449e47a3616455ec65909]) [#32 state:resolved][#33 state:resolved][#31 state:resolved] https://github.com/cwdesautels/sundae.js/commit/504585486a43d0ec892...
Please Sign in or create a free account to add a new ticket.
With your very own profile, you can contribute to projects, track your activity, watch tickets, receive and update tickets through your email and much more.
Create your profile
Help contribute to this project by taking a few moments to create your personal profile. Create your profile ยป
sundae.js is a reference testing framework for canvases designed in Javascript.
<br/>
Source:
https://github.com/cwdesautels/sundae.js
Blog:
http://cwdesautels.blogspot.com/
People watching this ticket
Attachments
Referenced by
-
31 Ref tests fail on XB PointStream (from [ccd699a050eecd85f929c675ff52d7ac63b20660]) WIP [#3...
-
33 Blur artifacting (from [504585486a43d0ec892449e47a3616455ec65909]) [#32 st...
-
31 Ref tests fail on XB PointStream (from [504585486a43d0ec892449e47a3616455ec65909]) [#32 st...
-
32 3D canvas (from [504585486a43d0ec892449e47a3616455ec65909]) [#32 st...