W3C home > Mailing lists > Public > whatwg@whatwg.org > April 2009

[whatwg] Canvas tests that fail across the board

From: Ian Hickson <ian@hixie.ch>
Date: Thu, 30 Apr 2009 21:59:38 +0000 (UTC)
Message-ID: <Pine.LNX.4.62.0904302149360.18851@hixie.dreamhostps.com>
On Thu, 3 Jul 2008, Simon Pieters wrote:
>
> Sorting Philip's canvas results table by failure score:
> 
>    http://simon.html5.org/dump/canvas-results-sorted.html
> 
> ...and quickly looking through the top ones, I make the following notes:
> 
> 
> 2d.transformation.scale.nonfinite
> 2d.transformation.rotate.nonfinite
> 2d.transformation.translate.nonfinite
> 2d.transformation.setTransform.nonfinite										
> Opera, Firefox and Konqueror throw an exception. Perhaps the spec should
> change to throw NOT_SUPPORTED_ERR instead of ignore?
> 2d.drawImage.nonfinite										
> All throw (see above on *.nonfinite).

I'd raher keep the nonfinite cases consistent throughout.


> 2d.fillStyle.parse.current.basic
> 2d.fillStyle.parse.current.changed
> 2d.fillStyle.parse.current.removed										
> Opera, Firefox and Konqueror seem to agree...

They seem to just not support currentColor.


> 2d.path.stroke.prune.arc										
> We seem to have interop here. Perhaps the spec should change?

This would make it inconsistent with the other pruning tests, some of 
which pass. Are we sure we want to change this?


> 2d.drawImage.negativesource
> 2d.drawImage.negativedest
> 2d.drawImage.outsidesource										
> All throw an exception. Perhaps the spec should change to throw
> INDEX_SIZE_ERR?

This now passes in at least one browser, so I haven't touched it.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 30 April 2009 14:59:38 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 16:59:11 UTC