43184 2010-07-29 02:35:49 -0700 WebKitCSSMatrix behaves differently on iOS 4.0 from Safari 5.0 2010-08-03 11:10:09 -0700 1 1 1 Unclassified WebKit CSS 528+ (Nightly build) All All RESOLVED INVALID InRadar P2 Normal --- 1 webkit webkit-unassigned cmarrin ddkilzer webkit oldest_to_newest 257597 0 webkit 2010-07-29 02:35:49 -0700 There is a bug in the JavaScript WebKitCSSMatrix object; the following example explains: var A = new WebKitCSSMatrix(); var B = new WebKitCSSMatrix(); var C = A.multiply(B); Result on Mobile Safari: C = B.A Result on Desktop Safar: C = A.B Notice that B.A is calculated different than A.B, and therefore the result is not likely to be the same! 258851 1 ap 2010-08-02 07:14:54 -0700 I think that this was an intentional change, see bug 38337. Apparently, this change hasn't been picked by iPhone version of WebKit yet - but it's not appropriate to discuss iPhone WebKit here, since it doesn't build from webkit.org sources. > Notice that B.A is calculated different than A.B, and therefore the result is not likely to be the same! One can multiply two known matrices to see what version of multiply() is implemented. 259557 2 ddkilzer 2010-08-03 11:08:43 -0700 <rdar://problem/7781582> 259559 3 ddkilzer 2010-08-03 11:10:09 -0700 Marking as RESOLVED/INVALID since we're tracking merging the fix for Bug38337 as part of <rdar://problem/7781582> for iOS. Thanks!