only in IE8 (Flash 10.1.102.64, SVG Web r1309) the following doesn't work
rightRect.style.display='block';
see reduced testcase here
http://vis.uell.net/gsvg/displayBlock_IE8.html
Compatibility View, IE7 and native renderers all work fine.
Comment #1
Posted on Dec 31, 2010 by Grumpy DogFixed in r1310. Tests in r1311.
This was an unreleased regression from Issue 573/r1273 and therefore will not be included in the release notes.
Comment #2
Posted on Jan 2, 2011 by Happy WombatThis is not yet working for me. I've updated the above testcase to r1317 but the problem persists in IE8 (and IE8 only)
http://vis.uell.net/gsvg/displayBlock_IE8.html
Flash 10.1.102.64 IE 8.0.6001.18702 (WinXP) IE 8.0.7600.16385 (Win7)
Comment #3
Posted on Jan 2, 2011 by Grumpy DogFixed in r1318.
Issue 598; Reduce compression of svg.js because this is causing a failure of the test file in Issue 598. Even --compiliation_level SIMPLE_OPTIMIZATIONS is failing. This increases compressed size by over 30K. Full compression will be restored if the problem with the Google closure compiler can be determined and resolved.
I am going to have to debug this failure in compressed mode and hopefully figure out how the compression process caused this new(?) failure. :(
Comment #4
Posted on Jan 2, 2011 by Grumpy DogCompression restored and compression friendly workaround implemented in r1319.
Comment #5
Posted on Jan 2, 2011 by Grumpy DogUpdated compiler and fixes for warnings in r1320.
Comment #6
Posted on Jan 2, 2011 by Happy WombatUps, I was of the opinion that the Google closure compiler was non-destructive. Sorry that I always tested with the closure compiled version.
Does comment #5 mean that the Google closure compiler at http://closure-compiler.appspot.com should not be used but one should use the Java version locally that you have improved?
Comment #7
Posted on Jan 3, 2011 by Grumpy DogThe update to the compiler was just one I obtained by taking a fresh download and I think it is the same one you are pointing to. I decided to check it in because it seemed to work to I figured it was a good idea to use the latest version. It did not fix any specific problem that I know of.
Comment #8
Posted on May 16, 2011 by Massive HippoBug doesn't seems to be fixed IMO. Just by checking the given url with IE8 (IETester in IE8 mode in fact), bugs.
Comment #9
Posted on Aug 9, 2011 by Grumpy OxI found the issue! It seems that in compatibility view, event if the page said that it IS compatible with IE8 (by adding a meta tag), the UA string is always MSIE 7.0 but documentMode is 8. If you change the only test of isIE >= 8 to isIE >= 7, it works, so I guess that when isIE and documentMode == 8, isIE should be set to 8.
Status: Fixed
Labels:
Type-Defect
Priority-Critical
regression