Comments: 1) It seems that __setCanvasAttributeQ is initialized for all runtimes, so I don't understand why the swf branch needs to test for it not?
2) If we really feel we need to save one array, I find it clearer to not optimize the case of pushing the first element. I know Adam did, but I think it makes the code harder to read. QA Report: I verified that the test case bombs in IE8/Win before the change, and runs successfully after. Approved for QA. On 2010-07-19, at 17:04, Max Carlson wrote: > Change 20100719-maxcarlson-Q by maxcarl...@friendly on 2010-07-19 13:17:33 PDT > in /Users/maxcarlson/openlaszlo/trunk-clean > for http://svn.openlaszlo.org/openlaszlo/trunk > > Summary: Add setCanvasAttribute() call queing for ls.embed.dhtml() > > Bugs Fixed: LPP-9207 - IE8 dhtml: webtop failing to start in IE 8 > > Technical Reviewer: hminsky > QA Reviewer: ptw > > Details: Add call queuing to lz.embed.dhtml() so early calls don't fail > before the canvas is created. > > Tests: See LPP-9207 > > Files: > M lps/includes/source/embednew.js > > Changeset: > http://svn.openlaszlo.org/openlaszlo/patches/20100719-maxcarlson-Q.tar
