From mark at novemberborn.net Mon Nov 3 21:57:17 2008 From: mark at novemberborn.net (Mark Wubben) Date: Mon, 3 Nov 2008 22:57:17 +0100 Subject: [sIFR3-dev] r436 Message-ID: Hey guys, I put r436 out a couple days ago. More here: . Perhaps the most interesting bit is that I changed the font-size detection, specifically for IE. Previously, I temporarily replaced the text inside to-be-replaced-elements by three lines, and then used htmlElement.getClientRects() to get coordinates for each line. The height of the middle line would then be used to as the font size. This required the line-height to be 1em. I've got a few reports of this breaking down in IE, so I started thinking about different ways of calculating font size. Also, the font size never was exactly correct. The new mechanism replaces the text by just one "x", removes padding and border, and sets the line-height (by JavaScript) not to 1em, but 2em. Then, calculating the offsetHeight of the element, and dividing by 2, gives the exact font size! Works in IE8, too. The 2em line- height did the trick for some reason, 1em line-height would still give an incorrect font size. Anyhow, wanted to explain the new mechanism, and request you guys to have a look a this new version. Potentially of course there may be issues, but it seems to be holding up quite well so far. (Although: A rare issue has been reported regarding IE, where unloading the page straight away, as it's loading, may cause IE to throw a JavaScript error. Harmless, but possibly annoying. This has not yet been fixed. Other than that, r436 seems to be holding up quite nicely.) Again, check it out here: . Would love to hear your feedback. Best, -- Mark Wubben +45 53 47 59 42 http://supercollider.dk