Are the different values, particularly the 'zeroes' for b1..b3 a 'browser issue' or a mootools issue?
http://jsfiddle.net/bBAe8/6/ As the fiddle was written while testing in Safari5, the 'expect' is equal to Safari5 / Chrome output due to copy&paste. Note the 0/0 entries in there for the B1..B3 <div>s: seems Safari+Chrome deliver w/h ~ 0 when the element does not have an explicit width or height CSS style attribute. Is this 'intended behaviour'? In reality, I did not 'expect' those 0/0 values in there, so is b1..b3 reporting width/height ~ 0 a webkit issue, which I should recover from in application level code? Then test on Opera 11.10 (which was really why I did this, having trouble with some milkbox internals on Opera delivering w/h == 0/0 output in Opera only; in this fiddle though, Opera does something else but still doing its own thing with the D2..D4 items, which have their CSS w/h set to 20px, but Opera reports 14px, which is ... weird. Opera scores 12 out of 12 different! Compare that to output from FF4: 3 out of 12 differ: apparently FF4 doesn't need a width/height CSS entry at all times; seems okay to me. IE9 of course is of a slightly different opinion (9 out of 12 differ), but it's all 'off by a bit' really when compared with the FF4 results and I have learned to expect IE to be different, so not really a surprise. Or should I be? It's the webkits and Opera that bother me: webkits delivering zeroes and Opera doing something magical to my D2..D4 div size reports (14 instead of 20) :-S (I have yet to reproduce the 0/0 failure of getComputedSize() as it happens in milkbox+Opera11 in a fiddle; that one will happen as soon as I get it occurring in fiddle isolation.) -- Met vriendelijke groeten / Best regards, Ger Hobbelt -------------------------------------------------- web: http://www.hobbelt.com/ http://www.hebbut.net/ mail: [email protected] mobile: +31-6-11 120 978 --------------------------------------------------
