Hello all, Whenever ex unit with Ahem are being used in a testcase, I suggest to use a font-size of 20px and not a font-size of 16px (or 12pt). I suggest to use a font-size which is a factor of 5: eg 20px, 25px, 30px, 35px, etc [Addendum Another idea would be to use value of ex so that when multiplied by 0.8, would result in an integer: 6.25ex, 12.5ex, 18.75ex, 25ex, etc seem to work. ] The reason is that this avoids fractional pixel when calculating a length. http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_17/border-spacing-083.xht http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_17/border-spacing-084.xht Font-size for those 2 testcases have been defined as 12pt, which is 16px. When doing this math, (left offset of the middle bar) 15ex = 15ex * 0.8em/ex * 16px/em some browsers will first calculate 0.8 * 16 = 12.8px and then will round up as 13px before multiplying by 15. So, in both border-spacing-083.xht and border-spacing-084.xht testcases, the expected calculation is 15ex * 0.8em/ex * 16px/em = 192px but in some browsers it will be 15ex * roundUp(0.8em/ex * 16px/em) = 195px This is the case for Firefox 3.6.6 and Konqueror 4.4.4. If you examine carefully border-spacing-083.xht and border-spacing-084.xht testcases with Firefox 3.6.6 or Konqueror 4.4.4 (with or without DOM inspector, DOM treeviewer), you will see that the middle bar overlaps a bit the last bar. My conclusion: all testcases involving ex unit with Ahem should be using a font-size of 20px or make appropriate adjustments to avoid fractional pixel issues. regards, Gérard -- Contributions to the CSS 2.1 test suite: http://www.gtalbot.org/BrowserBugsSection/css21testsuite/ CSS 2.1 test suite (alpha 3; June 15th 2010): http://test.csswg.org/suites/css2.1/20100615/html4/toc.html CSS 2.1 test suite contributors: http://test.csswg.org/source/contributors/Received on Wednesday, 30 June 2010 03:35:32 UTC
This archive was generated by hypermail 2.4.0 : Friday, 20 January 2023 19:58:14 UTC