- From: fantasai <fantasai.lists@inkedblade.net>
- Date: Mon, 22 Nov 2010 12:09:20 -0800
- To: css21testsuite@gtalbot.org
- CC: Arron Eicholz <Arron.Eicholz@microsoft.com>, "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
On 11/04/2010 08:21 AM, "Gérard Talbot" wrote: > > http://test.csswg.org/suites/css2.1/20101027/html4/numbers-units-008.htm > > http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_4/numbers-units-008.htm > > > Same reasoning as for numbers-units-014. If there is no specified, > normative scaling factor for x-large or larger (relative to the 'medium' > font size) or for adjacent indexes [1], then the used height of 1em of a > x-large font-size can not be predicted. Therefore this can not be > testcase-ed. > > CSS1 was specifying this: "On a computer screen a scaling factor of 1.5 > is suggested between adjacent indexes" [2] > > > Anyway.. the testcase compares a 1em wide by 1em tall box with a 1em > black "X" in ahem font: it's the exact same thing regardless of how > x-large is scaled, rendered in user agents. The testcase, as coded, can > not fail. If the scaling factor was specified by the spec and if it was > - say, for/from medium to x-large - 1.5 (150%), then the testcase would > still require adjustements. > > Again, numbers-units-014 and numbers-units-008 are not testing what they > were originally trying to test. It is not possible to test larger and > x-large if the scaling factor is unknown and unspecified by CSS 2.1 > spec. AFAICT from the assert, the test is trying to test that em sizing is supported even when the font-size is given as a keyword. In that respect, the test is correct. It might be easier to use if it were written as a red/green test, but I don't see how it's wrong. ~fantasai
Received on Monday, 22 November 2010 20:10:01 UTC