Re: [css-writing-modes-3] text-orientation-mixed-001 , text-orientation-010 , text-orientation-014 tests

2015-12-08 1:59 GMT+09:00 Gérard Talbot <css21testsuite@gtalbot.org>:

> Le 2015-12-07 10:26, Koji Ishii a écrit :
>
> [snipped]
>
> http://test.csswg.org/source/css-writing-modes-3/text-orientation-014.xht
>>>
>>> The second (bottom) part of the test is wrong, incorrect. The initial,
>>> default value of 'text-combine-upright' is 'none'; the initial, default
>>> value of 'text-orientation' is 'mixed'. So, 'sideways' text and 'mixed'
>>> text are compared.
>>>
>>>
>> This test is not good, but does not look wrong to me.
>>
>
>
> With Firefox 45.0a1 nightly, try
>
>
> http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/text-orientation-mixed-vs-sideways-002.xht
>
> where a) and b) correspond to the code involved in
> text-orientation-014.xht . Right now, I am not sure how should a
> sideways-ed latin text be rendered inside a line box whose dominant
> baseline is central.
>

Gecko looks incorrect to me. text-orientation should not change the line
height. Slight differences in baseline position can be possible, but not
line height.

text-orientation-014.xht also has another issue: the font used is not
> specified. So, the default system font on the tester's os is involved...
> while I might have another and different default system font for my os.
>

True, from that perspective, it's not really a good test. But the test
method being not good and incorrect test are two different things.

In this test
>
>
> http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/text-orientation-014-mplus-1p-GT.xht
>
> we "fix" (force) the font used. Chrome 49 Canary and Firefox 45 differ...
> and I think (albeit not sure) Firefox is closer to the correct rendering.
>

As mentioned above, varying line height isn't expected.

Koji, I've sent you another email on
> text-orientation-mixed-vs-sideways-002.xht
>

Sounds like I missed...will check.

BTW, do you have a github account? github issue tracker[1] helps unanswered
questions, if you can be used to it. If not, I'll try my best not to miss.

[1] https://github.com/w3c/csswg-test/issues

/koji