Re: CSS3-writing-modes text-orientation testsuite

(2013/02/21 15:22), "Gérard Talbot" wrote:
> Le Dim 17 février 2013 11:35, Kazuaki Takemura a écrit :
>> Hello.
>>
> Hello Kazuaki :)
>
>> You must be very busy because of the increasing number of committers
>> from Japan.
>
> I should be and will be in vacation in a few days. I'll notify all the
> committers from Japan.
Enjoy your vacation !

>
>> But, I have submitted the property of text-orientation.
>> http://test.csswg.org/shepherd/search/testcase/spec/css3-writing-modes/status/submitted/
>> I appreciate it very much if you could review it.
>>
>> Reviewing the site(http://http://wiki.csswg.org/test/format),
>> I did my best to describe it as accurately as possible, but if you find
>> any errors, please let me know.
>>
>> This property may be the same one as described in Mr. taka$B!G(Bs mail
>> (http://lists.w3.org/Archives/Public/public-css-testsuite/2013Feb/0039.html),
>>
>> but I submitted it anyway (since there are many kinds of properties in
>> Japan).
>>
>> It was very difficult to explain this property in words with as little
>> image as possible.
>> It was especially hard to explain rotations and positional relationship.
> I understand. It is difficult. One way is to use an image as reference
> or use CSS transforms rotate feature as reference.
>
> Example given of rotated text (45 degrees)
> http://test.csswg.org/suites/css3-transforms/nightly-unstable/html/transform-rotate-001.htm
OK.  I try to use CSS transforms rotate,but if it's difficult, I will 
use image as reference.


> This text
> "
> for reference:
> 		horizonal-only:(Includes: Latin, Arabic, Hebrew, Devanagari)
> 		vertical-only:(Includes: Mongolian, Phags Pa)
> 		bi-orientational:(Includes: Han, Hangul, Japanese Kana)
> "
> should NOT be read by testers. You can leave it as a <!-- comment --> in
> the test though.
OK. I will comment out it.

>
>> I tried to describe it as simply as possible to keep it short.
>> But, if you find it is difficult to understand or still too long, please
>> do not hesitate to point it out.
> I have been working on this.
>
> One way to make the pass/fail conditions sentence short and simple to
> understand is to create a test and a reference (image or SVG or
> something else) and then simply say:
>
> "Test passes if there are 2 <strong>identical</strong> [shape
> descriptor]s."
>
> where shape descriptor could be rectangle, stripe, bar, line, grid or
> even "Text sample", etc.. In this way, you avoid having to describe in
> great details the 2 shapes or text.
>
> That way, you can also reuse later the reference in the reftest.
I've got it.

>> By the way, I have tried several browsers and found that it works with
>> Safari and Chrome, though not perfectly.
> Yes. I agree.
>
>> Other browsers do not seem compatible with this property.
>>
>> And @font-face relationship is commented out.
>>
>>
>> Regards.
>> Takemura
>
> Here's what I've been working on:
>
> http://www.gtalbot.org/BrowserBugsSection/review/text-orientation-mixed-001-review.xht
>
> http://www.gtalbot.org/BrowserBugsSection/review/text-orientation-mixed-001-review-ref.xht
>
> http://www.gtalbot.org/BrowserBugsSection/review/text-orientation-mixed-123.xht
>
> http://www.gtalbot.org/BrowserBugsSection/review/text-orientation-mixed-123-ref.xht
>
> Another idea would be to use CSS transform rotate feature on <div
> id="reference">:
>
> "
>   rotate(<angle>)
>      specifies a 2D rotation by the angle specified in the parameter
> about the origin of the element, as defined by the
> 'transform-origin' property. For example, 'rotate(90deg)' would
> cause elements to appear rotated one-quarter of a turn in the
> clockwise direction.
> "
> coming from
> 15.1. 2D Transform Functions
> http://www.w3.org/TR/css3-transforms/#two-d-transform-functions
>
> as the reference. That way, no image to create.
>
> -----------
>
> <meta name="flags" content="font should" />
>
> You can remove the "should" flag from
> text-orientation-mixed-001.xht
> and
> text-orientation-sideways-001.xht
>
> "should" flag means that the property value or feature is recommended
> and not required by the spec. Here, text-orientation values are
> required, must be supported.
I've got it.

> More later.
>
> Gérard

Regards.
Takemura