Re: fixed new inline-block alignement test (complete review)

Gérard,

I have submitted.

https://hg.csswg.org/test/rev/74cf2d8eb799



> Hajime,
>
> Sorry for the long delay. Here is a preliminary review:
>
> 1-
>
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-003-ref.xht
>
> You need to swap ( 並び替えます ) the yellow and blue images, like this:
>
>
> <div>
>    <img src="support/swatch-orange.png" width="60" height="60" alt="Image
> download support must be enabled" /><br /><!--
> --><img class="left" src="support/swatch-yellow.png" width="120"
> height="120" alt="Image download support must be enabled" /><!--
> --><img class="right" src="support/swatch-blue.png" width="120"
> height="120" alt="Image download support must be enabled" /><br /><!--
> --><img src="support/swatch-orange.png" width="30" height="30" alt="Image
> download support must be enabled" /></div>


I've modified it.


> 2-
>
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-005.xht
>
> line 9: (...) and when 'text-orientation' is 'upright', then (...)
> line 18:   text-orientation: mixed;
>
> -- continuation --
>
>
I've modified it.


> 3-
>
>     span#orange30
>     {
>       display: inline-block;
>
>
>     span#fuchsia30
>     {
>       display: inline-block;
>
>
> In all your tests, you have been declaring 'display: inline-block' onto
> the smallest Ahem glyph (square) which follows the tested inline-block.
> Please explain why. I do not see the need to do this. It does not make your
> tests incorrect or unreliable ... but this declaration is extraneous to me.
> I would remove this.
>
>
I've deleted declaring 'display: inline-block'.

When I created inline-block test,  I thought that all elements must have
declaring 'display: inline-block' because the test case is for
'inline-block'...
However, now, I think that this is no need declaring. So I've deleted it.



> 4-
>
> <img class="left" src="support/swatch-blue.png" width="120" height="120"
> alt="Image download support must be enabled" /><!--
> --><img class="right" src="support/swatch-yellow.png" width="120"
> height="120" alt="Image download support must be enabled" /><br />
>
> Please remove class="left" and class="right" from the reference files
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-002-ref.xht
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-003-ref.xht
> as they are not defined and they are not needed.
>
>
I've deleted.


> 5-
>
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-009.xht
>
> I believe inline-block-alignment-new-009.xht's reference file should be
>
> inline-block-alignment-new-007-ref.xht and not
> inline-block-alignment-new-006-ref.xht
>
> and inline-block-alignment-new-009's pass-fail-conditions should be
>
> <p>Test passes if the <strong>right edge</strong> of an irregular polygon
> is straight and unbroken.</p>
>
> and not left edge.
>
>
> 6-
>
>
> http://test.csswg.org/source/css-writing-modes-3/inline-block-alignment-new-010.xht
>
> I believe inline-block-alignment-new-010.xht's reference file should be
>
> inline-block-alignment-new-006-ref.xht and not
> inline-block-alignment-new-007-ref.xht
>
> and inline-block-alignment-new-010's pass-fail-conditions should be
>
> <p>Test passes if the <strong>left edge</strong> of an irregular polygon
> is straight and unbroken.</p>
>
> and not right edge.


Now I have another thought. Now I think that the reference files must have
four pattern for alphabetical baseline testcase.

- (1) vertical-rl + sideways-right (= vertical-rl + sideways)
- (2) vertical-rl + sideways-left
- (3) vertical-lr + sideways-right
- (4) veritcal-lr + sideways-left (= vertical-lr + sideways)

I've attached the image which represents above all pattern.
In the image, "ABCDEFG" is the first line in inline-block, "HIJKLMN" is the
last line in inline-block.

What do you think about?


Hajime.

-- 
# 塩澤 元 (Shiozawa, Hajime)
# mail: hajime.shiozawa@gmail.com