Re: fixed new inline-block alignement test

Le 2015-07-05 02:25, 塩澤 元 a écrit :
> Gérard,
> I have fixed the inline-block alignement test in reference to your
> review^[1]
> Could you review it?
> After your review and approval, I will create other variation of
> inline-block alignment.
> [1]:
> Hajime.

Here is what I came up with:

Your test:

1- (line 15 in your test)
It's always safer to use a numerical line-height (1) instead of a 
font-size (1em) because computed line-height is inherited by default; a 
numerical line-height will scale with relevant font-size. In your test, 
1em was okay since other inline boxes were taller.

2- (lines 24 and 25 in your test)
Creating an asymetrical vertical padding on inline boxes can better 
reveal an implementation bug.
I've added a /* comment */ explaining the purpose of such logical 
vertical padding

3- (line 26 in your test)
I've removed color: fuchsia.

4- (line 33 in your test)
Since line-height is inherited, then you do not need to redeclare it for 
its descendants.

5- (lines 36 and 41 in your test)
I've used id instead of classes for first-line-box and last-line-box

6- (lines 49 and 50 in your test)
Asymetrical vertical padding on that inline plus a /* comment */

If you now load that test into the latest most recent Firefox 42 nightly 
build, you can see 2 bugs occuring. The left padding and right padding 
on the inline boxes should not affect baseline alignment of text (the 
horizontal position of those orange squares with respect to the blue 
square) on the dominant baseline... whatever such dominant baseline is 
and however how baseline-alignment is implemented.

Adapted reference file:

Test Format Guidelines

Test Style Guidelines

Test Templates

CSS Naming Guidelines

Test Review Checklist

CSS Metadata

Received on Sunday, 5 July 2015 19:30:56 UTC