Re: min-max

Ian Hickson wrote:
> On Sun, 15 Feb 2004, fantasai wrote:
> 
>>Sorry it took so long.
>>
>>http://fantasai.inkedblade.net/style/specs/css2.1/tests/min-max-replaced
>>
>>CSS2.1, in its current state, will not pass test 10 or similiar.
> 
> I don't think that test is valid. You are assuming that the 'height' and
> 'width' attributes set the intrinsic width and height of the image.
> However, they are simply being mapped to the 'width' and 'height'
> properties, as per section 6.4.4. However, assuming your images match the
> attributes you've given, just removing the attributes should solve that.

ok. I moved them into comments, like you did.
(I did that more to help me keep track of what was going on than
anything else.)

> Here is a fixed version:
> 
>    http://www.hixie.ch/tests/adhoc/css/box/replaced/intrinsic/001.html
> 
> However, I don't understand why you prefer case 10 to come out as 75x75
> rather than 150x75 (which is what the spec says now, and is what Gecko
> implements).

Gecko's implementation has problems. It's violating constraints in
tests 7 and 9 (among others), and it's not scaling per CSS2.1 in,
for example, test 13 (which should trigger rule 4).

> Both are skewed, why is your skew better than the already
> implemented skew? Just because yours is "less" skewed?

The rules in that section are designed to minimize skew. That was your
decision as the CSS WG. My skew is indeed better because it is less skewed
*because* that is the apparent goal of that set of rules.

~fantasai

Received on Tuesday, 17 February 2004 13:11:05 UTC