RE: [CSS3 print test suite] page-break-inside test inconsistency

Hello Ludger,

Yes, thanks, the test author in this case inadvertently ignored the suggested UA default for 'ol'.  (Guilty as sin, your honor.)

I've modified the test at http://www.w3.org/Style/CSS/Test/Print/current/t130301-pg-brki-00-b.htm to set 'page-break-before' for 'ol' to 'auto'.

So far as I can see, the CSS authoring guidelines don't cover this topic.  I'll bring it up with the WG to be sure others agree with this approach.

Thanks again for pointing out the problem.

Best wishes,

Melinda

> -----Original Message-----
> From: www-style-request@w3.org 
> [mailto:www-style-request@w3.org] On Behalf Of Ludger Buenger
> Sent: Thursday, April 12, 2007 3:58 AM
> To: www-style@w3.org
> Subject: [CSS3 print test suite] page-break-inside test inconsistency
> 
> 
> Dear CSS fellows.
> 
> I believe I found an inconsisteny in the way how the 
> following test suite is supposed to work:
> 
> http://www.w3.org/Style/CSS/Test/Print/20070131/t130301-pg-brk
> i-00-b.htm
> 
> This test suite implicitly assumes that there is an allowed 
> page break opportunity between the box with the green border 
> and the following ordered list. The result is supposed to 
> look like the attached pdf "assumed-result.pdf" with a page 
> break between the green bordered box and the ordered list.
> 
> However depending upon the user agent stylesheet this might 
> not be the case.
> 
> Using solely the HTML 4 default stylesheet supplied with CSS 
> 2.1 in http://www.w3.org/TR/CSS21/sample.html (even though 
> this section is informative, not normative) this implicit 
> assumption does not hold - see attached pdf 
> "result-with-css-default-stylesheet.pdf".
> 
> Reason for this page break behaviour is the following style 
> rule in the mentioned "informative default stylesheet for HTML 4":
> 
> @media print {
>   /* ... */
>   ul, ol, dl    { page-break-before: avoid }
> }
> 
> Now I tried to look through the HTML specs defining how page 
> breaks are supposed to be handled and to find some evidence 
> either to justify or de-ligitimate the rule from the HTML 4 
> default stylesheet in the CSS standard prohibiting page 
> breaks before lists.
> But I got utterly confused since it appears that page 
> breaking behaviour in HTML seems not to be well defined (or 
> at least I did not manage to figure it out).
> 
> So which behaviour is now right?
> Is the sample rule in the suggested default stylesheet wrong 
> and should be ignored (since it is not normative)?
> Or is the rule -though not normative- justified and the test 
> case should be adjusted to allow an explicit page break before lists?
> 
> best greetings,
> 
> Ludger
> --
> Ludger Bünger
> 
> RealObjects GmbH
>

Received on Wednesday, 18 April 2007 00:03:34 UTC