- From: <staffan.mahlen@comhem.se>
- Date: Thu, 02 Oct 2003 17:53:31 +0200
- To: "www-style@w3.org" <www-style@w3.org>
On 1 Oct 2003 at 18:47, fantasai wrote: > Have you looked carefully at CSS3 text? > http://www.w3.org/TR/css3-text/#text-decoration-overview I have to admit i did not. When looking at it, it seems to me it introduces differences to the CSS 2.1 model regarding how for instance the underline is drawn. "If an element contains no text (ignoring white space in elements that have 'all-space-treatment' set to 'collapse'), user agents MUST refrain from rendering text decorations on the element. " This means no underline crossing the inline images if i read it right? Another interesting point: "Names: text-underline-mode, text-line-through-mode, text-overline- mode Value: continuous | skip-white-space " How is that to work if an ancestor of the text is in fact drawing the line? It sounds to me like every element having text decoration and have descendant text may need to know where every space in every descendant text is? On 1 Oct 2003 at 22:59, Jukka K. Korpela wrote: > On Wed, 1 Oct 2003 staffan.mahlen@comhem.se wrote: > > Could you elaborate on when the common browser behavior is not > > suitable? > > When the "decorated" text contains superscripts, and especially when it > contains subscripts. The underlining is disturbingly broken. If the only real issue is the sub/superscripts then the choise to make the property: Inherited: yes Applies to: the text contents of all elements that have 'vertical- align' other than 'sub' or 'super'. seems like the obvious "minimal victory". This would cover the needs of authors and seems simple enough to implement. CSS3 can still apply all the new properties with the above restriction if i understand correctly. Even if the above is for some other reason not good enough, the models used in CSS 2.1 and CSS3 should be as similar as possible. /Staffan
Received on Thursday, 2 October 2003 11:53:21 UTC