- From: Jan Roland Eriksson <rex@css.nu>
- Date: Sun, 21 Feb 1999 06:52:40 -0500 (EST)
- To: "L. David Baron" <dbaron@fas.harvard.edu>
- Cc: "WWW-style-list" <www-style@w3.org>
On Sat, 20 Feb 1999 19:35:17 -0500 (EST), you wrote: >In the definition of line-height [1], the spec defines <number> values >to depend on the "element's font size" and <percentage> values to >depend on the "element's computed font size". This difference was >recently pointed out to me, and I'm wondering why it exists. >Furthermore, I also wonder if, instead of computed font size, actual >font size was meant. These terms are explained in the definition of >font-size [2]: > > The actual value of [font-size] may differ from the computed value > due a numerical value on 'font-size-adjust' and the unavailability of > certain font sizes. This passage in it self qualifies for a rewrite. To me it's not even clear if it is the "actual value" or the "computed value" that changes due to... "a numerical value on 'font-size-adjust' and the unavailability of certain font sizes" >It doesn't make sense to me that if a font was specified as 10px but only >available as 16px, then a line-height of 1.2 (or 120%) should result in >a line-height of 12px. My interpretation of this example would be that... line-height: 1.2 should result in 12px and, line-height: 120% would become ~19px but that's my view based on what I think I can understand from that previous passage. >I think both these statements should be changed so that line-height >depends on the actual font-size, not the computed font-size. (Even if >it should be computed, I definitely think they should be the same, >unless there was some reason I don't see.) I can second that, with a slight emphasis on going for the "computed value" (finally rendered value, as I read that) of font-size. It seems to me as a UA that finds reason to change a font-size due to suggested size not being available, should also use that new font-size as the base for calculating line-height. The alternative seems to lead to a possibility for overlapping lines of text. -- Jan Roland Eriksson <rex@css.nu> .. <URL:http://css.nu/>
Received on Sunday, 21 February 1999 20:10:38 UTC