Hi Andreas, Andreas L Delmelle a écrit : > On Mar 26, 2007, at 16:48, Vincent Hennebert wrote: > > Hi Vincent, > >> I'm having a lot of fun playing with border and padding >> conditionalities... > <snip /> >> I think the padding should actually not be discarded on the second page. >> It would if the corresponding edge were a leading edge in the >> region-body's reference area. However, it is not because there is no >> stacking constraint between the inner block's area and the (region >> body's) ref-area. Because the retained border on that inner block >> creates a fence which prevents rule 3.b (XSL 1.1, 4.2.5) to apply. > > Just a sanity check: aren't you confusing padding-* with space-* here?
No :-) > Comparing FO to PDF, my common sense jumps up and says: > "Why do you expect the padding to be retained at page-breaks, when > specifying .conditionality='discard'?" Precisely because of this notion of "fence". Actually it would even make sense to me that if the border-before is "retained", then the padding-before should also become retained. But it appears that the spec does not require that. In the case I've presented, the fact that the enclosing block has a retained border-before prevents the before-edge of the child block-area from being a leading edge in the page's ref-area. It results that the conditionality of its border and padding is not applicable. Follow me? ;-) I would be grateful if someone could confirm my thoughts. I'm wondering whether it's an intended behavior or an oversight resulting from the definition of leading edge. BTW, note the difference between "leading edge" (XSL-FO 1.1, 4.2.5) and "leading area" (XSL-FO 1.1, 4.8). The former applies to border- and padding-conditionality for non-first/last areas, the latter to rules for breaks and keeps. While there may be some relation between the two notions, this is generally not the case. As I was making the confusion until recently, I thought it might be worth telling it. ;-) Cheers, Vincent
