Oh, I found a good one in the 1.1 spec. Remove page-position=last/only; there is no way to guarentee that it can work. There is no algorithm that can make it work in the general case. Sure, if the last page and the page that would have been there had it not been the last page used the same region content rectangle, then it's fine.
But the user is not so constrained; the user can use any page. If the content is small enough that it fits on the not-last-page version, but is too big for the the last page version (or, in 1.1's case, doesn't even provide a region body that is in this flow's region map), then it can't be the actual last page, as you need another. It's fundamentally impossible in the general case. If they can't be removed, then at least provide some warning to the user about unresolvable cases. And provide some specified behavior from the FO processor in this case (chop off the content, etc). I suppose one could cheat by adding a blank last page ;) But I imagine the user would be pretty upset by that... -- View this message in context: http://www.nabble.com/XSL-FO-2.0-workshop-in-Heidelberg-next-week-tf2408511.html#a6751725 Sent from the FOP - Dev mailing list archive at Nabble.com.