A log message is not good enough. That needs to throw an exception. It's a bug if IPD and BPD are not set IMO. Would you write test cases for all the possible combinations after adding the exception and before fixing the problems? I can help you there, if you like.
On 25.08.2005 08:40:22 Manuel Mall wrote: > The safety check in addBackground is already there. This is how I > stumbled across it as it is triggered by one of the layout engine > tests. > > I'll look into it as part of the whole percentage stuff I'm currently > doing. > > On Thu, 25 Aug 2005 02:35 pm, Jeremias Maerki wrote: > > You are right. It seems like some calls to > > TraitSetter.addBackground() are issued before IPD and BPD of the area > > are set (list-block and list-item, for example). Yes, the call will > > need to be deferred until the BPD and IPD have been set on the area. > > A safety check in addBackground() will be a very good idea, too. You > > or me? :-) > > > > On 25.08.2005 05:14:04 Manuel Mall wrote: > > > When setting a relative background position the positioning is > > > relative to the size of the area the background is applied to. > > > Currently the position calculation is done when the area is > > > created, i.e. when the background trait is set. However, at that > > > point in time fop may not know the bpd and ipd of the area in > > > question. Therefore the calculated positioning will be wrong. Am I > > > correct in saying that the logic needs to be changed to do that > > > calculation (or even set the background trait) when the layout is > > > completed for that area and not when the area is created in the > > > layout process? > > > > Jeremias Maerki > Manuel Jeremias Maerki