Luca, I was intrigued by your idea though I see difficulties determining the right intra-page/column breaks for situations where the line height are not uniform and possibly have penalties that could interfere with later break decision already made. At least, mixed single-/multicolumn situations should be quite easy to handle since you could simply copy multiple footnote sections (one per span group) together. The current design should already help with that. Anyway, I'll defer this problem for later, since there are more important things to deal with. After all, we need to get a more or less usable FOP version out ASAP. I've documented my findings and the current status briefly on the Wiki:
http://wiki.apache.org/xmlgraphics-fop/MultiColumnLayout On 28.06.2005 18:45:06 Luca Furini wrote: > Jeremias Maerki wrote: > > > I just hope this breaking-up of the > > span-reference-areas won't cause too many problems with keeping the > > footnotes working. I'm a bit worried about that. We'll see. > > I was thinking about this too: the texbook, speaking about footnotes and > columns, says something interesting: paginating a page of height H, N > columns ans footnotes having the whole width is quite similar to > paginating a page of height N*H, with a single column, where the height of > each footnote line has a magnification factor = N. > > I think that this way the column break could be seen as "moving indexes", > dividing the normal content in pieces of equal length, that could be > updated while adding some more normal content. (this could mean that > having balanced columns could be even easier than having unbalanced ones!) > > These are just some thoughts, as I did not try and implement anything yet > ... > > Regards > Luca Jeremias Maerki
