You can see at http://wiki.apache.org/xmlgraphics-fop/CollapsingBorderModel to what conclusion I came.
I guess it's open for interpretation. In this scenario, we're not talking about rectangles anymore but the spec defines everything in rectangles. I wouldn't worry so much about that case because if something like that happens, it's bad/ugly table design IMO. The spec also doesn't say how exactly to paint the collapsed borders of different widths. I wouldn't waste time on such exotic setups. On 29.06.2007 11:30:59 Vincent Hennebert wrote: > Hi all, > > Another question regarding backgrounds. See the attached FO file and the > resulting PDF. Cell 1.1 is spanning over 2 rows. Its border-right is > composed of 2 segments, one thicker and one thinner. For the latter, the > gap is currently filled with the table's background. > > Question is: shouldn't that rather be the cell's background? Do we > consider that this gap should correspond to additional padding for the > cell, or to some transparent space showing the row's or the table's > background? > > The first answer seems the most logical to me, but that means that if > the background is an image, it may be cropped by the other borders. Is > everyone ok with that? > > Thanks, > Vincent Jeremias Maerki
