Problem with fo:marker and keep-with-next in tables

["Müller, Markus" <ma...@sdm.de>]

Hi *,

We want to list some items on an invoice using a table. It is possible, that
not all items fit into a single page. So we allow following pages, but then
we have to print a carry over at the end and the beginning of each page. The
computation is not the problem, we do this by adding appropriate attributes
to the items' corresponding xml-elements. The carry over values then are
placed into markers. Additionaly, always 3 items (= fo:table-rows) are set
to keep together using keep-with-next="always". Doing so, retrieving the
markers fails (first-starting and last-ending). FOP returns a marker NEAR
the correct marker, sometimes even the correct one. This feature seams to be
nondeterministic.

Jochen Unger asked the same question a few weeks ago, but there were no
answers in the mailing list. 

Is there a solution for this problem???

BTW, it seems that the markers' is_first / is_last attributes are set BEFORE
the keep-xxx directives are processed.

Markus	

---------------------------------------------------------------------
To unsubscribe, e-mail: fop-user-unsubscribe@xml.apache.org
For additional commands, e-mail: fop-user-help@xml.apache.org

Re: Problem with fo:marker and keep-with-next in tables

["J.Pietschmann" <j3...@yahoo.de>]

Müller, Markus wrote:
> We want to list some items on an invoice using a table. It is possible, that
> not all items fit into a single page. So we allow following pages, but then
> we have to print a carry over at the end and the beginning of each page. The
> computation is not the problem, we do this by adding appropriate attributes
> to the items' corresponding xml-elements. The carry over values then are
> placed into markers. Additionaly, always 3 items (= fo:table-rows) are set
> to keep together using keep-with-next="always". Doing so, retrieving the
> markers fails (first-starting and last-ending). FOP returns a marker NEAR
> the correct marker, sometimes even the correct one. This feature seams to be
> nondeterministic.

This is quite deterministic. The problem is that there is an attempt
at laying out some rows, then a conflict with the keep condition is
detected and the layout is rolled back, almost. This particular
problem with not resetting markers during rollback should be fixed
in CVS (see http://nagoya.apache.org/bugzilla/show_bug.cgi?id=16999)
Other similar problems remain, for example with footnotes.

> BTW, it seems that the markers' is_first / is_last attributes are set BEFORE
> the keep-xxx directives are processed.
I think the flags are reset during the rollback. If not, well, yet
another bug.

J.Pietschmann


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-user-unsubscribe@xml.apache.org
For additional commands, e-mail: fop-user-help@xml.apache.org