From: Corby Wilson <corby@ei...> - 2002-06-12 18:03:18
OK, I think I found the problem.
The table attached to the bug is divided into two tables.
The first table has three rows of one cell each.
This is pulled out correctly.
However: The second table (where the shaded area starts) is immediately
after the first table (so in fact it's one table, but treated as two).
There is no paragraph data between the two tables, so wvWare never gets
a chance to reset the ps->intable bit to zero. Instead it goes on to
the next paragraph and skips wvGetFullTableInit which would get the new
table information. This causes a segfault later on when the rows and
cols don't match up to the data in the text area.
How do I detect the end of a table? Or the beginning of one?
If I could detect these I could reset the ps->intable bit.
On Wednesday 12 June 2002 19:58, Corby Wilson wrote:
> How do I detect the end of a table? Or the beginning of one?
> If I could detect these I could reset the ps->intable bit.
Please look into the OpenOffice ww8*.cpp files: they contain the answer
to your question.
It's been a long time since then so I forgot the answer but I remember
there *was* one - after lots of painfull trying... :-D
Cheers and good luck!
Karl-Heinz Zimmer, Senior Software Engineer, Klar=E4lvdalens Datakonsult AB
Get latest updates about Open Source Projects, Conferences and News.