Comment 82 for bug 457220

BTW,(In reply to comment #8)
> Kohei, could you be a bit more specific (for those of us who are not Chris ;-)
> ) about what you mean with the latest build...?
> I cloned the git repository and switched to the 3.1.1 branch following this
> description: http://www.freedesktop.org/wiki/Software/ooo-build/GettingIt
> It seems that this gives exactly the same build number as the one packaged by
> ubuntu (ooo310m19, build: 9420).

Yup, that's what I mean by the 'latest build'. When someone says latest build, that normally means the latest development build, fresh from the git repository.

> After configure (--with-distro=UbuntuKarmic --disable-kde --disable-kde4 ),
> download and make I'm ending up with an Openoffice that shows the initially
> reported bug, i.e. crashes.

So, this eliminates the possibility that the problem was resolved in the latest build. This also hints strongly that the problem lies in somewhere in the distro-config difference.

Marcel, could you try rebuilding it with the '--without-stlport' build option? I've observed from my past experience that, many crash problems that occur on Ubuntu (or Debian) but do not occur on openSUSE were attributed to what appears to be a bug in the stlport itself. We do build ours without stlport for the openSUSE distro, so if you could try that, and see if that makes any difference, we will know where the problem likely is.

> As a sidenote: I tried the OpenSuse 11.2 Gnome Live CD, which seems to have
> ooo310m21, build 9319 -- and I *don't* see the crash. I'm not sure about the
> numbering scheme (why m21 instead of m19 but lower build No.?), but this seems
> to be an *older* version, right?

I'm not 100% sure, but I *think* m21 was used for 3.1, and m19 for 3.1.1. Now, I have no clue why we went backward in the milestone number when going from 3.1 to 3.1.1, but our 3.1.1 build AFAIK is based on ooo310-m19.

> If you need anything else, I'd be happy to do some more bug triaging.

Please try a new build with --without-stlport option to see if that makes any difference if it's not too much trouble. Also, the backtrace in launchpad suggests that the crash occurs in

WW8_SdrAttrIter::OutParaAttr ()

but it would help us if you could somehow find out exactly on which line the crash occurs. Thanks!