[Paraview] [Xdmf] xdmf2 and paraview troubles

[Alin M Elena]

Hi Berk,

Personally I do not use the xdmf and I am interested only in the molecular 
plots.

The 1.8 version is the one that came default on my opensuse 11.1 box. 
In the face of the issues may be a good idea to bundle, as an option, a 
version that is known to work with paraview.

I try to compile paraview with all plugins as we intend here at ICHEC to offer 
it to the users as a complement/replacement of their visualization tools.

regards,

Alin


-- 
 ______________________________________________________________________
 "If the Universities will not study useless subjects, who will?"
                             G. F. FitzGerald, Nature, 45/46, 392 (1892)
 ______________________________________________________________________
 Mr  Alin M ELENA
 Irish Centre for High-End Computing -- www.ichec.ie
 The Design Tower, Trinity Technology & Enterprise Campus
 Grand Canal Quay,  Dublin 2, Ireland
 Tel:   +353 (0) 1 5241608 ext 29
 Fax: +353 (0) 1 7645845
 http://alin.elenaworld.net
 alin.elena at ichec.ie
 alinm.elena at gmail.com
 ______________________________________________________________________
On Wednesday 25 March 2009 10:40:25 you wrote:
> Is there a particular reason why you are using hdf5 1.8? Things would
> probably go smoother if you used 1.6.X.
>
> -berk
>
> On Wed, Mar 25, 2009 at 5:01 AM, Alin M Elena
>
> <alinm.elena at googlemail.com> wrote:
> > Hi Berk,
> >
> > Here is the answer that I got from John on the xdmf list.
> > This is after some header problems were solved plus the hdf5 1.8 version
> > problem solved too,
> >
> > Now the problem seem to be related with some casting of uint64. Anyhow it
> > looks like me related with the problem that Mike Jackson fixed last
> > summer in June.
> >
> > I will not have time to follow John's advice until next week.
> >
> > regards,
> >
> > Alin
> >
> > --
> >  ______________________________________________________________________
> >  "If the Universities will not study useless subjects, who will?"
> >                             G. F. FitzGerald, Nature, 45/46, 392 (1892)
> >  ______________________________________________________________________
> >  Mr  Alin M ELENA
> >  Irish Centre for High-End Computing -- www.ichec.ie
> >  The Design Tower, Trinity Technology & Enterprise Campus
> >  Grand Canal Quay,  Dublin 2, Ireland
> >  Tel:   +353 (0) 1 5241608 ext 29
> >  Fax: +353 (0) 1 7645845
> >  http://alin.elenaworld.net
> >  alin.elena at ichec.ie
> >  alinm.elena at gmail.com
> >  ______________________________________________________________________
> >
> >
> >
> > ---------- Forwarded message ----------
> > From: John Biddiscombe <biddisco at cscs.ch>
> > To: alin.elena at ichec.ie
> > Date: Tue, 24 Mar 2009 18:37:45 +0100
> > Subject: Re: [Xdmf] xdmf2 and paraview troubles
> > Alin
> >
> >> [ 46%] Building CXX object
> >> Utilities/Xdmf2/libsrc/CMakeFiles/Xdmf.dir/XdmfH5Driver.o
> >> /home/alin/ParaView3/Utilities/Xdmf2/libsrc/XdmfH5Driver.cxx(165):
> >> error: a value of type "haddr_t={uint64_t={unsigned long}} (*)(H5FD_t
> >> *)" cannot be used to initialize an entity of type "herr_t={int}
> >> (*)(H5FD_t *, H5FD_mem_t={H5FD_mem_t}, hid_t={int},
> >> haddr_t={uint64_t={unsigned long}}, hsize_t={unsigned long long})"      
> >>                                                H5FD_dsm_get_eoa,        
> >>   /*get_eoa       */                                                    
> >>                                                                        
> >>              ^
> >
> > Look carefully and you can see a 64bit int is being cast to a standard
> > int (or vice versa). Have a play with Enable SIZE_T and 64 bit
> > compilation options. This error clearly should not occur, but since
> > you're using an intel compiler, some combination might not have been
> > discoverd. (possibly even a CMake issue causing types not to be detected
> > properly, but unlikely)
> >
> > See if you can find what settings make the error go away and post the
> > results back. Perhaps it'll help provide a fix for the future.
> >
> > NB. I'm guessing, so if I am barking up the wrong tree. Apologies.
> >
> > JB