[MOAB-dev] hdf5 file read

[Lukasz Kaczmarczyk]

I spotted that if I write meshset to hdf file and tag is not set for elements of that set and default value is not defined, then error occurs when I read that file. This problem exists for h5minfo as well.
I don't now if I should back to older revisions, or is better to wait for update.

Kind regards,
Lukasz

On 17 Nov 2009, at 14:14, Lukasz Kaczmarczyk wrote:

> I tracked error to that point:
> Breakpoint 2, ReadHDF5::create_tag (this=0x100877460, info=@0x101080522, handle=@0x7fff5fbfe680, hdf_type=@0x7fff5fbfe68c) at ReadHDF5.cpp:2640
> 2640	                                  hid_t& hdf_type )
> 3: info.size = 1
> 2: info.type = mhdf_FLOAT
> 1: info.storage = 1
> (gdb) c
> Continuing.
> 
> Breakpoint 3, ReadHDF5::create_tag (this=0x100877460, info=@0x101080522, handle=@0x7fff5fbfe680, hdf_type=@0x7fff5fbfe68c) at ReadHDF5.cpp:2818
> 2818	                                handle, info.default_value );
> 3: info.size = 1
> 2: info.type = mhdf_FLOAT
> 1: info.storage = 1
> (gdb) s                      
> MBCore::tag_create (this=0x100866980, name=0x1010811f2 "AREA_FE", size=8, storage=MB_TAG_SPARSE, data=MB_TYPE_DOUBLE, handle=@0x7fff5fbfe680, def_val=0xfffffffffffd1000, use_existing=false) at MBCore.cpp:1771
> 1771	  if (def_val && MB_VARIABLE_LENGTH == size)
> (gdb) n
> 1774	  MBErrorCode rval = tagServer->add_tag( name, size, storage, data, handle, def_val );
> (gdb) 
> \
> Program received signal EXC_BAD_ACCESS, Could not access memory.
> Reason: KERN_INVALID_ADDRESS at address: 0xfffffffffffd1000
> 0x00007fffffe007bf in __memcpy ()
> 
> 
> On 17 Nov 2009, at 01:33, Lukasz Kaczmarczyk wrote:
> 
>> Hello
>> I have problem with reading attached file in moab format. I created file with
>> mbImpl->write_file(file, "MOAB" , "PARALLEL=WRITE_PART", &PostProcTrefftzMeshSet,1)
>> 
>> Couple days ago, before svn update everything works ok. Can someone help me find the problem,
>> <AAA.h5m>
>> 
>> 
>> I made quite big analysis, and I could not open hdf files with results. I hope that I don't have to recompute it once again.
>> 
>> Kind regards,
>> Lukasz
> 
> Lukasz Kaczmarczyk
> Lecturer
> Department of Civil Engineering,
> University of Glasgow,
> GLASGOW, G12 8LT
> Tel: +44 141 3305348
> email: likask at civil.gla.ac.uk
> web: http://www.civil.gla.ac.uk/~kaczmarczyk/
> web: http://code.google.com/p/yaffems/
> 
> 

Lukasz Kaczmarczyk
Lecturer
Department of Civil Engineering,
University of Glasgow,
GLASGOW, G12 8LT
Tel: +44 141 3305348
email: likask at civil.gla.ac.uk
web: http://www.civil.gla.ac.uk/~kaczmarczyk/
web: http://code.google.com/p/yaffems/