GitLab

Previously, for HDF5 datasets, the number of dimensions in the light data
description had to match the number of dimensions in the heavy data file in
order to properly select a hyperslab. To support older files and be more
flexible, we now allow the number of dimensions to be different as long as
we are reading the entire dataset.

Add HDF5 tree writer test.

made changes for preparations to push to master; still need to finish documentation changes and line length changes

based on connectivity.

Previously, nodes added to a mesh in XdmfTopologyConverter were
compared using position to ensure nodes added to element boundaries
were not duplicated (added more than once). This failed in some
cases, especially for highly refined meshes, because of incorrect
floating point comparisons. Now, shared nodes are identified using
connectivity, ensuring that nodal addition should work in all cases,
even for highly refined meshes.

Add tests for I/O of strings (to hdf5 and xml).
Fix bug in XdmfExodusReader when reading grid centered data.
Add string I/O in python and add tests.

Modified writer to utilize hyperslabs to write in parallel to file.
Updated DSM to newest version (0.9.9).
Modified DSM test to write in parallel to DSM buffer.
Add test for new hyperslab writing mode.

Using statements have been added to XdmfSharedPtr.hpp. All used of
boost::shared_ptr and boost::shared_dynamic_cast have been replaced
by shared_ptr and shared_dynamic_cast. Slight modification to XdmfCore.i
to get swig wrapping working again.

relying on the hdf5 data type in the file to always correspond to the
XdmfArrayType. This will allow hdf5 to handle datatype conversion (e.g. from
int to double) if this is required.  It also further decouples light and heavy
data descriptions.

Still an issue with invalid read of size 1 in hdf5 in valgrind with the new test.
This has to do with overwriting a dataset with a smaller sized dataset when calling
H5Dset_extent(). It is deep within hdf5 when dealing with chunked data. No errors
from hdf5 returned however.

same dimension.

ENH: Modify HDF5Controllers to accept start, stride, and count to pull hyperslabs from hdf5 data sets.

Hopefully this will become a standard style for Xdmf development.
All spaces - 80 character line wrapping.

ENH: Add XdmfHeavyDataController base class and make XdmfArray use it.  Add XdmfHeavyDataWriter base class and make XdmfWriter use it.  This is in preparation to add additional heavy data formats.

ENH: Add XdmfGridRegular for CoRectilinear (Regular) Structured Grids.  Add test for interaction and output.

ENH: Change some method names in XdmfArray to be clearer and less verbose.  Modify files for documentation clarity.  Modify some minor things to prevent -Wall warnings in gcc.  Add python install rules and rpathing.

ENH: Add XdmfExodusReader for reading ExodusII files into Xdmf.  Add ability for XdmfArray::initialize() to take a size value to initialize the array to.

ENH: Modify all filesystem paths used in xdmf to use the absolute real path - aka "canonical path".  Modify tests to ensure this is working.

ENH: Add Mode to XdmfWriter.  Default mode ensures that all heavy data referenced by the XML file being written
exists in a single heavy data file.  A DistributedHeavyData mode can be set to allow heavy data referenced by
the XML file to exist in more than one hdf5 file.

ENH: Move XdmfCore into its own subdirectory.  Move XdmfCore specific tests into core subdirectory and modify
swig wrapping to create two separate libraries (Xdmf and XdmfCore).

destructors to be publicly accessible.  Modify tests to work with new wrapping.

between XdmfArray and XdmfHDF5Controller a bit more robust.

location previously written to) and append (write data to end of previously written
dataset).