* commit 'bf570b1a7ca3b9cbd4a59f0933a19ce1bcc99103': (71 commits)
Amend tests to explicitly use H5Dcreate2 and H5Dopen2
Revert malloc's back to using hard-coded type for sizeof
Minor comment refactoring
Update documentation
Finish up Parallel Compression test scaling work
Partial update for scaling parallel filters tests
Modify t_dset.c in lieu of Parallel Compression changes
Updated H5C__flush_single_entry() in H5C.c to correct duplicate metadata write bug observed in 1.10.1.
Amend MANIFEST
Add test for write parallel; read serial case
Fix uninitialized array issue in test
Test updates
Fix bug where incorrect amount of data was being read from the file
Add data verification for first half of tests
Start adding data verification
Switch tests over to use testing macros
Updates to parallel filters tests
Move test files to testpar directory
Add test file to build process
Suggested changes from code review
...
* commit 'b95986e50d1d1043952b00cad6595e8ecbff414f': (71 commits)
Amend tests to explicitly use H5Dcreate2 and H5Dopen2
Revert malloc's back to using hard-coded type for sizeof
Minor comment refactoring
Update documentation
Finish up Parallel Compression test scaling work
Partial update for scaling parallel filters tests
Modify t_dset.c in lieu of Parallel Compression changes
Updated H5C__flush_single_entry() in H5C.c to correct duplicate metadata write bug observed in 1.10.1.
Amend MANIFEST
Add test for write parallel; read serial case
Fix uninitialized array issue in test
Test updates
Fix bug where incorrect amount of data was being read from the file
Add data verification for first half of tests
Start adding data verification
Switch tests over to use testing macros
Updates to parallel filters tests
Move test files to testpar directory
Add test file to build process
Suggested changes from code review
...
* commit 'abcf30f232914c1ec9b8c5f2a3bb3a621af7bfe4':
HDFFV-10284 Add VS2017 support
Add commentary around calling H5Screate_simple with a single element
Miscellaneous improvements (cont.) Description: Put back H5Location::getNumObjs and marked as deprecated in favor of Group::getNumObjs. Platforms tested: Linux/32 2.6 (jam) Darwin (osx1010test)
Miscellaneous improvements Description: Moved H5Location::getNumObjs to Group::getNumObjs (i.e., H5Gget_info) Switched reinterpret_cast to static_cast in H5Object::iterateAttrs Miscellaneous cleanup Platforms tested: Linux/32 2.6 (jam) Linux/64 (platypus) Darwin (osx1010test)
HDFFV-10282 refactor out assert calls in tools
Fix the t_bigio test so that it runs on 32bit machines
* commit '289007567e6ab5e0893af787b1870f27f10644c7':
Add commentary around calling H5Screate_simple with a single element
Miscellaneous improvements (cont.) Description: Put back H5Location::getNumObjs and marked as deprecated in favor of Group::getNumObjs. Platforms tested: Linux/32 2.6 (jam) Darwin (osx1010test)
Miscellaneous improvements Description: Moved H5Location::getNumObjs to Group::getNumObjs (i.e., H5Gget_info) Switched reinterpret_cast to static_cast in H5Object::iterateAttrs Miscellaneous cleanup Platforms tested: Linux/32 2.6 (jam) Linux/64 (platypus) Darwin (osx1010test)
Fix for daily test failure Fix for the compilation error from the PGI compiler.
Modifications based on comments from pull request review (1) Remove unnecessary asserts (2) Add code to insert bad offset values to the test file in gen_bad_offset.c
Fix for HDFFV-10216 segfault in H5G_node_cmp3 with corrupt h5 file Fix H5HL_offset_into() to return error when offset exceeds heap data block size. Also fix other places that call this routine to detect error return.
Fix the t_bigio test so that it runs on 32bit machines
* commit '883a235d5ca81770228c5cb5f9d162afda0b411f':
Miscellaneous improvements (cont.) Description: Put back H5Location::getNumObjs and marked as deprecated in favor of Group::getNumObjs. Platforms tested: Linux/32 2.6 (jam) Darwin (osx1010test)
Miscellaneous improvements Description: Moved H5Location::getNumObjs to Group::getNumObjs (i.e., H5Gget_info) Switched reinterpret_cast to static_cast in H5Object::iterateAttrs Miscellaneous cleanup Platforms tested: Linux/32 2.6 (jam) Linux/64 (platypus) Darwin (osx1010test)
Added some commentary on the fact that 32 bit compilers appear to require actual array arguments to call H5Screate_simple rather than a pointer to a single size_t element.
* commit '22baab211d912f1012057b3a7ae217abd2a44a38':
Add commentary around calling H5Screate_simple with a single element
Fix the t_bigio test so that it runs on 32bit machines
Description:
Put back H5Location::getNumObjs and marked as deprecated in favor of
Group::getNumObjs.
Platforms tested:
Linux/32 2.6 (jam)
Darwin (osx1010test)
* commit '593e4038b3ab474a47e468bb3478d4ae3a6820e6':
Fix for daily test failure Fix for the compilation error from the PGI compiler.
Modifications based on comments from pull request review (1) Remove unnecessary asserts (2) Add code to insert bad offset values to the test file in gen_bad_offset.c
HDFFV-10188 force non-native type description
HDFFV-10188 emu insists on printing non-native description
Fix for HDFFV-10216 segfault in H5G_node_cmp3 with corrupt h5 file Fix H5HL_offset_into() to return error when offset exceeds heap data block size. Also fix other places that call this routine to detect error return.
HDFFV-10188 fix typo
HDFFV-10188 add missing copy commands
HDFFV-10188 Add tests and files
HDFFV-10188 enable null space test
HDFFV-10188 add release note
HDFFV-10188 - Check for empty string first
HDFFV-10188 error on NULL dataspace
Description:
Moved H5Location::getNumObjs to Group::getNumObjs (i.e., H5Gget_info)
Switched reinterpret_cast to static_cast in H5Object::iterateAttrs
Miscellaneous cleanup
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
* commit '1ef8577a4aaaa4120a819ff36747fdca99abdc26':
Modifications based on comments from pull request review (1) Remove unnecessary asserts (2) Add code to insert bad offset values to the test file in gen_bad_offset.c
Fix for HDFFV-10216 segfault in H5G_node_cmp3 with corrupt h5 file Fix H5HL_offset_into() to return error when offset exceeds heap data block size. Also fix other places that call this routine to detect error return.
* commit '678c3ae1535b9b3b9faa53bc48250f1635778dd7':
HDFFV-10188 force non-native type description
HDFFV-10188 emu insists on printing non-native description
duplicate metadata write bug observed in 1.10.1.
Tested parallel/production on jelly
parallel/debug & parallel/production on charis (in develop branch)