Fix tests accordingly and fix misuse of hbool_t in various places
Fix initialization of H5Pgcpl/ocpl structs in property decoding routines
Tested on:
Linux/32 (jam)
Linux/64 (platypus)
Linux/PPC64 (ostrich)
MacOSX/64 10.11
NOTES:
- Developers will have to run autogen.sh before building with the autotools.
- autogen.sh takes the -p option to mimic the old bin/reconfigure behavior.
- The generated error, overflow and version headers have been left in place.
- The generated H5LT parser code has also been left in place.
- There are no changes for CMake users at this time.
Tested on: h5committest
Complete revamp of package initialization/shutdown mechanism in the library.
Each package now has a single init/term routine.
This new way should avoid packages being re-initialized during library
shutdown and is also be _much_ more proactive about giving feedback for
resource leaks internal to the library.
Introduces a new "module" header file for packages in the library
(e.g src/H5Fmodule.h) which sets up some necessary package configuration macros
for the FUNC_ENTER/LEAVE macros. (The VFL drivers have their own slightly
modified version of this header, src/H5FDdrvr_module.h)
Also cleaned up a bunch of resources leaks all across the library and tests,
along with addressing many warnings, as I encountered them.
Tested on:
MacOSX/64 10.10.5 (amazon) w/serial & parallel
Linux/64 3.10.x (kituo) w/serial & parallel
Linux/64 2.6.x (ostrich) w/serial
The only remaining code consists of a few floating-point tests
that rely on pre-generated and checked-in VMS files. These have
been left alone, even though they will not be possible to
recreate, since testing VMS float behavior is still important.
Tested on: h5committest
Tested w/ h5committest
NOTES: - The manifest may still be messed up.
- Cmake fails since the dual binary work needs to be merged with
this repo's CMake externals.
Description:
- Put back the UNUSED parameters in dsets test because the change to remove
the warning last time caused failure in setting filter, in turn, caused
failure in the test with such obscure/unrelated errors!
- Added incRefCount() to other constructors that missed from last time.
Platforms tested:
Linux/64 (platypus)
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Description:
Per user Jason Newton request, the following constructor is added:
H5File(hid_t existing_id);
Also, fixed H5File::openFile to close current file first before re-using
the object.
Platforms tested:
Linux/64 (platypus)
Linux/32 2.6 (jam gnu and Intel 15.0)
SunOS 5.11 (emu)
Description:
When copy constructor or constructor that takes an existing id is invoked,
the C ref counter stays the same but there is an extra C++ object which
later is destroyed and may cause the HDF5 id to be closed prematurely. The
C++ library needs to increment the ref counter in these situations, so that
the C library will not close the id when it is still being referenced.
However, the incrementing of ref count left some objects opened at the end
of the program, perhaps, due to compiler's optimization on cons/destructors. The constructor, that takes an existing id, needs to increment the counter
but it seems that the matching destructor wasn't invoked. The workaround
is to have a function for each class that has "id" that only sets the id
and not increment the ref count for the library to use in these situations.
These functions are "friend" and not public.
The friend functions are:
void f_Attribute_setId(Attribute *, hid_t)
void f_DataSet_setId(DataSet *, hid_t)
void f_DataSpace_setId(DataSpace *, hid_t)
void f_DataType_setId(DataType *, hid_t)
Platforms tested:
Linux/64 (platypus)
Linux/32 2.6 (jam gnu and Intel 15.0)
SunOS 5.11 (emu)
Description:
Added wrappers for C functions H5P[s/g]et_libver_bounds and wrappers
for getting object header version
// Sets bounds on versions of library format to be used when creating
// or writing objects.
void setLibverBounds(H5F_libver_t libver_low, H5F_libver_t libver_high) const;
// Gets the current settings for the library version format bounds.
void getLibverBounds(H5F_libver_t& libver_low, H5F_libver_t& libver_high) const;
// Returns the object header version of an object in a file or group,
// given the object's name.
unsigned childObjVersion(const char* objname) const;
unsigned childObjVersion(const H5std_string& objname) const;
Platforms tested:
Linux/64 (platypus)
Linux/32 2.6
SunOS 5.11
Description:
- Changed DataType::operator= to simply copy the id of rhs instead of
calling H5Tcopy because, when the operator= is invoked, a different
datatype id is created and it won't have the same characteristics as
rhs', specifically, if the rhs represents a named datatype, "this"
would still be a transient datatype.
- Added a DataType constructor that takes a PredType object, and this
constructor will cause H5Tcopy to generate another datatype id, from a
predefined datatype.
- Fixed various mistakes in tests.
Platforms tested:
Linux/64 (platypus)
Linux/32 2.6 (jam/gnu and jam/icc 15)
SunOS 5.11 (emu development/production)
CXX_SHARED_CONDITIONAL does not appear in AM_CONDITIONAL
This was due to a removed AM_CONDITIONAL macro in configure.ac.
The CXX_SHARED_CONDITIONAL code was removed from the Makefile.am
files. Additionally, the C++ shared library build settings field in
libhdf5.settings file was updated.
Tested on: local linux VM w/ C++
--enable-shared --enable-static
--enable-shared --disable-static
--disable-shared --enable-static
branch.
Removed the configure option that allows selective disabling of individual
internal filters (fletcher32, shuffle, etc.).
This feature mucked up the code with a lot of #ifdefs, saved very little space,
and was not scalable to a general scheme for library size reduction. We've
decided to remove the feature while we investigate a more general scheme for
decreasing the library size.
Part of: HDFFV-9086
Tested on: h5committest
- Remove Infering parallel compilers (C and Fortran) from configure.ac
- Remove restriction to build shared with parallel
- Cleanup parallel sections in configure.ac
- remove large file support checks
- MPE fixes.
tested with h5committest.
Developers will now have to run bin/reconfigure (deprecated, will be removed
soon) or autogen.sh after checking out code.
Part of: HDFFV-9120
Tested on: jam with Fortran and C++
Change AC_TRY_RUN TO AC_TRY_COMPILE
Changed all the instances of AC_TRY_RUN for the Fortran and C++ tests to use
AC_COMPILE_IFELSE and moved all the tests to new files in m4 directory:
aclocal_cxx.m4 (for C++ tests)
aclocal_fc.m4 (for Fortran tests)
tested: jam
Description:
Followed hints on the JIRA issue to remove several potential memory
leaks.
Platforms tested:
Linux/ppc64 (ostrich)
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Description:
H5F_ACC_CREAT was included in the C++ API while the C library doesn't
allow it yet. Possibly, in the future, but not now. In addition, the
two flags H5F_ACC_RDONLY and H5F_ACC_RDWR were missing from the
documentation, causing confusion that appending is not supported.
Solution:
- Removed H5F_ACC_CREAT from the function until the C library support it
- Added H5F_ACC_RDONLY and H5F_ACC_RDWR to the comments to update the
documentation
Platforms tested:
Linux/ppc64 (ostrich)
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
