Description:
- Removed calls to H5Tget_native_type from PT APIs because it is up to the
application, whether it wants the buffer to be read into memory in the
machine’s native architecture. Currently, however, the PT doesn't
allow an application to specify memory datatype. Perhaps, a new API can
be added to provide that capability.
- Added calls to H5Tcopy to H5PTcreate/H5PTcreate_fl/H5PTopen to save a
copy of the application's datatype or the dataset's datatype.
- Added various missing H5Tclose to the packet table tests, and various
error checkings.
Note: leave out changes to test_packet_vlen.c for 1.8 to wait on QAK about
merging the commit ec2fbe0883
[svn-r30158] first.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Tested platform:
Kagiso only since it is only a comment block change. If it works in one
machine, it should work in all, I hope. Still need to check the parallel
build on copper.
have an extra parameter that sets the deflate filter.
Added tests, made examples use the new APIs.
Cleaned up include files a little and removed some casts that I hope are
superfluous. If anybody encounters strange errors in the packet table
tests where the chunk size is set to some extremely large value, please
let me know, but I was unable to reproduce this error on any system.
Tested on mir, heping, sol, juniper, VS 6.0. Looked for the chunk size
error on several other systems a month or two ago.
Purpose:
Added C++ wrapper for Packet Table API.
Description:
Added macro for high-level C++ library (LIBH5CPP_HL), which changes every
Makefile.in.
Added directories for high-level C++ library (though currently only Packet
Table API is supported).
Added both C++ source and tests.
Platforms tested:
sleipnir, mir, modi4
Misc. update:
