Correct H5I use of skip list to acquire the 'next' pointer in the skip
list after the ID's 'free' callback has been called, since it occasionally
deletes the 'next' node.
Also a little bit of code cleanup in other modules.
Tested on:
FreeBSD/32 8.2 (loyalty) w/gcc4.6, w/C++ & FORTRAN, in debug mode
FreeBSD/64 8.2 (freedom) w/gcc4.6, w/C++ & FORTRAN, in debug mode
Linux/32 2.6 (jam) w/PGI compilers, w/default API=1.8.x,
w/C++ & FORTRAN, w/threadsafe, in debug mode
Linux/64-amd64 2.6 (koala) w/Intel compilers, w/default API=1.6.x,
w/C++ & FORTRAN, in production mode
Solaris/32 2.11 (emu) w/deprecated symbols disabled, w/C++ & FORTRAN,
w/szip filter, w/threadsafe, in production mode
Linux/PPC 2.6 (ostrich) w/C++ & FORTRAN, w/threadsafe, in debug mode
Switch H5I internal data structure from using hash table to skip lists.
Tested on:
Mac OSX 10.8.4 (amazon) w/C++ & FORTRAN
FreeBSD, Linux, etc. forthcoming
- Appended h5tutr_ to names of data files to avoid name clashing with
existing examples.
- Added example dependencies to Makefile.*
(pointed out by Allen)
Platforms tested:
SunOS 5.11 (emu) with gmake
Linux/ppc64 (ostrich)
Linux/32 2.6 (jam)
Linux/64 2.6 (koala)/PGI compilers
The reason was typo in the index variable in the newly added test_get_obj_ids().
Solution: fixed the typo.
Tested: AIX machine. (Did not run h5committest because the fix is in a remote
machine and the fix is quite obvious.)
Mailing lists instructions are incorrect.
Fix by removing hdf5dev which has long retired (2009). Subscription requests
should be sent to new address due to server change.
Tested: tried to subscribe myself and it worked.
Description:
There is a bug in the current big.c tests. In a nut-shell,
the writing part randomly picks 50 locations and write 4KB
from that location. Random location range is from byte 0
to end of dataset. There is the problem--if the random location
is greater than (end of dataset - 4KB), then error.
Solution:
Adjust the random number to be at least 4KB less than the end of the
dataset. That allows it write 4KB more from the generated starting
position.
Tested: h5comittest plus jam serial build and test.
Description:
Changed header guards from single underscore to double underscore.
Platforms tested:
Linux/32 2.6 (jam) - very minor
Linux/64 2.6 (koala)/PGI compilers
Description:
- The failure in daily test was caused by missing initialization of member
"id" in a few constructors. This is now fixed.
- Added two overloaded H5Location::setComment
- Improved some error reporting in H5Location
- Improved error reporting in tests
Platforms tested:
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Linux/64 2.6 (koala)/PGI compilers
Description:
Added tutorial examples that Barbara made following the C tutorial examples.
They will be configured for daily test.
Platforms tested:
Linux/32 2.6 (jam) - tested by running a script that compiled them with
h5c++ and executing the examples; output are checked manually
Description:
With optimization enabled, gcc 4.8 inserts garbage into the padding bytes of
floating point types when assigning from a literal constant. This caused
problems when H5detect.c scanned the bits in floating point types to determine
their properties.
Modified H5detect.c to scan for padding before further analyzing the type, and
to ignore all information in the padding areas. Also removed code that
temporarily disabled optimization.
Tested: jam, koala, ostrich, platypus (h5committest)
Description:
- Improved the changes in revisions r22836 and r23438, mainly on the
wrappers of reference, dereference, get region, and set/getcomment.
- Added more tests.
Platforms tested:
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Linux/ppc64 (ostrich)
Description:
- Completed the changes in revisions r22836 and r23438 with improvement
on wrappers such as reference, dereference, get region, and set/getcomment.
- Added more tests.
Platforms tested:
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Linux/ppc64 (ostrich)
Description:
- Passing the c_str() of an std string into a C function caused failure
on OpenVMS. Added a work around using temporary string. (th5s.cpp)
- Passing incorrect file access property list caused test_datasize() to
fail. Fixed.
- Close a group in test_dset, before the file can be properly closed.
This should fix the problem on OpenVMS.
Platforms tested:
Linux/32 2.6 (jam)
SunOS 5.11 (emu)
Linux/ppc64 (ostrich)
Description:
H5Fis_hdf5 uses H5F_locate_signature to check if the file is hdf5. If it does
not locate the signature, H5F_locate_signature would issue an error and return
HADDR_UNDEF. H5Fis_hdf5 does not consider it an error if the signature is not
found, so it does not issue an error or clear the stack. The filled stack could
then cause issues later on.
Changed H5F_locate_signature to return herr_t, not issue an error if the
signature is not found, and added a parameter for a pointer to the signature
address that the function fills in.
Tested: jam, koala, ostrich (h5committest)
