[svn-r20406] Description: Refactor root group routines to centralize and simplify them some. Also patch root group's file pointer when it doesn't match the querying file, from H5G_rootof.Tested on:
FreeBSD/32 6.3 (duty) in debug mode
FreeBSD/64 6.3 (liberty) 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 (amani) w/Intel compilers, w/default API=1.6.x,
w/C++ & FORTRAN, in production mode
Solaris/32 2.10 (...
[svn-r20384] Purpose: Fixed Bug 2184 - GMQS: h5diff - incorrect calculation code for --use-system-epsilon optionDescription:
Merged from HDF5 1.8 branch r20369.
Fixed h5diff for --use-system-epsilon option: the calculation changed
from ( |a - b| / b ) to ( |a - b| ). This was decided for better
performance and was corrected only in HDF5 trunk, so 1.8 got updated.
Also comments for equal_XXX() function were updated correctly.
Also help page and RM got updated correctly.
Also ...
[svn-r20379] Missed test file from previous checkin r20378.Bug 2182 - h5ls and h5dump: update to use --enable-error-stack among tools, also update related descriptions
[svn-r20378] Purpose: Bug 2182 - h5ls and h5dump: update to use --enable-error-stack among tools, also update related descriptionsDescription:
Decided to use --enable-error-stack long option name only to
display error stack messages from HDF5 lib among tools.
Updated to unify option name to '--enable-error-stack' for printing
HDF5 error stack messages for HDF5 tools. h5ls and h5dump for now.
For h5ls, this replaces "-e/--errors" option, which is deprecated.
For h5dump, remove -E from help page an...
[svn-r20353] Bug 1752 - H5Lite Fortran APIs do not allow you to create 4 dimensional datasets (or greater) * Fixed and tested the integer, real, double routines for creating and reading dimensional dataset with ranks 4-7Tested: jam (intel, gnu compilers)
[svn-r20334] Purpose: Fixed CHICAGO: Bug 2121 - h5diff - incorrect and lack of output for the different set of attributes (different number and names)Description:
This is related to previous checkin r20294.
Add help page update for verbose option level feature, which displays
useful information about attribute difference with -v1, --verbose=1 -v2
or --verbose=2 option.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE)
[svn-r20312] Change incorrect use of "struct stat" to use "h5_stat_t". Also removed includes from top of file - they are included with proper config checks in h5private.h. Changed name of Windows only gettimeofday function to avoid a define loop according to VS2008Tested: local linux, windows
[svn-r20310] Change incorrect use of "struct stat" to use "h5_stat_t". On windows, this used the wrong structure and caused crashes. Also removed includes from top of file - they are included with proper config checks in h5private.h.Tested: local linux, windows
[svn-r20294] Purpose: Fixed CHICAGO: Bug 2121 - h5diff - incorrect and lack of output for the different set of attributes (different number and names)Description:
Previously h5diff compared attributes correctly only when two objects have
the same number of attributes and attribute names are identical.
This fix covers all other cases.
Also didn't display useful information about attribute difference.
This fixes both issues.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), lin...
[svn-r20285] Purpose: Improve the previous fix for Bug 2216 - GMQS: h5diff - memory leak when compares vlen string in dataset or attributesDescription:
Related to the previous checkin r20270 and r20266.
Improve h5tools_detect_vlen() code for better performance. H5Tdetect_class
already recusive on given type so don't need to be part of recusive call
again. Also improve error handlings in h5tools_detect_vlen and
h5tools_detect_vlen_str functions.
Also updated h5ls and h5dump code accordingly.
Tested:
...
[svn-r20270] Purpose: Improve the previous fix for Bug 2216 - GMQS: h5diff - memory leak when compares vlen string in dataset or attributesDescription:
Improve the fix along with the previous checkin r20266.
Add a new function to tool lib, h5tools_detect_vlen_data() which return
TRUE if include any kind of vlen data all at once, either VLEN-data or
VLEN-string and so on.
Also updated h5ls and h5dump code accordingly.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE...
[svn-r20266] Purpose: Fixed Bug 2216 - GMQS: h5diff - memory leak when compares vlen string in dataset or attributesDescription:
Test for dataset : valgrind --leak-check=full ./h5diff -v h5diff_dset1.h5 h5diff_dset2.h5 /g1/VLstring
Test for attr : valgrind --leak-check=full ./h5diff h5diff_attr1.h5 h5diff_attr2.h5
Both test cases are in testing script.
Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), tejeda (mac32-LE), linew (solaris-BE), Cmake - jam
[svn-r20254] Description: h5ls is used to verify for the result, so change the two tests not to run h5diff like others. This is related to the previous checkin SVN r18398 for Bug#1733.Tested:
jam (linux32-LE), amani (linux64-LE), heiwa (linuxppc64-BE), cmake (jam)
[svn-r20247] Change name of new VLEN function from H5Tdetect_vlen_str to h5tools_detect_vlen_str to match other functions in tools lib. Added back test for H5Tdetect_class of H5T_VLEN after each instance of above function to catch all VLEN types in h5dump.Tested: local linux
[svn-r20241] Corrected an error in the comments for H5Oget_comment(_by_name) and H5G_loc_get_comment_cb: the return value is the size of the comment excluding NULL terminator.No test is needed.
Quincey Koziol
Allen Byrne
Jonathan Kim
Scot Breitenfeld
Raymond Lu