summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r1071] Moved the MPI test to a file of its own (t_mpi.c) for future ↵Albert Cheng1999-02-176-257/+259
| | | | | | | | | addition of other MPI tests. Changed return code tests from comparing with FAIL to with 0. Updated MANIFEST for the addition of a new file. Tested in O2k.
* [svn-r1070] Updated to do auto-configuration for parallel tests (testpar)Albert Cheng1999-02-163-16/+39
| | | | | too. Also used mpi function calls to test mpi libraries. Tested on O2K platform.
* [svn-r1069] SnapshotRobb Matzke1999-02-162-2/+2
|
* [svn-r1068] Completed the previous to do auto-configure for parallel tests too.Albert Cheng1999-02-151-7/+6
| | | | | | Also removed the pre-defined "USE_PAUSE" since it should be invoked only in individual cases, rather than as defaults. Tested in O2K.
* [svn-r1067] fixed a problem with the enumdll proj(release)Patrick Lu1999-02-151-0/+0
|
* [svn-r1066] added 3 new projects into the file. enum, enumdll and h5toh4.Patrick Lu1999-02-151-0/+0
| | | | | currently the h5toh4 is not included in the all project since it has some compilation problems.
* [svn-r1065] Changes since 19990121Robb Matzke1999-02-1521-243/+410
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./configure.in ./acconfig.h ./configure [REGENERATED] ./src/H5config.h.in [REGENERATED] ./src/H5public.h ./src/H5Omtime.c Check for <stddef.h> Checks for `__tm_gmtoff' in `struct tm' because old versions of GNU libc are different than recent versions. This fixes the failing mtime test. ./bin/config.guess ./config/freebsd2.2.7 [REMOVED] ./config/freebsd [ADDED] Changed the name so it works with all versions of FreeBSD. ./src/H5.c Moved H5F after H5T and H5G in H5_term_library() to satisfy dependencies. ./src/H5G.c Fixed a bug that caused H5Gcreate() to fail if the group name had trailing slashes. ./src/H5Gpublic.h Changed `group_name' to `name' in a prototype. ./src/Makefile.in Dynamic library on Linux, but needs for work to be generally useful. ./src/H5HG.c ./src/H5HGprivate.h Fixed alignment problems when using old GCC compilers (like the one shipped with RedHad Linux). ./tools/h5ls.c Fixed a bug where the contents of the root group could be listed twice if there was a link back to the root group. Similarly for groups that are mentioned on the command line. Fixed a bug where unknown types were printed with a random type class number. ./src/H5T.c ./src/H5Tconv.c ./src/H5Tprivate.h Fixed O(log N) conversion bugs.
* [svn-r1063] Added a feature to dump all the shell variables initial valuesAlbert Cheng1999-02-112-242/+254
| | | | | to the config.log file for debugging. Made change in configure.in and created configure via autoconf. Tested okay in an O2K.
* [svn-r1062] Adjusted for v1.1 new configuration.Albert Cheng1999-02-101-7/+5
|
* [svn-r1061] Adjusted for v1.1 new configuration.Albert Cheng1999-02-101-1/+11
| | | | | | Hard coded against intN_t types. They are not supported by Intel Red but configure thinks they are by default for cross-compiler. (Need a better solution for this.)
* [svn-r1060] Updated the example in ddl.html and supported features in ↵Ruey-Hsia Li1999-02-092-30/+93
| | | | Tools.html.
* [svn-r1059] Purpose:Paul Harten1999-02-051-7/+30
| | | | | | | | | | | | | | | Bug fix Problem: Did a H5get_type() on a dataset of committed type without doing the associated H5Tclose(). This caused an abort to occur during the atexit() phase of the h5toh4 converter. Solution: Execute the associated H5Tclose(). Platform tested: Solaris2.5
* [svn-r1058] Added a section of SUPPORTED PLATFORMS. Edited in a few platformsAlbert Cheng1999-02-051-2/+14
| | | | that I know for sure.
* [svn-r1057] Updated gcc flags for 2.7.x compilersQuincey Koziol1999-02-031-1/+1
|
* [svn-r1056] Removed carriage returns inserted by a "broken" operating system.Robb Matzke1999-02-033-82/+168
|
* [svn-r1055] Purpose:Paul Harten1999-02-028-40/+55
| | | | | | | | | | | | | | | | Bug fix Problem: The hdp dumper output files used in the h5toh4 converter test program have the result of the latest changes to the HDF4 library, but the converter was to maintain compatability with HDF4.1r2. Solution: Regenerate the HDF4.1r2 version of the hdp dumper ouput files to be used for the h5toh4 converter test program. Platform tested: Solaris2.5
* [svn-r1054] Added H5api_adpt.h tstr*.dmp and tstr2.h5Robb Matzke1999-02-022-1/+5
|
* [svn-r1053] changed the macro for using the dll.Patrick Lu1999-02-021-0/+0
| | | | | HDF5Global and HDF5DLL --> __DLLVAR__ and __DLL__ also added HDF5USEDLL into the projects that use the dll.
* [svn-r1052] changed the HDF5GLOBAL and HDF5DLL to __DLLVAR and __DLL__Patrick Lu1999-02-025-48/+51
| | | | also exported all the non static functions and globals variables to the dll
* [svn-r1051] changed the HDF5DLL and HDF5GLOBAL to __DLL__ and __DLLVAR__Patrick Lu1999-02-0220-526/+526
| | | | also exported all globals and non static functions to the dll
* [svn-r1050] made all non static functions and globals accessible to the dll.Patrick Lu1999-02-0212-149/+149
| | | | changed the HDF5DLL and HDF5GLOBAL to __DLL__ and __DLLVAR__
* [svn-r1049] Corrected a typo in the definition of $AR.Albert Cheng1999-02-021-1/+1
|
* [svn-r1047] Entered the bug fix for IRIX 6.x configure in the CHANGES log.Albert Cheng1999-02-011-1/+6
|
* [svn-r1046] Purpose:Paul Harten1999-02-014-5/+580
| | | | | | | | | | | New feature Solution: Modified h5dumptst.c to generate more single dimension datasets of H5T_STRING type held in file tstr2.h5. Platform tested: Solaris2.5
* [svn-r1045] Purpose:Paul Harten1999-02-011-0/+11
| | | | | | | | | | New feature Solution: Add string conversion testers Platform tested: Solaris2.5
* [svn-r1044] Purpose:Paul Harten1999-02-011-9/+341
| | | | | | | | | | | | | New feature Solution: Changes to support conversion of 1-dimensional HDF5 datasets of H5T_STRING type into HDF4 Vdatas. Also, support conversion of HDF5 attribute of string type into HDF4 attribute of INT8 type. Dimensional information is lost. Platform tested: Solaris2.5
* [svn-r1043] Purpose:Paul Harten1999-02-011-5/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Bug fix Problem: Depending on how the string size changes, different buffers are used when copying from one array of strings to another array of strings. An ASSERTION() is set up to check the validity of the buffer being used for each element of the string array being copied. This ASSERTION() fails for a 10 element string array of string length 42 being copied to a 10 element string array of string length 50. Solution: The overlap variable (olap) is calculated slightly differently. Also, since olap and nelmts are unsigned types, the conditional assignment: d = elmtno >= nelmts-olap ? dbuf : dp should be rewitten as: d = elmtno+olap >= nelmts ? dbuf : dp This same problem/solution may exist in H5T_conv_i_i(), and H5T_conv_f_f(). Platform tested: Solaris2.5
* [svn-r1042] added ./tools/testfiles/tstr-1.ddl and ./tools/testfiles/tstr.h5.Ruey-Hsia Li1999-02-011-0/+2
|
* [svn-r1041] Added the string test file.Ruey-Hsia Li1999-02-011-0/+0
|
* [svn-r1040] Added the expected output for string test.Ruey-Hsia Li1999-02-011-0/+435
|
* [svn-r1039] Added string tests and temporarily disabled test_compound_dt2() ↵Ruey-Hsia Li1999-02-017-90/+397
| | | | | | test. Reflected the format changes to the expected output.
* [svn-r1038] Did some work on the output format for compound data and string.Ruey-Hsia Li1999-02-014-478/+524
| | | | Also added test script for string.
* [svn-r1037] All IRIX (old 32bit OS) version 6.x can use the same configuration.Albert Cheng1999-01-293-2/+5
| | | | Replaced irix6.2 by irix6.x.
* [svn-r1036] Updated compiler flagsQuincey Koziol1999-01-282-4/+4
|
* [svn-r1035] t_file.c:Albert Cheng1999-01-272-2/+143
| | | | | | | | | | | Added a new test to verify the MPIO can support independent overlapping writes. testphdf5.c: Bracketed the MPIO write test with a #ifdef so that it is tested only when the explicitely turned on. Platform tested: O2K.
* [svn-r1034] Changed spelling of extendable to extendible just for ↵Albert Cheng1999-01-271-8/+8
| | | | consistency sake.
* [svn-r1033] added some calls to get the time elapsed on WIN NTPatrick Lu1999-01-251-2/+105
|
* [svn-r1032] Changes since 19990118Robb Matzke1999-01-2129-276/+766
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./tools/h5tools.c Strings are not converted to null-padding before being printed; they are printed with whatever byte values appear in the file. ./tools/h5ls.c Now able to display attribute data type and data. Added a `-f' or `--full' switch which causes the full name of each object to be displayed instead of just the base name. Added a `-r' or `--recursive' switch that recursively prints the contents of groups, avoiding cycles. More bulletproofing for non-printable characters in things like object names, attribute names, and comment strings. We don't want listing a file to send termal escape sequences because it's sometimes possible to execute commands that way. Since h5ls doesn't usually use quotes around object names we must sometimes escape space characters. External files are listed in a table to make the output less confusing. ./tools/h5tools.c ./tools/h5tools.h Changed h5dump() to h5dump_dset() and added h5dump_mem(). Also make h5dump_fixtype() public. ./test/dtypes.c Wrote some data to an attribute to test h5ls attribute printing. ./src/H5ACprivate.h ./src/H5Apublic.h ./src/H5Dprivate.h ./src/H5Dpublic.h ./src/H5Epublic.h ./src/H5Fprivate.h ./src/H5Fpublic.h ./src/H5Gprivate.h ./src/H5HLprivate.h ./src/H5Oprivate.h ./src/H5Ppublic.h ./src/H5RApublic.h ./src/H5Rpublic.h ./src/H5Spublic.h ./src/H5Tpkg.h ./src/H5Tpublic.h ./src/H5Vprivate.h ./src/H5Zpublic.h ./src/H5private.h ./src/H5public.h Reindented function prototypes after `HDF5API' was added. Also rewrapped long lines. ./src/H5Flow.c Added an `#ifdef WIN32' around an unused variable. ./src/H5api_adpt.h Removed extra carriage returns inserted by "broken" operating system. ./src/H5Dprivate.h ./src/H5Oprivate.h ./src/H5Vprivate.h ./src/H5private.h Removed extraneous inclusion of H5api_adpt.h since it's included in H5public.h which is included by everything. ./src/Makefile.in Added H5api_adpt.h to the list of public header files to fix broken `make install'.
* [svn-r1031] added necessary info to export info to the dllPatrick Lu1999-01-2113-252/+252
|
* [svn-r1030] added necessary info to export functions to dllPatrick Lu1999-01-2111-65/+65
|
* [svn-r1029] added the necessary information to make a dllPatrick Lu1999-01-211-1/+1
|
* [svn-r1028] new include file that is needed on NT to export functions and ↵Patrick Lu1999-01-211-0/+25
| | | | gloabls in a dll
* [svn-r1027] Added CVS Revision and ID tags so we can track the snapshots ↵Quincey Koziol1999-01-201-0/+7
| | | | more effectively.
* [svn-r1026] Changes since 19990118Robb Matzke1999-01-201-1/+0
| | | | | | | | | ---------------------- ./tools/h5tools.c Strings are not converted to null-padding before being printed; they are printed with whatever byte values appear in the file.
* [svn-r1025] added new project files for the alpha versions and the dll versionPatrick Lu1999-01-191-0/+0
|
* [svn-r1024] Purpose:Paul Harten1999-01-191-18/+19
| | | | | | | | | | | | | | | Bug fix Problem: When -with-hdf4 option is selected during configure, no machine type found for HDF4 header file hdfi.h. Solution: Modify hdf5/config/linux file: -D__i386 to be added to CPPFLAGS. Platform tested: linux(penfold)
* [svn-r1023] Changes since 19990105Robb Matzke1999-01-189-71/+121
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | ---------------------- ./INSTALL ./doc/html/Compression.html ./doc/html/Filters.html Disassociated GNU from zlib. Did not change the introduction which incorrectly states that compression in hdf5 is available only if GNU gzip is available (should read zlib). Changed a typo in INSTALL to match what is documented in various other places: that zlib 1.1.2 (not 1.0.2) or better is required for compression. Added a reference to the main zlib web page in various places. ./src/H5Fistore.c Fixed a typo in a comment. ./src/H5Flow.c Removed carriage returns from the end of lines. ./test/dsets.c The deflate compression tests are still run but `-SKIP-' is printed instead of `PASSED' if the zlib library was not detected at configuration time. The tests check that the compression layer (H5Z) is correctly bypassed.
* [svn-r1022] changed the macro for FILE to be FILENAME since it conflicted ↵Patrick Lu1999-01-121-12/+12
| | | | | | with a macro in a windows header.
* [svn-r1021] Purpose:Paul Harten1999-01-122-18/+40
| | | | | | | | | | | | | | | | Bug fix Problem: Segmentation fault when attempting to free NULL or uninitialized pointers. Problem noticed on Linux and HPUX10.20 platforms. Solution: Initialialize the pointer to NULL. Execute the "free" statement upon the condition that the pointer is not equal to NULL (i.e. after the allocation has taken place.) Platform tested: Linux, HPUX10.20, Solaris2.5
* [svn-r1020] I messed the last one up again. sorry this should be the last time.Patrick Lu1999-01-111-6/+4
|
generated by cgit v0.12 at 2024-11-09 13:05:04 (GMT)