summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [svn-r1689] Mainly adding support for "native" variable-length strings (C ↵Quincey Koziol1999-09-2918-630/+824
| | | | | | | | only currently), but I fixed lots of misc. compiler warnings in other code and also tracked down the memory overwrite bug that was causing the development branch to core dump on most machines.
* [svn-r1688] Added missing vector Cray (x90 series) configuration file.Quincey Koziol1999-09-291-0/+138
|
* [svn-r1686] updated the testing section to reflect the changes in the ↵Patrick Lu1999-09-281-16/+9
| | | | | | testing with the new bat files
* [svn-r1685] removed the bat files that were used for testing and replaced ↵Patrick Lu1999-09-281-0/+0
| | | | | | them with 2 bat files that take command line arguments to determine which test to run
* [svn-r1683] Did some cleanup.Albert Cheng1999-09-281-18/+4
|
* [svn-r1682] Added the ability to accept command line arguements.Albert Cheng1999-09-281-1/+1
|
* [svn-r1672] bin/ltconfig:Albert Cheng1999-09-282-5/+9
| | | | | | | | | | | | | ltconfig used just an empty source file to test the share lib compiler options. SUNSpro CC returns succeed though it issues a warning too. ltconfig did not like the warning. Changed it to test with a file containing a simple dummy program. Platform tested: solaris 2.6 and 2.7 config/solaris2.x: Changed default compiler to "cc" now that Sunpro CC can produce static and shared codes correctly. Also added -s to PROD_CFLAGS to produce leaner binary files.
* [svn-r1670] same changes as ont he release sidePatrick Lu1999-09-271-7/+7
|
* [svn-r1668] same fixes that were done on the release branch.Patrick Lu1999-09-271-2/+2
| | | | just casted the isdigit and isprintf parameter
* [svn-r1667] same fixes from the release branch were done here.Patrick Lu1999-09-272-4/+4
| | | | just casts with the parameters for the isdigit and isprint functions
* [svn-r1665] same changes i just made to the release versionPatrick Lu1999-09-251-2/+2
| | | | changed sdims to int
* [svn-r1664] still getting some warnings. changed the sdims variable to intPatrick Lu1999-09-251-2/+2
|
* [svn-r1663] added the include for fcntl.h here tooPatrick Lu1999-09-251-1/+1
|
* [svn-r1661] removed some of the warningsPatrick Lu1999-09-251-14/+14
|
* [svn-r1659] brought changes from the release branch over to the developementPatrick Lu1999-09-251-11/+13
| | | | removed some of the warnings.
* [svn-r1657] added the 4 new batch files and removed the install instructions ↵Patrick Lu1999-09-241-0/+0
| | | | | | from the zip file
* [svn-r1655] added some lines in the test section about the dumptest batch filesPatrick Lu1999-09-241-9/+60
|
* [svn-r1653] Removed unneeded sys/fcntl.h include line, which was causing J90 ↵Quincey Koziol1999-09-241-1/+0
| | | | problems.
* [svn-r1651] Fold in J90 changesQuincey Koziol1999-09-246-567/+595
|
* [svn-r1645] conclude.in:Albert Cheng1999-09-223-3/+3
| | | | | | | | | | | | | | | | | | | Problems: libtool generated compiler linking commands with dependency_libs appended to any dynamic library used but only "-lz -lmfhdf ..." were appended but "-L...", if used, are not appended. So, the "-lz ..." information is incomplete and solaris cc does not allow even if the "-L..." information is provide in later part of the command. Solution: Included $LDFLAGS in the building of $(LIB). irix5.x: irix6.x: Added "-s" to PROD_CFLAGS so that production code are striped for smaller sizes. Platform Tested: Solaris 2.6
* [svn-r1643] Tweaks to run on baldric better.Quincey Koziol1999-09-221-4/+8
|
* [svn-r1642] Snapshot version 1.3 release 0Quincey Koziol1999-09-212-2/+2
|
* [svn-r1638] forgot to print the oid for the groups.hdf5-1_3_0Patrick Lu1999-09-171-13/+63
|
* [svn-r1636] Patrick Lu1999-09-152-21/+8
| | | | | | | | h5tools.h- added a define for OBJECTID-->"OBJECTID" h5dump.c - added an option in there to print object id's still needs some testing but I just wanted the changes to be put up there before i forget
* [svn-r1634] fixed the compilation error where the H5P_DATASET_CREATE and ↵Patrick Lu1999-09-081-2/+2
| | | | | | H5P_DATA_XFER were using the old names.
* [svn-r1633] added instructions for removing or replacing the zlibsPatrick Lu1999-09-081-3/+47
|
* [svn-r1632] updated the proj files so it uses the zlib and zlib.dll.Patrick Lu1999-09-081-0/+0
| | | | | | updated the zlib.h and zconf.h added the zlib dir to the src dir for nt update install instructions
* [svn-r1629] Fixed error in point selections which was incorrectly computing ↵Quincey Koziol1999-09-061-2/+2
| | | | | | the offsets of array elements for most datatypes.
* [svn-r1627] Cleanup some minor bugs in the MPIO file-driver ported by Robb.Albert Cheng1999-09-035-33/+23
| | | | It is working now. Still need to tie up some loose ends.
* [svn-r1625] update the H5Tinit.cPatrick Lu1999-09-011-0/+0
|
* [svn-r1624] forgot to actually chance the open to HDopen last time!Patrick Lu1999-09-011-1/+1
|
* [svn-r1623] included the H5private.h file and replace calls to open() with ↵Patrick Lu1999-08-311-5/+1
| | | | HDopen().
* [svn-r1622] Updated with new GASS driver files.Albert Cheng1999-08-311-0/+5
|
* [svn-r1621] Added GASS driver. Coded by Saurabh Bagchi, bagchi@uiuc.edu.Albert Cheng1999-08-3112-4/+1233
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Minor changes done to test/gass_xxx.c so that they print the test skip message when GASS driver is not available. This change is the implementation of GASS within HDF5-1.3 (HDF5 with Virtual File Layer). The GASS driver gives the facility of accessing HDF files on remote ftp servers. To use the GASS driver, the option --with-gass=<GASS path> shoud be specified with configure. An example of the command line used to test the distribution was: ./configure --disable-shared --without-hdf4 --with-gass=/afs/ncsa/projects/hdf/v5/bagchi/globus/GLB/development/sparc-sun-solaris2.6_nothreads_standard_debug/include,/afs/ncsa/projects/hdf/v5/bagchi/globus/GLB/development/sparc-sun-solaris2.6_nothreads_standard_debug/lib --disable-parallel The user should change the path to point to his local GASS installation. Documentation about the features of GASS and the HDF-GASS design is available separately and till it is put up on the official web site, anyone interested may contact me. Test programs to read, write or append remote files have been provided in the test directory as "gass_read.c", "gass_write.c", "gass_append.c". The test programs have the ftp site to access #define-d at the top of the file which the user can change accordingly. ./src/H5Epublic. Added new error type for file close. ./src/H5F.c Added hooks for the GASS driver. ./src/H5public.h Added header files for GASS & Globus. ./src/Makefile.in Added dependancy on GASS driver in LIB_SRC. ./src/hdf5.h Included header file for GASS driver. ./src/H5FDgass.c [NEW] Routines for the GASS driver. ./src/H5FDgass.h [NEW] Header file for the GASS driver. ./test/Makefile.in Added dependancy on the gass test routines. ./test/gass_read.c File to test remote file reading using GASS. ./test/gass_write.c File to test remote file writing using GASS. ./test/gass_append.c File to test remote file appending using GASS.
* [svn-r1620] Fixed a typo in the configure.in. configure is derived from it.Albert Cheng1999-08-312-2/+2
|
* [svn-r1619] ./src/H5O.c [1.2, 1.3]Robb Matzke1999-08-309-1720/+250
| | | | | | | | | | | | | | | | Fixed a read-uninitialized-memory error that resulted in the file containing small amounts (<8 bytes each) of uninitialized data. These padding areas (which are never read by hdf5) are initialized to zero now. ./MANIFEST ./src/H5Z.c ./src/H5Zdeflate.c [NEW] ./src/H5Zprivate.h ./src/H5Zpublic.h ./src/Makefile.in The zlib filter was moved to a separate file but is still automatically defined.
* [svn-r1616] added __DLL__ in front of all the function prototypes so the dll ↵Patrick Lu1999-08-302-32/+32
| | | | can see it:
* [svn-r1615] put the __DLL__ in front of the function prototypes so the dll ↵Patrick Lu1999-08-305-28/+28
| | | | could see them
* [svn-r1614] put a #ifdef HAVE_UNISTD_H around the #include for the it.Patrick Lu1999-08-301-4/+32
| | | | | | since on NT I took out the device and inode members in the H5FD_sec2_t and replaced them with 2 fields more appropriate for NT I had to replace all occurences of the device and inode with the correct member
* [svn-r1613] Added check for reading an attribute immediately after writing it.Quincey Koziol1999-08-271-0/+13
|
* [svn-r1612] Fixed bug in attribute writing which was causing core dump if an ↵Quincey Koziol1999-08-271-3/+0
| | | | | | | attribute was read back in immediately after being written out (without closing the attribute in between).
* [svn-r1608] ./src/libhdf5.settings.in [1.2, 1.3]Robb Matzke1999-08-272-4/+4
| | | | | | | | | Added C proprocessor flags and non-library linker flags to the description. ./src/H5T.c [1.2, 1.3] Changed return type in tracing macros to H5Tenum_nameof() and H5Tenum_valueof().
* [svn-r1607] corrected an error in the hdf5dll project filePatrick Lu1999-08-271-0/+0
|
* [svn-r1606] Patrick Lu1999-08-271-0/+0
| | | | update the H5Tinit.c file in there and the proj file for the lib and the dll
* [svn-r1602] Changes since 19990825Robb Matzke1999-08-262-2/+10
| | | | | | | | | | | | | | ---------------------- ./src/H5O.c [1.2, 1.3] Initialzed the padding area between the end of a message and the end of the space allocated for that message. This hopefully fixes an UMR error found by Quincey and prevents the file from containing unwanted data. ./tools/h5ls.c [1.2, 1.3] Fixed a bug where the `Data:' tag didn't get printed for attributes that have more than five values.
* [svn-r1599] Changed return type of H5Tenum_nameof and H5Tenum_valueof from ↵Quincey Koziol1999-08-252-5/+11
| | | | hid_t to herr_t
* [svn-r1597] Extended space for string read in.Quincey Koziol1999-08-251-1/+1
|
* [svn-r1595] Patrick Lu1999-08-251-2/+2
| | | | | | | | not sure if this is what was wanted but in the code where the file drivers are registered the end_registration: label was giving me some trouble with compilation on NT. I think the end_registration label was supposed to be empty and but on NT it didn't like that so i had to put a semi colon after the label. if this is wrong let me know and i'll fix it someother way
* [svn-r1594] Patrick Lu1999-08-251-0/+3
| | | | | | | added a conditional for __INT64 in the part of the code where the file addresses are given types. so now if the sizeof__int64 is >=8 then haddr_t is typedefed to an unsigned __int64 and the HADDR_UNDEF = ((haddr_t)(__int64)(-1)). I used the unsigned long long as my example.
* [svn-r1593] *** empty log message ***Robb Matzke1999-08-251-0/+85
|
generated by cgit v0.12 at 2024-10-09 00:28:59 (GMT)