hdf5/test/fillval.c

748 lines
21 KiB
C
Raw Normal View History

[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
/*
* Copyright (C) 1998 NCSA
* All rights reserved.
*
* Programmer: Robb Matzke <robb@arborea.spizella.com>
* Thursday, October 1, 1998
*
* Purpose: Tests dataset fill values.
*/
#include <h5test.h>
[svn-r859] Changes since 19981030 ---------------------- ./MANIFEST Added new Pablo files HDF5record_RT.h and ProcIDs.h ./acconfig.h ./configure [REGENERATED] ./configure.in ./src/H5.c ./src/H5Vprivate.h ./src/H5config.h.in [REGENERATED] ./src/H5private.h ./src/H5public.h ./test/big.c Added more configuration stuff for the Win32 environment. Removed all the #ifdef WIN32 from the source and replaced them with OS-independent stuff. Specifics follow: Check for non-Posix.1 `st_blocks' field in `struct stat' which is used by the big file test to decide if the file system supports holes. If the st_blocks field isn't present then we just skip the test. Configure checks for <io.h> <sys/resource.h> <sys/time.h> and <winsock.h> and defines HAVE_IO_H, HAVE_SYS_RESOURCE_H, HAVE_SYS_TIME_H and HAVE_WINSOCK_H when they're found. Configure checks whether both <sys/time.h> and <time.h> can be included and defines SYS_TIME_WITH_TIME if so. Otherwise include only <sys/time.h> or <time.h> even if both exist. Configure checks sizeof(__int64) and defines SIZEOF___INT64 to the result or to zero if __int64 isn't defined. The source uses `long long' in preference to `__int64'. Removed null WIN32 definition for `inline' since such a definition already exists in H5config.h Protected gettimeofday() calls in debugging code with HAVE_GETTIMEOFDAY instead of WIN32. ./src/H5F.c ./src/H5Flow.c ./src/H5Fmpio.c ./src/H5Fsec2.c ./src/H5Fstdio.h ./src/H5P.c ./src/H5Tconv.c ./src/H5private.h Removed #include of system files from library source files and consolodated them into H5private.h where they're protected by various configuration macros (most of them were duplicated there already anyway). ./test/big.c ./test/chunk.c ./test/cmpd_dset.c ./test/dsets.c ./test/dtypes.c ./test/extend.c ./test/external.c ./test/fillval.c ./test/flush1.c ./test/flush2.c ./test/iopipe.c ./test/links.c ./test/mount.c ./test/mtime.c ./test/overhead.c ./test/ragged.c ./test/shtype.c ./test/unlink.c Protected system #include's with #ifdef's from H5config.h. Undefined NDEBUG since some of the tests rely on assert() to check return values. Removed WIN32 definitions for __unused__ since this can be controlled by the definition of HAVE_ATTRIBUTE in H5config.h ./test/testhdf5.h Removed the CLEAN_CMD definition because we no longer use it. Albert's cleanup() functions replaced it. ./test/fillval.c Initialized auto hid_t variables to fix warnings in error recovery code when data flow analysis is turned on in compilers. ./test/h5tools.c Initialized an auto variable to fix a compiler warning. ./test/chunk.c ./test/ragged.c The WIN32 had some unsigned variables changed to signed because the compiler generates warnings when coercing unsigned to double(?). I changed them back to unsigned because they really are unsigned quantities. If this the change was just to shut up extraneous warnings then perhaps a compiler flag can do the same; otherwise if the compiler generates bad code then we should supply a patch file instead messing up source code with bug work-arounds. ./src/H5detect.c Protected system #include's with #ifdef's from H5config.h thereby removing a WIN32. If getpwuid() doesn't exist (HAVE_GETPWUID) then we assume that `struct passwd' doesn't exist either (we don't really need it in that case). The H5T_NATIVE_LLONG and H5T_NATIVE_ULLONG are defined in terms of `long long' or else `__int64' or else `long' depending on what's available. ./src/H5Flow.c ./src/H5Ofill.c Added __unused__ to some function arguments that aren't used when assertions are turned off. ./src/H5V.c Changed an auto variable name in some hand-inlined code to get rid of a warning about the variable shadowing a previous auto.
1998-11-03 01:58:28 +08:00
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
/*
* Define NO_FILLING if you want to compare how this test works when there is
* no fill value (that is, when the fill value is zero).
*/
/* #define NO_FILLING */
const char *FILENAME[] = {
"fillval_1",
"fillval_2",
"fillval_3",
"fillval_4",
"fillval_5",
"fillval_6",
NULL
};
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#define FILE_NAME_RAW "fillval.raw"
/*-------------------------------------------------------------------------
* Function: test_getset
*
* Purpose: Tests the H5Pget_fill_value() and H5Pset_fill_value()
* functions.
*
* Return: Success: 0
*
* Failure: number of errors
*
* Programmer: Robb Matzke
* Thursday, October 1, 1998
*
* Modifications:
*
*-------------------------------------------------------------------------
*/
static int
test_getset(void)
{
herr_t status;
hid_t dcpl=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
int fill_i;
hid_t type_ss=-1, type_si=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
struct fill_si {
int v1, v2;
} fill_si;
struct fill_ss {
short v1, v2;
} fill_ss, fill_ss_rd;
TESTING("property lists");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
/*
* Create the dataset creation property list and the data types that will
* be used during this test.
*/
if ((dcpl=H5Pcreate(H5P_DATASET_CREATE))<0) goto error;
if ((type_ss=H5Tcreate(H5T_COMPOUND, sizeof fill_ss))<0 ||
H5Tinsert(type_ss, "v1", HOFFSET(struct fill_ss, v1),
H5T_NATIVE_SHORT)<0 ||
H5Tinsert(type_ss, "v2", HOFFSET(struct fill_ss, v2),
H5T_NATIVE_SHORT)<0) {
goto error;
}
if ((type_si=H5Tcreate(H5T_COMPOUND, sizeof fill_si))<0 ||
H5Tinsert(type_si, "v1", HOFFSET(struct fill_si, v1),
H5T_NATIVE_INT)<0 ||
H5Tinsert(type_si, "v2", HOFFSET(struct fill_si, v2),
H5T_NATIVE_INT)<0) {
goto error;
}
/*
* Reading the fill value from a dataset creation property list that has
* no fill value should result in a failure.
*/
H5E_BEGIN_TRY {
status = H5Pget_fill_value(dcpl, H5T_NATIVE_INT, &fill_i);
} H5E_END_TRY;
if (status>=0) {
FAILED();
puts(" H5Pget_fill_value() should have been negative");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
/*
* Set the fill value using a struct as the data type.
*/
fill_ss.v1 = 1111;
fill_ss.v2 = 2222;
if (H5Pset_fill_value(dcpl, type_ss, &fill_ss)<0) goto error;
/*
* Get the fill value using the same data type that was used to set it.
*/
if (H5Pget_fill_value(dcpl, type_ss, &fill_ss_rd)<0) goto error;
if (fill_ss.v1!=fill_ss_rd.v1 || fill_ss.v2!=fill_ss_rd.v2) {
FAILED();
puts(" Failed to get fill value using same data type that was ");
puts(" used to set the fill value.");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
/*
* Get the fill value using some other data type.
*/
if (H5Pget_fill_value(dcpl, type_si, &fill_si)<0) goto error;
if (fill_ss.v1!=fill_si.v1 || fill_ss.v2!=fill_si.v2) {
FAILED();
puts(" Failed to get fill value using a data type other than what");
puts(" was used to set the fill value.");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
/*
* Reset the fill value
*/
if (H5Pset_fill_value(dcpl, type_si, &fill_si)<0) goto error;
if (H5Pget_fill_value(dcpl, type_ss, &fill_ss)<0) goto error;
if (fill_si.v1!=fill_ss.v1 || fill_si.v2!=fill_ss.v2) {
FAILED();
puts(" Resetting the fill value was unsuccessful.");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
/* Success */
if (H5Pclose(dcpl)<0) goto error;
if (H5Tclose(type_si)<0) goto error;
if (H5Tclose(type_ss)<0) goto error;
PASSED();
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 0;
error:
H5E_BEGIN_TRY {
H5Pclose(dcpl);
H5Tclose(type_si);
H5Tclose(type_ss);
} H5E_END_TRY;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 1;
}
/*-------------------------------------------------------------------------
* Function: test_create
*
* Purpose: Tests creating datasets that have fill values.
*
* Return: Success: 0
*
* Failure: number of errors
*
* Programmer: Robb Matzke
* Thursday, October 1, 1998
*
* Modifications:
*
*-------------------------------------------------------------------------
*/
static int
test_create(hid_t fapl, const char *base_name, H5D_layout_t layout)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
{
[svn-r859] Changes since 19981030 ---------------------- ./MANIFEST Added new Pablo files HDF5record_RT.h and ProcIDs.h ./acconfig.h ./configure [REGENERATED] ./configure.in ./src/H5.c ./src/H5Vprivate.h ./src/H5config.h.in [REGENERATED] ./src/H5private.h ./src/H5public.h ./test/big.c Added more configuration stuff for the Win32 environment. Removed all the #ifdef WIN32 from the source and replaced them with OS-independent stuff. Specifics follow: Check for non-Posix.1 `st_blocks' field in `struct stat' which is used by the big file test to decide if the file system supports holes. If the st_blocks field isn't present then we just skip the test. Configure checks for <io.h> <sys/resource.h> <sys/time.h> and <winsock.h> and defines HAVE_IO_H, HAVE_SYS_RESOURCE_H, HAVE_SYS_TIME_H and HAVE_WINSOCK_H when they're found. Configure checks whether both <sys/time.h> and <time.h> can be included and defines SYS_TIME_WITH_TIME if so. Otherwise include only <sys/time.h> or <time.h> even if both exist. Configure checks sizeof(__int64) and defines SIZEOF___INT64 to the result or to zero if __int64 isn't defined. The source uses `long long' in preference to `__int64'. Removed null WIN32 definition for `inline' since such a definition already exists in H5config.h Protected gettimeofday() calls in debugging code with HAVE_GETTIMEOFDAY instead of WIN32. ./src/H5F.c ./src/H5Flow.c ./src/H5Fmpio.c ./src/H5Fsec2.c ./src/H5Fstdio.h ./src/H5P.c ./src/H5Tconv.c ./src/H5private.h Removed #include of system files from library source files and consolodated them into H5private.h where they're protected by various configuration macros (most of them were duplicated there already anyway). ./test/big.c ./test/chunk.c ./test/cmpd_dset.c ./test/dsets.c ./test/dtypes.c ./test/extend.c ./test/external.c ./test/fillval.c ./test/flush1.c ./test/flush2.c ./test/iopipe.c ./test/links.c ./test/mount.c ./test/mtime.c ./test/overhead.c ./test/ragged.c ./test/shtype.c ./test/unlink.c Protected system #include's with #ifdef's from H5config.h. Undefined NDEBUG since some of the tests rely on assert() to check return values. Removed WIN32 definitions for __unused__ since this can be controlled by the definition of HAVE_ATTRIBUTE in H5config.h ./test/testhdf5.h Removed the CLEAN_CMD definition because we no longer use it. Albert's cleanup() functions replaced it. ./test/fillval.c Initialized auto hid_t variables to fix warnings in error recovery code when data flow analysis is turned on in compilers. ./test/h5tools.c Initialized an auto variable to fix a compiler warning. ./test/chunk.c ./test/ragged.c The WIN32 had some unsigned variables changed to signed because the compiler generates warnings when coercing unsigned to double(?). I changed them back to unsigned because they really are unsigned quantities. If this the change was just to shut up extraneous warnings then perhaps a compiler flag can do the same; otherwise if the compiler generates bad code then we should supply a patch file instead messing up source code with bug work-arounds. ./src/H5detect.c Protected system #include's with #ifdef's from H5config.h thereby removing a WIN32. If getpwuid() doesn't exist (HAVE_GETPWUID) then we assume that `struct passwd' doesn't exist either (we don't really need it in that case). The H5T_NATIVE_LLONG and H5T_NATIVE_ULLONG are defined in terms of `long long' or else `__int64' or else `long' depending on what's available. ./src/H5Flow.c ./src/H5Ofill.c Added __unused__ to some function arguments that aren't used when assertions are turned off. ./src/H5V.c Changed an auto variable name in some hand-inlined code to get rid of a warning about the variable shadowing a previous auto.
1998-11-03 01:58:28 +08:00
hid_t file=-1, space=-1, dcpl=-1, dset1=-1, dset2=-1, dset3=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
hsize_t cur_size[5] = {32, 16, 8, 4, 2};
hsize_t ch_size[5] = {1, 1, 1, 4, 2};
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
short rd_s, fill_s = 0x1234;
long rd_l, fill_l = 0x4321;
char filename[1024];
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if (H5D_CHUNKED==layout) {
TESTING("chunked dataset creation");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
} else {
TESTING("contiguous dataset creation");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/*
* Create a file and three datasets. The three datasets test three fill
* conversion paths: small to large, large to small, and no conversion.
* They depend on `short' being smaller than `long'.
*/
h5_fixname(base_name, fapl, filename, sizeof filename);
if ((file=H5Fcreate(filename, H5F_ACC_TRUNC, H5P_DEFAULT, fapl))<0)
goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if ((space=H5Screate_simple(5, cur_size, cur_size))<0) goto error;
if ((dcpl=H5Pcreate(H5P_DATASET_CREATE))<0) goto error;
if (H5D_CHUNKED==layout) {
if (H5Pset_chunk(dcpl, 5, ch_size)<0) goto error;
}
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/* Small to large fill conversion */
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#ifndef NO_FILLING
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
if (H5Pset_fill_value(dcpl, H5T_NATIVE_SHORT, &fill_s)<0) goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#endif
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
if ((dset1=H5Dcreate(file, "dset1", H5T_NATIVE_LONG, space, dcpl))<0)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/* Large to small fill conversion */
#ifndef NO_FILLING
if (H5Pset_fill_value(dcpl, H5T_NATIVE_LONG, &fill_l)<0) goto error;
#endif
if ((dset2=H5Dcreate(file, "dset2", H5T_NATIVE_SHORT, space, dcpl))<0)
goto error;
/* No conversion */
#ifndef NO_FILLING
if (H5Pset_fill_value(dcpl, H5T_NATIVE_LONG, &fill_l)<0) goto error;
#endif
if ((dset3=H5Dcreate(file, "dset3", H5T_NATIVE_LONG, space, dcpl))<0)
goto error;
/* Close everything */
if (H5Dclose(dset1)<0) goto error;
if (H5Dclose(dset2)<0) goto error;
if (H5Dclose(dset3)<0) goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if (H5Sclose(space)<0) goto error;
if (H5Pclose(dcpl)<0) goto error;
if (H5Fclose(file)<0) goto error;
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/* Open the file and get the dataset fill value from each dataset */
if ((file=H5Fopen(filename, H5F_ACC_RDONLY, fapl))<0)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/* Large to small conversion */
if ((dset1=H5Dopen(file, "dset1"))<0) goto error;
if ((dcpl=H5Dget_create_plist(dset1))<0) goto error;
if (H5Dclose(dset1)<0) goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#ifndef NO_FILLING
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
if (H5Pget_fill_value(dcpl, H5T_NATIVE_SHORT, &rd_s)<0) goto error;
if (rd_s!=fill_s) {
FAILED();
puts(" Got a different fill value than what was set.");
printf(" Got %d, set %d\n", rd_s, fill_s);
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
#endif
if (H5Pclose(dcpl)<0) goto error;
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
/* Small to large conversion */
if ((dset2=H5Dopen(file, "dset2"))<0) goto error;
if ((dcpl=H5Dget_create_plist(dset2))<0) goto error;
if (H5Dclose(dset2)<0) goto error;
#ifndef NO_FILLING
if (H5Pget_fill_value(dcpl, H5T_NATIVE_LONG, &rd_l)<0) goto error;
if (rd_l!=fill_l) {
FAILED();
puts(" Got a different fill value than what was set.");
printf(" Got %ld, set %ld\n", rd_l, fill_l);
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
goto error;
}
#endif
if (H5Pclose(dcpl)<0) goto error;
/* No conversion */
if ((dset3=H5Dopen(file, "dset3"))<0) goto error;
if ((dcpl=H5Dget_create_plist(dset3))<0) goto error;
if (H5Dclose(dset3)<0) goto error;
#ifndef NO_FILLING
if (H5Pget_fill_value(dcpl, H5T_NATIVE_LONG, &rd_l)<0) goto error;
if (rd_l!=fill_l) {
FAILED();
puts(" Got a different fill value than what was set.");
printf(" Got %ld, set %ld\n", rd_l, fill_l);
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
goto error;
}
#endif
if (H5Pclose(dcpl)<0) goto error;
if (H5Fclose(file)<0) goto error;
PASSED();
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 0;
error:
H5E_BEGIN_TRY {
H5Pclose(dcpl);
H5Sclose(space);
[svn-r745] Changes since 19981002 ---------------------- ./doc/html/H5.format.html ./src/H5HG.c Fixed a bug in the global heap that caused H5HG_read() to write past the end of the buffer in certain cases. ./test/big.c The test is skipped if hdf5 was configured with `--disable-hsizet'. ./src/H5Ofill.c Data type conversions are implemented for the fill value. ./src/H5.c Tracing prints one of H5P_FILE_CREATE, H5P_FILE_ACCESS, H5P_DATASET_CREATE, H5P_DATASET_XFER, or H5P_MOUNT instead of the more cryptic H5I_TEMPLATE_* constants. ./src/H5D.c Removed prototype for H5D_find_name(). ./src/H5I.c The GROUP_MASK and ID_MASK are both calculated from GROUP_BITS instead of being set by hand. We don't use the sign bit of hid_t; all valid hid_t values are positive so we can say things like `if ((file=H5Fopen(...))<0)'. Changed `(int)pow(2.0,x)' to `1<<x' so we don't have to worry about rounding. Fixed H5I_get_type() so it doesn't always fail an assertion. ./src/H5E.c ./src/H5Epublic.h Added minor error H5E_MOUNT ./src/H5F.c ./src/H5Fprivate.h Added H5Fmount() and H5Funmount(). Mounting and unmounting works as documented but some of the other things aren't implemented yet, the biggest being current working groups always acting on the root of the mount tree, and H5Fclose() closing the entire tree. The rest of the stuff will be added shortly... ./src/H5P.c ./src/H5Ppublic.h Added the H5P_MOUNT property list but haven't implemented any particular properties for it yet. ./src/H5Gstab.c Hard links across files return an error instead of failing an assertion.
1998-10-09 01:13:14 +08:00
H5Dclose(dset1);
H5Dclose(dset2);
H5Dclose(dset3);
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
H5Fclose(file);
} H5E_END_TRY;
return 1;
}
/*-------------------------------------------------------------------------
* Function: test_rdwr
*
* Purpose: Tests fill values for chunked datasets.
*
* Return: Success: 0
*
* Failure: number of errors
*
* Programmer: Robb Matzke
* Thursday, October 1, 1998
*
* Modifications:
*
*-------------------------------------------------------------------------
*/
static int
test_rdwr(hid_t fapl, const char *base_name, H5D_layout_t layout)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
{
[svn-r859] Changes since 19981030 ---------------------- ./MANIFEST Added new Pablo files HDF5record_RT.h and ProcIDs.h ./acconfig.h ./configure [REGENERATED] ./configure.in ./src/H5.c ./src/H5Vprivate.h ./src/H5config.h.in [REGENERATED] ./src/H5private.h ./src/H5public.h ./test/big.c Added more configuration stuff for the Win32 environment. Removed all the #ifdef WIN32 from the source and replaced them with OS-independent stuff. Specifics follow: Check for non-Posix.1 `st_blocks' field in `struct stat' which is used by the big file test to decide if the file system supports holes. If the st_blocks field isn't present then we just skip the test. Configure checks for <io.h> <sys/resource.h> <sys/time.h> and <winsock.h> and defines HAVE_IO_H, HAVE_SYS_RESOURCE_H, HAVE_SYS_TIME_H and HAVE_WINSOCK_H when they're found. Configure checks whether both <sys/time.h> and <time.h> can be included and defines SYS_TIME_WITH_TIME if so. Otherwise include only <sys/time.h> or <time.h> even if both exist. Configure checks sizeof(__int64) and defines SIZEOF___INT64 to the result or to zero if __int64 isn't defined. The source uses `long long' in preference to `__int64'. Removed null WIN32 definition for `inline' since such a definition already exists in H5config.h Protected gettimeofday() calls in debugging code with HAVE_GETTIMEOFDAY instead of WIN32. ./src/H5F.c ./src/H5Flow.c ./src/H5Fmpio.c ./src/H5Fsec2.c ./src/H5Fstdio.h ./src/H5P.c ./src/H5Tconv.c ./src/H5private.h Removed #include of system files from library source files and consolodated them into H5private.h where they're protected by various configuration macros (most of them were duplicated there already anyway). ./test/big.c ./test/chunk.c ./test/cmpd_dset.c ./test/dsets.c ./test/dtypes.c ./test/extend.c ./test/external.c ./test/fillval.c ./test/flush1.c ./test/flush2.c ./test/iopipe.c ./test/links.c ./test/mount.c ./test/mtime.c ./test/overhead.c ./test/ragged.c ./test/shtype.c ./test/unlink.c Protected system #include's with #ifdef's from H5config.h. Undefined NDEBUG since some of the tests rely on assert() to check return values. Removed WIN32 definitions for __unused__ since this can be controlled by the definition of HAVE_ATTRIBUTE in H5config.h ./test/testhdf5.h Removed the CLEAN_CMD definition because we no longer use it. Albert's cleanup() functions replaced it. ./test/fillval.c Initialized auto hid_t variables to fix warnings in error recovery code when data flow analysis is turned on in compilers. ./test/h5tools.c Initialized an auto variable to fix a compiler warning. ./test/chunk.c ./test/ragged.c The WIN32 had some unsigned variables changed to signed because the compiler generates warnings when coercing unsigned to double(?). I changed them back to unsigned because they really are unsigned quantities. If this the change was just to shut up extraneous warnings then perhaps a compiler flag can do the same; otherwise if the compiler generates bad code then we should supply a patch file instead messing up source code with bug work-arounds. ./src/H5detect.c Protected system #include's with #ifdef's from H5config.h thereby removing a WIN32. If getpwuid() doesn't exist (HAVE_GETPWUID) then we assume that `struct passwd' doesn't exist either (we don't really need it in that case). The H5T_NATIVE_LLONG and H5T_NATIVE_ULLONG are defined in terms of `long long' or else `__int64' or else `long' depending on what's available. ./src/H5Flow.c ./src/H5Ofill.c Added __unused__ to some function arguments that aren't used when assertions are turned off. ./src/H5V.c Changed an auto variable name in some hand-inlined code to get rid of a warning about the variable shadowing a previous auto.
1998-11-03 01:58:28 +08:00
hid_t file=-1, fspace=-1, mspace=-1, dcpl=-1, dset=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
hsize_t cur_size[5] = {32, 16, 8, 4, 2};
hsize_t ch_size[5] = {1, 16, 8, 4, 2};
hsize_t one[5] = {1, 1, 1, 1, 1};
hsize_t hs_size[5], hs_stride[5];
hssize_t hs_offset[5], nelmts;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#ifdef NO_FILLING
int fillval = 0;
#else
int fillval = 0x4c70f1cd;
#endif
int val_rd, should_be;
int i, j, *buf=NULL, odd;
char filename[1024];
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if (H5D_CHUNKED==layout) {
TESTING("chunked dataset I/O");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
} else {
TESTING("contiguous dataset I/O");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
/* Create a file and dataset */
h5_fixname(base_name, fapl, filename, sizeof filename);
if ((file=H5Fcreate(filename, H5F_ACC_TRUNC, H5P_DEFAULT, fapl))<0)
goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if ((fspace=H5Screate_simple(5, cur_size, cur_size))<0) goto error;
if ((dcpl=H5Pcreate(H5P_DATASET_CREATE))<0) goto error;
if (H5D_CHUNKED==layout) {
if (H5Pset_chunk(dcpl, 5, ch_size)<0) goto error;
}
#ifndef NO_FILLING
if (H5Pset_fill_value(dcpl, H5T_NATIVE_INT, &fillval)<0) goto error;
#endif
if ((dset=H5Dcreate(file, "dset", H5T_NATIVE_INT, fspace, dcpl))<0)
goto error;
/* Read some data and make sure it's the fill value */
if ((mspace=H5Screate_simple(5, one, NULL))<0) goto error;
for (i=0; i<1000; i++) {
for (j=0; j<5; j++) {
hs_offset[j] = rand() % cur_size[j];
}
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, NULL,
one, NULL)<0) goto error;
if (H5Dread(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
&val_rd)<0) goto error;
if (val_rd!=fillval) {
FAILED();
puts(" Value read was not a fill value.");
printf(" Elmt={%ld,%ld,%ld,%ld,%ld}, read: %u, "
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
"Fill value: %u\n",
(long)hs_offset[0], (long)hs_offset[1],
(long)hs_offset[2], (long)hs_offset[3],
(long)hs_offset[4], val_rd, fillval);
goto error;
}
}
if (H5Sclose(mspace)<0) goto error;
/* Write to all odd data locations */
for (i=0, nelmts=1; i<5; i++) {
hs_size[i] = cur_size[i]/2;
hs_offset[i] = 0;
hs_stride[i] = 2;
nelmts *= hs_size[i];
}
if ((mspace=H5Screate_simple(5, hs_size, hs_size))<0) goto error;
assert((nelmts*sizeof(int))==(hssize_t)((size_t)(nelmts*sizeof(int)))); /*check for overflow*/
buf = malloc((size_t)(nelmts*sizeof(int)));
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
for (i=0; i<nelmts; i++) buf[i] = 9999;
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, hs_stride,
hs_size, NULL)<0) goto error;
if (H5Dwrite(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
buf)<0) goto error;
free(buf);
buf = NULL;
H5Sclose(mspace);
/* Read some data and make sure it's the right value */
if ((mspace=H5Screate_simple(5, one, NULL))<0) goto error;
for (i=0; i<1000; i++) {
for (j=0, odd=0; j<5; j++) {
hs_offset[j] = rand() % cur_size[j];
odd += hs_offset[j]%2;
}
should_be = odd ? fillval : 9999;
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, NULL,
one, NULL)<0) goto error;
if (H5Dread(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
&val_rd)<0) goto error;
if (val_rd!=should_be) {
FAILED();
puts(" Value read was not correct.");
printf(" Elmt={%ld,%ld,%ld,%ld,%ld}, read: %u, "
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
"should be: %u\n",
(long)hs_offset[0], (long)hs_offset[1],
(long)hs_offset[2], (long)hs_offset[3],
(long)hs_offset[4], val_rd, should_be);
goto error;
}
}
if (H5Sclose(mspace)<0) goto error;
if (H5Dclose(dset)<0) goto error;
if (H5Sclose(fspace)<0) goto error;
if (H5Pclose(dcpl)<0) goto error;
if (H5Fclose(file)<0) goto error;
PASSED();
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 0;
error:
H5E_BEGIN_TRY {
H5Dclose(dset);
H5Sclose(fspace);
H5Sclose(mspace);
H5Pclose(dcpl);
H5Fclose(file);
} H5E_END_TRY;
return 1;
}
/*-------------------------------------------------------------------------
* Function: test_extend
*
* Purpose: Test that filling works okay when a dataset is extended.
*
* Return: Success: 0
*
* Failure: number of errors
*
* Programmer: Robb Matzke
* Monday, October 5, 1998
*
* Modifications:
*
*-------------------------------------------------------------------------
*/
static int
test_extend(hid_t fapl, const char *base_name, H5D_layout_t layout)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
{
[svn-r859] Changes since 19981030 ---------------------- ./MANIFEST Added new Pablo files HDF5record_RT.h and ProcIDs.h ./acconfig.h ./configure [REGENERATED] ./configure.in ./src/H5.c ./src/H5Vprivate.h ./src/H5config.h.in [REGENERATED] ./src/H5private.h ./src/H5public.h ./test/big.c Added more configuration stuff for the Win32 environment. Removed all the #ifdef WIN32 from the source and replaced them with OS-independent stuff. Specifics follow: Check for non-Posix.1 `st_blocks' field in `struct stat' which is used by the big file test to decide if the file system supports holes. If the st_blocks field isn't present then we just skip the test. Configure checks for <io.h> <sys/resource.h> <sys/time.h> and <winsock.h> and defines HAVE_IO_H, HAVE_SYS_RESOURCE_H, HAVE_SYS_TIME_H and HAVE_WINSOCK_H when they're found. Configure checks whether both <sys/time.h> and <time.h> can be included and defines SYS_TIME_WITH_TIME if so. Otherwise include only <sys/time.h> or <time.h> even if both exist. Configure checks sizeof(__int64) and defines SIZEOF___INT64 to the result or to zero if __int64 isn't defined. The source uses `long long' in preference to `__int64'. Removed null WIN32 definition for `inline' since such a definition already exists in H5config.h Protected gettimeofday() calls in debugging code with HAVE_GETTIMEOFDAY instead of WIN32. ./src/H5F.c ./src/H5Flow.c ./src/H5Fmpio.c ./src/H5Fsec2.c ./src/H5Fstdio.h ./src/H5P.c ./src/H5Tconv.c ./src/H5private.h Removed #include of system files from library source files and consolodated them into H5private.h where they're protected by various configuration macros (most of them were duplicated there already anyway). ./test/big.c ./test/chunk.c ./test/cmpd_dset.c ./test/dsets.c ./test/dtypes.c ./test/extend.c ./test/external.c ./test/fillval.c ./test/flush1.c ./test/flush2.c ./test/iopipe.c ./test/links.c ./test/mount.c ./test/mtime.c ./test/overhead.c ./test/ragged.c ./test/shtype.c ./test/unlink.c Protected system #include's with #ifdef's from H5config.h. Undefined NDEBUG since some of the tests rely on assert() to check return values. Removed WIN32 definitions for __unused__ since this can be controlled by the definition of HAVE_ATTRIBUTE in H5config.h ./test/testhdf5.h Removed the CLEAN_CMD definition because we no longer use it. Albert's cleanup() functions replaced it. ./test/fillval.c Initialized auto hid_t variables to fix warnings in error recovery code when data flow analysis is turned on in compilers. ./test/h5tools.c Initialized an auto variable to fix a compiler warning. ./test/chunk.c ./test/ragged.c The WIN32 had some unsigned variables changed to signed because the compiler generates warnings when coercing unsigned to double(?). I changed them back to unsigned because they really are unsigned quantities. If this the change was just to shut up extraneous warnings then perhaps a compiler flag can do the same; otherwise if the compiler generates bad code then we should supply a patch file instead messing up source code with bug work-arounds. ./src/H5detect.c Protected system #include's with #ifdef's from H5config.h thereby removing a WIN32. If getpwuid() doesn't exist (HAVE_GETPWUID) then we assume that `struct passwd' doesn't exist either (we don't really need it in that case). The H5T_NATIVE_LLONG and H5T_NATIVE_ULLONG are defined in terms of `long long' or else `__int64' or else `long' depending on what's available. ./src/H5Flow.c ./src/H5Ofill.c Added __unused__ to some function arguments that aren't used when assertions are turned off. ./src/H5V.c Changed an auto variable name in some hand-inlined code to get rid of a warning about the variable shadowing a previous auto.
1998-11-03 01:58:28 +08:00
hid_t file=-1, fspace=-1, mspace=-1, dcpl=-1, dset=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
hsize_t cur_size[5] = {32, 16, 8, 4, 2};
hsize_t max_size[5] = {128, 64, 32, 16, 8};
hsize_t ch_size[5] = {1, 16, 8, 4, 2};
hsize_t one[5] = {1, 1, 1, 1, 1};
hsize_t hs_size[5], hs_stride[5];
hssize_t hs_offset[5], nelmts;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
#ifdef NO_FILLING
int fillval = 0;
#else
int fillval = 0x4c70f1cd;
#endif
int val_rd, should_be;
int i, j, *buf=NULL, odd, fd;
char filename[1024];
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if (H5D_CHUNKED==layout) {
TESTING("chunked dataset extend");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
} else {
TESTING("contiguous dataset extend");
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
if ((dcpl=H5Pcreate(H5P_DATASET_CREATE))<0) goto error;
if (H5D_CHUNKED==layout) {
if (H5Pset_chunk(dcpl, 5, ch_size)<0) goto error;
}
#ifndef NO_FILLING
if (H5Pset_fill_value(dcpl, H5T_NATIVE_INT, &fillval)<0) goto error;
#endif
#if 1
/*
* Remove this once contiguous datasets can support extensions in other
* than the slowest varying dimension. The purpose of this block is to
* make only the slowest varying dimension extendible and yet have the
* same total number of elements as originally.
*
* If this is removed prematurely then you will get an error `only the
* first dimension can be extendible' as long as the test isn't skipped
* below.
*/
if (H5D_CONTIGUOUS==layout) {
max_size[0] = (max_size[0]*max_size[1]*max_size[2]*
max_size[3]*max_size[4]) /
(cur_size[1]*cur_size[2]*cur_size[3]*cur_size[4]);
max_size[1] = cur_size[1];
max_size[2] = cur_size[2];
max_size[3] = cur_size[3];
max_size[4] = cur_size[4];
}
#endif
#if 1
/*
* Remove this once internal contiguous datasets can support
* extending. If it's removed prematurely you will get an error
* `extendible contiguous non-external dataset' as long as the test isn't
* skipped below.
*/
if (H5D_CONTIGUOUS==layout) {
nelmts = max_size[0]*max_size[1]*max_size[2]*max_size[3]*max_size[4];
if ((fd=open(FILE_NAME_RAW, O_RDWR|O_CREAT|O_TRUNC, 0666))<0 ||
close(fd)<0) goto error;
if (H5Pset_external(dcpl, FILE_NAME_RAW, (off_t)0, (hsize_t)nelmts*sizeof(int))<0)
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
goto error;
}
#endif
#if 1
/*
* Remove this when contiguous datasets can be exended to some
* predetermined fininte size, even if it's just in the slowest varying
* dimension. If it's removed prematurely then you'll get one of the
* errors described above or `unable to select fill value region'.
*/
if (H5D_CONTIGUOUS==layout) {
SKIPPED();
puts(" Not implemented yet -- needs H5S_SELECT_DIFF operator");
goto skip;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
#endif
/* Create a file and dataset */
h5_fixname(base_name, fapl, filename, sizeof filename);
if ((file=H5Fcreate(filename, H5F_ACC_TRUNC, H5P_DEFAULT, fapl))<0)
goto error;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if ((fspace=H5Screate_simple(5, cur_size, max_size))<0) goto error;
if ((dset=H5Dcreate(file, "dset", H5T_NATIVE_INT, fspace, dcpl))<0)
goto error;
/* Read some data and make sure it's the fill value */
if ((mspace=H5Screate_simple(5, one, NULL))<0) goto error;
for (i=0; i<1000; i++) {
for (j=0; j<5; j++) {
hs_offset[j] = rand() % cur_size[j];
}
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, NULL,
one, NULL)<0) goto error;
if (H5Dread(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
&val_rd)<0) goto error;
if (val_rd!=fillval) {
FAILED();
puts(" Value read was not a fill value.");
printf(" Elmt={%ld,%ld,%ld,%ld,%ld}, read: %u, "
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
"Fill value: %u\n",
(long)hs_offset[0], (long)hs_offset[1],
(long)hs_offset[2], (long)hs_offset[3],
(long)hs_offset[4], val_rd, fillval);
goto error;
}
}
if (H5Sclose(mspace)<0) goto error;
/* Write to all odd data locations */
for (i=0, nelmts=1; i<5; i++) {
hs_size[i] = cur_size[i]/2;
hs_offset[i] = 0;
hs_stride[i] = 2;
nelmts *= hs_size[i];
}
if ((mspace=H5Screate_simple(5, hs_size, hs_size))<0) goto error;
assert((nelmts*sizeof(int))==(hssize_t)((size_t)(nelmts*sizeof(int)))); /*check for overflow*/
buf = malloc((size_t)(nelmts*sizeof(int)));
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
for (i=0; i<nelmts; i++) buf[i] = 9999;
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, hs_stride,
hs_size, NULL)<0) goto error;
if (H5Dwrite(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
buf)<0) goto error;
free(buf);
buf = NULL;
H5Sclose(mspace);
/* Read some data and make sure it's the right value */
if ((mspace=H5Screate_simple(5, one, NULL))<0) goto error;
for (i=0; i<1000; i++) {
for (j=0, odd=0; j<5; j++) {
hs_offset[j] = rand() % cur_size[j];
odd += hs_offset[j]%2;
}
should_be = odd ? fillval : 9999;
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, NULL,
one, NULL)<0) goto error;
if (H5Dread(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
&val_rd)<0) goto error;
if (val_rd!=should_be) {
FAILED();
puts(" Value read was not correct.");
printf(" Elmt={%ld,%ld,%ld,%ld,%ld}, read: %u, "
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
"should be: %u\n",
(long)hs_offset[0], (long)hs_offset[1],
(long)hs_offset[2], (long)hs_offset[3],
(long)hs_offset[4], val_rd, should_be);
goto error;
}
}
if (H5Sclose(mspace)<0) goto error;
/* Extend the dataset */
if (H5Dextend(dset, max_size)<0) goto error;
if (H5Sclose(fspace)<0) goto error;
if ((fspace=H5Dget_space(dset))<0) goto error;
/* Read some data and make sure it's the right value */
if ((mspace=H5Screate_simple(5, one, NULL))<0) goto error;
for (i=0; i<1000; i++) {
for (j=0, odd=0; j<5; j++) {
hs_offset[j] = rand() % max_size[j];
if ((hsize_t)hs_offset[j]>=cur_size[j]) {
odd = 1;
} else {
odd += hs_offset[j]%2;
}
}
should_be = odd ? fillval : 9999;
if (H5Sselect_hyperslab(fspace, H5S_SELECT_SET, hs_offset, NULL,
one, NULL)<0) goto error;
if (H5Dread(dset, H5T_NATIVE_INT, mspace, fspace, H5P_DEFAULT,
&val_rd)<0) goto error;
if (val_rd!=should_be) {
FAILED();
puts(" Value read was not correct.");
printf(" Elmt={%ld,%ld,%ld,%ld,%ld}, read: %u, "
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
"should be: %u\n",
(long)hs_offset[0], (long)hs_offset[1],
(long)hs_offset[2], (long)hs_offset[3],
(long)hs_offset[4], val_rd, should_be);
goto error;
}
}
if (H5Sclose(mspace)<0) goto error;
if (H5Dclose(dset)<0) goto error;
if (H5Sclose(fspace)<0) goto error;
if (H5Pclose(dcpl)<0) goto error;
if (H5Fclose(file)<0) goto error;
PASSED();
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 0;
error:
H5E_BEGIN_TRY {
H5Dclose(dset);
H5Sclose(fspace);
H5Sclose(mspace);
H5Pclose(dcpl);
H5Fclose(file);
} H5E_END_TRY;
return 1;
skip:
H5E_BEGIN_TRY {
H5Dclose(dset);
H5Sclose(fspace);
H5Sclose(mspace);
H5Pclose(dcpl);
H5Fclose(file);
} H5E_END_TRY;
return 0;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
/*-------------------------------------------------------------------------
* Function: main
*
* Purpose: Tests fill values
*
* Return: Success:
*
* Failure:
*
* Programmer: Robb Matzke
* Thursday, October 1, 1998
*
* Modifications:
*
*-------------------------------------------------------------------------
*/
int
main(int argc, char *argv[])
{
int nerrors=0, argno, test_contig=1, test_chunk=1;
hid_t fapl=-1;
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
if (argc>=2) {
test_contig = test_chunk = 0;
for (argno=1; argno<argc; argno++) {
if (!strcmp(argv[argno], "contiguous")) {
test_contig = 1;
} else if (!strcmp(argv[argno], "chunked")) {
test_chunk = 1;
} else {
fprintf(stderr, "usage: %s [contiguous] [chunked]\n", argv[0]);
exit(1);
}
}
}
h5_reset();
fapl = h5_fileaccess();
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
nerrors += test_getset();
/* Chunked storage layout tests */
if (test_chunk) {
nerrors += test_create(fapl, FILENAME[0], H5D_CHUNKED);
nerrors += test_rdwr (fapl, FILENAME[2], H5D_CHUNKED);
nerrors += test_extend(fapl, FILENAME[4], H5D_CHUNKED);
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
/* Contiguous storage layout tests */
if (test_contig) {
nerrors += test_create(fapl, FILENAME[1], H5D_CONTIGUOUS);
nerrors += test_rdwr (fapl, FILENAME[3], H5D_CONTIGUOUS);
nerrors += test_extend(fapl, FILENAME[5], H5D_CONTIGUOUS);
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
}
if (nerrors) goto error;
puts("All fill value tests passed.");
if (h5_cleanup(FILENAME, fapl)) remove(FILE_NAME_RAW);
[svn-r735] Changes since 19981002 ---------------------- ./src/H5D.c Fill values are working for contiguous datasets now except there are two things that need more support from the data space layer, specifically the ability to form a selection from the difference of two selections. They are (1) extending an external contiguous dataset, (2) optimization by delaying the fill until after the first H5Dwrite(). Renamed H5D_allocate() to H5D_init_storage() since allocation is only part of the story. Added a data space argument so it doesn't have to query the space from the object header -- the space is always available in the caller anyway. Removed `#ifdef HAVE_PARALLEL' from a few places where it wasn't necessary. We don't need it around code that doesn't compile anything from mpi.h or mpio.h. ./src/H5Fistore.c Uncommented H5F_istore_alloc() for non-parallel and moved the `#ifdef HAVE_PARALLEL' just around Kim's barrier. ./src/H5Fmpio.c Wrapped a couple long lines. Got rid of two signed vs. unsigned comparison warnings. ./MANIFEST ./test/Makefile.in ./test/fillval.c [NEW] Added tests for fill values. The contiguous dataset extend test is disabled until H5S_SELECT_DIFF is implemented. ./tools/Makefile.in Fixed a bug where `make test' didn't build the executables first. This should cause the snapshots to start up again. ./Makefile.in Changed to build in `test' directory before `tools' directory. We want the library tests to pass before we even start considering the tools. You can still build and/or test the tools independent of the library tests passing.
1998-10-06 05:01:10 +08:00
return 0;
error:
puts("***** FILL VALUE TESTS FAILED *****");
return 1;
}