* commit '089afc48561ba8838d6a515c6b00fc6f7032ca13':
Updated the developer flags so that they respect gcc versions.
Description: Added data files from the packet table vlen tests to the CHECK_CLEANFILES list. Platforms tested: Linux/64 (jelly)
- In the test for HDFFV-9758, a pragma pack(1) caused failure on Emu
because Sparc cannot access misaligned data. Changed it to pack() to
do the default alignment.
- Added DOXYGEN_SHOULD_SKIP_THIS blocks to private elements to prevent
warnings from doxygen.
Platforms tested:
SunOS 5.11 (emu)
Linux/32 2.6 (jam)
- Removed calls to H5Tget_native_type in the PT code. The application would
need to do that if desired.
- Added Abhi's program to tests to verify the fix.
- This fix might have fixed HDFFV-9927, HDFFV-9042, and the issue reported
by Barbara Jones from Ametek as well.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (jelly)
Darwin (osx1010test)
* commit 'd817b63020310b6f275a569529075551dd2b756e':
Fix copyright to point to cmake
Fix cyclic depends due to typo
Correct how include is used when located inside project
HDFFV-9961: Add GIT option to ext lib macros
Update copyright to latest from kitware
* commit 'f9364c0080405bb36d704eb3f9505029d3da41f4':
removed the use of C_SIZEOF for non BIND(C) derived type
Fixed test to use storage_size instead of c_sizeof when available.
fixed missing closing bracket
Removed unused variables.
Fixed: Fortran_DOUBLE was being set to C_LONG_DOUBLE when C_LONG_DOUBLE is not available.
Removed duplicate FCFLAG
Removed duplicate FCFLAG.
Added number of integer KINDs found to the header files.
Added path to source include files when building buidiface.
Added rule to build buildiface program, without a rule, build would add repeated compile options when using the NAG compiler.
Misc. cleaning up of the program.
Added SEQUENCE to derived types for NAG:
misc. format code-cleanup
Removed the use of hard-coded integer KINDs.
Code clean-up.
HDFFV-9973 Fortran library fails to compile and fails tests with NAG compiler
* commit '5a7880183025f56421cf6f2274d9f1ac36f59641':
Clean up hardcoded constants and check return values better. (Comments from group code review)
Description: Cleanups from Dana's review.
Description: Further warning cleanups: from 667 warnings to 503.
Section 81 of the compiler manual, e.g. at
http://www.nag.co.uk/nagware/np/r61_doc/np61_manual.pdf
covers details about the compiler's internal representations, including
"
Fortran derived types are translated into C structs.
If BIND(C) or SEQUENCE is used, the order of the items within
the struct is the same as the order within the derived type definition.
Otherwise, the order of items is permuted to
put larger types at the front of the struct so as to improve alignment
and reduce storage; the C output code can be
inspected to determine the ordering used.
"
Removed INTENT(IN) to fix segfaults in C APIs for TYPE(C_PTR).
Tested: NAG (gnu)
More warning cleanups, bringing the build down to 25 unique types of
warnings, with 550 warnings in 122 files (down from 28, 770, and 134).
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial & parallel
(h5committest forthcoming)
Clean up more warnings: drop the warning count from ~1310 down to ~940,
with only 31 types of warnings in 148 files (down from 38 types in 167 files).
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial & parallel
(h5committest forthcoming)
More warning cleanups. This brings us down to ~1300 warnings in 167 files
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial & parallel
(h5committest forthcoming)
Clean up warnings (from 2774 -> 1560, with my standard debug build)
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial, parallel & production
(h5committest forthcoming)
Bring object/dataset/group/named datatype features from revise_chunks
branch to trunk. Also CMake support for h5format_convert and a bunch of
misc. cleanups.
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial, parallel & production
(h5committest forthcoming)
Bring r29992 from revise_chunks branch to trunk:
Turn off error reporting for h5debug and hdf2gif.
Tested on:
MacOSX/64 10.11.5 (amazon) w/serial, parallel & produciton
(h5committest forthcoming)
Bring H5LD* routines from revise_chunks branch to trunk.
Tested on:
MacOSX/64 10.11.4 (amazon) w/serial, parallel & production
(h5committest forthcoming)
Bring H5DOappend(), H5P[s|g]et_object_flush_cb, and H5P[s|g]et_append_flush
from revise_chunks branch to trunk. Brings along updated metadata cache
entry tagging, and the internal object flush routine.
Tested on:
MacOSX/64 10.11.4 (amazon) w/serial, parallel & production
(h5committest forthcoming)
Description:
The constructor FL_PacketTable added in 1.10.0 did not have good
prototype; it didn't allow the property list to be default. Marked
it deprecated and added this one:
FL_PacketTable(hid_t fileID, const char* name, hid_t dtypeID,
hsize_t chunkSize = 0, hid_t plistID = H5P_DEFAULT)
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Description:
The "warning: deprecated conversion from string constant to âchar*â
[-Wwrite-strings]" was already removed by adding const to char*
parameter in the FL_PacketTable. This change is to remove the use
of literal strings in code for maintainability.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Description:
- Replaced "goto out" and "out:" with "goto error" and "error:" to indicate
the failure situations.
- Replaced old-style casts with static_cast to remove warnings.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Description:
Performed similar improvement as the C Packet Table library.
- Removed the VLPT_REMOVED ifdef around class VL_PacketTable
- Moved VL_PacketTable::IsVariableLength to PacketTable
- Moved VL_PacketTable::FreeReadBuff to PacketTable::FreeBuff
- Added overloaded constructor to FL_PacketTable that takes a
property list ID and call the new C API H5PTcreate.
- Added accessor wrappers to class PacketTable, per HDFFV-8623/patch 003.
/* Returns the ID of the dataset associated with the packet table */
hid_t GetDataset();
/* Returns the ID of the datatype the packet table uses */
hid_t GetDatatype();
- Replaced appropriate calls in tests
- Various cleanup
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Description:
Added several tests for packet table that were missing from the
previous update.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
Description:
The Packet Table API used to have fixed-length and variable-length functions
(i.e., H5PTcreate_fl and H5PTcreate_vl), but in 2006, the vlen functions
were ifdef'ed out with VLPT_REMOVED, see HDFFV-442. The majority of this
revision includes removing unused vlen stuff, adding various vlen tests,
initial code cleanup, and first pass of commenting.
Details:
- Removed the VLPT_REMOVED ifdef, but keep H5PTis_varlen and the free
function because they are useful or even needed, but the free
function's name is changed to be more appropriate for both
read/write buffers, i.e., H5PTfree_vlen_buff.
- The new test file, test_packet_vlen.c, contains tests showing packet
tables, containing various types of vlen data, can be created with
the fixed-length function
- Two accessor functions were added per HDFFV-8623/patch 003.
/* Returns the ID of the dataset associated with the packet table */
hid_t H5PTget_dataset(hid_t table_id);
/* Returns the ID of the datatype the packet table uses */
hid_t H5PTget_type(hid_t table_id);
- Various cleanup: replacing 0/-1 with SUCCEED/FAIL and H5I_BADID with
H5I_INVALID_HID, and removing many warnings.
Platforms tested:
Linux/32 2.6 (jam)
Linux/64 (platypus)
Darwin (osx1010test)
This was giving a lot of users headaches and the parser code rarely
changes. There is now a new script (bin/genparser) which can be
run as needed.
Tested on: Ubuntu 15.10 (x86_64 Linux 4.2.0)
gcc 5.2.1, flex 2.5.39, bison 3.0.2
NOTES:
- Developers will have to run autogen.sh before building with the autotools.
- autogen.sh takes the -p option to mimic the old bin/reconfigure behavior.
- The generated error, overflow and version headers have been left in place.
- The generated H5LT parser code has also been left in place.
- There are no changes for CMake users at this time.
Tested on: h5committest
Tested w/ h5committest
NOTES: - The manifest may still be messed up.
- Cmake fails since the dual binary work needs to be merged with
this repo's CMake externals.
