Issue #195: Actually call uselocale() in the new locale handling code in json_tokener.
Also, be sure the right locale_t is freed if we fail on the second uselocale.
Finally, fix test_locale so it *doesn't* use json_object_to_json_string as
that will simple re-emit the original parsed string values.
It's possible (e.g. by using json_object_from_file() on an empty file)
to get json-c to try to use a printbuf that has never had anything
written to it. Before this change, it could access a string that
should be length zero, but was never initialized, and could
theoretically have an unexpected string.
Issue #113: add "new" files to appropriate variables in Makefile, and note the need to run "make distcheck" as part of the release process.
Fix issues with the test script so it works under distcheck too.
Issue #189: Eliminate use of MC_ERROR from json_util.c, and add a json_util_get_last_err() function to retrieve the error for those callers that care about it.
Add tests and descriptions for the functions in json_util.c
Issue #236: Add -Wcast-qual and fix casts to retain constness.
To better distinguish between entry->k and entry->v being const within linkhash, but non-const outside, add lh_entry_v() and lh_entry_k() accessors.
Make lh_entry->k const.
Check the __GCC_HAVE_SYNC_COMPARE_AND_SWAP_{2,4,8} defines to decide whether to use __sync_val_compare_and_swap(), as described at https://gcc.gnu.org/onlinedocs/cpp/Common-Predefined-Macros.html
Also, fix the types of the variables when building on Windows.
Also, should address issue #214.
Emmanuele Bassi [Mon, 23 May 2016 09:18:59 +0000 (10:18 +0100)]
Use stdint.h to check for size_t size
If we use json_inttypes.h we are going to fail when builddir != srcdir,
and since JSON_C_HAVE_INTTYPES_H is defined after the configure script
has run, including json_inttypes.h would be the equivalent on including
stdint.h anyway.
This fixes the build of json-c in the GNOME Continuous builder.
Sevan Janiyan [Tue, 5 Jan 2016 11:43:17 +0000 (11:43 +0000)]
Use a macro to indicate C99 to the compiler
The [AC_PROG_CC_C99 macro](https://www.gnu.org/software/autoconf/manual/autoconf-2.64/html_node/C-Compiler.html) sets the complier to the C99 standard if the compiler does not default to such behaviour (albeit gnu99). This is better than specifically hardcoding flags
Jehan [Fri, 1 Jan 2016 17:35:14 +0000 (18:35 +0100)]
configure: check realloc with AC_CHECK_FUNCS() to fix cross-compilation.
AC_FUNC_REALLOC is messed up when cross-compiling, ending up in failed
build with "undefined reference to `rpl_realloc'" error.
AC_CHECK_FUNCS will work both with native and cross builds.
That commit introduced read-only keys, but when the hash table
was resized, that attribute was not preserved. This resulted in
an invalid free at time of table destruction.
Remove the AC_FUNC_MALLOC check, since we don't depend on the malloc(0) behavior it checks for, and we don't provide a rpl_malloc() implementation anyway.
more efficient handling for smalls strings inside json_object
smalls strings inside json_objects had a high overhead because dynamic
memory allocation was needed for each of them. This also meant that the
pointer needed to be updated. This is now changed so that small strings
can directly be stored inside the json_object. Note that on the regular
64 bit machines a pointer takes 8 bytes. So even without increasing
memory, we could store string up to 7 bytes directly inside the object.
The max size is configurable. I have selected up to 31 bytes (which
means a buffer of 32 including the NUL byte). This brings a 24-bytes
memory overhead, but I consider that still useful because the memory
allocator usually also has quite some overhead (16 bytes) for
dyn alloced memory blocks. In any case, the max buffer size can be
tweaked via #define.
These items were used for statistics tracking, but no code at all
exists to consume them. By removing them we save
a) space
because they counters required space, and did so in each and every
json object
b) performance
because calloc() needs to write less data and the counters are
no longer maintained; cache performance can be better, load
on OS main memory is lighter
We could conditionally enable/disable these counters, but I have not
done this they were really nowhere used and it looked more like a
left-over from the import of hashtable code.
This also adds a new API json_global_set_string_hash() which permits
to select the hash function. The default one is the only one that was
previously present. So there are no changes to existing apps, and the
new hash function needs to be explicitely be opted in. Especially for
smaller strings, the perllike functions seems to be around twice as
fast as the other one, with similarly good results in value distribution.
In certain C libraries (e.g uClibc), isnan() and related functions are
implemented in libm, so json-c needs to link against it. This commit
therefore adds an AC_TRY_LINK() test to check whether a program
calling isnan() can be properly linked with no special flags. If not,
we assume linking against libm is needed.
The json-c.pc.in file is also adjusted so that in the case of static
linking against json-c, -lm is also used.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
projects / json-c / log