]>
granicus.if.org Git - json-c/log
Alexandru Ardelean [Wed, 19 Jul 2017 12:22:36 +0000 (15:22 +0300)]
build,travis: drop `-enable-strerror-override` argument (no longer exists)
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 19 Jul 2017 12:10:10 +0000 (15:10 +0300)]
strerror_override: add extern "C" and JSON_EXPORT specifiers for Visual C++ compilers
Fixes build on AppVeyor.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Sat, 15 Jul 2017 14:12:44 +0000 (07:12 -0700)]
PR #336: since we can't use function overriding (due to problems with it on
OSX) always include the _json_c_strerror function but only enable it with a flag
during tests.
Eric Haszlakiewicz [Sat, 15 Jul 2017 14:07:28 +0000 (07:07 -0700)]
PR #336: since we can't use function overriding (due to problems with it on OSX) always include the _json_c_strerror function but only enable it with a flag during tests.
Eric Haszlakiewicz [Sat, 15 Jul 2017 14:03:18 +0000 (07:03 -0700)]
Allow USE_VALGRIND to be set to anything starting with 0, N or n to disable valgrind during tests.
Eric Haszlakiewicz [Fri, 14 Jul 2017 02:39:01 +0000 (22:39 -0400)]
Merge pull request #336 from commodo/fix-tests
tests: fix tests in travis-ci.org
Alexandru Ardelean [Thu, 13 Jul 2017 07:24:27 +0000 (10:24 +0300)]
build,travis: enable strerror override option in build
To get consistent output between Linux & OS X.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Thu, 13 Jul 2017 07:11:15 +0000 (10:11 +0300)]
build: make `strerror()` override-able
If we want to override `strerror()` in libjson-c
to make tests consistent across platforms, we
need to do it build-wide as configure/build
option.
Apple linkers make it really hard to override functions
at link-time, and this seems to be locked down on travis-ci.org
[ for security reasons I assume ].
While I got it to work locally, it did not work
when running on travis.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Thu, 13 Jul 2017 06:33:39 +0000 (09:33 +0300)]
configure.ac: check for `uselocale` function only on Linux platforms
On Apple this seems to fail the `test_locale` test,
which would imply that the `uselocale` function
does not behave as expected.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 10 Jul 2017 15:00:41 +0000 (18:00 +0300)]
tests: fix leak in `test_util_file` ; found by cppcheck
Which now seems to fail the build.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 10 Jul 2017 14:47:00 +0000 (17:47 +0300)]
travis,tests: run cppcheck only if it exists
ugh... seems cppcheck is not packaged for OS X
And `set -e` exposes this.
And also `cppcheck` seems to exit with non-zero
exit codes by default [even if errs found].
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 10 Jul 2017 10:30:49 +0000 (13:30 +0300)]
tests: compress test_utile_file with test_basic
More code compression/de-duplication.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 10 Jul 2017 09:01:56 +0000 (12:01 +0300)]
tests: add `set -e` specifier to bail early on build run
Seems that test1 is failing, but travis is not catching it.
Likely, this is because the `cppcheck` returns success
and we need to bail on the `make check` step.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Tue, 11 Jul 2017 04:16:12 +0000 (00:16 -0400)]
Merge pull request #335 from commodo/build_status_travis
README.md: show build status tag from travis-ci.org
Alexandru Ardelean [Mon, 10 Jul 2017 08:28:10 +0000 (11:28 +0300)]
README.md: show build status tag from travis & appveyor.
Looks nice to see a `Build Passing` tag when you
scroll to the README.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Sun, 9 Jul 2017 23:08:29 +0000 (16:08 -0700)]
Move a variable declaration to the start of the block to work better with older compilers.
Eric Haszlakiewicz [Sun, 9 Jul 2017 23:00:32 +0000 (16:00 -0700)]
Remove a spurious comma in configure.ac
Eric Haszlakiewicz [Sun, 9 Jul 2017 22:22:53 +0000 (15:22 -0700)]
Updated expected test1 output which should have been included in commit
fd9b3b2 .
Eric Haszlakiewicz [Sun, 9 Jul 2017 22:13:02 +0000 (15:13 -0700)]
Comment out the warning about racy random seed initialization in lh_char_hash(), if you're on a platform where it'll be triggered it just makes it a pain to build.
Eric Haszlakiewicz [Sun, 9 Jul 2017 22:08:21 +0000 (15:08 -0700)]
Add extra casts to void * to squash some warning on certain systems (e.g. CentOS w/ gcc 4.1.2).
Eric Haszlakiewicz [Sun, 9 Jul 2017 22:04:18 +0000 (15:04 -0700)]
Fix bad usage of strncat introduces in
1a94c70 . Pointed out by @rouault in PR #331.
Eric Haszlakiewicz [Sun, 9 Jul 2017 21:56:18 +0000 (14:56 -0700)]
Eliminate static qualifiers on a couple local variables that were causing thread safety issues. Suggested by @rouault in PR #331.
Eric Haszlakiewicz [Sun, 9 Jul 2017 03:33:28 +0000 (20:33 -0700)]
Reformat json_object_double_to_json_string_format() to have consistent spacing.
Eric Haszlakiewicz [Sun, 9 Jul 2017 02:35:06 +0000 (19:35 -0700)]
Issue #308: improve the build instructions in README.md to include the exact commands to run for installing prerequisites, as mentioned earlier on Issue #308.
Eric Haszlakiewicz [Sun, 9 Jul 2017 02:04:35 +0000 (19:04 -0700)]
Issue #332: fix a long-standing bug in array_list_put_idx() where it would attempt to free previously free'd entries due to not checking the current array length.
Add a test that triggers the problem to ensure it stays fixed.
Eric Haszlakiewicz [Sun, 9 Jul 2017 01:25:48 +0000 (21:25 -0400)]
Merge pull request #312 from Tailmon/master
Fix CMake Build process improved for MinGW and MSYS2
Eric Haszlakiewicz [Sun, 9 Jul 2017 01:24:55 +0000 (21:24 -0400)]
Merge pull request #319 from Dashlane/visual-studio-build
Windows: Fix dynamic library build with Visual Studio
Eric Haszlakiewicz [Sat, 8 Jul 2017 19:44:27 +0000 (15:44 -0400)]
Merge pull request #329 from commodo/rename-static-lib
build,cmake: build,cmake: rename libjson-c-static.a to libjson-c.a
Eric Haszlakiewicz [Sat, 8 Jul 2017 19:43:45 +0000 (15:43 -0400)]
Merge pull request #330 from commodo/symlink_some_tests
tests: symlink basic tests to a single file that has the common code
Eric Haszlakiewicz [Sat, 8 Jul 2017 19:17:17 +0000 (15:17 -0400)]
Merge pull request #333 from besser82/bugfix/obsolete_macros_autotools
Replace obsolete AM_PROG_LIBTOOL
Björn Esser [Mon, 3 Jul 2017 10:17:36 +0000 (12:17 +0200)]
Replace obsolete AM_PROG_LIBTOOL
Alexandru Ardelean [Thu, 22 Jun 2017 06:53:30 +0000 (09:53 +0300)]
tests: symlink basic tests to a single file that has the common code
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 19 Jun 2017 15:12:03 +0000 (18:12 +0300)]
build,cmake: set C_STANDARD 99 property to libjson-c-static
Apply the same property as to json-c.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Mon, 19 Jun 2017 15:08:13 +0000 (18:08 +0300)]
build,cmake: rename libjson-c-static.a to libjson-c.a
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Mon, 19 Jun 2017 04:34:44 +0000 (00:34 -0400)]
Merge pull request #321 from commodo/fix-cmake-vasprintf
build,cmake: fix vasprintf implicit definition and generate both static & shared libs
Eric Haszlakiewicz [Sun, 18 Jun 2017 18:44:45 +0000 (18:44 +0000)]
Issue #161: add a json_object_to_fd() function.
Eric Haszlakiewicz [Sun, 18 Jun 2017 18:34:41 +0000 (18:34 +0000)]
Fix test_double_serializer expected output.
Eric Haszlakiewicz [Sun, 18 Jun 2017 18:04:49 +0000 (18:04 +0000)]
Add a json_c_set_serialization_double_format() function to set the *library-wide* format for how doubles are written to a serialized JSON output.
Eric Haszlakiewicz [Sun, 4 Jun 2017 18:25:51 +0000 (18:25 +0000)]
Make _set_last_err() non-static so it can be used outside of json_util.c
Eric Haszlakiewicz [Mon, 22 May 2017 01:35:16 +0000 (21:35 -0400)]
Merge pull request #325 from rouault/fix_stack_overflow_in_json_object_double_to_json_string_format
Fix stack buffer overflow in json_object_double_to_json_string_format()
Even Rouault [Thu, 18 May 2017 20:36:35 +0000 (22:36 +0200)]
Fix stack buffer overflow in json_object_double_to_json_string_format()
Issue originally found in the json-c 0.11 internal copy in GDAL but also found
in latest git version.
If doing things like
json_object* obj = json_object_new_double(1e300);
json_object_set_serializer(obj, json_object_double_to_json_string, "%f", NULL);
json_object_to_json_string(obj)
size = snprintf(buf, sizeof(buf),
format ? format :
(modf(jso->o.c_double, &dummy) == 0) ? "%.17g.0" : "%.17g",
jso->o.c_double);
will return a value greater than 128 since at least 300 characters are needed.
This value is then passed to printbuf_memappend(pb, buf, size); that tries to
read size bytes in buf.
So we should clamp size to sizeof(buf). And on Windows, _snprintf() returns -1
in that situation, so deal also with this case.
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=1682
Credit to OSS-Fuzz
Alexandru Ardelean [Wed, 3 May 2017 08:30:38 +0000 (11:30 +0300)]
.gitignore: add cmake generated files
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 3 May 2017 08:59:31 +0000 (11:59 +0300)]
build,cmake: generate both static and shared libjson
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 3 May 2017 08:29:12 +0000 (11:29 +0300)]
build,cmake: add _GNU_SOURCE to CFLAGS
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
David Henot [Wed, 26 Apr 2017 12:08:31 +0000 (14:08 +0200)]
Fix library build with Visual Studio
Pablo Sanabria [Sat, 25 Mar 2017 22:18:29 +0000 (19:18 -0300)]
Fixed linkhash.c for MinGW
Added #ifdef conditions for MinGW because this compiler doesn't define
gcc atomic builtins, so __sync_val_compare_and_swap was not found
neither any of any __GCC_HAVE_SYNC_COMPARE_AND_SWAP_* definition
Pablo Sanabria [Sat, 25 Mar 2017 19:31:22 +0000 (16:31 -0300)]
Fix CMake Build process for MinGW and MSYS2
This fix includes some fixes for config.h.win32 that was generating
some compiling errors on MinGW64 and also fix CMakeLists.txt in order to
give better support to MinGW and MSYS2
Eric Haszlakiewicz [Sun, 26 Mar 2017 03:52:57 +0000 (03:52 +0000)]
Merge pull request #311 from jj1118/master
Fix error C3688 when compiling on Visual Studio 2015
Jason Li [Fri, 24 Mar 2017 06:53:25 +0000 (14:53 +0800)]
Fix error C3688 when compiling on Visual Studio 2015: invalid literal suffix 'PRId64'; literal operator or literal operator template 'operator ""PRId64' not found
Eric Haszlakiewicz [Thu, 23 Mar 2017 03:37:14 +0000 (23:37 -0400)]
Merge pull request #310 from marxin/fix-gcc7-warnings
Add FALLTHRU comment to handle GCC7 warnings.
marxin [Tue, 21 Mar 2017 07:42:11 +0000 (08:42 +0100)]
Add FALLTHRU comment to handle GCC7 warnings.
Eric Haszlakiewicz [Sat, 4 Mar 2017 04:12:56 +0000 (23:12 -0500)]
Merge pull request #305 from alexanderst/master
Fix compilation without C-99 option
Alex [Thu, 2 Mar 2017 08:57:49 +0000 (10:57 +0200)]
Fix compilation without C-99 option
Eric Haszlakiewicz [Sun, 26 Feb 2017 16:22:45 +0000 (16:22 +0000)]
Add a few more cases to test_parse to provide some examples of how parsing works; should help address the questions raised in Issue #302.
Eric Haszlakiewicz [Sat, 25 Feb 2017 03:34:36 +0000 (22:34 -0500)]
Merge pull request #303 from yogo1212/tokener_h
fix doc in tokener header file
Leon M. George [Mon, 20 Feb 2017 03:41:55 +0000 (04:41 +0100)]
fix doc in tokener header file
Eric Haszlakiewicz [Sun, 5 Feb 2017 17:00:35 +0000 (17:00 +0000)]
Issue #300: Cast size_t values to int instead of unsigned long long to work around broken behavior of MinGW.
Eric Haszlakiewicz [Sun, 5 Feb 2017 04:43:54 +0000 (04:43 +0000)]
Issue #300: Some platforms, such as MinGW, don't have ETXTBSY and ENOTBLK, so skip those there.
Eric Haszlakiewicz [Sun, 5 Feb 2017 04:37:27 +0000 (04:37 +0000)]
Issue #300: Use %p instead of %lx in test_util_file since some platforms' pointers are larger than "unsigned long".
Also, there's no need to worry about output consistency here, since it'll be
different anyway due to different pointer values.
Eric Haszlakiewicz [Sun, 5 Feb 2017 04:34:05 +0000 (04:34 +0000)]
Add cast to int in calls to isdigit() since some platforms complain about char being used as an array index.
Eric Haszlakiewicz [Sat, 4 Feb 2017 16:13:14 +0000 (11:13 -0500)]
Merge pull request #299 from qlyoung/perf-improvements
Improve json_object -> string performance
Quentin Young [Fri, 3 Feb 2017 21:43:59 +0000 (21:43 +0000)]
Restore sprintbuf(), add macro for string literals
Hawciz pointed out that the previous commit modifies the public
interface of printbuf. Per his suggestion, sprintbuf() was restored
and a new pair of macros was added that wraps printbuf_memappend().
Using a wrapper macro instead of modifying sprintbuf() also reduces
function call overhead, bringing total performance gains to
approximately 400%.
Eric Haszlakiewicz [Fri, 3 Feb 2017 17:10:27 +0000 (17:10 +0000)]
Use strdup instead of strndup in test1.c, there's no need for the latter because json_object_to_json_string_length() will always return a properly sized string.
Quentin Young [Thu, 2 Feb 2017 17:16:26 +0000 (17:16 +0000)]
Improve json_object -> string performance
Removes variadic prints for ~3x performance improvement.
Eric Haszlakiewicz [Sun, 8 Jan 2017 03:55:31 +0000 (22:55 -0500)]
Issue #295: also check if size_t is the size of long long, to help support 64-bit Windows platforms.
Eric Haszlakiewicz [Sun, 18 Dec 2016 19:35:06 +0000 (14:35 -0500)]
Ignore tests/test_float
Eric Haszlakiewicz [Sun, 18 Dec 2016 19:33:41 +0000 (14:33 -0500)]
Change a memcpy that should be a memmove within json_pointer_get, and fix memory leaks in of one the json_pointer tests.
Eric Haszlakiewicz [Tue, 13 Dec 2016 04:01:20 +0000 (23:01 -0500)]
Note some minimum versions needed for autoconf, etc...
Eric Haszlakiewicz [Mon, 12 Dec 2016 03:18:33 +0000 (19:18 -0800)]
Issue#292: bump the required version of configure listed in configure.ac so a more useful error message is generated, rather than failing later because the extra macros from ./autoconf-archive/m4/ didn't get loaded.
Eric Haszlakiewicz [Mon, 5 Dec 2016 01:52:33 +0000 (20:52 -0500)]
Merge pull request #289 from jgerhards/i-278
bugfix: floating point representaion without fractional part
Jan Gerhards [Sun, 27 Nov 2016 10:50:48 +0000 (11:50 +0100)]
testbench: add test for floating point representation
Jan Gerhards [Sun, 27 Nov 2016 10:47:00 +0000 (11:47 +0100)]
bugfix: floating point representaion without fractional part
closes https://github.com/json-c/json-c/issues/278
Eric Haszlakiewicz [Thu, 24 Nov 2016 20:03:50 +0000 (15:03 -0500)]
Merge pull request #287 from commodo/json_pointer_va_args
json_pointer: extend setter & getter with printf() style arguments
Alexandru Ardelean [Wed, 16 Nov 2016 15:04:41 +0000 (17:04 +0200)]
vasprintf_compat.h: spin-off this compat header ; use math_compat.h as template
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 16 Nov 2016 15:00:45 +0000 (17:00 +0200)]
strdup_compat.h: re-spin this compat header ; use math_compat.h as template
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 16 Nov 2016 14:22:13 +0000 (16:22 +0200)]
test_json_pointer: update test with a few printf variants
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 16 Nov 2016 09:55:41 +0000 (11:55 +0200)]
json_pointer: add json_pointer_getf/setf() function variants
These include support for printf() style args for path.
Adds support for calling with 'json_pointer_getf(obj, &res, "/foo/%d/%s", 0, bar)'
style args.
Makes it easier for doing more dynamic stuff/magic, without
needing to use vasprintf() externally.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 16 Nov 2016 09:34:29 +0000 (11:34 +0200)]
json_pointer.c: fix whitespace
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Wed, 9 Nov 2016 14:56:58 +0000 (09:56 -0500)]
Merge pull request #285 from ebassi/revert-strdup-compat
Revert "compat/strdup.h: move common compat check for strdup() to own…
Emmanuele Bassi [Mon, 7 Nov 2016 20:51:11 +0000 (20:51 +0000)]
Revert "compat/strdup.h: move common compat check for strdup() to own file"
This reverts commit
aaba8c1080533133329086e47f4f45981dfdd1ee .
This commit breaks builddir != srcdir build, but, more importantly, it
also adds a dependency on a header, "config.h", which is not installed
and it's supposed to be private — since it's generated at configuration
time and it's not meant to be used by projects compiling against a
library.
Eric Haszlakiewicz [Fri, 4 Nov 2016 15:48:30 +0000 (11:48 -0400)]
Merge pull request #283 from commodo/json_pointer
json_pointer: add first revision based on RFC 6901
Alexandru Ardelean [Tue, 1 Nov 2016 16:07:03 +0000 (18:07 +0200)]
tests: add test_json_pointer test
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Tue, 1 Nov 2016 15:49:28 +0000 (17:49 +0200)]
tests/strerror_override.c: fix compilation error
I got this on Mac OS X at least.
Not sure if it shows up in other envs
error:
```
strerror_override.c:53:13: error: incompatible redeclaration of library function 'strerror' [-Werror,-Wincompatible-library-redeclaration]
const char *strerror(int errno_in)
```
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Tue, 25 Oct 2016 14:39:16 +0000 (17:39 +0300)]
json_pointer: add first revision
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Alexandru Ardelean [Wed, 26 Oct 2016 07:37:50 +0000 (10:37 +0300)]
compat/strdup.h: move common compat check for strdup() to own file
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Eric Haszlakiewicz [Sat, 29 Oct 2016 19:42:36 +0000 (15:42 -0400)]
Handle NULL objects in json_object_get_userdata() by returning NULL, but abort in json_object_set_userdata() since we can't actually do anything with the userdata.
Eric Haszlakiewicz [Sat, 29 Oct 2016 19:31:31 +0000 (15:31 -0400)]
Issue #280: add a few assert() calls to ensure objects are of the appropriate type.
Eric Haszlakiewicz [Sat, 29 Oct 2016 19:13:16 +0000 (15:13 -0400)]
Use json_visit() to clear the serializer data set when parsing so the output from test_parse reflects the actual values parsed.
Eric Haszlakiewicz [Sat, 29 Oct 2016 19:01:20 +0000 (15:01 -0400)]
Add a json_c_visit() function to provide a way to iterate over a tree of json-c objects.
Eric Haszlakiewicz [Mon, 10 Oct 2016 04:13:39 +0000 (00:13 -0400)]
Fix broken key-order test in test_compare.
Eric Haszlakiewicz [Sat, 8 Oct 2016 17:13:33 +0000 (13:13 -0400)]
Merge pull request #272 from sgerbino/master
Allows for compilation with CMake on macOS
Eric Haszlakiewicz [Sat, 8 Oct 2016 16:08:53 +0000 (12:08 -0400)]
Merge pull request #277 from ALLTERCO/json_object_set_xxx
Json object set xxx
Stoian Ivanov [Fri, 7 Oct 2016 19:54:06 +0000 (22:54 +0300)]
try restore windows automated builds
Stoian Ivanov [Thu, 6 Oct 2016 22:07:34 +0000 (01:07 +0300)]
fix compiler warning for int sizes
Stoian Ivanov [Thu, 6 Oct 2016 21:51:24 +0000 (00:51 +0300)]
string set and tests
Stoian Ivanov [Thu, 6 Oct 2016 20:32:19 +0000 (23:32 +0300)]
gitignore test; add double value set checks in test
Stoian Ivanov [Thu, 6 Oct 2016 20:16:29 +0000 (23:16 +0300)]
some simple tests
Eric Haszlakiewicz [Thu, 6 Oct 2016 03:15:51 +0000 (23:15 -0400)]
Issue #275: fix out of bounds read when handling unicode surrogate pairs.
Stoian Ivanov [Wed, 5 Oct 2016 20:55:46 +0000 (23:55 +0300)]
some basic set
Eric Haszlakiewicz [Wed, 28 Sep 2016 12:08:01 +0000 (08:08 -0400)]
Merge pull request #273 from konoal/master
fixed wrong object name in json_object_all_values_equal