Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2013-06-12 | Tweak the JSON_DECODE_INT_AS_REAL test introduced in #123 | Petri Lehtinen | 1 | -6/+10 | |
Only run the imprecision part if json_int_t is long long, otherwise the imprecision cannot be tested against json_int_t. Also, convert the double value to json_int_t before checking the imprecision, because otherwise the json_int_t is converted to double and the imprecision happens in the "wrong direction". | |||||
2013-06-11 | Using jsonp_strtod instead of casting to double to catch double overflow | Jason Choy | 1 | -3/+14 | |
2013-06-11 | Renamed flag to JSON_DECODE_INT_AS_REAL and added documentation | Jason Choy | 1 | -1/+1 | |
2013-06-09 | Added a simple test for the JSON_DECODE_NO_INT option | Jason Choy | 1 | -0/+12 | |
2013-06-07 | Added missing rv check - caught by clang | Jason Choy | 1 | -2/+2 | |
2013-05-12 | Fixed "comparison between signed and unsigned" warning | Daniel Griscom | 1 | -1/+1 | |
2013-05-12 | Removed (non-functioning) printfs from test code | Daniel Griscom | 1 | -6/+0 | |
2013-05-12 | Added test cases and documentation for json_array_foreach() | Daniel Griscom | 2 | -2/+28 | |
2013-04-29 | test:suites:api: modified error message | Andrei Epure | 1 | -1/+1 | |
In the "/* perform the same update again */" area the error message should be "unable to update an non-empty object" instead of "unable to update an empty object". Signed-off-by: Andrei Epure <epure.andrei@gmail.com> | |||||
2013-04-15 | Merge branch '2.4' | Petri Lehtinen | 1 | -1/+4 | |
2013-04-15 | Fix the signature again, and also rest of the errors | Petri Lehtinen | 1 | -1/+4 | |
Next time, remember to check with -Werror first... | |||||
2013-04-15 | Merge branch '2.4' | Petri Lehtinen | 1 | -1/+1 | |
2013-04-15 | Fix the signature of a test function | Petri Lehtinen | 1 | -1/+1 | |
2013-04-15 | Merge branch '2.4' | Petri Lehtinen | 1 | -0/+21 | |
2013-04-15 | json_dump*(): Don't crash if json is NULL and JSON_ENCODE_ANY is set | Petri Lehtinen | 1 | -0/+21 | |
2013-03-14 | Fix a refleak in test/bin/json_process.c | Petri Lehtinen | 1 | -0/+1 | |
2013-03-14 | Remove trailing whitespace | Petri Lehtinen | 1 | -16/+16 | |
2013-03-13 | Got rid of some warnings for unused variables. | Joakim Söderberg | 1 | -6/+6 | |
2013-03-13 | Add back the old ENV variabe approach to json_process. | Joakim Söderberg | 6 | -19/+141 | |
Enable using the old envrionment variable passing of settings for json_process so that the current autoconf test suite scripts can run nicely. json_process now takes an optional command line parameter --env which causes it to use the old method of reading the settings from environment variables instead of directly from the "env" file. Also added a --strip command line option, this will be used to run the stripped tests with CMake as well. | |||||
2013-03-07 | Get rid of type warning on MSVC. | Joakim Soderberg | 1 | -1/+1 | |
We should use json_int_t instead of int, since this can be different depending on platform. Also MSVC warns that this can cause "loss of information" since it's converting long long to int. | |||||
2013-03-07 | Fix incorrect cast to ssize_t. | Joakim Söderberg | 1 | -1/+0 | |
gcc 4.2.1 warns about a possible incorrect cast to ssize_t when comparing against refcount, which is of type size_t. signed vs unsigned comparison. Sinc warnings are treated as errors (as in the autoconf project) this would cause a compile error. | |||||
2013-03-07 | Always use LF as line ending so that we don't get any problems with line | Joakim Soderberg | 1 | -0/+2 | |
endings in any tests. | |||||
2013-03-07 | Consolidated the CMake project. | Joakim Soderberg | 3 | -102/+154 | |
- Moved everything to one CMakeLists.txt - Added support for the json_process test suites (instead of just the API tests). - Changed to use the modified json_process version that does away with the environment variables (originally written by DanielT). - Had to exclude "test_memory_funcs" on MSVC, since void pointer arithmetics are not allowed as it is done in secure_malloc and secure_free. - Had to add a check for "ssize_t". This is not available on Windows and maybe on some other platforms (used in test_pack.c) - Result from running ctest (The failure seems unrelated to CMake, it's just that the expected result is in a different order): 99% tests passed, 1 tests failed out of 121 Total Test time (real) = 1.31 sec The following tests FAILED: 24 - valid__complex-array (Failed) | |||||
2013-03-07 | Add CMake build system. | Paul Harris | 1 | -0/+31 | |
Added multiple CMake-related files to project. Supports building the library and the tests. See CMakeLists.txt for notes on how it works. I had to adjust 3 existing files in order to disable some configuration that should be taken care of by cmake/automake anyway. I also added jansson.def from a future jansson version, to test cmake's support for .def files (which works fine). | |||||
2013-02-05 | check-exports test: Filter out symbols that start with an underscore | Petri Lehtinen | 1 | -1/+1 | |
These are usually internal to libc, and such symbols are exported in the shared library on some platforms. | |||||
2013-02-04 | Enable gcc specific flags in tests correctly | Petri Lehtinen | 4 | -4/+2 | |
... and fix issues found. | |||||
2012-12-04 | Really fix the off-by-one error in json_array_remove() | Petri Lehtinen | 1 | -2/+2 | |
It didn't affect only the last element but all of them. One item too much was always moved. | |||||
2012-12-04 | Fix an off-by-one error in json_array_remove() | Petri Lehtinen | 1 | -0/+14 | |
Uninitialized memory was read when the last item of a "full" array was removed. | |||||
2012-10-26 | Support building and testing with VPATH. | Bert Wesarg | 2 | -2/+2 | |
2012-09-13 | Disallow NaN or Inf real values | Petri Lehtinen | 1 | -0/+31 | |
2012-08-11 | Merge branch '2.3' | Petri Lehtinen | 1 | -1/+1 | |
2012-08-11 | fix check-exports test on ppc642.3 | Alessandro Ghedini | 1 | -1/+1 | |
2012-07-30 | Add json_boolean() macro | Petri Lehtinen | 1 | -0/+16 | |
Mostly for symmetry reasons. Makes it easier e.g. to: int ok = 0; if(something) ok = 1; json_object_set_new(obj, "ok", json_boolean(ok)); Fixes #86. | |||||
2012-06-29 | Merge branch 'fix-slash' of git://github.com/jrbasso/jansson | Petri Lehtinen | 1 | -0/+25 | |
Fixes #81. | |||||
2012-06-28 | Created flag to dump escaping slash | Juan Basso | 2 | -1/+26 | |
2012-06-27 | Escaping the slash when dump | Juan Basso | 1 | -1/+1 | |
2012-06-06 | Merge branch '2.3' | Petri Lehtinen | 2 | -14/+14 | |
2012-06-06 | Bugfix: must cast void* to do pointer math | Paul Harris | 1 | -2/+2 | |
Fixes #78. | |||||
2012-06-06 | Bugfix: Changed all use of ssize_t to size_t. | Paul Harris | 1 | -12/+12 | |
ssize_t is only used in a test, and its comparing to a size_t variable. Fixes #77. | |||||
2012-04-29 | Support building on Windows | Petri Lehtinen | 1 | -91/+9 | |
Jansson now builds correctly with Visual C++ Express 2010. | |||||
2012-04-26 | Merge branch '2.3' | Petri Lehtinen | 1 | -1/+12 | |
2012-04-26 | Check for missing args and envvars in run-tests.sh | Petri Lehtinen | 1 | -1/+12 | |
Fixes #71. | |||||
2012-03-26 | Document and tweak json_load_callback() | Petri Lehtinen | 1 | -1/+1 | |
Change the return value of json_load_callback_t to size_t, as it's feels more correct. The callback should return (size_t)-1 on failure. Issue #57. | |||||
2012-03-25 | Ignore the binary test_load_callback | Rogerz | 1 | -0/+1 | |
2012-03-22 | Merge branch '2.3' | Petri Lehtinen | 1 | -2/+4 | |
2012-03-22 | Add json_load_callback to the list of exported symbols | Petri Lehtinen | 1 | -1/+2 | |
2012-03-22 | Add json_load_callback() | Rogerz Zhang | 2 | -0/+76 | |
2012-03-22 | Make test stripping locale independent | Petri Lehtinen | 1 | -2/+4 | |
2012-03-20 | Update copyright notices for 2012 | Petri Lehtinen | 21 | -23/+23 | |
2012-03-08 | Fix tests on shells that don't support the export FOO=bar syntax | Petri Lehtinen | 11 | -17/+28 | |