-
matt_hargett committed revision 172 to the bugreport SVN repository, changing 1 files.
2009-11-15 01:00:00 UTC in bugreport
-
matt_hargett committed revision 171 to the bugreport SVN repository, changing 38 files.
2009-11-15 00:53:42 UTC in bugreport
-
matt_hargett committed revision 170 to the bugreport SVN repository, changing 29 files.
2009-09-11 22:28:12 UTC in bugreport
-
To keep it as simple (and compatible with current sources) as possible, an environment variable called CGREEN_PER_TEST_TIMEOUT is checked. (Because users write their own main function, this seemed the best way to make sure the feature works consistently even if they forget to add/call argv parsing.) If the variable is present, and the value is valid, it is passed to die_in() before each test...
2009-08-07 23:29:58 UTC in Cgreen
-
On Ubuntu 9.04 on amd64, with current SVN, we are getting odd output when using assert_equal() on uint64_t variables (as defined in stdint.h)
This code:
uint64_t long1 = 0x100000000;
assert_equal(long1, 0);
produced this output:
test.cpp:355 - Failure!: tests -> uint64_test -> [0] should match [0] at [test.cpp] line [55]
There are other variations as well,
2009-06-16 20:18:58 UTC in Cgreen
-
In order to drive our implementation of a SocketStream class with tests, we needed to be able to go from
"foo(NULL);"
to
"foo(&localVariable);"
by driving it with a failing test. To do this, we needed to be able to set our mocked "foo" function with the expectation that its argument would be non-NULL. Since the argument is a local variable reference, we can't use a specific hard-coded...
2009-04-27 20:43:01 UTC in Cgreen
-
We put a patch in the tracker that fixed a few compiler warnings (gcc 4.2.4), but here are a few whose fix might be non-trivial. Someone who knows has better knowledge of the code will probably be able to fix these in a simpler/saner way than in our attempts.
src/reporter.c: In function ‘create_reporter’:
src/reporter.c:34: warning: assignment from incompatible pointer type...
2009-04-07 23:13:05 UTC in Cgreen
-
forgot to add #include at the top of messaging.c. please correct when/if you apply the patch.
thanks!
2009-04-07 23:04:43 UTC in Cgreen
-
This may not be the 100% correct fix for this memory leak. It does fix the leak in our tests, as well as cgreen's own tests. We have a minor concern about some usages of adding multiple suites that might introduce a double-free scenario. That being said, we really tried to evoke that situation and never could.
2009-04-07 23:02:12 UTC in Cgreen
-
This is to enable expected exceptions patterns of this style:
try {
somethingThatShouldThrow();
fail();
}
catch(SpecificException& e) {
assert_string_equals(e.Message(), "invalid operation");
}.
2009-04-07 22:59:59 UTC in Cgreen
