Re: [Boost-users] [UTF] Strange behaviour of unit_test_examples

Hi Gennadiy, Gennadiy Rozental wrote:

"Mateusz Loskot" wrote in message news:44064B20.4050504@loskot.net...

...

Then, I tried to run rest of examples, unit_test_example_02.cpp and unit_test_example_03.cpp, as explained above. And here comes my problems I noticed with unit_test_example_02 and unit_test_example_03. When I try to run compiled example from unit_test_example_02.cpp I get memory leaks detected: [...]

I couldn't explain what you are expiriencing. Just one suggestion: make sure you link with proper library. Autolinking inforce proper name. Try to disable it (define BOOST_TEST_NO_LIB) and name library explicetly.

I tested with autolinking disabled and I tried to use libboost_unit_test_framework-vc80-mt-*.lib libraries explicitly. I still get the same memory leaks dump. The only difference is that when I use UTF library built in debug mode (-gd) I get more info (here I test UTF example 03): D:\temp\utf_test>debug\utf_test.exe Running 2 test cases... ./utf_test.cpp(11): error in "force_division_by_zero": check false failed Detected memory leaks! ... ... here memory dump but when I use UTF library built in non-debug mode I get only: D:\temp\utf_test>debug\utf_test.exe Running 2 test cases... In both cases I get Windows dialog box with information about crash. I also rebuilt UTF library using -d2 bjam option and I compared compiler options used by bjam with options I use to build examples in VC++ 8.0. In both cases I use same options, double checked. Do you have any idea where may be the problem that I get this memory dump when I use UTF lib built with bjam? When I build UTF library with VC++ 8.0 project everything works well. Simply, UTF seems to not to catch bugs in test examples properly. Here, in example 03, I checked where debuger stops when I push Debug button in the crash dialog box: // reports 'unknown location(0): fatal error in "force_division_by_zero": integer divide by zero' i = i / j; // <------ HERE DEBUGER STOPS but it should be handled by UTF as described in comments above, shouldn't it? Thanks for any help Cheers -- Mateusz Łoskot http://mateusz.loskot.net