In my particular case, I am working with a domain specific language of sorts implemented in C++. Lots of operator overloading, including relational ops. These will inevitably propagate exceptions in certain cases.
In a general case and IMHO , I think that coding catch(...) without a preceding catch(std::exception & e) is always bad form. On Dec 14, 2009, at 2:30 PM, Patrick Johnmeyer wrote: > > On Mon, Dec 14, 2009 at 1:45 PM, liam mail <liam.l...@googlemail.com> wrote: > > 2009/12/14 Vlad <v...@demoninsight.com> > > In my case, the message would be simply > > Unhandled exception in CHECK_EQUAL(A,B) > > IIRC That is an exception being thrown from the equality check not the actual > operation under test. The operation under test is scoped inside a try catch > block which does output the result of what if it is derived from > std::exception. > > That is correct, this is the error you get if the comparison operation throws. > > The error "Failure in VerifyBlahBlah: Unhandled exception: HOO-HA!" results > from the test-wide try/catch block when evaluating A or B throws, but has the > disadvantage of not knowing the file / line of the failure. > > The error "Unhandled exception in CHECK_EQUAL(A,B)" when operator== throws is > both simultaneously more informative (includes file, line, expected, and > actual -- unless of course streaming expected or actual crashes as well) and > less informative (doesn't include the exception string). > > It would be nice if the std::exception special case was in all UT++ try / > catch blocks so that in the latter case you'd get more info. Sounds like a > feature request / patch. > > All that being said, typically operator== is written to be a non-throwing > function, so I would think the latter message would be more than sufficient > to indicate that there's a serious problem. > > > ------------------------------------------------------------------------------ > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/google-dev2dev > _______________________________________________ > unittest-cpp-devel mailing list > unittest-cpp-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/unittest-cpp-devel
------------------------------------------------------------------------------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev
_______________________________________________ unittest-cpp-devel mailing list unittest-cpp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/unittest-cpp-devel
