> On 04/24/2012 11:59 AM, Daniel d'Andrada wrote:
> > ADD_FAILURE() does print the file name and line number but it does *not*
> return. Therefore a wrapping macro is still needed. So I removed the printf()
> command and renamed the macro.
>
> Ahh, I remembered wrong. You want FAIL() << [message]; It's just like
> ADD_FAILURE() except that it causes the function to return. It's
> analogous to ASSERT_*() vs EXPECT_*().
>
> http://code.google.com/p/googletest/wiki/AdvancedGuide#Explicit_Success_and_Failure
FAIL(), like ASSERT_*(), can only be used from within functions that return void. That's why I have to stick with ADD_FAILURE().
>
> > I've removed the --std=c++0x flag from gtest
>
> gtest_geis2_grail_backend_CPPFLAGS still has --std=c++0x.
I needs --std=c++0x because it uses c++0x stuff (like nullptr and range-based for loops)
> On 04/24/2012 11:59 AM, Daniel d'Andrada wrote: code.google. com/p/googletes t/wiki/ AdvancedGuide# Explicit_ Success_ and_Failure
> > ADD_FAILURE() does print the file name and line number but it does *not*
> return. Therefore a wrapping macro is still needed. So I removed the printf()
> command and renamed the macro.
>
> Ahh, I remembered wrong. You want FAIL() << [message]; It's just like
> ADD_FAILURE() except that it causes the function to return. It's
> analogous to ASSERT_*() vs EXPECT_*().
>
> http://
FAIL(), like ASSERT_*(), can only be used from within functions that return void. That's why I have to stick with ADD_FAILURE().
> grail_backend_ CPPFLAGS still has --std=c++0x.
> > I've removed the --std=c++0x flag from gtest
>
> gtest_geis2_
I needs --std=c++0x because it uses c++0x stuff (like nullptr and range-based for loops)