Unexpected exceptions no longer cause abort and there should be no more
potential for false negatives.
The trade-off now is that exceptions are no longer translated.
This is another warning that follows test macros, making it painful to
suppress without leaking outside. Luckily clang's `_Pragma`
implementation works.
Should fix#308
Effectively a revert of previous commit, fixing #542, where this was
added to stop linters complaining about `REQUIRE_THROWS_AS` used like
`REQUIRE_THROWS_AS(expr, std::exception);`, which would be slicing the
caught exception. Now it is user's responsibility to pass us proper
exception type.
Closes#833 which wanted to add `typename`, so that the construct works
in a template, but that would not work with MSVC and older GCC's, as
having `typename` outside of a template is allowed only from C++11
onward.
This seems to give about 15% speedup when compiling tests using GCC.
The tradeoff is that under certain circumstances, there is a chance for
false negative result, when the expression under test throws exception
and the test code catches it before it gets to the test runner.
Example:
``` cpp
TEST_CASE("False negative") {
try {
REQUIRE(throws() == "");
} catch (...) {}
}
```
This test case will succeed, reporting no assertions checked, instead of
failing as it would with `CATCH_CONFIG_FAST_COMPILE` disabled. However,
just removing the try-catch block inside client's code will fix this, so
it is worthwhile.
This change does not apply to CHECK* macros, because these are currently
specified as continuing on exception and thus need the local try-catch
to work as intended.
std::ifstream in libstdc++ contains a bug, where it sets errno to zero.
To work around it, we manually save the errno before using std::ifstream
in debugger check, and reset it after we are done.
We also preventively save errno before using sprintf.
Fixes#835
In some cases, like when given
```cpp
std::vector<char>* str =
reinterpret_cast<std::vector<char>*>(0x1234458);
CHECK(*str == std::vector<char>());
```
reconstructing the expression to report it would cause another fatal
error. Instead we just put together an AssertionResult without
reconstructing the expression fully.
This should fully fix#810
If the gcc version supports `_Pragma` properly, we use that to disable
it locally inside assertions.
Otherwise we disable it for the entire TU.
Fixes#674
Some versions of MinGW do not support enough of Win32 API to let us work
with SEH, so SEH is now MSVC only (+ configurable toggle).
Also made use of gmtime_s MSVC only (as oposed to Windows only).
Fixes#805
Because the signal changes were in a different branch from the windows.h
related changes, the SEH handling code included the header directly.
Fixes#803
This prevents Clang from complaining about unused value in expressions
containing explicit casts used in the THROW assertion macro family.
Example:
`REQUIRE_THROWS_AS(static_cast<bool>(object), std::bad_cast);` would
trigger `-Wunused-value` warning. Now it does not.
Credits to Arto Bendiken, who submitted a PR almost 3 years ago, but his
branch has since died and I was unable to merge it.
The integrated assembler segment was missing an underscore:
"_asm__". Also we remove the "DEBUG" macro check, so we are consistent
with the linux and windows variant.
Now breaking into gdb on failure should work via:
gdb --args test_executable --break
Using sizeof(expr) can trigger a compile-time error,
"lambda expressions are not allowed in an unevaluated context", when passing
expression containing lambda, like a std algorithm. This error is considered
a standard defect, as it is meant to prevent lambdas in decltype
or templates, but not in sizeof.
This reverts commit 227598af47.
In reality, this is a relatively small performance improvement,
especially with the previous improvements removing lots of superfluous
string handling, but still was measurable.
This is sane, because those `const char*`s are given to us by compiler,
from the text area and thus we do not have to manage their lifetimes. We
also never want to change them.
Also moved copy constructor to compiler-generated methods, not sure why
it wasn't -- even before it was the same as a compiler would generate.
This means that all tabs used in indentation are now 4 spaces and that
there should be no more trailing whitespace.
Ill also look into creating a pre-commit hook that will prevent this
from happening in the future.
Fixes#105
