I messed up the order of the different brackets and did not preview the
markdown before pushing. I'm very sorry.
This fix is tested in Atom's Markdown preview now and really fixes the
issue.
- added '--version' that prints information equivalent to "--build_info" and then exits
- prints the Boost.Test variant in use
- moved the definition of the BOOST_TEST_INCLUDED macro on the top of the files such that CLA can safely print the Boost.Test variant in use
- added smoke test
- Module name printing should use BOOST_TEST_MODULE (and not BOOST_TEST_MAIN): adding compilation check
* topic/11907-string-as-collection-error:
Change log and doc update
String comparison can now be used with BOOST_TEST additional compare options
Making std::string compatible with string comparison backend
# Conflicts:
# doc/closing_chapters/change_log.qbk
* topic/12531-run-test-with-colons:
Change log and breaking change annoucement
Apparently the name appears as "test<struct_ns__X>" in Visual (instead of "test<ns__X>" on gcc like)
Cleaning names with colons and fixing the proper demangling on all platforms
- replacing colons with underscores in the names cleaning
- names generated by template elements were not properly demangled in all platforms: using now the boost.core demangling functions (same as execution monitor)
- adding smoke tests and changing existing ones accordingly
- promoting std::string as a type convertible to basic_cstring
- changing the string comparison operator to account for basic_cstring convertible elements
- avoiding basic_cstring convertible elemets in collection comparison
- additional checks of the use of the adequate string comparator
- declaring string as a sequence for non C++11 compilers
* topic/11756-floatingpoint-exception-macros:
Some cleanup on the fpe symbols
Updated change log
trac 11756: fix usage of floating point exception macros
# Conflicts:
# doc/closing_chapters/change_log.qbk
* topic/12507-broken-report-sink:
Changelog update
More smoke tests
Bug fix default argument
Adding smoke tests for the loggers issue
Bugfix: cast to wrong type
A fully standards conforming fenv.h does not have to define any FE_* macros, and if it does define them,
then it defines macros only for the FP exceptions it actually supports.
Correct usage requires a triple check:
1) Check BOOST_NO_FENV_H to see if the header is supported.
2) Include the header and then check FE_ALL_EXCEPT to see
if any FP exceptions are supported.
3) Before using the individual FE_* macros, you need to
check for their existence too as not all may be
supported.
The reason is that some (notably FPU-less) architectures,
including mips*-nf, don't define/implement some of the
floating point constants, even though fenv.h is
available.
Other projects have similar issues, e.g. pixman, and apply similar work-arounds:
https://lists.freedesktop.org/archives/pixman/2014-February/003172.html
Architectures are notably also allowed to define FE_ALL_EXCEPT to 0!
Keeping this in mind, and knowing that the compiler will eliminate
code that can't be executed, we can change BOOST_FPE_ALL to be 0 for
the case of compiling using Clang and/or fenv.h being unavailable
as well, which allows simplification of the #ifdef's in
execution_monitor.ipp a bit.
