Boost logo

Boost :

From: John Maddock (john_at_[hidden])
Date: 2003-10-31 06:19:57


> //force the definition of INT64_C macro used in
> posix_time_system
> #ifndef __STDC_CONSTANT_MACROS
> #define __STDC_CONSTANT_MACROS
> # if __GNUC__ >= 2
> # undef BOOST_HAS_STDINT_H
> # endif
> #endif
> #include "boost/cstdint.hpp"
>
> Undefining BOOST_HAS_STDINT_H leads to numerous macros redefinition
warnings
> (essentially macros defined in stdint, are redefined in boost/cstdint).

Undefining BOOST_HAS_STDINT_H will surely have a negative impact on other
Boost libraries, if they are mixed with date_time: If there is a bug in
boost/stdint.hpp or boost/config.hpp then it should be fixed, not hacked
around like this. What was the original problem that led to this - the
regression tests are all passing for boost/stdint.hpp aren't they?

John.


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk