Hi, The header <boost/config/select_platform_config.hpp> contains the lines #elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp" Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over? Jonathan
Jonathan Turkanis wrote:
Hi,
The header <boost/config/select_platform_config.hpp> contains the lines
#elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp"
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Jonathan _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost Some cross-compilers, I think, depending on the configuration. Probably only historical reasons now.
Simon Buchan wrote:
Jonathan Turkanis wrote:
Hi,
The header <boost/config/select_platform_config.hpp> contains the lines
#elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp"
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Some cross-compilers, I think, depending on the configuration.
Thanks, I didn't think of that.
Probably only historical reasons now.
-- Jonathan Turkanis www.kangaroologic.com
The header <boost/config/select_platform_config.hpp> contains the lines
#elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp"
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Borland. John.
John Maddock wrote:
The header <boost/config/select_platform_config.hpp> contains the lines
#elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp"
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Borland.
John.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Last I checked (BCC 4.something!) it did as well.
John Maddock wrote:
The header <boost/config/select_platform_config.hpp> contains the lines
#elif defined(_WIN32) || defined(__WIN32__) || defined(WIN32) // win32: # define BOOST_PLATFORM_CONFIG "boost/config/platform/win32.hpp"
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Borland.
Right, I forgot to mention it. But I also tested it (5.6.4) and it defines _WIN32 too.
John.
Jonathan
Jonathan Turkanis wrote:
John Maddock wrote:
Does anyone know why it's necessary to test __WIN32__ and WIN32? A quick test showed that vc6-7.1, cw8-9.4, como 4.3.3, gcc 3.4.2 (minGW), dmc, bcc 5.6.4, and intel 7.1-8.0 all define _WIN32. What's left over?
Borland.
Right, I forgot to mention it.
Hmmm .... No I didn't: it's in my original list ;-)
But I also tested it (5.6.4) and it defines _WIN32 too.
Jonathan
participants (3)
-
John Maddock -
Jonathan Turkanis -
Simon Buchan