Re: [boost] Proposed Boost config macro

2002-12-09 Thread John Maddock
BOOST_UNREACHABLE_RETURN(0) or BOOST_UNREACHABLE_RETURN() I like it! Ditto, added: can you add a define for BOOST_NO_UNREACHABLE_RETURN_DETECTION to the compiler configs for any compilers that require the BOOST_UNREACHABLE_RETURN workaround to be activated (none of those I have access

Re: [boost] Proposed Boost config macro

2002-12-07 Thread David Abrahams
John Maddock [EMAIL PROTECTED] writes: If this fails to compile, we may need to add a dummy return 0; at the end. I know of two compilers which act this way: Metrowerks and HP. What do you want to call it? I'm not happy with it, but... BOOST_NO_NON_RETURN_PATH_COMPREHENSION is the best I

Re: [boost] Proposed Boost config macro

2002-12-07 Thread Mark Rodgers
From: David Abrahams [EMAIL PROTECTED] John Maddock [EMAIL PROTECTED] writes: If this fails to compile, we may need to add a dummy return 0; at the end. I know of two compilers which act this way: Metrowerks and HP. What do you want to call it? I'm not happy with it, but...

Re: [boost] Proposed Boost config macro

2002-12-07 Thread David Abrahams
Mark Rodgers [EMAIL PROTECTED] writes: From: David Abrahams [EMAIL PROTECTED] John Maddock [EMAIL PROTECTED] writes: If this fails to compile, we may need to add a dummy return 0; at the end. I know of two compilers which act this way: Metrowerks and HP. What do you want to call it?

[boost] Proposed Boost config macro

2002-12-03 Thread David Abrahams
I'd like a macro to detect that a compiler requires a return value even when no paths return from a function. Example: int f() { throw hello; }; If this fails to compile, we may need to add a dummy return 0; at the end. I know of two compilers which act this way: Metrowerks