- Boost - lists.boost.org

Re: [errorinfo] A proposal for better error handling
by Don G 05 Apr '05

05 Apr '05

Jeff, thanks for the info. I've put a copy in the sandbox as well. Don --- Jeff Garland <jeff(a)crystalclearsoftware.com> wrote: > On Sun, 3 Apr 2005 22:44:40 -0700 (PDT), Don G wrote > > I have posted errorinfo.zip on the Yahoo files section for those > > wanting more detail. > > Hi Don - > > We are trying to get away from using Yahoo files since it > requires people to register. Could you put your files up > in the sandbox instead? You will have to register to post > files, but downloaders don't need to register. > The sandbox vault also has other nice features like a > download counter. > > Thx! > > Jeff > > http://boost-sandbox.sourceforge.net/vault/ __________________________________ Do you Yahoo!? Yahoo! Personals - Better first dates. More second dates. http://personals.yahoo.com

1 0

function arguments question
by Brock Peabody 05 Apr '05

05 Apr '05

I'm trying to build a library where the user provides a list of functions objects (or plain functions) and I need to extract the argument information from the types of these functions. Am I correct in believing that there is no generic way to do that with function objects? It will be easy to extract the signature from the boost::function objects, but I can't figure out how to extract argument information from function objects created by boost::bind. Is it possible? I guess it wouldn't be too hard to fall back on the old STL style typedefs (first_argument_type, second_argument_type) when possible and require the user to wrap functions with more parameters in a boost::function object otherwise. Would it be desirable to make the proposed function_types library work on function objects whose argument information can be queried, like the result_of template does for return types? Thanks, Brock

1 0

[filesystem] Can't compile on Tru64/gcc
by Markus Schöpflin 05 Apr '05

05 Apr '05

Hello, currently the filesystem library doesn't compile on Tru64 5.1 using gcc-3.4.3 as the compiler. schoepf@area51:~/net/src/boost/boost-HEAD/tools/regression/build> bjam -sTOOLS=gcc ...found 752 targets... ...using 3 temp targets... ...updating 4 targets... ...using <@boost!libs!filesystem!build/libboost_filesystem.a/gcc/release>exception.o... gcc-C++-action ../../../bin/boost/libs/filesystem/build/libboost_filesystem.a/gcc/release/operations_posix_windows.o /net/users/schoepflin/src/boost/boost-HEAD/libs/filesystem/build/../src/operations_posix_windows.cpp:123:39: operator '&&' has no right operand set -e "g++" -c -Wall -ftemplate-depth-255 -DNDEBUG -DNDEBUG -O3 -finline-functions -Wno-inline -I"../../../bin/boost/libs/filesystem/build" -I"/net/users/schoepflin/src/boost/boost-HEAD" -I "/net/users/schoepflin/src/boost/boost-HEAD" -o "../../../bin/boost/libs/filesystem/build/libboost_filesystem.a/gcc/release/operations_posix_windows.o" "/net/users/schoepflin/src/boost/boost-HEAD/libs/filesystem/build/../src/operations_posix_windows.cpp" The code in question is: # if defined(_POSIX_THREAD_SAFE_FUNCTIONS) \ && defined(_SC_THREAD_SAFE_FUNCTIONS) \ && _POSIX_THREAD_SAFE_FUNCTIONS >= 0 if ( ::sysconf( _SC_THREAD_SAFE_FUNCTIONS ) >= 0 ) { return ::readdir_r( dirp, entry, result ); } # endif Seems like _POSIX_THREAD_SAFE_FUNCTIONS but doesn't expand to anything. Markus

6 10