[C++-sig] Re: building Boost 1.30 with gcc 3.2.2 on Linux (Mandrake 9.1), w/Python 2.3

David Abrahams dave at boost-consulting.com
Fri Jul 25 19:46:42 CEST 2003


Alex Martelli <aleaxit at yahoo.com> writes:

> I'm trying -- and failing miserably.  According to bjam -d+2 -a -q (run from 
> boost_1_30_0) the C++ compilation command being used is:
>
> g++  -c -Wall -ftemplate-depth-100  -DBOOST_PYTHON_DYNAMIC_LIB 
> -DBOOST_PYTHON_SOURCE  -g -O0 -fno-inline -fPIC   -I"libs/python/build"  -I 
> "/usr/local/include/python2.3" -I "/home/alex/pyr/boost_1_30_0"  -o 
> "libs/python/build/bin/libboost_python.so/gcc/debug/runtime-link-dynamic/shared-linkable-true/numeric.o"  
> "libs/python/build/../src/numeric.cpp"
>
> This first warns about a redefinition of _POSIX_C_SOURCE (in 
> /usr/local/include/python2.3/pyconfig.h:844:1 , already defined in 
> /usr/include/features.h:131:1 -- hopefully a bening warning...?).
>
> Then the killers - a slew of error messages about 
> /home/alex/pyr/boost_1_30_0/boost/python/converter/builtin_converters.hpp:106:
> of "parse error before & token" followed by "syntax error before & token", 
> then again the same two messages for three more tokens (const > }), then
> again for lines 107, then for line 112 (and also 113, 114, 115, ...) errors 
> that "to_python_value" is not a template, "detail" is not a class or 
> namespace, etc etc.  I've tried peering into the macros and templates 
> involved but so far haven't come anywhere close to any understanding.
>
> I hope it's just some silly error on my part -- any suggestions...?

Get the CVS;  1.30.0 doesn't work with Python 2.3.

HTH,
-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com





More information about the Cplusplus-sig mailing list