From: Vladimir Prus (ghost_at_[hidden])
Date: 2005-04-28 09:13:46
On Thursday 28 April 2005 17:53, David Abrahams wrote:
> > alias python
> > : <os>MACOSXX <toolset>darwin
> > :
> > : <framework>$(PYTHON_FRAMEWORK)
> > ;
> You do something similar in the NT case, so that probably needs to be
> fixed too. I checked in something that got things partially
> "working," but it's such a mess that I have no clue really what's
> supposed to be happening here. Also, the current testing/Jamfile.v2
> doesn't take into account the py-unit-test template from the v1
> Jamfile, resulting in, e.g.:
> call "C:\Program Files\Microsoft Visual Studio .NET
> 2003\Vc7\bin\vcvars32.bat" > nul cl /Zm800 -nologo -TP /Z7 /Od /Ob0 /EHsc
> /GR /MDd /Zc:forScope /Zc:wchar_t
> ..\..\..\libs\python\src\converter\type_id.cpp(121) : error C2491:
> 'boost::python::operator`<<'' : definition of dllimport function not
I think that eventually py-unit-test tempate from V1 should become
usage-requirements on '/python//python' target or on /boost//python target.
> > At least on Linux, as I understand, extensions should include Python
> > headers but don't link to Python interpreter.
> Sort of. Extension .so's will pick the interpreters symbols out of
> the executable itself.
> > So, the python_extension rule will add
> > /python//python_for_extensions to the passed sources.
> And what is the python alias supposed to do, then?
One can use it to link to the interpreter. Say, the 'embedding' test would
link to /python//python
-- Vladimir Prus http://vladimir_prus.blogspot.com Boost.Build V2: http://boost.org/boost-build2
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk