User reports showing regressions...
...for newly added compilers! Boost.Python has apparently regressed for vc-8.0 How can that be? -- Dave Abrahams Boost Consulting www.boost-consulting.com
David Abrahams writes:
...for newly added compilers! Boost.Python has apparently regressed for vc-8.0
How can that be?
Because they are not newly added? VC 8.0 was tested (and even specified as required) in 1.32.0: http://engineering.meta-comm.com/boost-regression/1_32_0/developer/summary_r... -- Aleksey Gurtovoy MetaCommunications Engineering
At 19:00 2005-08-12, you wrote:
....for newly added compilers! Boost.Python has apparently regressed for vc-8.0
How can that be?
that's getting some error I've never seen before... headers with and withOUT _DEBUG ??
-- Dave Abrahams Boost Consulting www.boost-consulting.com _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
At 01:40 2005-08-13, you wrote:
At 19:00 2005-08-12, you wrote:
....for newly added compilers! Boost.Python has apparently regressed for vc-8.0
How can that be?
that's getting some error I've never seen before... headers with and withOUT _DEBUG ??
to be a little more precise, there is (what is now) a logic error in boost/python/detail/wrap_python.hpp lines 23ff: #ifdef _DEBUG # ifndef BOOST_DEBUG_PYTHON # undef _DEBUG // Don't let Python force the debug library just because we're debugging. # define DEBUG_UNDEFINED_FROM_WRAP_PYTHON_H # endif #endif appears to include several of the "standard includes for the vc runtime libraries" notably crtdefs.h (indirectly). this header appears to check for inconsistancy in #including files with _DEBUG defined and not... the error shows uplater in use_ansi.h it appears as if one can mess around with a symbol called: _CRT_NOFORCE_MANIFEST in wrap_python.hpp and it will get rid of the error, I have no opinion yet whether the resulting code will actually work
-- Dave Abrahams Boost Consulting www.boost-consulting.com _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
At 04:16 2005-08-13, Victor A. Wagner Jr. wrote:
At 01:40 2005-08-13, you wrote:
At 19:00 2005-08-12, you wrote:
....for newly added compilers! Boost.Python has apparently regressed for vc-8.0
How can that be?
that's getting some error I've never seen before... headers with and withOUT _DEBUG ??
to be a little more precise, there is (what is now) a logic error in boost/python/detail/wrap_python.hpp
lines 23ff: #ifdef _DEBUG # ifndef BOOST_DEBUG_PYTHON # undef _DEBUG // Don't let Python force the debug library just because we're debugging. # define DEBUG_UNDEFINED_FROM_WRAP_PYTHON_H # endif #endif
appears to include several of the "standard includes for the vc runtime libraries" notably crtdefs.h (indirectly). this header appears to check for inconsistancy in #including files with _DEBUG defined and not... the error shows uplater in use_ansi.h
it appears as if one can mess around with a symbol called:
_CRT_NOFORCE_MANIFEST
in wrap_python.hpp and it will get rid of the error, I have no opinion yet whether the resulting code will actually work
Well, the tests apparently ran (see regression from RudbekAssociates Sat, 13 Aug 2005 09:48:32 +0000) it's now green for python under VC-8_0. I'm a bit concerned about the change for actual "release", it appears that a conflict will arise if the user defines _CRT_FORCE_MANIFEST, but I'll commit my changes anyhow so the tests start running when run from regression.py. Obviously, if you don't like them, you can change them to suit. Checking in wrap_python.hpp; /cvsroot/boost/boost/boost/python/detail/wrap_python.hpp,v <-- wrap_python.hpp new revision: 1.21; previous revision: 1.20 done
-- Dave Abrahams Boost Consulting www.boost-consulting.com _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"
participants (3)
-
Aleksey Gurtovoy -
David Abrahams -
Victor A. Wagner Jr.