- Boost-users - lists.preview.boost.org

Problems compiling boost.python with Boost 1.31.0
by Markus Sch�pflin 17 Apr '04

17 Apr '04

I'm trying to compile boost.python from Boost 1.31.0 on Tru64 5.1 with Compaq C++ V6.5-033. Python has been built like this: ./configure --with-cxx=cxx --prefix=/opt --exec-prefix=/opt/exec/OSF1-V5.1-a lpha --without-gcc The command line used to build boost.python looks like this: bjam -sTOOLS=tru64cxx65 -sPYTHON_ROOT="/opt" -sPYTHON_VERSION="2.3" stage When build boost.python I get the following error message: tru64cxx65-C++-action bin/boost/libs/python/build/libboost_python.so/tru64cxx65/debug/numeric.o cxx: Error: /net/camos/opt/include/python2.3/Python.h, line 8: #5 could not open source file "pyconfig.h" #include "pyconfig.h" ---------------------^ It looks like the exec prefix is not taken into account because the file is located here: /opt/exec/OSF1-V5.1-alpha/include/python2.3/pyconfig.h Is this not supported by the current boost build process or is there any option I missed? Or is there anything else I'm doing wrong? Thanks, Markus

3 3

broken link in boost.org FAQ
by Mike Morearty 17 Apr '04

17 Apr '04

Minor problem: on the FAQ at http://www.boost.org/more/faq.htm , the very first question is, "How do I download the libraries?" and the answer is, "See Download instructions." But "Download" is a broken link. Points to http://www.boost.org/libs/libraries.htm#Download ; although http://www.boost.org/libs/libraries.htm does exist, it does not contain an anchor called "Download". The FAQ should probably be changed to point to http://www.boost.org/more/getting_started.html , or maybe to http://sourceforge.net/project/showfiles.php?group_id=7586 - Mike Morearty

1 0

[MPL] Runtime Interaction
by Debian User 16 Apr '04

16 Apr '04

[Sent via gmane but it never appeared] Hi gang, Suppose I have a number of types that each have a "value" enum, much like MPL's constant classes. Now suppose that I want to construct a runtime-indexable array of those values, according to the position of each type in an MPL vector. For example: int values[3]; typedef boost::mpl::vector_c<int,10, 11, 12> values; typedef some_magic_metafunction<values>::type runtime_function; // values[0] = 10, values[1] = 11, values[2] = 12 runtime_function::fill_array(values); The idea is that some_magic_metafunction<values> would generate a type kind of like this: template<typename TypeValueList, typename Index> struct magic { static void fill_array(int *v) { // Non-MPL type vector indexing -- could use an // MPL iterator here v[Index::value] = TypeValueList<Index::value>::value; if (Index::value < distance<begin<TypeValueList>::type, end<TypeValueList>::type>::type::value) { magic<TypeValueList, Index::next>::fill_array(v); } }; }; Get it? The problem is that I don't know how to make MPL transition from strictly compile-time to something in-between. It can statically know the size of the value array and the values themselves but it has to fill the array "values" at runtime so that the values may be indexed by runtime computations. Any guidance? Thanks! -Dave

2 1

problems using VC 7.1 and boost regex library
by Keith Smith 15 Apr '04

15 Apr '04

I have been trying to compile the following code with VC 7.1 ( which compiled fine with VC 6 ) with no luck. std::string m_rBuffer; boost::regex m_rStartExp; string::const_iterator rRawDataBegin = m_rBuffer.begin(); string::const_iterator rRawDataEnd = m_rBuffer.end(); boost::cmatch rStartMessage; unsigned int flags = boost::match_default; while ( boost::regex_search(rRawDataBegin, rRawDataEnd, rStartMessage, m_rStartExp, flags) ) Included is the compiler error report. Please throw me a bone.

2 1

BGL linking on Win32
by maxx＠dgap.mipt.ru 15 Apr '04

15 Apr '04

Hi! whenever I use libbgl-viz.lib to read graphviz files, I get a linker errors like this: msvcprt.lib(MSVCP71.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@ABV01@@Z) already defined in Graphviz.obj It was Ok to add "/NODEFAULTLIB:msvcrt" to the compiler options, but now I need that library! So if I don't add this option, I get errors like the one above, if I do add it, I get errors "unresolved symbols"... What is probably(or definitely) wrong? Thanks in advance, -- Maxim

2 1

RE: [Boost-users] Changing the heap allocator
by Jim.Hyslop 14 Apr '04

14 Apr '04

David Barrett-Lennard wrote: > Yes HeapSetInformation() will do it once you have the heap handle. > However, the CRT creates its own heap with a call to HeapCreate(), and > stores the heap handle in a global variable called _crtheap > that is not > exported by the library. Right, of course - I thought that GetProcessHeap() would return the correct heap, but it does not. -- Jim Hyslop Senior Software Designer Leitch Technology International Inc. (http://www.leitch.com) Columnist, C/C++ Users Journal (http://www.cuj.com/experts)

1 0

RE: [Boost-users] Changing the heap allocator
by Jim.Hyslop 14 Apr '04

14 Apr '04

David Barrett-Lennard wrote: > Does anyone know how to enable the LFH (low fragmentation > heap) for the > heap created by the VC CRT? Have you tried HeapSetInformation()? -- Jim Hyslop Senior Software Designer Leitch Technology International Inc. (http://www.leitch.com) Columnist, C/C++ Users Journal (http://www.cuj.com/experts)

2 1

Changing the heap allocator
by David Barrett-Lennard 13 Apr '04

13 Apr '04

Hi, I'm working on a project where I've replaced Microsoft's (terrible) heap allocator with the Doug Lea allocator. There seems to be some cases where the Windows XP allocator is 500 times slower than it should be. This made my program take minutes rather than seconds to run. To try to understand what kind of allocation patterns cause the behaviour I wrote a small console application (see below). After playing around with different allocation patterns for half an hour I found a case where the pathological behaviour appeared. I'm none the wiser about the problem. For example, moving the malloc(301) statement before the 7 x malloc(1000) makes the problem go away! However, given the ease with which I rediscovered the problem, I wonder why there aren't more complaints about Microsoft's heap allocator. With some allocation patterns I have seen the allocator fall below 1kHz on a Pentium IV 2GHz machine! I've created a library that exports alternative versions of operator new and operator delete, by using a def file as follows EXPORTS ??2@YAPAXI@Z ??3@YAXPAX@Z I've had to do some hacky things to make it work with the STL (because of msvcp60.dll) , but that's another story! Now my question : I want to use boost, but need to work out how to make it use a different allocator. Will I need to rebuild the boost dlls (such as boost_python.dll), against my static library? Regards, David Barrett-Lennard /////////////////////// heap allocation test /////////////////////////// #include <iostream> #include <time.h> int main(int argc, char* argv[]) { for (int count=1 ; count <= 5 ; ++count) { int n1 = 0, n2 = 0; clock_t start = clock(); { const int NUM = 10000; void* L[NUM*7]; for (int i=0 ; i < NUM ; ++i) { for (int j=0 ; j < 7 ; ++j) L[n1++] = malloc(1000); void* leak = malloc(301); ++n2; } for (int k=n1-1 ; k >= 0 ; --k) free(L[k]); } clock_t elapsed = clock()-start; if (elapsed) { std::cout << "Time = " << elapsed << " ms" << " Rate = " << 1000 * (n1 + n2) / elapsed << " Hz" << std::endl; } } return 0; }

3 4

Using boost::regex 1.31.0 on vc6
by Chris Branch 10 Apr '04

10 Apr '04

Hi, I'm trying to use boost::regex 1.31.0 (the v4 source code) on MSVC++6 (sp5), but whenever I compile a project using the new code I get an internal compiler error in perl_matcher_common.hpp and no object is built. I want to use the new v4 code (the older v3 code works, and I'll use that if I have to) but this file is quite important... is there something I can do, or does it not work with VC6? Thanks, Chris

2 2

boost::thread and boost::bind
by GoochRules! 10 Apr '04

10 Apr '04

I am attempting to do this in vs.net 2003: int f(HANDLE, HANDLE, HANDLE); int main(int argc, char* argv[]) { HANDLE x; HANDLE y; HANDLE z; boost::thread* p = new boost::thread(boost::bind(f, x, y, z); p->join(); delete p; } As a contrived example it works. In my project it does not, spewing out 5 errors: 1) c:\src\include\boost\bind.hpp(62) : error C2825: 'F::result_type': cannot form a qualified name c:\src\include\boost\bind\bind_template.hpp(16) : see reference to class template instantiation 'boost::_bi::result_traits<R,F>' being compiled with [ R=boost::_bi::unspecified, F=int (__cdecl *)(HANDLE,HANDLE) ] c:\src\ansa-1.0\src\main\main.cpp(369) : see reference to class template instantiation 'boost::_bi::bind_t<R,F,L>' being compiled with [ R=boost::_bi::unspecified, F=int (__cdecl *)(HANDLE,HANDLE), L=boost::_bi::list3<boost::_bi::list_av_3<fd_xpt,fd_xpt,fd_xpt>:: B1,boost::_bi::list_av_3<fd_xpt,fd_xpt,fd_xpt>::B2,boost::_bi:: list_av_3<fd_xpt,fd_xpt,fd_xpt>::B3> ] 2) c:\src\include\boost\bind.hpp(62) : error C2039: 'result_type' : is not a member of 'operator``global namespace''' 3) c:\src\include\boost\bind.hpp(62) : error C2146: syntax error : missing ';' before identifier 'type' 4) c:\src\include\boost\bind.hpp(62) : error C2955: 'boost::_bi::type' : use of class template requires template argument list c:\src\include\boost\bind.hpp(94) : see declaration of 'boost::_bi::type' 5) c:\src\include\boost\bind.hpp(62) : fatal error C1903: unable to recover from previous error(s); stopping compilation note: fd_xpt is declared as "typedef HANDLE fd_xpt;", and c:\src\ansa-1.0\src\main\main.cpp(369) is similar to "boost::thread* p = new boost::thread(boost::bind(f, x, y, z);" in the example. It appears that the compiler thinks my first argument is a "int (__cdecl *)(HANDLE,HANDLE)" when it is most certainly a "int (__cdecl *)(HANDLE,HANDLE,HANDLE)". It used to take two arguments, but I've since added a third (it was at this very point that these errors crashed the party). I am sure that there is no declaration that specifies only two arguments: I've had two other developers verify this. I have cleaned my project multiple times, even by hand, as well as getting a clean source tree from cvs, and yet I still get these errors. Any suggestions on how to silence these errors are greatly appreciated. --mpeltzer goochrules(a)hotmail.com

3 3