- Boost-users - lists.boost.org

[Python] rule python-extension unknown
by Robert MacHardy 09 Oct '08

09 Oct '08

Boost Users, I am trying to build the Hello World python example. When I execute bjam I get the "rule python-extension unknown" error. I believe this to be because my user-config.jam file is never being used. The tutorial states to configure the rules in the user-config.jam file in my home directory. My home directory is "C:\Documents and Settings\robert.machardy". I did not find the file there, and this does not appear to be the place for jam files. I found the file user-config.jam in "C:\Program Files\boost\boost_1_36_0\tools\build\v2", I edited it and copied to my home directory. I configured it for the following: # Configure specific msvc version (searched for in standard locations and PATH). using msvc : 7.1 ; using python : 2.6 : C:\Program Files\Python26; Please help. Thank you, Robert MacHardy

4 7

Re: [Boost-users] boost::variant construction,assignment
by Hicham Mouline 09 Oct '08

09 Oct '08

Thank you. I apologize, one often fails to grasp the obvious, regards, > ----- Original Message ----- > From: "Steven Watanabe" <watanabesj(a)gmail.com> > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] boost::variant construction,assignment > Date: Thu, 09 Oct 2008 10:06:05 -0700 > > > AMDG > > Hicham Mouline wrote: > > const R* r_; > > > > > > *r_ = rhs; > > > > > > There is no operator= for variant<> it seems? > > > > Sure there is, but only works for non-const variant<>s. > > In Christ, > Steven Watanabe

1 0

boost::program_options
by Duy Nguyen 09 Oct '08

09 Oct '08

When I try to use the boost::program_options library to read a property file, I got the error when I run "make" on Cygwin (I installed Cygwin on WIndows XP). > make g++.exe -I /cygdrive/c/MyTest/Test1/include/boost-1_34_1/ -c TestInit.cpp Test.cpp TestDriver.cpp g++.exe -I /cygdrive/c/MyTest/Test1/include/boost-1_34_1/ -c TestInit.cpp Test.cpp TestDriver.cpp g++.exe -I /cygdrive/c/MyTest/Test1/include/boost-1_34_1/ -c TestInit.cpp Test.cpp TestDriver.cpp g++.exe TestInit.o Test.o TestDriver.o -o "TestDriver" -L/cygdrive/c/MyTest/Test1/lib -lboost_program_options-gcc34-mt-d-1_34_1 -lg2c Info: resolving vtable for boost::program_options::variables_mapby linking to __imp___ZTVN5boost15program_options13variables_mapE (auto-import) Then when I run my program, it does not work. If I comment out the "try ... catch" block which uses the boost::propram_options library then it compiles and works fine. Please show me how to fix it. Thank you.

6 5

shared_ptr and std::set::erase
by Javier Jerónimo 09 Oct '08

09 Oct '08

Hello, I have: std::set< boost::shared_ptr< A > > m_set; and when I try to do: m_set.erase( somesharedptr ); the shared_ptr is not removed from the set. There is already a shared_ptr object pointing to the same A object inside the set. I think it's because the definition of "shared_ptr<>::operator<" and its use in std::set<>::erase. What's the correct way of using a shared_ptr inside a set? Thanks.

3 2

Re: [Boost-users] [boost-users][serialization] porting to windows ce?
by Bobby Nations 09 Oct '08

09 Oct '08

Ack, I found a spot where I fat-fingered the code at fault when writing the original message. Please replace ::boost::detail::false_t value; With typename ::boost::detail::false_t value; It doesn't seem to make a difference in the actual error messages themselves, but typos are never good. In case anyone asks, I'm trying to compile against the 1.34.1 release using Visual Studio 2005 SP1 to compile the source via the solution and project files found in "boost\libs\serialization\vc7ide". I also have a patch file that has all of my changes if it might help. Granted it's pretty big (~12 meg) because VS2005 upgraded all of the project files, but I can make it available if it would help. Thanks, Bobby Nations

1 0

[multi-index] Segmentation fault during erase
by Joshua Moore-Oliva 09 Oct '08

09 Oct '08

(I am using 64-bit boost 1.33.1, gcc version 4.1.2) I am trying to track down an elusive segmentation fault that occurs after days of running a program. I am wondering if I am using the boost multi-index library incorrectly. I have a coredump that gives the following backtrace (full backtrace at the end of this email). at /usr/include/boost/multi_index/detail/ord_index_node.hpp:274 a274 while(x!=root&&(x==0 || x->color()==black)){ having looked at the code in that function, I can only assume it is called as a result of these two following lines: (Attempting to erase the first member of the index). CallQueue::nth_index<CALL_NEXT_CALL_TIME>::type & timeIndex = this->m_calls.get<CALL_NEXT_CALL_TIME>(); timeIndex.erase( timeIndex.begin() ); I am certain that the queue is not empty, as the following methods guard against an empty queue. (Shown in an excerpt of my relevant code at the bottom of this email). Finally, my definition for the m_calls structure and #defines for the indexes follow at the end of this email. My program is very clean memory wise, having run a fairly extensive test suite through valgrind, so the possibility of other memory errors compounding are low. My main question is if I am using the multi-index library incorrectly, or if there is a bug in the multi-index library. Failing that, any insight as to why I would be getting this error is appreciated :). Thanks, Josh %%%START m_calls structure%%% typedef boost::multi_index::multi_index_container< Call * , boost::multi_index::indexed_by< boost::multi_index::ordered_unique< BOOST_MULTI_INDEX_CONST_MEM_FUN( Call , int , getId ) > , boost::multi_index::ordered_non_unique< BOOST_MULTI_INDEX_CONST_MEM_FUN( Call , const boost::posix_time::ptime & , getNextCallTime ) > > > CallQueue; CallQueue m_calls; %%%END m_calls structure%%% %%%START #defines for m_calls indexes%%% #define CALL_ID 0 #define CALL_NEXT_CALL_TIME 1 %%%END #defines for m_calls indexes%%% %%%START MY CODE%%% Call * PhoneList::getNextCall() { if ( this->m_calls.empty() ) { throw NoAvailableCallException(); } CallQueue::nth_index<CALL_NEXT_CALL_TIME>::type::iterator pos = this->m_calls.get<CALL_NEXT_CALL_TIME>().begin(); return *pos; } Call * PhoneList::popAvailableCall() { Call * c = this->getNextCall(); if ( c->getNextCallTime() > LocalTime::getInst()->getCachedTime() ) { throw NoAvailableCallException(); } CallQueue::nth_index<CALL_NEXT_CALL_TIME>::type & timeIndex = this->m_calls.get<CALL_NEXT_CALL_TIME>(); timeIndex.erase( timeIndex.begin() ); /*************BELIEVED THAT ERROR ORIGINATES HERE****************/ return c; } %%%END MY CODE%%% %%%START GDB BACKTRACE%%% #0 0x000000000044a7fd in Database::AQM::QueueEngine::PhoneList::popAvailableCall (this=<value optimized out>) at /usr/include/boost/multi_index/detail/ord_index_node.hpp:274 a274 while(x!=root&&(x==0 || x->color()==black)){ (gdb) backtrace #0 0x000000000044a7fd in Database::AQM::QueueEngine::PhoneList::popAvailableCall (this=<value optimized out>) at /usr/include/boost/multi_index/detail/ord_index_node.hpp:274 #1 0x000000000044664c in Database::AQM::QueueEngine::Controller::getCall (this=0x6d54d0) at Controller.cpp:83 #2 0x0000000000411d55 in Database::AQM::Controller::getCall (this=<value optimized out>) at Controller.cpp:75 #3 0x000000000041169b in Control::AQM::Controller::getCall (this=<value optimized out>) at Controller.cpp:27 #4 0x000000000040a01b in Boundary::AQM::RequestSocket::Controller::getCall (this=<value optimized out>) at Controller.cpp:43 #5 0x000000000045a703 in Boundary::AQM::RequestSocket::GetCallRequest::perform (this=<value optimized out>) at GetCallRequest.cpp:17 #6 0x0000000000410564 in Boundary::AQM::RequestSocket::RequestSocket::OnLine (this=0x6d31f0, l=@0x6d3340) at RequestSocket.cpp:30 #7 0x0000000000483829 in TcpSocket::ReadLine (this=0x6d31f0) at TcpSocket.cpp:810 #8 0x0000000000480a9c in SocketHandler::Select (this=0x42802b90, tsel=0x42802b70) at SocketHandler.cpp:395 #9 0x0000000000482e25 in SocketHandler::Select (this=0x2aaab04c8b98, sec=<value optimized out>, usec=1297040160) at SocketHandler.cpp:227 #10 0x000000000040ee6b in Boundary::AQM::RequestSocket::RequestPoller::run (this=0x2aaab0000e40) at RequestPoller.cpp:84 #11 0x00002aaaaad15c3f in boost::function0<void, std::allocator<boost::function_base> >::operator() () from /usr/lib64/libboost_thread.so.2 #12 0x00002aaaaad156bf in boost::thread_group::join_all () from /usr/lib64/libboost_thread.so.2 #13 0x0000003aa56062f7 in start_thread () from /lib64/libpthread.so.0 #14 0x0000003aa4ace86d in clone () from /lib64/libc.so.6 #15 0x0000000000000000 in ?? () %%%END GDB BACKTRACE%%%

2 1

Question on boost multi-index container erase method
by Ramesh 09 Oct '08

09 Oct '08

Hello Joaquin, I am trying to erase a record in the container based on keys - in my case both the keys are ordered and unique and I need to have ways to delete records based on either of the keys. I tried to erase an element based on key2 & I got the following error: "error: no matching function for call to 'boost::multi_index ....<Error runs for more than a page> :erase(boost::multi_index::detail::index_iterator <boost::multi_index::detail::ordered_index_node<boost::multi_index::detail ::index_node_base<MyMap> > >&)' The compiler is perhaps complaining that I havent provided a prototype for erase in the map? In my case I wish to erase the record pointed by the iterator (In another class I had a single index and I didnt have the need to provide a prototype, is it because I have two 'unsigned long' keys that the compiler expects a prototype? )- Kindly help me fix this error. My code is posted as below: Thanks /R --- //fwd decl class A; struct MyMap { unsigned long key1; unsigned long key2; A *pObj; MyMap(unsigned long param_key1, unsigned long param_key2, A *param_pObj): key1(param_key1), key2(param_key2), pObj(param_pObj){} ~LCMap(){ key1 = 0; key2 = 0; pObj = 0; } }; struct key1{}; struct key2{}; typedef multi_index_container< MyMap, indexed_by< ordered_unique< tag <key1>, BOOST_MULTI_INDEX_MEMBER(MyMap, unsigned long, key1)>, ordered_unique< tag <key2>, BOOST_MULTI_INDEX_MEMBER(MyMap, unsigned long, key2)> > >tMapDef; class MyWrap { private: MyMap MyContainer; .. public: insert(UINT32, UINT32, A *); erasebykey2(unsigned long); }; bool MyWrap::erasebykey2(unsigned long param_key2) { bool ret = FALSE; typedef tMapDef::index<key2>::type tKey2Type; tKey2Type& key2_index = MyContainer.get<key2>(); tMapDef::index_iterator<key2>::type it = key2_index.find(param_key2); if (it == key2_index.end()) { cout << "No matching Recs with key2" <<endl<<; return FALSE; } tMapDef::size_type status = MyContainer.erase(it); }

2 3

[python] Adding a path to sys.path
by Robert Dailey 08 Oct '08

08 Oct '08

Hi, I was wondering if the code below is correct for adding a path to sys.path in Python. Note that this will occur before any modules have been executed/imported through Boost.Python. I'm also embedding the interpreter, which is why I am doing this. I'm not sure how it can obtain the sys.__dict__ when I haven't imported sys, so explanation on this would be helpful for me. using namespace boost::python; using namespace boost::filesystem; object sys = GetNamespace( "sys" ); list syspath( sys["path"] ); syspath.insert( 0, complete( path ).file_string() ); sys["path"] = syspath; Note that "path" is a boost::filesystem::path object containing the path I wish to add to sys.path.

2 4

The Placeholders _1 and _2
by Missing Rainbow 08 Oct '08

08 Oct '08

Hi All, I am going through Chapter 3 of the MPL book ("A Deeper Look at Metafunctions") and have some questions on the placeholders _1 and _2. They are first introduced in the section 3.1.5 as follows: typename mpl::transform<D1, D2, mpl::minus<_1, _2> >::type I understand from the context that _1 represents D1 and _2 represents D2. From this I understood that _1 will represent the first template argument and _2 will represent the 2nd template argument. This made sense in this context. But later I came across the following in section 3.3: template <class X> struct two_pointers : twice<boost::add_pointer<_1>, X> {}; In this context, it looks like _1 represents X, which is the 2nd template argument. So what exactly _1 and _2 represents? The book doesn't seem to define them. While I haven't completely read the book yet, if it is defined at some other place, it would have been helpful to provide a cross-reference when they are first introduced in chapter 3. Can anybody provide a reference that explains what is a placeholder and what _1 and _2 mean? And what is the maximum number of placeholders? Thank you. Rgds, anna -- Two Party Fraud: Staged Presidential Debates http://missingrainbow.blogspot.com/2008/10/two-party-fraud-staged-president…

2 1

[boost-users][serialization] porting to windows ce?
by Bobby Nations 08 Oct '08

08 Oct '08

All, I've been working to get boost-serialization running on a Windows CE (actually Pocket PC 2003) device these last few days, and have hit a snag. Could anyone help me with what might be causing the following errors in the operators.hpp file? c:\Documents and Settings\388572\Desktop\sandbox\Boost_1_34_1\boost\libs\serialization\vc 7ide\..\..\..\boost/operators.hpp(699) : error C3083: 'detail': the symbol to the left of a '::' must be a type It happens at this section of code: namespace boost { // A type parameter is used instead of a plain bool because Borland's compiler // didn't cope well with the more obvious non-type template parameter. namespace detail { struct true_t {}; struct false_t {}; } // namespace detail // Unspecialized version assumes that most types are not being used for base // class chaining. We specialize for the operator templates defined in this // library. template<class T> struct is_chained_base { ::boost::detail::false_t value; }; The error happens at the '::boost::detail::false_t value' line. So far, I've been surrounding any mentions of the locale header file or it's attendant classes with '#ifndef BOOST_NO_STD_LOCALE ... #endif' blocks, which allowed me to compile about half of the .cpp files. I also added code to use Microsoft's WideCharToMultiByte() and MultiByteToWideChar() methods rather than using the std::wctomb() and std::mbtowc() . Can anyone explain why the above code throws errors? It seems to be te final remaining hurdle to getting this working (well, sort of working anyway). Thanks, Bobby Nations

1 0