- Boost-users - lists.boost.org

[mpi] Sending derived classes via MPI and serialize
by Number Cruncher 08 Mar '08

08 Mar '08

I'm trying to send polymorphic objects using boost::mpi, but I'm getting extensive "undefined reference" errors for the simple test case below. If I #include <boost/archive/text_oarchive.hpp> instead of the mpi/communicator, everything compiles fine. Any help greatly appreciated, Simon #include <boost/mpi/communicator.hpp> #include <boost/serialization/export.hpp> struct Base { virtual ~Base() {} virtual int id() const { return 1; } template<class Archive> void serialize(Archive &ar, const unsigned int) { ar & data; } int data; }; struct Derived : public Base { virtual int id() const { return 2; } }; BOOST_CLASS_EXPORT(Base) BOOST_CLASS_EXPORT(Derived) int main(int argc, char *argv[]) { Base *b = new Base; Base *d = new Derived; delete b; delete d; return 0; }

2 5

[boost graph] : trouble with the add_edge function
by Bernhard Lippmann 08 Mar '08

08 Mar '08

Hi, I use the add_edge () function to build my graph. Works almost perfect. While analysing my graph I found there have been add vertices created, I have have no idea where is is going to happen. Any helpmpossible ? I add a code snippet for the add_edge() function and my data log. vertices 12-15 are never created by me, but the are suddenly in my graph.... Best Bernhard >>>>>>>>>>>><SOURCE CODE Snippet >>>>>>>> iter = this->Elements.begin(); while(iter != this->Elements.end()) { aAtomicElement = (*iter).second; for (str_iter=aAtomicElement.AE_PinList.begin(); str_iter!= aAtomicElement.AE_PinList.end(); ++str_iter) { cout << "\n Params before call Add_edge(...) function (Bipartit Graph Type) \n "; cout << "edge [<std::string>]: "<< aAtomicElement.AE_Name << " ::" << *str_iter << " \n "; cout << "edge [<int> <int>]: "<< index_of_element( this->VertexNamesList, aAtomicElement.AE_Name) << " :: " << index_of_element( this->VertexNamesList, *str_iter) << " \n "; OneEdge.first = index_of_element( this->VertexNamesList, aAtomicElement.AE_Name); OneEdge.second = index_of_element( this->VertexNamesList, *str_iter); tie(e, inserted) = add_edge(OneEdge.first, OneEdge.second, g); if (inserted) { v = source(e, g); VertexNamesMap[v] = aAtomicElement.AE_Name ; v = target(e, g); VertexNamesMap[v] = *str_iter; EdgeNamesMap[e] = aAtomicElement.AE_Name + "-->" +*str_iter; cout << " by edge :" << EdgeNamesMap[e] << "\n"; WeightMap[e] = 1 ; } } iter ++; } // while ... <<<<<<< data log >>>>>>>>> Drücken Sie eine beliebige Taste . . . -------------------->>>>>>>>>><<<<<<<<<<------------------------- = Create the Edge Array and the Graph (V2).. Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans299 ::Z edge [<int> <int>]: 0 :: 1 by edge :ptrans299-->Z Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans299 ::A edge [<int> <int>]: 0 :: 2 by edge :ptrans299-->A Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans299 ::vdd! edge [<int> <int>]: 0 :: 3 by edge :ptrans299-->vdd! Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans299 ::vdd!s edge [<int> <int>]: 0 :: 4 by edge :ptrans299-->vdd!s Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans38 ::Z edge [<int> <int>]: 5 :: 1 by edge :ptrans38-->Z Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans38 ::B edge [<int> <int>]: 5 :: 6 by edge :ptrans38-->B Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans38 ::vdd! edge [<int> <int>]: 5 :: 3 by edge :ptrans38-->vdd! Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ptrans38 ::vdd!s edge [<int> <int>]: 5 :: 4 by edge :ptrans38-->vdd!s Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans35 ::Z edge [<int> <int>]: 7 :: 1 by edge :ntrans35-->Z Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans35 ::A edge [<int> <int>]: 7 :: 2 by edge :ntrans35-->A Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans35 ::net25 edge [<int> <int>]: 7 :: 8 by edge :ntrans35-->net25 Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans35 ::vss!s edge [<int> <int>]: 7 :: 9 by edge :ntrans35-->vss!s Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans36 ::net25 edge [<int> <int>]: 10 :: 8 by edge :ntrans36-->net25 Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans36 ::B edge [<int> <int>]: 10 :: 6 by edge :ntrans36-->B Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans36 ::vss! edge [<int> <int>]: 10 :: 11 by edge :ntrans36-->vss! Params before call Add_edge(...) function (Bipartit Graph Type) edge [<std::string>]: ntrans36 ::vss!s edge [<int> <int>]: 10 :: 9 by edge :ntrans36-->vss!s Drücken Sie eine beliebige Taste . . . -------------------->>>>>>>>>><<<<<<<<<<------------------------- = http://www.graphviz.org/Download_windows.php write graph wizard file ... -------------------->>>>>>>>>><<<<<<<<<<------------------------- = vertices(g) = 0 : ptrans299 1 : Z 2 : A 3 : vdd! 4 : vdd!s 5 : ptrans38 6 : B 7 : ntrans35 8 : net25 9 : vss!s 10 : ntrans36 11 : vss! 12 : 13 : 14 : 15 : Drücken Sie eine beliebige Taste . . . -------------------->>>>>>>>>><<<<<<<<<<------------------------- = edges(g) = (0,1) =ptrans299-->Z (0,2) =ptrans299-->A (0,3) =ptrans299-->vdd! (0,4) = ns299-->vdd!s (5,1) =ptrans38-->Z (5,6) =ptrans38-->B (5,3) =ptrans38-->v (5,4) =ptrans38-->vdd!s (7,1) =ntrans35-->Z (7,2) =ntrans35-->A (7,8) =n s35-->net25 (7,9) =ntrans35-->vss!s (10,8) =ntrans36-->net25 (10,6) =ntra -->B (10,11) =ntrans36-->vss! (10,9) =ntrans36-->vss!s in-edges: (1,0) (2,0) (3,0) (4,0) in-edges: (0,1) (5,1) (7,1) in-edges: (0,2) (7,2) in-edges: (0,3) (5,3) in-edges: (0,4) (5,4) in-edges: (1,5) (6,5) (3,5) (4,5) in-edges: (5,6) (10,6) in-edges: (1,7) (2,7) (8,7) (9,7) in-edges: (7,8) (10,8) in-edges: (7,9) (10,9) in-edges: (8,10) (6,10) (11,10) (9,10) in-edges: (10,11) in-edges: in-edges: in-edges: in-edges: Drücken Sie eine beliebige Taste . . . -------------------->>>>>>>>>><<<<<<<<<<------------------------- No virus found in this outgoing message. Checked by AVG. Version: 7.5.518 / Virus Database: 269.21.6/1318 - Release Date: 07.03.2008 14:01

1 0

Platform-specific Boost filesystem warning
by Caleb Gingles 08 Mar '08

08 Mar '08

(Please forgive me if this has already been discussed, I'm new to the list and wasn't able to find it.) I'm using Boost 1.34.1 under MSVC8 and am experiencing a compile warning that I thought was a bit odd. Whenever I include convenience.hpp, I and up with the following warning: d:\prgrmmng\libraries\boost_1_34_1\boost\filesystem\convenience.hpp(46) : warning C4245: 'argument' : conversion from 'int' to 'boost::filesystem::system_error_type', signed/unsigned mismatch I eventually tracked this down to the following code in path.hpp: # ifdef BOOST_WINDOWS_API typedef unsigned system_error_type; BOOST_FILESYSTEM_DECL errno_type lookup_errno( system_error_type sys_err_code ); # else typedef int system_error_type; inline errno_type lookup_errno( system_error_type sys_err_code ) { return sys_err_code; } # endif Why is system_error_type given a different type depending on the platform?

1 0

Extending Multi Array
by Klett, Stefan 07 Mar '08

07 Mar '08

Hi list members, This is a more or less a general kind of " how to extend container classes" question - I'm trying to write a class which inherits from boost::multi_array - the to get a kind of matrix class with some additional service routines. Maybe it is a lack of experience - but I for the time being I don't know how to solve the problem to access the constructor of multi_array in my own class to initialize my matrix class - my current guess was to use the "this" pointer but it had not worked out like I supposed it to do. class NetworkMultiArray: public boost::multi_array<double,3> { public: NetworkMultiArray(int); virtual ~NetworkMultiArray(); }; Here is the constructor code: NetworkMultiArray::NetworkMultiArray(int length) { this::iterator iter3; typedef this::template subarray<2>::type::iterator iter2; typedef this::template subarray<1>::type::iterator iter1; for( iter1 i = *this.begin(); iter1 != *this.end(); iter1++){ for(iter2 ii = (*i).begin(); ii != (*i).end(); ii++){ for(iter3 iii = (*ii).begin(); iii != (*ii).end(); iii++){ *iii = 0.0 ; } } } } sure this does not work - and is the result from trying to adapt the iterator example from the documentation. my aim is to handle the NetworkMultiArray like a boost::multi_array and initialize it to zero - further it would be nice if somebody could tell me how I could assign a length to the array and the subarrays. The key-point is that I want to use my class (NetworkMultiArray) like a multi_array by inheritance of the multi_array features of boost:multi_array. Thank you very much in advance. Best regards Stefan -------------------------------------------------------- Stefan Klett (Dipl. Informatiker) | E-mail: stefan.klett(a)kit.edu Forschungszentrum Karlsruhe / KIT | Web: http://www.fzk.de SCC Steinbuch Centre for Computing| Phone: +49 (0)7247 828649 Hermann-von-Helmholtz-Platz 1 | Fax: +49 (0)7247 824972 D-76344 Eggenstein-Leopoldshafen | Mobile: +49 (0)176 20317250 -------------------------------------------------------------------

3 2

[python] redirecting PyErr_Print()
by Robert Dailey 07 Mar '08

07 Mar '08

Hi, Right now when an exception occurs in boost.python, I simply call PyErr_Print() to print the message to a console window. However, I would like to be able to have the option of redirecting the output of PyErr_Print() to a string, file, or any other ostream interface compatible object I choose. Is there an easy way to do this (possibly using boost)? My searching on the internet has not really shown any useful results. The answers I've seen involve creating pipes, which I don't want to do because it is non trivial. I was hoping boost could shine in this area since it seems to be a common issue.

2 2

ASIO: ssl - load_verify_file
by j.c. 07 Mar '08

07 Mar '08

Hello, I have implemented a TLS client/server application and will be using self-assigned certificates due to the small scale of distribution. I am currently using the function load_verify_file("ca.pem"), however I don't want to write the certificate to end-users disk. What would be the proper way to use an "in-memory or compiled" certificate for peer verification? Thanks, JC

2 4

boost::function "could not deduce template argument ..."
by Derrick Hathaway 07 Mar '08

07 Mar '08

I apologize if this isn't the sort of question this list is intended for, but if anyone could help me with my lack of sufficient template howto skill I would appreciate it. Essentially what I have is class template for boost::function1<bool, const ValueType &> like so: template <typename ValueType> class predicate { public: typedef boost::function1<bool, const ValueType&> type; struct and { bool operator()(const ValueType &)() { return left_() && right(); } type left_; type right_; }; struct or { ... }; }; And what I want is to overload the & and | operators for predicate<ValueType>::type like so: template <typename ValueType> typename predicate<ValueType>::type operator &( const typename predicate<ValueType>::type &left, const typename predicate<ValueType>::type &right) { return typename predicate<ValueType>::and(left, right); } The problem is when I compile I get the error message "could not deduce template argument for 'ValueType'" when I compile. Any idea what I'm doing wrong? I have attached example code. I am using MS Visual Studio 8, and boost 1.34.1. Thanks in advance. -Derrick

2 1

automake + boost lib versions
by Klett, Stefan 07 Mar '08

07 Mar '08

Hi altogether, I've got a problem with rolling a packet to distribute a library source - first I tried to do this with boost-build , but later on I had some difficulties and fell back to automake. What I need is a (hopefully convenient) way to deal with different versions of the boost libraries - because I have to expect that the names of these libraries will differ on nearly all machines a potential User of "my" library is trying to build the source. For example the date-time lib appears as: /usr/local/lib/libboost_date_time-gcc42-1_34_1.a /usr/local/lib/libboost_date_time-gcc42-1_34_1.so /usr/local/lib/libboost_date_time-gcc42-1_34_1.so.1.34.1 /usr/local/lib/libboost_date_time-gcc42.a /usr/local/lib/libboost_date_time-gcc42-d-1_34_1.a /usr/local/lib/libboost_date_time-gcc42-d-1_34_1.so /usr/local/lib/libboost_date_time-gcc42-d-1_34_1.so.1.34.1 /usr/local/lib/libboost_date_time-gcc42-d.a /usr/local/lib/libboost_date_time-gcc42-d.so /usr/local/lib/libboost_date_time-gcc42-mt-1_34_1.a /usr/local/lib/libboost_date_time-gcc42-mt-1_34_1.so /usr/local/lib/libboost_date_time-gcc42-mt-1_34_1.so.1.34.1 /usr/local/lib/libboost_date_time-gcc42-mt.a /usr/local/lib/libboost_date_time-gcc42-mt-d-1_34_1.a /usr/local/lib/libboost_date_time-gcc42-mt-d-1_34_1.so /usr/local/lib/libboost_date_time-gcc42-mt-d-1_34_1.so.1.34.1 /usr/local/lib/libboost_date_time-gcc42-mt-d.a /usr/local/lib/libboost_date_time-gcc42-mt-d.so /usr/local/lib/libboost_date_time-gcc42-mt.so /usr/local/lib/libboost_date_time-gcc42.so in the case of my library using the multithreaded version is preferred. Hardwiring fails by the obvious reason that not everyone else is using gcc42 Is there anyone of you who has some experience in writing a proper configure.ac Makefile.am that is aware of these different boost-library names ? If so please give me a hint ... Thank you very much in advance Best regards Stefan -------------------------------------------------------- Stefan Klett (Dipl. Informatiker) | E-mail: stefan.klett(a)kit.edu Forschungszentrum Karlsruhe / KIT | Web: http://www.fzk.de SCC Steinbuch Centre for Computing| Phone: +49 (0)7247 828649 Hermann-von-Helmholtz-Platz 1 | Fax: +49 (0)7247 824972 D-76344 Eggenstein-Leopoldshafen | Mobile: +49 (0)176 20317250 -------------------------------------------------------------------

2 1

Please help: Seg fault with threads
by Nishant Mittal 07 Mar '08

07 Mar '08

Hi All, I started on boost today and wrote my first hello world application. #include <boost/thread/thread.hpp> #include <iostream> using namespace std; void hello_world() { cout << "Hello world, I'm a thread!" << endl; } int main(int argc, char* argv[]) { boost::thread my_thread(&hello_world); my_thread.join(); return 0; } I compiled it using the command. g++ helloworld.cpp -lboost_thread-gcc41-mt-1_34_1 However, when I try to run it, I keep getting a segmentation fault. I tried to debug with gdb and this is what I get. (gdb) run Starting program: boost_test/a.out (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread 46912498370912 (LWP 21199)] [New Thread 1084229952 (LWP 21202)] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 1084229952 (LWP 21202)] 0x00002aaaaaab4932 in boost::thread_group::add_thread () from /usr/local/lib/libboost_thread-gcc41-mt-1_34_1.so.1.34.1 (gdb) Please let me know as to what I am doing wrong. Any help is appreciated. Thanks Nishant

2 2

[spirit] Segmentation fault when parsing grammar with threads
by Andreas Bruckmeier 06 Mar '08

06 Mar '08

Hi out there, after searching for the cause of segmentation faults in my program i figured out that it happens while parsing my grammar with spirit. But this happens only if i run multiple threads, with only one thread all works as expected. I used the current boost trunk, gcc 4.1.2 and option D_REENTRANT Does anybody have a clue what happens inside parse(...) that could cause this segfault? Best regards Andi calling code: ('buf' is a zero-terminated string) -------------------------------------------------------- HttpRequestGrammar parser; parse_info<const char*> parserInfo = parse(buf, parser); my grammar: (cutted some rules to limit length ;-)) -------------------------------------------------------- #ifndef BASIC_HTTP_GRAMMAR_H_ #define BASIC_HTTP_GRAMMAR_H_ #include <boost/spirit/core.hpp> #include <boost/spirit/actor/assign_actor.hpp> #include <boost/spirit/actor/insert_at_actor.hpp> using namespace boost::spirit; struct HttpRequestGrammar : public grammar<HttpRequestGrammar> { template <typename ScannerT> struct definition { definition(HttpRequestGrammar const& self) { first = ( request = *( SP | CRLF ) >> requestLine >> CRLF >> *(( generalHeader | requestHeader | entityHeader ) >> CRLF) >> CRLF, requestLine = method[insert_at_a(self.headerMap, "method")] >> SP >> requestUri[insert_at_a(self.headerMap, "url")] >> SP >> httpVersion[insert_at_a(self.headerMap, "version")], method = str_p("GET") | "HEAD", ... .. token = +graph_p, SP = str_p(" "), CRLF = str_p("\r\n") ); } subrule<0> request; subrule<1> requestLine; subrule<2> generalHeader; ... .. subrule<1000> SP; subrule<1001> CRLF; rule<ScannerT> first; rule<ScannerT> const& start() const { return first; } }; mutable std::map<std::string, std::string> headerMap; }; #endif /*BASIC_HTTP_GRAMMAR_H_*/

2 1