February 2009 - Boost-users - lists.preview.boost.org

boost::serialize stream error
by ppatel＠efs-us.com 12 Feb '09

12 Feb '09

Hi all, I am having following problem while reading the data out of the file using one of the sample program provided in boost serialization example online. I am trying to read the file having multiple gps_position objects but getting following exception. Can someone please let me know what might be the problem here. Let me know if you need any more information: #include <fstream> #include <string> // include headers that implement a archive in simple text format #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/list.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/base_object.hpp> ///////////////////////////////////////////////////////////// // gps coordinate // // illustrates serialization for a simple type // class gps_position { private: friend class boost::serialization::access; // When the class Archive corresponds to an output archive, the // & operator is defined similar to <<. Likewise, when the class Archive // is a type of input archive the & operator is defined similar to >>. template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & degrees; ar & minutes; ar & seconds; ar & day; } public: int degrees; int minutes; float seconds; std::string day; gps_position(){}; gps_position(int d, int m, float s, std::string inDay) : degrees(d), minutes(m), seconds(s), day(inDay) {} }; int main() { // create and open a character archive for output std::ofstream ofs("filename"); // create class instance const gps_position g(35, 59, 24.567f, "Monday_Pointer"); const gps_position g1(45, 69, 34.567f, "Tuesday_Pointer"); const gps_position * g2 = new gps_position(1,2,3.3, "WEDNESDAY_PTR"); { boost::archive::text_oarchive oa(ofs); oa << g; oa << g1; oa << (*g2); } { /**gps_position g; std::vector<gps_position> vec; std::ifstream ifs("filename"); boost::archive::text_iarchive ia(ifs); while (!ifs.eof()) { std::cout << " ++++++++++++++++ " << std::endl; ia >> g; std::cout << g.degrees << " : " << g.minutes << " : " << g.seconds << " : " << g.day <<std::endl; std::cout << " **************** " << std::endl; }*/ } { // Using boost vector std::vector<gps_position> gps; std::ifstream ifs("filename"); boost::archive::text_iarchive ia(ifs); ia >> gps; for (int i = 0; i < gps.size(); ++i) { std::cout << gps[i].degrees << " : " << gps[i].minutes << " : " << gps[i].seconds << " : " << gps[i].day <<std::endl; } } return 0; } // File data that is being read $ cat filename 22 serialization::archive 4 0 0 35 59 24.566999 14 Monday_Pointer 45 69 34.567001 15 Tuesday_Pointer 1 2 3.3 13 WEDNESDAY_PTR // Output of program with while loop, don't know why eof is not being detected after reading three gps position objects ++++++++++++++++ 35 : 59 : 24.567 : Monday_Pointer **************** ++++++++++++++++ 45 : 69 : 34.567 : Tuesday_Pointer **************** ++++++++++++++++ 1 : 2 : 3.3 : WEDNESDAY_PTR **************** ++++++++++++++++ terminate called after throwing an instance of 'boost::archive::archive_exception' what(): stream error Aborted (core dumped) //Output of program with using vector terminate called after throwing an instance of 'boost::archive::archive_exception' what(): stream error Aborted (core dumped) Thanks Priyank

2 3

[test] BOOST_CHECK_EQUAL and wchar_t
by Alexander 12 Feb '09

12 Feb '09

The following code: BOOST_CHECK_EQUAL("boo", "boo"); BOOST_CHECK_EQUAL("boo", "bar"); BOOST_CHECK_EQUAL(L"boo", L"boo"); BOOST_CHECK_EQUAL(L"boo", L"bar"); Gives the following results test.cpp(368): error in "Strings": check "boo" == "bar" failed [boo != bar] test.cpp(370): error in "Strings": check L"boo" == L"bar" failed [00B6FAA4 != 00B6FA90] Can this be fixed ?

4 6

[asio] Poll without read
by Oleg Malashenko 12 Feb '09

12 Feb '09

Hi everyone, I am trying to get video stream from Video4Linux2 device (/dev/video*) and faced with a interesting boost::asio usecase. V4L2 device implements poll calls, so one can poll() on specific device file to wait for next portion of data, but read itself is made via ioctl. I suppose, boost::asio is a natural choice in this situation and I am wondering if there is a method to run my callback when "the data is ready to read", but without calling read()? (on the same manner as ACE::Reactor does)? Thank you.

2 2

[serialization] problem using BOOST_CLASS_EXPORT macro in a static library
by Renaud Lepere 12 Feb '09

12 Feb '09

I have some problems with BOOST_CLASS_EXPORT macro, it seems that this macro have no effect if i put it in a separate .cpp file of a static library. I have a simple class Element that contains a pointer to a curve that can be of different types (geom2d.hpp). struct Curve { Curve() {}; virtual ~Curve() {}; }; struct Line : public Curve { Line(int y = 0): m_y(y) {}; int m_y; }; typedef boost::shared_ptr< Curve > CurvePtr; struct Element { Element(const CurvePtr & curve): m_curve(curve) {} boost::shared_ptr< Curve > m_curve; }; I would like to be able to serialize Element, so i wrote a serialize.hpp file (attached) containing code for serializing those structures. I also wrote export.cpp using #include "geom2d.hpp" #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include "serialize.hpp" #include <boost/serialization/export.hpp> #include <iostream> BOOST_CLASS_EXPORT(Curve); BOOST_CLASS_EXPORT(Line); I put the export.cpp file in a library named geom2d, but when trying to serialize my structures using something (attached main.cpp). CurvePtr c(new Line(10)); const Element e(c); std::ofstream ofs("filename"); { boost::archive::text_oarchive oa(ofs); oa << e; } I received an exception "unregistered class". If i dont put export.cpp in a library but in the same compilation unit as main.cpp, the error disappear (visual studio 8.0 and 9.0). What is the good way of doing this, should i wrote an export.hpp file containing the macros and include this file in a .cpp of the main compilation unit ? Thanks, Renaud

4 5

Re: [Boost-users] how to make a vector to derivedClass::create() methods
by Victor Whiskey Yankee 12 Feb '09

12 Feb '09

Victor.Whiskey.Yankee wrote: > > > > Thank you for your patience. > > I am really wanting to make a factory that holds a > > std::map where the pair.first is a std::string and > > the pair.second is somehow a pointer to a constructor or create() > > method to any of several derived classes>. > > > > I thought expressing in terms of vector might be simplified and enough > > for my understanding. > > > > I see. How about using Boost.Function like this: > > std::map<std::string, boost::function<Base*()> > factory_impl; > > //... > > factory_impl["Derived1"] = boost::lambda::new_ptr<Derived1>(); > > // ... > > Base* b = factory_impl["Derived1"]() Thank you Steve for your solution, it is working fine. I have a nice thread-safe singleton factory now, and it is generic using templates. I am now trying to make my derived classes register with the factory automatically, but cannot figure it out. I basically have this: template<typename BaseType, typename keyType> class GenericFactory { public: GenericFactory(); boost::shared_ptr<BaseType> create(const keyType& typeName); template<typeName DerivedType> void registrate(const keyType& _key); }; As said, I can register my derived classes with the above just fnie. I have tried writing a help class like I've seened "RegisterInFactory", but it escapes me. Thanks, Vic

2 4

Gomory Hu algorithm (another time)
by Gianni Loiacono 11 Feb '09

11 Feb '09

Hi all, sorry for my english. I'm desperate :(.... From about 15 days I'm trying to make the Gomory-Hu algorithm: http://www.corelab.ntua.gr/~syrganis/sources/Gomory-Hu.pdf I'm using succesfully the Boost library to manage and build the graph but unfortunately the Gomory-Hu algorithm was not implemented :( Into the boost library there is only the mincut function and I have trying to use it (edmunds_karp_max_flow) for to make the Go-Hu function but the result was disastrous (the code is very long and I have lost the "control" of the program so at the and the program works only with some examples :(.....It's unusable) On the net there isn't any source code :( I have found only a pseudo-code here: http://www.prefield.com/algorithm/graph/gomory_hu.html But honestly I don't know how to convert it in c++ for to use with boost library. Please, somebody can help me? I'm very very desperate :( . Thanks a lot all for the help. Bye :) Passa a Yahoo! Mail. La webmail che ti offre GRATIS spazio illimitato, antispam e messenger integrato. http://it.mail.yahoo.com/

3 2

boost::thread performance questions
by Alfonso Urdaneta 11 Feb '09

11 Feb '09

I'm looking for information about boost::thread switch time and mutex lock/release times (trying to track down a performance issue). If someone could point me in the right direction it would be greatly appreciated. -- alfonso e. urdaneta _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ This e-mail and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this e-mail in error please notify the originator of the message. This footer also confirms that this e-mail message has been scanned for the presence of computer viruses. Any views expressed in this message are those of the individual sender, except where the sender specifies and with authority, states them to be the views of Indra Systems, Inc.

1 0

What is wrong building a debug build of boost?
by Hansi 11 Feb '09

11 Feb '09

Hi, Til now I have used the following 2 commands to build the boost library: bjam.exe --toolset=msvc --build-type=debug link=static threading=multi stage --stagedir=. bjam.exe --toolset=msvc --build-type=release link=static threading=multi stage --stagedir=. and it has worked with boost 1.37. With boost 1.38 I don't get anymore the debug builds of the library. What it is wrong with this command? I use 3.1.17 of bjam.exe Thanks for any hint! Regards Hansjörg

3 3

Re: [Boost-users] [multi_index] Problem of scope with the replace in static data structure
by Rodrigo Dias Ferreira 11 Feb '09

11 Feb '09

Hi Joaquim, > > It's hard to know whether you got copy semantics right without knowing > the types > of _rangeBegin, _rangeEnd and the rest of Graph data members, but this > has prompted > me to suspect that maybe there's something wrong with the handler type > Graph_h (which > I assume is some sort of ref-counted reference to Graph, right?). Can > you please > test the following? In your //Test of the commit section, rather than > reuse graph_h please > use a different variable to do the after-commitChanges check: > > //Test of the commit > if (graph_h.commitChanges()) > cout << "Commit was successful!\n"; > else > cout << "Commitretr was unsuccessful!\n"; > // Debug > > > //Graph_h graph_h; > //NodeIterator nodeIt; > //Node_h node_h; > > graphIt it = DynamicGraphs::_graphRep.find("COI:0351-123-4555"); > Graph_h graph_h2 = Graph_h(it); // WE DON'T REUSE graph_h > //propertystream_h propS; > propS = graph_h2->lookupLocalProp("name"); > p_h = propS.next(); > cout << "DEBUG@main:right after the commit:_git " << p_h.value() <<endl; > nodeIt = graph_h2->nodeIterator("Node:0351-123-4569"); > if (!nodeIt) > cout << "Node version was not found!"<<endl; > else //Works for both: one match and range lookup > while (!nodeIt.end()){ > node_h = nodeIt.next(); > cout << "Found it: " << *node_h; > } > //Debug > > Does this make any difference? Yes, it worked, if I instantiate another object (like graph_h2). After that I have done some more Debug to try to find the reason, but I still don't understand,because inside the constructor it finds the Nodes, when I use the same object (graph_h)... Inside the overloaded operator-> does not work anymore, even I am using the _localGraph in both stretch of codes. So if it was the assignment of the _localGraph, it should not work with the constructor also, isn't true? And the commitChanges, that I told before, it is for replacing this _localGraph in the repository of graphs graphIndex _graphRep; For better explanation, here it is the code: Remark: the Graph_h it will be using referencing count, but since I don't know how yet, it is not implemented. So the Graph_h I use as way of the user handle the Graph, so it has as members: - the iterator of the lookup for the Graph - _localGraph: it is where I can change the Graph, because in the index it is not possible, since their elements are constant - and bool to indicate the creation success /// Graph Handler: Wrapper for manipulating the Graph objects class Graph_h : virtual Handler { private: graphIt _git; Graph _localGraph; bool _createSuccess; public: /// Default Constructor Graph_h() : _createSuccess(false){} /// Constructor with Parameters: for creation of Graphs Graph_h (const pair<graphIt,bool> &creation) : _git(creation.first), _createSuccess(creation.second){ _localGraph = _git->_graph; } ///////////////The below constructor it is where it works,//////////// ///////////////the retrieve of nodes with _localGraph///////////// /// Constructor with Parameters: for lookup of Graphs Graph_h (const graphIt &it) : _git(it), _createSuccess(false){ _localGraph = it->_graph; //#ifndef NDEBUG_DG //Graph_h graph_h; NodeIterator nodeIt; Node_h node_h; //graphIt it = DynamicGraphs::_graphRep.find("COI:0351-123-4555"); //Graph_h //graph_h = Graph_h(it); propertystream_h propS; property_h p_h; propS = _localGraph.lookupLocalProp("name"); p_h = propS.next(); cout << "DEBUG@Graph_hConstructor(it) " << p_h.value() <<endl; nodeIt = _localGraph.nodeIterator("Node:0351-123-4569"); if (!nodeIt) cout << "Node version was not found!"<<endl; else //Works for both: one match and range lookup while (!nodeIt.end()){ node_h = nodeIt.next(); cout << "Found it: " << *node_h; } //#endif //NDEBUG_DG } /// Copy Constructor Graph_h (const Graph_h &h) : _git(h._git), _createSuccess(h._createSuccess){ _localGraph = h._localGraph; } Graph_h& operator=(const Graph_h &h) { _git = h._git; _createSuccess = h._createSuccess; _localGraph = h._git->_graph; return *this; } bool commitChanges(); bool operator!(){ return !_createSuccess; } ///////////////The below operator-> it is where it does not works,////// ///////////////the retrieve of nodes with _localGraph////////////////// /// Operator-> overloading: allows access to graph members Graph* operator->() { //#ifndef NDEBUG_DG //Graph_h graph_h; NodeIterator nodeIt; Node_h node_h; //graphIt it = DynamicGraphs::_graphRep.find("COI:0351-123-4555"); //Graph_h //graph_h = Graph_h(it); propertystream_h propS; property_h p_h; propS = _localGraph.lookupLocalProp("name"); p_h = propS.next(); cout << "DEBUG@Graph::operator-> " << p_h.value() <<endl; nodeIt = _localGraph.nodeIterator("Node:0351-123-4569"); if (!nodeIt) cout << "Node version was not found!"<<endl; else //Works for both: one match and range lookup while (!nodeIt.end()){ node_h = nodeIt.next(); cout << "Found it: " << *node_h; } //#endif //NDEBUG_DG return &_localGraph; } }; Output from these strech of code: The node are not being printed because I am changing some stuff in other part of the code... but the important is that it is found the nodes in the constructor, and in the operator-> the node were not found: Constructor Output and operator-> output: DEBUG@Graph_hConstructor(it) COI:0351-123-4555 Found it: Name: Timestamp: -1080916036 Found it: Name: Timestamp: -1080916036 DEBUG@Graph::operator-> COI:0351-123-4555 Node version was not found! Code in the main which generate that: graphIt it = DynamicGraphs::_graphRep.find("COI:0351-123-4555"); graph_h = Graph_h(it); propS = graph_h->lookupLocalProp("name"); As you asked before, Here are the type of the Graph: class Graph { private: /// This variables represent the range of time that one would like to retrieve from the API. time_t _rangeBegin, _rangeEnd; /// Represents the repository of nodes in a graph nodeIndex _nodeRep; /// Represents the Node's name repository in a graph (in order to not have redundant data) uniqueNameTable _nodeNameRep; /// Represents the Node's view with index in the timestamp (in order to filter the data) nodeTimestampView _nodeTimeStampView; /// Represents the properties of a graph propIndex _localPropRep; /// Repository for the properties of the global properties related to the graph; grobalProperties _globalPropRep; The main types here that I think it could be giving me problems it would be the ones related to the nodes, here are the definitions: typedef multi_index_container< nodeMap, indexed_by< hashed_non_unique< tag<nodeName>, key_from_key< BOOST_MULTI_INDEX_MEMBER(uniqueNodeNameStruct,const string,_name), BOOST_MULTI_INDEX_MEMBER(nodeMap, const uniqueNodeNameStruct *,_uniqueName) > >, ordered_non_unique< tag<nodeTimeStamp>,BOOST_MULTI_INDEX_MEMBER(nodeMap,time_t,_timestamp) > > > nodeIndex; /// Defining an iterator for the nodeMapIndex typedef nodeIndex::iterator nodeMapIt; struct nodeMap { const uniqueNodeNameStruct* _uniqueName; time_t _timestamp; Node _node; /// Parameter nodeMap(const uniqueNodeNameStruct* name, time_t t, Node node): _uniqueName(name),_timestamp(t), _node(node) {} /// Copy Constructor nodeMap(const nodeMap &nm): _uniqueName(nm._uniqueName),_timestamp(nm._timestamp), _node(nm._node) {} //TODO: Delete this. /// For debugging friend std::ostream& operator<<(std::ostream& os,const nodeMap& nm) { os<<"Name: " << nm._uniqueName << " Timestamp: " << nm._timestamp << std::endl; return os; } }; struct uniqueNodeNameStruct { string _name; /// Parameter Constructor uniqueNodeNameStruct(const string& name):_name(name){} /// Copy Constructor uniqueNodeNameStruct(const uniqueNodeNameStruct &nodeName): _name(nodeName._name){} uniqueNodeNameStruct* operator=(uniqueNodeNameStruct& st){ _name = st._name; return &st; } friend std::ostream& operator<<(std::ostream& os,const uniqueNodeNameStruct& nodeName) { os<<"uniqueName: " << nodeName._name << endl; return os; } // TODO: Verify the necessity of change a name of the node struct change_name { change_name(const string& new_name):_new_name(new_name){} void operator()(uniqueNodeNameStruct& n) { n._name=_new_name; } private: std::string _new_name; }; }; /** * \brief uniqueNodeNames are store in this table and they are unique. * The uniqueNodeNames will be reference in this table, in order to * associate then with the different versions, in that way it's possible * to maintain just one name for all its versions. */ typedef multi_index_container< uniqueNodeNameStruct, indexed_by< hashed_unique< BOOST_MULTI_INDEX_MEMBER(uniqueNodeNameStruct,string,_name) > > > uniqueNameTable; typedef uniqueNameTable::iterator nameMapIt; I don't know how did you find that could be something in the Graph_h, perhaps you can understand that, because I don't understand what is wrong there... Looking foward for a feedback, and thanks for the help so far... best regards, Rodrigo -- ************************************************* Rodrigo Dias Ferreira Master Student of Computational Engineering Technische Universität Dresden - Germany Mobile Phone:+49 171 3158797 E-mail: rodrigodias(a)gmx.de ************************************************* Jetzt 1 Monat kostenlos! GMX FreeDSL - Telefonanschluss + DSL für nur 17,95 Euro/mtl.!* http://dsl.gmx.de/?ac=OM.AD.PD003K11308T4569a

2 1

Re: [Boost-users] for_each_until like algorithm
by Robert Jones 11 Feb '09

11 Feb '09

On Wed, Feb 11, 2009 at 2:24 PM, Marshall Clow <marshall(a)idio.com> wrote: > You might look in the sandbox, specifically in boost/algorithms/all.hpp. > What you call "exists" is called "any" there - which follows the proposal > for TR2. Oh yes, so it is! With your name on the file I see. I prefer exists to any, but OTOH I can see the symetry with none, so maybe your version is best. When is this likely to emerge from the shadows? Also, I see you have range versions too - does this need the RangeEx stuff to be in Boost too? Do you think your algorithms should have Boost concept checks on their template parameters? Cheers, Rob.

1 0