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

[bind] & [function] How do I store member functions with generalised arguments for later invocation?
by Steve Lorimer 14 Jun '10

14 Jun '10

I am wanting to write library code that will allow a user to create an "event" which is associated with a thread, a class instance, and a member function of that class. The user can then "post" data (generalised number of arguments) - which results in the class instance's member function being called in the context of the associated thread. For example (pseudo code): event { thread* thr; class &that; mem_fun* fun; void post(); void post(a1); void post(a1, a2); void post(a1, a2, a3, ...etc); } mem_fun* fun stores a member function pointer which is of the form void class::fun(...0, 1 or more arguments...) Here is some example code which uses boost::bind to bind arguments to member functions and posts them to be called on another thread. However, it does not go far enough. #include <boost/function.hpp> > #include <boost/bind.hpp> > #include <boost/enable_shared_from_this.hpp> > #include <boost/thread/thread.hpp> > #include <boost/thread/mutex.hpp> > #include <boost/thread/condition.hpp> > #include <list> > #include <stdio.h> > > // cross-thread event queue. multiple threads can post jobs, one or many > threads can execute > // jobs. > class EventQueue > { > public: > typedef boost::function<void()> Job; > > // puts a jon into the queue > void post(Job job) > { > boost::mutex::scoped_lock lock(mtx_); > jobs_.push_back(job); > cnd_.notify_one(); > } > > // pulls one job from the queue, returns false when stopped > bool pull(Job* job) > { > boost::mutex::scoped_lock lock(mtx_); > for(;;) { // handle spurious wake-ups > while(!stop_ && jobs_.empty()) > cnd_.wait(lock); > if(!jobs_.empty() && 2 != stop_) { > job->swap(jobs_.front()); // move efficiently, avoiding > *job = jobs.front() > jobs_.pop_front(); > return true; > } > else if(stop_) { > return false; > } > } > } > > // make pull() return false > void stop(bool cancel_jobs) > { > boost::mutex::scoped_lock lock(mtx_); > stop_ = 1 + cancel_jobs; // 1 - complete jobs, 2 - cancel jobs > cnd_.notify_all(); > } > > EventQueue() : stop_() {} > ~EventQueue() { this->stop(true); } > > private: > boost::mutex mtx_; > boost::condition cnd_; > typedef std::list<Job> Jobs; > Jobs jobs_; > int stop_; > }; > > struct JobX > // micro-oprimization, embed the reference counter into the job to avoid an > extra memory > // allocation by boost::shared_ptr ctor > : boost::enable_shared_from_this<JobX> > { > void foo() > { > printf("%p foo()\n", this); > } > > void bar(int a) > { > printf("%p bar(%d)\n", this, a); > } > }; > > void anotherThread(EventQueue* queue) > { > EventQueue::Job job; > // wait and execute jobs till stopped > while(queue->pull(&job)) > job(); // execute the job > } > > int main() > { > EventQueue queue; > // start another thread and pass an argument > boost::thread another_thread(boost::bind(anotherThread, &queue)); > > // post jobs, allocate in this thread, deallocate in the other > boost::shared_ptr<JobX> job_0(new JobX); > queue.post(boost::bind(&JobX::foo, job_0)); > // post several jobs to the same object, deallocated when no longer in > use > boost::shared_ptr<JobX> job_1(new JobX); > queue.post(boost::bind(&JobX::foo, job_1)); > queue.post(boost::bind(&JobX::bar, job_1, 1)); // pass an extra arg 1 > queue.post(boost::bind(&JobX::bar, job_1, 2)); // pass an extra arg 2 > > // stop the queue and let it complete all jobs > queue.stop(false); > another_thread.join(); > } > And this is where I fall down. I am able to bind arguments to JobX::bar above, at the point at which I create my boost::function<void()>, but if I want to bind _1, _2 etc arguments for calling later, I'm not sure what to do. So my question is, how should I form class Event such that I can store member functions of the form void T::fn(), void T::fn(a1), void T::fn(a1, a2, etc...), and then allow a generalised form of post(), post(a1), post(a1, a2, etc...) to bind the arguments given in post to the callback and post it to the queue? I have some code below which clearly won't compile, but should illustrate to some degree what I'm looking for. template <typename T> class Event { EventQueue &queue_; // the event queue which will process the job boost::shared_ptr<T> that_; // the object whose mem_fun the event queue will call typedef boost::function<void (T*, void) > CB_0 ; // used for callbacks which are of the form void T::fn(); typedef boost::function<void (T*, _1) > CB_1 ; // used for callbacks which are of the form void T::fn(1 arg); typedef boost::function<void (T*, _1, _2) > CB_2 ; // used for callbacks which are of the form void T::fn(1 arg); CB_0 cb_0; CB_1 cb_1; CB_2 cb_2; public: Event(boost::shared_ptr<T> &that, CB &cb, EventQueue &queue) : that_(that), cb_(cb), queue_(queue) {} void post() { queue_.post(boost::bind(&cb_0, that)); template <typename arg_1> void post(arg_1 &arg1) { queue_.post(boost::bind(&cb_1, that)); template <typename arg_1, typename arg_2> void post(arg_1 &arg1, arg_2 &arg2) { queue_.post(boost::bind(&cb_2, that)); // TODO: args should be shared_ptr }; TIA Steve

4 16

[units] initialization from zero magic
by alfC 14 Jun '10

14 Jun '10

Hi, I am curious. In Boost.Units it is not possible to *initialize* a quantity from a plain value, the closest thing to it is the ::from_value static member function. But ... What is the magic that makes this particular initialization compile? boost::units::quantity<si::length> L(0); //compiles and works as expected but *not* these ones: boost::units::quantity<si::length> L(1); //does not compile or boost::units::quantity<si::length> L(0.); //does not compile (gives a compile error, "no matching function for call to '...::quantity(int/double)') By the way, I think it is a great feature! I am interested to know how is this achieved. Thanks, Alfredo

2 1

Serializing to .csv files
by Sean Farrow 14 Jun '10

14 Jun '10

Hi: Has anyone ever written extra classes to serialize c++ datatypes to .csv files using boost.Serialization. If so what is the best way of going about this? Cheers Sean

1 0

boost MPI error
by Jack Bryan 13 Jun '10

13 Jun '10

Hi, all: I need to install boost library (/home/myname/boost_2010_6_8/boost_1_43_0). http://www.boost.org/doc/libs/1_43_0/more/getting_started/index.html It has MPI library : http://www.boost.org/doc/libs/1_42_0/doc/html/mpi.html After installing it, I compile some c++ program with calling for boost.MPI class, the compiler cannot find the header files in the program even though I have included the boost path in the compile command. example: ------------------------------------------------------------------------ #include </boost/mpi/environment.hpp> #include <boost/mpi/communicator.hpp> #include <iostream> namespace mpi = boost::mpi; int main(int argc, char* argv[]) { mpi::environment env(argc, argv); mpi::communicator world; std::cout << "I am process " << world.rank() << " of " << world.size() << "." << std::endl; return 0; } -------------------------------------------------------------------------- mpic++ -o testp2p test_p2p.cpp I/home/myname/boost_2010_6_8/boost_1_43_0 g++: I/home/myname/boost_2010_6_8/boost_1_43_0: No such file or directory test_p2p.cpp:7:38: error: /boost/mpi/environment.hpp: No such file or directory test_p2p.cpp:11:38: error: boost/mpi/communicator.hpp: No such file or directory Actually, I have installed these header files in the directory. I find a possible reason: When I built the package, it tells me that : ----------------------------------------------------------------- -bash-3.2$ ./bjam Building the Boost C++ Libraries. warning: Building Boost.Regex with the optional Unicode/ICU support disabled. note: Please refer to the Boost.Regex documentation for more information note: this is a strictly optional feature. warning: Graph library does not contain MPI-based parallel components. note: to enable them, add "using mpi ;" to your user-config.jam warning: skipping optional Message Passing Interface (MPI) library. note: to enable MPI support, add "using mpi ;" to user-config.jam. note: to suppress this message, pass "--without-mpi" to bjam. note: otherwise, you can safely ignore this message. Component configuration: - date_time : building - filesystem : building - graph : building - graph_parallel : building - iostreams : building - math : building - mpi : building - program_options : building - python : building - random : building - regex : building - serialization : building - signals : building - system : building - test : building - thread : building - wave : building ...patience... ...patience... ...patience... ...found 6754 targets... The Boost C++ Libraries were successfully built! The following directory should be added to compiler include paths: /lustre/jxding/boost_2010_6_8/boost_1_43_0 The following directory should be added to linker library paths: /lustre/jxding/boost_2010_6_8/boost_1_43_0/stage/lib ------------------------------------------------------------------------------------------ It seems that I need to specify I need to use mpi for the bjam builder. But, I have added "using mpi ;" and "using mpi : /opt/openmpi-gnu/bin/mpic++ ;" in the file /home/myname/boost_2010_6_8/boost_1_43_0/user-config.jam. I have also added it in the file /home/myname/boost_2010_6_8/boost_1_43_0/tools/build/v2/user-config.jam. But, I still got the same error. It seems that there is something wrong with the package ? or the compiler cannot understand the "using mpi ;" in the user-config.jam ? Any help is appreciated. Jack June 13 2010 _________________________________________________________________ The New Busy is not the too busy. Combine all your e-mail accounts with Hotmail. http://www.windowslive.com/campaign/thenewbusy?tile=multiaccount&ocid=PID28…

3 3

dynamic_properties and std::vector attributes
by Dominik Dahlem 13 Jun '10

13 Jun '10

Hi guys, the boost graph library has an example on using bundled properties. I would like to serialise the resulting graph into graphml. Using dynamic_properties for the std::vector attribute fails to compile with boost 1.41. Is there a way to get this example below to work? I would greatly appreciate it. Thanks, Dominik #include <vector> #include <string> #include <boost/graph/adjacency_list.hpp> #include <boost/property_map/dynamic_property_map.hpp> struct City { std::string name; int population; std::vector<int> zipcodes; }; struct Highway { std::string name; double miles; int speed_limit; int lanes; bool divided; }; int main() { typedef boost::adjacency_list< boost::listS, boost::vecS, boost::bidirectionalS, City, Highway> Graph; Graph g; Graph::vertex_descriptor v = *vertices(g).first; g[v].name = "Troy"; g[v].population = 49170; g[v].zipcodes.push_back(12180); Graph::edge_descriptor e = *out_edges(v, g).first; g[e].name = "I-87"; g[e].miles = 10; g[e].speed_limit = 65; g[e].lanes = 4; g[e].divided = true; boost::dynamic_properties dp; // dp.property("zip_codes", boost::get(&City::zipcodes, g)); dp.property("name", boost::get(&City::name, g)); return EXIT_SUCCESS; } -- Dominik Dahlem <Dominik.Dahlem(a)scss.tcd.ie> ------------------------------------------------------- Trinity College Dublin School of Computer Science and Statistics, Dublin 2, Ireland tel: (+353) 1-896 3789 fax: (+353) 1-677 2204 www: http://www.cs.tcd.ie/Dominik.Dahlem/ -------------------------------------------------------

2 3

boost lib error
by Jack Bryan 13 Jun '10

13 Jun '10

Hi, all: I need to install boost library (/home/myname/boost_2010_6_8/boost_1_43_0). http://www.boost.org/doc/libs/1_43_0/more/getting_started/index.html It has MPI library : http://www.boost.org/doc/libs/1_42_0/doc/html/mpi.html After installingit, I compile some c++ program with calling for boost.MPI class, the compiler cannot find the header files in the program even though I have included the boost path in the compile command. example: ------------------------------------------------------------------------ #include </boost/mpi/environment.hpp> #include <boost/mpi/communicator.hpp> #include <iostream> namespace mpi = boost::mpi; int main(int argc, char* argv[]) { mpi::environment env(argc, argv); mpi::communicator world; std::cout << "I am process " << world.rank() << " of " << world.size() << "." << std::endl; return 0; } -------------------------------------------------------------------------- mpic++ -o testp2p test_p2p.cpp I/home/myname/boost_2010_6_8/boost_1_43_0 g++: I/home/myname/boost_2010_6_8/boost_1_43_0: No such file or directory test_p2p.cpp:7:38: error: /boost/mpi/environment.hpp: No such file or directory test_p2p.cpp:11:38: error: boost/mpi/communicator.hpp: No such file or directory Actually, I have installed these header files in the directory. I find a possible reason: When I built the package, it tells me that : ----------------------------------------------------------------- -bash-3.2$ ./bjam Building the Boost C++ Libraries. warning: Building Boost.Regex with the optional Unicode/ICU support disabled. note: Please refer to the Boost.Regex documentation for more information note: this is a strictly optional feature. warning: Graph library does not contain MPI-based parallel components. note: to enable them, add "using mpi ;" to your user-config.jam warning: skipping optional Message Passing Interface (MPI) library. note: to enable MPI support, add "using mpi ;" to user-config.jam. note: to suppress this message, pass "--without-mpi" to bjam. note: otherwise, you can safely ignore this message. Component configuration: - date_time : building - filesystem : building - graph : building - graph_parallel : building - iostreams : building - math : building - mpi : building - program_options : building - python : building - random : building - regex : building - serialization : building - signals : building - system : building - test : building - thread : building - wave : building ...patience... ...patience... ...patience... ...found 6754 targets... The Boost C++ Libraries were successfully built! The following directory should be added to compiler include paths: /lustre/jxding/boost_2010_6_8/boost_1_43_0 The following directory should be added to linker library paths: /lustre/jxding/boost_2010_6_8/boost_1_43_0/stage/lib ------------------------------------------------------------------------------------------ It seems that I need to specify I need to use mpi for the bjam builder. But, I have added "using mpi ;" and "using mpi : /opt/openmpi-gnu/bin/mpic++ ;" in the file /home/myname/boost_2010_6_8/boost_1_43_0/user-config.jam. I have also added it in the file /home/myname/boost_2010_6_8/boost_1_43_0/tools/build/v2/user-config.jam. But, I still got the same error. It seems that there is something wrong with the package ? or the compiler cannot understand the "using mpi ;" in the user-config.jam ? Any help is appreciated. Jack June 11 2010 _________________________________________________________________ The New Busy is not the too busy. Combine all your e-mail accounts with Hotmail. http://www.windowslive.com/campaign/thenewbusy?tile=multiaccount&ocid=PID28…

3 3

Boost.Test XML Output formatter generates Invalid XML file.
by Gaurav N Shah 13 Jun '10

13 Jun '10

Hi, I have written my own struct to customize logging from struct "boost::unit_test::output::xml_log_formatter" However, the XML file generated by log Formatter is Invalid. it adds invalid node end = "</>" , by the end of each test case logging. Please check below XML. <TestLog> <TestSuite name="Master Test Suite"> <TestSuite name="test_suite1"> <TestSuite name="MarketDataMasterTestSuite"> <TestSuite name="PostMarketData"> <TestSuite name="CTA"> <TestCase name="LevelOneData">error #LevelOneData#LevelOneData#STOCK$121@4106!132^SYMBOL&C#CLS_PRC_TD#0#NA</> <TestingTime>16000</TestingTime> </TestCase> <TestCase name="LevelOneData">error #LevelOneData#LevelOneData#STOCK$121@4106!132^SYMBOL&CHL#CLS_PRC_TD#0#NA</><TestingTime>0</TestingTime> </TestCase> </TestSuite> </TestSuite> </TestSuite> </TestSuite> </TestSuite> </TestLog> Has anyone faced scenario like this earlier?? Please let me know how to resolve this. Thank You, Gaurav This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase & Co., its subsidiaries and affiliates. This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by JPMorgan Chase & Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. Please refer to http://www.jpmorgan.com/pages/disclosures for disclosures relating to European legal entities.

6 8

count compressed bytes
by Kraus Philipp 13 Jun '10

13 Jun '10

Hi, I have implemented this litte code to compress a file bio::filtering_streambuf< bio::input > l_deflate; switch (m_compress) { case gzip : l_deflate.push( bio::gzip_compressor() ); break; case bzip2 : l_deflate.push( bio::bzip2_compressor() ); break; } l_deflate.push( bio::file_source(p_str, BOOST_IOS::binary) ); I need only the number of compressed bytes of l_deflate. I try to use l_deflate.size() but it returns only a fixed number. How I can read the number of compressed bytes? Thanks Phil

3 7

How to figure out if a boost class does lazy copy or not?
by Peng Yu 13 Jun '10

13 Jun '10

Hi, I need to construct some container in a function and return it. I think that some container in boost follow lazy copy sematics but some don't. For example, if tr1::unordered_map uses lazy copy, then I can return an object of tr1::unordered_map. Otherwise, I should return boost::shared_ptr<tr1::unordered_map>. Although this information may be documented for any container that I will use, I still think that it is still worthwhile to having my own test case to figure it out, because it may not easy to find such information in the document. I'm wondering if there is a generic way to test if a container use lazy copy or not without reading the actual implementation of the class? -- Regards, Peng

2 2

building 1.43 for iPhone
by Igor R 13 Jun '10

13 Jun '10

Hello, Few months ago I succesfully built 1.41 for iPhone, but it seems that the things changed since then. I'm trying to follow these recommendations (for 1.42): http://iphone.galloway.me.uk/2009/11/compiling-boost-for-the-iphone/ I define "iphone3.1.2" version in my user-config.jam, but bjam says: "feature.jam:485: in feature-validate-value-string from module feature error: 'iphone3.1.2' is not know value of feature <macosx-version>". Does it mean that the building system doesn't find user-config.jam? Thanks.

3 10

[bind] counterintuitive behavior with vector and auto_ptr
by Christian Henning 13 Jun '10

13 Jun '10

Hi there, I'm doing some job interview preparations. Always a good idea to go back to the basics and tinker around a bit. I came across a surprising behavior when using std::vector, std::auto_ptr, and boost::bind. Basically, I have a vector of auto_ptrs which is copied into a second vector, thus the first vector's auto_ptr loss their resource and point to nowhere. Basic stuff, as I said. Now, when I try to use an auto_ptr from the first vector using boost::bind everything still works. Meaning the code doesn't fail, although it should. I find that surprising. Why is boost::bind recreating the resource? Here is a my example code: #include <algorithm> #include <iostream> #include <vector> #include <boost/bind.hpp> using namespace std; struct A { A() { cout << "A::A();"; } ~A() { cout << "A::~A();"; } void foo() { cout << "A::foo();"; } }; auto_ptr< A > allocate_A() { return auto_ptr<A>( new A() ); } void call_foo( auto_ptr< A > p ) { p->foo(); } int main() { std::vector< auto_ptr< A > > v( 100 ); generate_n( v.begin(), 100, allocate_A ); for_each( v.begin(), v.end(), boost::bind( &A::foo, _1 )); std::vector< auto_ptr< A > > v2( 100 ); copy( v.begin(), v.end(), v2.begin() ); // auto_ptr in v should be empty now. // this fails //v[0]->foo(); // this fails //for_each( v.begin(), v.end(), call_foo ); // this doesn't fail for_each( v.begin(), v.end(), boost::bind( &A::foo, _1 )); return 0; } I'm using MSVC10 Express. Regards, Christian

5 6

Re: [Boost-users] downloading boost 1.43 using svn
by Steven Watanabe 13 Jun '10

13 Jun '10

AMDG Sean Farrow wrote: > Could someone give me the http path to download the 1.43 release from svn. > http://svn.boost.org/svn/boost/tags/release/Boost_1_43_0/ In Christ, Steven Watanabe

1 0

downloading boost 1.43 using svn
by Sean Farrow 13 Jun '10

13 Jun '10

Hi: Could someone give me the http path to download the 1.43 release from svn. Regards Sean.

1 0

concat files for compression
by Brian Wood 12 Jun '10

12 Jun '10

Am 12.06.2010 um 23:12 schrieb Brian Wood: Kraus Philipp: > Thanks I will take a look at your classes, biut I think, that you > "seperate" files. > I need this one: > > cat first_file > temp > cat second_file >> temp > gzip temp > > The zip functions is implementated but I don't know how to create the >"cat" into one > stream. You're welcome. With my implementation, it depends on the size of the sending buffer. If data from more than one file fits into the buffer, that data is added to the buffer. -- Brian Wood Ebenezer Enterprises http://webEbenezer.net (651) 251-9384 "The L-RD is my light and my salvation: whom shall I fear? the L-RD is the strength of my life; of whom shall I be afraid?"

1 0

Null Stream
by Kraus Philipp 12 Jun '10

12 Jun '10

Hi, I'm new at boost, but I don't know how to use the http://beta.boost.org/doc/libs/1_43_0/boost/iostreams/device/null.hpp device. I must use the boost::iostreams::copy to copy a stream to the /dev/ null, but I don't know to declare my out null-output stream. Can anyone help me to declare the stream? Thanks Phil

2 1

Re: [Boost-users] concat files for compression
by Brian Wood 12 Jun '10

12 Jun '10

Kraus Philipp: > On 2010-06-11 09:34:27 +0200, Roland Bock said: >> >> Hi, >> >> I see two options: >> >> a) You write a new multi_file_source class which can any number of input >> files. >> >> b) Instead of working with a source, you could work with a sink at the >> other end of the pipeline. Then just copy the content of the two files >> into the stream. > > Hi, > > I think I would like to use the second idea, but how I can do this? > I have some classes -- SendCompressedBuffer and ReceiveCompressedBuffer -- that may be helpful. SendCompressedBuffer uses bzip to compress data before it forwards the data. It is able to marshall files and other data. There's an archive -- fear not, it's all of 17,517 bytes -- that you can download and has the classes I mentioned. Brian Wood Ebenezer Enterprises http://webEbenezer.net (651) 251-9384 "The kingdom of heaven is like a treasure hidden in the field, which a man found and hid again; and from joy over it he goes and sells all that he has and buys that field."

2 1

mpi reduce on a vector
by Prevrhal, Sven 12 Jun '10

12 Jun '10

I have a std::vector in each process and would like to sum them up by element in a master process such that v_master[0] = v_worker1[0] + v_worker2[0] +... v_master[1] = v_worker1[1] + v_worker2[1] +... ... boost::mpi::reduce(world, v_worker, v_master, std::plus<int>(), masterID) doesn't cut it because std::add only works for individual vector elements but not for a whole vector. So I could define my own functor struct vplus : public std::binary_function<std::vector<int>, std::vector<int>, std::vector<int>> { // functor for operator+ std::vector<int> operator()(const std::vector<int>& _Left, const std::vector<int>& _Right) const { // apply operator+ to operands std::vector<int> v(_Left.size()); std::transform(_Left.begin(), _Left.end(), _Right.begin(), v.begin(), std::plus<int>()); return (v); } }; and apply it like so in the distributor boost::mpi::reduce(world, a, b, vplus(), 0); This works, but here is my question: the functor makes a new vector v that gets returned and (do I say this right?) passed on to boost::mpi::reduce and goes out of scope. This making of a new vector strikes me as superfluous and wasteful (think big vectors) - should I avoid it, and if so, how? Cheers,Sven working on CUDA+Thrust+MPI for medical image processing The information contained in this message may be confidential and legally protected under applicable law. The message is intended solely for the addressee(s). If you are not the intended recipient, you are hereby notified that any use, forwarding, dissemination, or reproduction of this message is strictly prohibited and may be unlawful. If you are not the intended recipient, please contact the sender by return e-mail and destroy all copies of the original message.

1 0

Type arithmetic library?
by Peng Yu 12 Jun '10

12 Jun '10

Hello, boost typeof library can automatically deduce the type. But I'm interested in finding the nested type. For example, if I have a template S<T1, T2>. I'd like to deduce that it has two template arguments, I'd like to access its template argument by index (say, 0 and 1 for this case). I did a preliminary search, but found no such facilities. I may overlook something. Would you please let me know if there is such a facility in boost? -- Regards, Peng

3 2

[program_options] linking to specific library version requested?
by Diederick C. Niehorster 12 Jun '10

12 Jun '10

Hi All, I use program_options happily. One minor annoyance is that the headers somewhere seems to specify which version of the library (the exact file name) they want to link with. I've tried setting "#define BOOST_PROGRAM_OPTIONS_NO_LIB ON" that I found by googling, but it made no difference. I always get "LINK : fatal error LNK1104: cannot open file 'libboost_program_options-vc100-mt-gd-1_44.lib'". Ideally, I'd like to link with libboost_program_options-vc100-mt-gd.lib so that when I upgrade boost I don't have to change the includes in my project files. More strangely, on one of the three computers I work with, next to the library corresponding to the newest boost version, it always requests to link to **-1_41.lib files as well, complete cleaning and rebuilding from scratch does not solve this. I have observed this with both MSVC2008 and MSVC2010. How can I configure program_options such that I can do the linking myself? Best and thanks! Dee

2 2

Adding include information the reference of typeof
by Peng Yu 12 Jun '10

12 Jun '10

Hi, I feel that that the reference page of typeof can be improved by adding what header file should be included. Could the authors add such information in the future? -- Regards, Peng

1 0

[boost/Quickbook] Build documentation with several files
by Carlos Guimarães 12 Jun '10

12 Jun '10

Hi, I have several *.qbk files and I want to build documentation of all of them at one time. They are all in the same folder, so I use the same Jamfile for each one: *using quickbook ; import boostbook : boostbook ; # Intro xml intro : src/intro.qbk ; # Libraries xml base_lib : src/libs/base.qbk ; # App xml client : src/app/client.qbk ; boostbook documentation : intro : <dependency>css ; boostbook lib : base_lib : <dependency>css ; boostbook client : client : <dependency>css ; install css : [ glob ./*.css ] : <location>../html ; explicit css ;* However, they have been building at the same root folder. That fact left me only one index file, because the previous one have been overwrite. How can I resolve it? Thanks

2 3

BOOST_FOREACH infinite loop issue
by yabcok 11 Jun '10

11 Jun '10

Hi, I have a problem with BOOST_FOREACH never ending under specific conditions described here: http://stackoverflow.com/questions/2881491/need-help-with-boost-foreach-com… Could anyone help me understand what's going on? Best regards, Jacek Ławrynowicz

3 6

boost::function, pointer to a member function.
by Mikhail Tatarnikov 11 Jun '10

11 Jun '10

Hi guys, Is there a way to make boost::function to be passed to a function required a pointer to a member function as parameter: struct CFoo { int foo1(int){} int foo2(int, double){} }; int g(int (CFoo::*)(int)); int main() { g(&CFoo::foo1); g(boost::bind(&CFoo::foo2, _1, _2, 666)); // Doesn't compile - boost::bind returns an object instead of function pointer // The following doesn't compile - can't convert from int (*)(CFoo*, int) to int (CFoo::*)(int) g(boost::function<int (CFoo*, int)>( boost::bind(&CFoo::foo2, _1, _2, 666) ).target<int (CFoo*, int)>() ) } The initial problem is to bind a parameter to a member function. It's easy to solve the problem by changing g(), e.g.: template<typename Functor> int g(Functor fun); but is it possible to solve the problem is g() can't be changed? Regards, Mike.

3 2

[thread] mingw32: multiple definition of __tls_used
by Frédéric Bron 11 Jun '10

11 Jun '10

I have some troubles with mingw32 (compiling from cygwin with g++ 3.4.4): The code below fails at link with following error message: /usr/lib/gcc/i686-pc-mingw32/3.4.4/../../../../i686-pc-mingw32/lib/libmingw32.a(tlssup.o):tlssup.c:(.tls+0x0): multiple definition of `__tls_used' /cygdrive/d/Softs/1.7/mgw-release/boost/lib/libboost_thread_win32.a(tss_pe.o):tss_pe.cpp:(.rdata$T+0x0): first defined here collect2: ld returned 1 exit status #include <boost/thread/tss.hpp> int main() { boost::thread_specific_ptr<int> tsp; return 0; } Compiling command was: $ g++-3 -o toto -O3 -DNDEBUG -mno-cygwin -mthreads -DBOOST_WINDOWS_API -I/cygdrive/d/Softs/1.7/mgw-release/boost/include toto.cc -L/cygdrive/d/Softs/1.7/mgw-release/boost/lib -lboost_thread_win32 Recently, mingw32 runtime was updated for cygwin with the following message: > Implement TLS Callback. > > * tlsmcrt.c: New file. > * tlsmthread.c: Ditto. > * tlssup.c: Ditto. > * tlsthrd.c: Ditto. > * Makefile.in: Include new files. > * crt1.c: Implement TLS Callback. > * dllcrt1.c: Ditto. > * mthr_stub.c: Remove. Any clue? Frédéric

2 1

utf8_codecvt_facet.cpp
by John Dlugosz 11 Jun '10

11 Jun '10

Why does the common file utf8_codecvt_facet.cpp include utf8_codecvt_facet.hpp =before= including boost/config.hpp? The wrapper boost_1_43_0\libs\program_options\src\utf8_codecvt_facet.cpp must itself include config first or it won't work. So putting the config include here is useless as it can never be used there. --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 3

[MultiIndex] As if one iterator
by Dan Bloomquist 11 Jun '10

11 Jun '10

I have a muti_index for a database. First I'll say, this will be so much better than my old way of running an SQL every time I wanted to change the order. Thanks. I never see a database bigger than some 1200 records, usually <200, so I'm not worried about memory bloat. So I've wrapped the multi_index in a class that will hide the boost includes from the rest of my app. I'll include the work below. I'd like to treat some kind of interface, like a generic iterator, depending on what order I have set. I would only use one iterator at a time. But the iterators are templated and have no base so that I can easily keep a set of iterators handy, say in a container, and then supply the appropriate iterator through an interface. In my class, RaOrderRecords, I'd something like: set.GetIt( ) set.IncIt( ) set.IsEndIt( ) etc. Or even overload so I could set++ And a member like SetOrder( [enum of new order] ) to have the internal iterator switched. I'm sure there is an elegant way to deal with this, but it escapes me. Thanks, Dan. ~~~~~ .hpp #pragma once // // ............................................................ namespace RsOrderIndex { struct ord_key{ }; struct ord_id{ }; struct ord_desc{ }; struct ord_sevice{ }; struct ord_remain{ }; struct ord_cat{ }; struct ord_group{ }; enum RecOrderIndex { ordNull, ordKey, ordId, ordDesc, ordService, ordRemain, ordCat, ordGroup, }; }; // ............................................................ class RaOrderRecords; struct ord_record_item { //properties RaOrderRecords* pParent; size_t absPos; //indexed properties long key; std::string strId; std::string strDescription; COleDateTime dtInService; long remaining; long catKey; long groupKey; //one and only constructor ord_record_item( RaOrderRecords *pInParent, long inKey, CString& inId, CString& inDesc, COleDateTime& indtIns, long inRemain, long inCatKey, long inGroupKey, size_t inPos ) :pParent( pInParent ) ,key( inKey ) ,strId( inId ) ,strDescription( inDesc ) ,dtInService( indtIns ) ,remaining( inRemain ) ,catKey( inCatKey ) ,groupKey( inGroupKey ) ,absPos( inPos ) { } //useful members size_t GetAbsPos( ) const { return absPos; } //indexing members bool operator < ( const ord_record_item& e ) const { return key < e.key; } unsigned long cat_pos( ) const; unsigned long group_pos( ) const; }; // ............................................................ #ifdef INCLUDE_ORDRECORD_BASE //to isolate boost headers typedef boost::multi_index::multi_index_container < ord_record_item, boost::multi_index::indexed_by < // sort by primary database key, ord_record_item comp boost::multi_index::ordered_unique < boost::multi_index::tag< RsOrderIndex::ord_key >, boost::multi_index::identity< ord_record_item > >, // sort by less<string> on strId boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_id >, BOOST_MULTI_INDEX_MEMBER( ord_record_item, std::string, strId ) >, // sort by less<string> on strDescription boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_desc >, BOOST_MULTI_INDEX_MEMBER( ord_record_item, std::string, strDescription ) >, // sort by less<COleDateTime> on dtInService boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_sevice >, BOOST_MULTI_INDEX_MEMBER( ord_record_item, COleDateTime, dtInService ) >, // sort by less<COleDateTime> on remaining, Description boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_remain >, boost::multi_index::composite_key < ord_record_item, BOOST_MULTI_INDEX_MEMBER( ord_record_item, long, remaining ), BOOST_MULTI_INDEX_MEMBER( ord_record_item, std::string, strDescription ) > >, // sort by category, description boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_cat >, boost::multi_index::composite_key < ord_record_item, BOOST_MULTI_INDEX_CONST_MEM_FUN( ord_record_item, unsigned long, cat_pos ), BOOST_MULTI_INDEX_MEMBER( ord_record_item, std::string, strDescription ) > >, // sort by group, description boost::multi_index::ordered_non_unique < boost::multi_index::tag< RsOrderIndex::ord_group >, boost::multi_index::composite_key < ord_record_item, BOOST_MULTI_INDEX_CONST_MEM_FUN( ord_record_item, unsigned long, group_pos ), BOOST_MULTI_INDEX_MEMBER( ord_record_item, std::string, strDescription ) > > > >ord_record_set; typedef ord_record_set::index< RsOrderIndex::ord_key >::type::iterator It_key_ord; typedef ord_record_set::index< RsOrderIndex::ord_id >::type::iterator It_id_ord; typedef ord_record_set::index< RsOrderIndex::ord_desc >::type::iterator It_desc_ord; typedef ord_record_set::index< RsOrderIndex::ord_sevice >::type::iterator It_service_ord; typedef ord_record_set::index< RsOrderIndex::ord_remain >::type::iterator It_remain_ord; typedef ord_record_set::index< RsOrderIndex::ord_cat >::type::iterator It_cat_ord; typedef ord_record_set::index< RsOrderIndex::ord_group >::type::iterator It_group_ord; nth_index_iterator #else // INCLUDE_ORDRECORD_BASE //forward declared for general use, no multi record boost headers class ord_record_set; #endif // INCLUDE_ORDRECORD_BASE // ............................................................ class RaOrderRecords { ord_record_set* pOrder; RaCatGroupMap catSet; RaCatGroupMap groupSet; //RSKeyPosSet keyPosSet; public: RaOrderRecords( ); ~RaOrderRecords( ); bool ReadRecordSet( RaRecordset* pRS ); unsigned long GetCatPos( long inCatKey ) { RaCatGroupMapIt it= catSet.find( inCatKey ); if( it != catSet.end( ) ) return it->second.ordered; else return -1;//max unsigned } unsigned long GetGroupPos( long inGroupKey ) { RaCatGroupMapIt it= groupSet.find( inGroupKey ); if( it != groupSet.end( ) ) return it->second.ordered; else return -1;//max unsigned } #ifdef _DEBUG void Dump( RsOrderIndex::RecOrderIndex index ); #else void Dump( RsOrderIndex::RecOrderIndex index ) { } #endif }; ~~~~~~.cpp #include "stdafx.h" #if !defined(NDEBUG) #include <boost/multi_index/safe_mode_errors.hpp> #define BOOST_MULTI_INDEX_ENABLE_INVARIANT_CHECKING #define BOOST_MULTI_INDEX_ENABLE_SAFE_MODE #endif #include <boost/multi_index_container.hpp> #include <boost/multi_index/ordered_index.hpp> #include <boost/multi_index/identity.hpp> #include <boost/multi_index/member.hpp> #include <boost/multi_index/mem_fun.hpp> #include <boost/multi_index/composite_key.hpp> #include "records.h" #define INCLUDE_ORDRECORD_BASE #include "RecordsOrd.hpp" // ......................................................... using namespace boost::multi_index; using namespace RsOrderIndex; unsigned long ord_record_item::cat_pos( ) const { return pParent->GetCatPos( catKey ); } unsigned long ord_record_item::group_pos( ) const { return pParent->GetGroupPos( groupKey ); } RaOrderRecords::RaOrderRecords( ) { pOrder= new ord_record_set; } RaOrderRecords::~RaOrderRecords( ) { delete pOrder; } bool RaOrderRecords::ReadRecordSet( RaRecordset* pRS ) { pOrder->clear( ); LoadCategories( catSet ); LoadGroups( groupSet ); COleDateTime dtNo( 10000.0 );// TODO !!!! try { int i= 1; for( pRS->MoveFirst( ); ! pRS->IsEOF( ); pRS->MoveNext( ) ) { pOrder->insert( ord_record_item( this, pRS->ds_id, pRS->ds_number, pRS->ds_description, pRS->IsFieldNull( &pRS->ds_inService ) ? dtNo : pRS->ds_inService, pRS->GetRemaining( ), pRS->ds_catKey, pRS->ds_groupKey, i ) ); // keyPosSet.SetKeyPos( pRS->ds_id, i++ ); } } catch( CDBException* e ) { e->Delete( ); return false; } return true; } #ifdef _DEBUG void RaOrderRecords::Dump( RecOrderIndex index ) { std::string strT; strT.c_str( ); It_cat_ord it= pOrder->get< ord_cat >( ).begin( ); for( int i= 1; it != pOrder->get< ord_cat >( ).end( ); ++it, ++i ) { CString strT; RaCatGroupMapIt cat= catSet.find( it->catKey ); if( cat != catSet.end( ) ) strT= cat->second.title; else strT= _T("(null)"); TRACE( "%3.0d %s---%s\n", i, strT, it->strDescription.c_str( ) ); } //It_service_ord sit= pOrder->get< ord_sevice >( ).begin( ); //for( ; sit != pOrder->get< ord_sevice >( ).end( ); ++sit ) // TRACE( "%s---%s\n", sit->strDescription.c_str( ), sit->dtInService.Format( _T("%m/%d/%y") ) ); } #endif //_DEBUG

4 9

[Filesystem] create dir tree
by John Dlugosz 11 Jun '10

11 Jun '10

In the past, I've used a function like "assure_path_exists" which will create it if necessary, multiple levels deep if the parent(s) don't already exist. Does the creation function in this library create intermediate directories? The docs are not clear. I also want to efficiently and silently accept the case where the directory already exists. Does the creation function do that (how to tell real errors?), or do I need to check first? --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 1

Filesystem question
by John Dlugosz 11 Jun '10

11 Jun '10

Is there a function in the filesystem (TR2 proposal) that will take one path and return the fully qualified form? That is, what it resolves to now, applying any current-working-directory concepts. The Win32 API function I'm using now (with plain strings) is GetFullPathName. --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 1

[Filesystem] remove (p,ec);
by John Dlugosz 10 Jun '10

10 Jun '10

I call remove (p,ec); and get a compile-time error that no form takes two arguments. Looking in the header, that seems to be the case. The Filesystem Reference (<http://www.boost.org/doc/libs/1_43_0/libs/filesystem/doc/reference.html>) clearly shows an error code parameter. I understand this module is tracking the standardization process. So which is newer? Should I be anticipating the code change to match the docs, or vice versa? How do I find out if there was an error and report the details? Note that I want to consider the case of the file not existing in the first place as OK. --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

1 0

[Filesystem] what is 'singular'?
by John Dlugosz 10 Jun '10

10 Jun '10

template <class Path> bool remove(const Path& p, system::error_code & ec = singular ); * if ec != singular, then ec.clear(). I can't find any definition for 'singular'. What is it? Besides acting as a flag to know if the argument was defaulted, it seems like it's doing some kind of one-shot behavior? --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

1 0

Filesystem Path question
by John Dlugosz 10 Jun '10

10 Jun '10

The basic_path constructor that takes two InputIterators: what kind of iterator is that? It doesn't seem to want characters, but an iterator of parts like produced by the begin/end functions. But the iterator's underlying type is not defined, so you can't make your own collection of "parts". So why doesn't the constructor just take its own iterator type, but is a template? What else could you give it? The documentation is not clear. --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

1 0

bz2 & gzip support - ld can't find symboles
by Kraus Philipp 10 Jun '10

10 Jun '10

Hi, I try to use the bz2 and gzip iostreams from Boost 1.43 on my OSX Snow Leopard. The code is compiled without any errors, but the ld returns: Undefined symbols: "boost::iostreams::detail::zlib_base::do_init(boost::iostreams::zlib_params const&, bool, void* (*)(void*, unsigned int, unsigned int), void (*)(void*, void*), void*)", referenced from: boost::iostreams::detail::zlib_compressor_impl<std::allocator<char> >::zlib_compressor_impl(boost::iostreams::zlib_params const&)in ncd.o [...] I had setup the -l G++ option to "z" and "bz2" but I can't link my code. I had compiled the Boost from sources with bjam -sHAVE_ICU=1 -sZLIB_BINARY -sBZIP2_BINARY --with-mpi toolset=darwin install --prefix= Binaries for the zip algorithms are under /usr/lib After compiling with bjam I get 4 libs: libbost_mpi.dylib, libbost_serialization.dylib and both as *.a files. Do I have forget anything on compile / link option? I had found an articel for Windows in which bjam should create some boost zip libraries. I try the example on http://www.boost.org/doc/libs/1_43_0/libs/iostreams/doc/classes/zlib.html#i… but it run into errors. Thanks for help Phil

3 2

Program Options and utf8_codecvt_facet.hpp
by John Dlugosz 10 Jun '10

10 Jun '10

The Program Options library source includes the common utf8_codecvt_facet stuff. But, I don't see any use of it. Why is it in this lib/dll? --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 2

[proto] Matching grammars in the presence of adapted types
by Sebastian Redl 10 Jun '10

10 Jun '10

Hi, First off, this whole thing uses Boost 1.42. I'm trying to define arithmetic types where some expression patterns will be matched and replaced by something else (e.g. a & ~b, the andnot pattern). The whole thing is a proof-of-concept and teach-myself-proto thing that should eventually do SIMD operations. Right now, though, it simply wraps integers. First I tried defining my type as a proto type this way: ----------------- template <typename Expr> class simd_expr : public proto::extends<Expr, simd_expr<Expr>, simd_domain> { /* as in the tutorial */ }; class smartint : public simd_expr<proto::terminal<smartint>::type> { int value; // ... }; inline smartint ilit(int i) { return smartint(i); } ----------------- However, this didn't work, because at the instantiation point of simd_expr<proto::terminal<smartint>::type>, smartint is incomplete, and something in proto doesn't like that. So I used the only thing in the docs that looked like my use case, and did the non-intrusive adapting thing demonstrated on the matrix and vector classes. ----------------- class smartint { ... }; template <typename T> struct is_terminal : mpl::false_ {}; template <> struct is_terminal<smartint> : mpl::true_ {}; BOOST_PROTO_DEFINE_OPERATORS(is_terminal, simd_domain) ----------------- This works nicely for normal evaluations. I have a simd_context, which knows how to evaluate smartint, and forwards everything else to default_context: ----------------- struct simd_context : proto::callable_context<simd_context const, proto::default_context> { typedef int result_type; int operator ()(proto::tag::terminal, smartint const &si) const; }; inline int simd_context::operator ()( proto::tag::terminal, smartint const &si) const { return si.get(); } ----------------- Now I can do fun stuff like this: ----------------- int i = ilit(7) + 3; // simd_expr has 'operator int' ----------------- Now I want to match specific expression patterns and do something special for them. For example, in this expression: ----------------- i = 1 * (0xF0 & ~ilit(0x70)); ----------------- I want to match the 'a & ~b' pattern and treat it specially. So first, I define this pattern as a grammar: ----------------- struct andnot_pattern : proto::bitwise_and<proto::_, proto::complement<proto::_>> {}; ----------------- And then I add another function to my context that applies only to expressions matching this pattern: ----------------- template <typename Tag, typename Expr> typename boost::enable_if< proto::matches<Expr, andnot_pattern>, int>::type operator ()(Tag, Expr const &e) const { std::cerr << "Found andnot pattern\n"; return 0xFF; } ----------------- However, when I try to compile this, VC++ 2010 complains: boost_1_42_0\boost\proto\matches.hpp(502): error C2039: 'proto_base_expr' : is not a member of 'smart::smartint' Pointing here: ----------------- template<typename Expr, typename Grammar> struct matches : detail::matches_< typename Expr::proto_base_expr , typename Grammar::proto_base_expr > {}; ----------------- So apparently, while callable_context is trying to find an appropriate overload for some stuff, it tries out my template operator, passing a smartint as Expr, and matches<> can't handle adapted types. Because the error is within matches<>, it's not SFINAEable and the compiler errors out instead of eliminating the function. What can I do? Can I teach matches<> to handle adapted types? Is there a "right" way to implement smartint so that it needn't be an adapted type, which would work around the whole issue? Is this a bug in Proto that is fixed in trunk, or can be fixed? Sebastian

3 5

Testing private methods in Boost
by hmiller＠hiwaay.net 10 Jun '10

10 Jun '10

Is their a good strategy for testing private methods in a class using Boost? I am looking for something that is non-intrusive to the developed code, but I can't think of any good ways without modifying the code under test. Thanks, Herb Miller

6 6

boost.MPI test error
by Jack Bryan 10 Jun '10

10 Jun '10

Hi All: I have installed boost lib also including mpi. I am trying to run some simple mpi program to test it. my program: (test_p2p.cpp)------------------------------------------------------- #include <boost/mpi/environment.hpp>#include <boost/mpi/communicator.hpp>#include <iostream>namespace mpi = boost::mpi; int main(int argc, char* argv[]) { mpi::environment env(argc, argv); mpi::communicator world; std::cout << "I am process " << world.rank() << " of " << world.size() << "." << std::endl; return 0;}-------------------------------------------------------- my compile command: g++ -o testp2p test_p2p.cpp I got error : /boost/mpi/environment.hpp:16:32: error: boost/mpi/config.hpp: No such file or directory But , I have checked the directory: boost/mpi/, it actually have config.hpp. I have added the path for these files into my PATHPATH=/home/myfolder/boost_2010_6_8/boost_1_43_0 also LD_LIBRARY_PATH=/home/myfolder/boost_2010_6_8/boost_1_43_0 why I still have this error. Any help is really appreciated. thanks Jack June 9 2010 _________________________________________________________________ The New Busy is not the too busy. Combine all your e-mail accounts with Hotmail. http://www.windowslive.com/campaign/thenewbusy?tile=multiaccount&ocid=PID28…

3 6

shared_ptr + bind, is it ok?
by Matheus Araújo Aguiar 10 Jun '10

10 Jun '10

Hello list, In my application i am using a shared_ptr to manage the lifetime of some important objects. Lately i've been noticing that they are not being destroyed (their destructor is never called) and their use count is 9. I've started debugging and noticed that if i pass the shared_ptr to boost::bind as a parameter when the binded function is executed the use_count is greater than what would be expected (in my particular case 4 instead of one). As i make use of boost::bind, the shared_ptr reaches a point where its use_count is way out of expectations, getting to 9. So is this expected? Am i incorrectly using boost::bind or shared_ptr in this situation? Sorry not giving more detailed info or code, for now this is the info i can give. Thank you all for your time, -- Matheus Araújo Aguiar Computer Scientist matheus.pit(a)gmail.com

2 2

Re: [Boost-users] STL and boost
by yushang 10 Jun '10

10 Jun '10

But after add path to cstdlib I got following error: 1>d:\boostlib\boost\tr1\detail\config_all.hpp(150) : error C1083: Cannot open include file: '../include/utility': No such file or directory I've checked config_all.hpp found following line generate the error #define BOOST_TR1_STD_HEADER(name) <../include/name> and tried to modify this line , make it point to my utility file but got more errors 1>errors in directory e:\src\matrix\test 1>d:\winddk\7600.16385.0\inc\api\crt\stl70\iosfwd(202) : error C2144: syntax err or : 'int' should be preceded by ';' 1>d:\winddk\7600.16385.0\inc\api\crt\stl70\iosfwd(202) : error C2143: syntax err or : missing ';' before '*' 1>d:\winddk\7600.16385.0\inc\api\crt\stl70\iosfwd(202) : error C2061: syntax err or : identifier '_Elem' 1>d:\winddk\7600.16385.0\inc\api\crt\stl70\iosfwd(209) : error C2143: syntax err or : missing ';' before '*' 1>d:\winddk\7600.16385.0\inc\api\crt\stl70\iosfwd(209) : error C2086: 'int std:: char_traits<_Elem>::_Elem' : redefinition ... I'm totally confused

1 0

STL and boost
by yushang 10 Jun '10

10 Jun '10

Hi dear all, What is the boost STL including logic ? I'm using WDK build enveriment which include a STL70 of MS . Following is the detail message 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\warning.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\suppress.h 1>Note: including file: D:\boostlib\boost\tr1\tr1\iostream 1>Note: including file: D:\boostlib\boost/tr1/detail/config_all.hpp 1>Note: including file: D:\WinDDK\7600.16385.0\inc\crt\cstddef 1>Note: including file: D:\WinDDK\7600.16385.0\inc\crt\stddef.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\crt\crtdefs.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\specstrings.h 1>Note: including file: d:\winddk\7600.16385.0\inc\api\sal_supp.h 1>Note: including file: d:\winddk\7600.16385.0\inc\api\specstrings_supp.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\sal.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\specstrings_strict.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\specstrings_undef.h 1>Note: including file: D:\WinDDK\7600.16385.0\inc\api\driverspecs.h 1>Note: including file: d:\winddk\7600.16385.0\inc\api\sdv_driverspecs.h 1>d:\boostlib\boost\tr1\detail\config_all.hpp(48) : fatal error C1083: Cannot open include file: 'cstdlib': No such file or directory Thanks in advance

2 1

Re: [Boost-users] issues with boost 1.43
by Jeremiah Willcock 10 Jun '10

10 Jun '10

On Wed, 9 Jun 2010, Sean Farrow wrote: > > Hi: > > When compiling boost 1.43.0 (release) from the website, I get the following errors with the GraphMl library: This is <URL:https://svn.boost.org/trac/boost/ticket/4092> which is fixed in the trunk. You should be able to get a patch at <URL:https://svn.boost.org/trac/boost/changeset/61606?format=diff&new=61606>. -- Jeremiah Willcock

2 1

Re: [Boost-users] [Program Options] Requiring command-line options
by Jeremy Abramson 10 Jun '10

10 Jun '10

Since there isn't any documentation about this, what's the expected behavior here? Does it require that if you include the argument, you must include a parameter, or does it mean that you have to require the argument itself? The latter is the preferred behavior (for me) but from some brief experimentation, that doesn't seem to be the case. Of course, I'm not even sure I'm using it correctly, since it's not documented anywhere. Any tips? Thanks! J. > Message: 5 Date: Fri, 4 Jun 2010 11:40:00 +0800 From: "Diederick C. > Niehorster" <dcnieho(a)gmail.com> To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] [Program Options] Requiring command-line > options Message-ID: > <AANLkTilD9kvnz0sU03y5kh5CUwWKNLCXxWid-zwAicot(a)mail.gmail.com> > Content-Type: text/plain; charset=UTF-8 So this is not in the > documentation? Could you add a ticket in track to fix the > documentation? Best, Dee On Thu, Jun 3, 2010 at 9:09 PM, immanuel > litzroth <ilitzroth(a)gmail.com> wrote: >> > This seems to work for me: >> > ?("namespace", po::value<std::string>(&args.name_space)->required(),"namespace >> > to scrub") >> > throws an exception when appropriate: >> > ERROR - main: Caught an exception exiting main: missing required >> > option namespace >> > I remember having to hunt for it too. >> > Greetz, >> > Immanuel >> > >> > On Thu, Jun 3, 2010 at 9:41 AM, Jeremy Abramson <abramson(a)isi.edu> wrote: >> >>> >> Heya! ?I just thinking about switching from simpleOpt to BoostPO for command >>> >> line handling, mainly because of its support of configuration files. ?This >>> >> may be a stupid question, but I haven't seen anything about making a command >>> >> line parameter *required*. ?You'd think this is a pretty common thing, but I >>> >> didn't see any mention of it in the tutorials or anything. ?I guess I could >>> >> just make certain parameters default -1/NULL or something, and then test at >>> >> the end if they're still equal to that, but that seems a bit ugly. >>> >> Is there built in support (a flag in the "add_options" function or >>> >> something) for making certain parameters required? >>> >> >>> >> Thanks a lot! >>> >> >>> >> _______________________________________________ >>> >> Boost-users mailing list >>> >> Boost-users(a)lists.boost.org >>> >> http://lists.boost.org/mailman/listinfo.cgi/boost-users >>> >> >>> >> > _______________________________________________ >> > Boost-users mailing list >> > Boost-users(a)lists.boost.org >> > http://lists.boost.org/mailman/listinfo.cgi/boost-users >> > >> > >

2 1

issues with boost 1.43
by Sean Farrow 09 Jun '10

09 Jun '10

Hi: When compiling boost 1.43.0 (release) from the website, I get the following errors with the GraphMl library: compile-c-c++ bin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static\gr aphml.obj graphml.cpp C:\Program Files\boost\boost/property_tree/detail/ptree_implementation.hpp(30) : error C2440: 'specialization' : cannot convert from 'const std::string std::_Pair_base<_Ty1,_Ty2>::* ' to 'const std::basic_string<_Elem,_Traits,_Ax> std::pair<_Ty1,_Ty2>::* ' with [ _Ty1=const std::string, _Ty2=boost::property_tree::basic_ptree<std::string,std::string> ] and [ _Elem=char, _Traits=std::char_traits<char>, _Ax=std::allocator<char>, _Ty1=const std::string, _Ty2=boost::property_tree::basic_ptree<std::string,std::string> ] Standard conversion from pointer-to-member of base to pointer-to-member of derived is not applied for template arguments C:\Program Files\boost\boost/property_tree/detail/ptree_implementation.hpp(78) : see reference to class template instantiation 'boost::property_tree::basic_ptree<Key,Data>::subs' being compiled with [ Key=std::string, Data=std::string ] C:\Program Files\boost\boost/foreach.hpp(265) : see reference to class template instantiation 'boost::property_tree::basic_ptree<Key,Data>::const_iterator' being compiled with [ Key=std::string, Data=std::string ] libs\graph\src\graphml.cpp(37) : see reference to class template instantiation 'boost::foreach_detail_::auto_any<T>' being compiled with [ T=boost::property_tree::basic_ptree<std::string,std::string>::const_iterat or ] call "C:\Program Files\Microsoft Visual Studio 10.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static\ graphml.obj.rsp" ...failed compile-c-c++ bin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static\gr aphml.obj... ...skipped <pbin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static> libboost_graph-vc100-sgd-1_43.lib for lack of <pbin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static> graphml.obj... ...skipped <pstage\lib>libboost_graph-vc100-sgd-1_43.lib for lack of <pbin.v2\libs\graph\build\msvc-10.0\debug\link-static\runtime-link-static> libboost_graph-vc100-sgd-1_43.lib... ...skipped <pstage\lib>libboost_graph-vc100-sgd.lib for lack of <pstage\lib>libboost_graph-vc100-sgd-1_43.lib... Visual studio 2010, windows xp 32-bit. Could a patch be provided, or is there a fix. I've not checked with the latest svn trunk, as I'm restricted to building with 1.43. Cheers Sean.

1 0

[Property Tree] Reading & Writing XML
by Brent Arias 09 Jun '10

09 Jun '10

Regarding the write_xml routine used with the boost Property Tree, it evidently does not utilize the underlying RapidXML tool? Said differently, is it true Property Tree relies on RapidXML for the reading step, but not the writing step? I ask because I may be swapping Property Tree's dependence on RapidXML, for a Unicode branch of pugiXML<http://code.google.com/p/pugixml/source/browse/#svn/branches/unicode> instead, so that I get free conversions to-and-from UTF-16/UTF-8. But if Property Tree does not use the underlying XML parser for the writing step, then I won't get that free conversion when writing back to disk. As for the read_xml func, is it also true that "twice" the memory is needed, because RapidXML will load a document into a DOM, and then the Property Tree will allocate more memory to convert the DOM content into its own format? Brent Arias Software Architect Lumension Security(tm) Tel ~ (480) 444-1607 Fax ~ (480) 970-6323 brent.arias(a)lumension.com<mailto:brent.arias@lumension.com>

2 1

Boost lambda bind and boost smart_ptr issue
by Brian Martin 09 Jun '10

09 Jun '10

Hi all, I have a problem involving Boost Lambda and shared_ptr. Although this has come up before, my problem is a little different. Essentially, I have classes derived from bases that contain pure virtuals. These are stored in STL containers as shared_ptrs (to the base) and to avoid having to write all those pesky functions/functors for STL algorithms, it would be nice to use Lambda instead. The following is similar to the sort of code I'm writing: class shape { public: virtual void draw(const int)=0; }; class rectangle : public shape { public: virtual void draw(const int x){std::cout << "rectangle::draw() with x = " << x << "\n";} }; class circle : public shape { public: virtual void draw(const int x) {std::cout << "circle::draw() with x = " << x << "\n";} }; vector< shared_ptr<shape> > shapes; // assume I stick stuff into shapes... for_each(shapes.begin(), shapes.end(), bind(&shape::draw, _1, 2)); The last statement fails under VC++ 2008 with error C2665: 'boost::lambda::function_adaptor<Func>::apply' : none of the 2 overloads could convert all the argument types This works if you use boost::bind instead of the lambda bind. The rest of the error message (too lengthy to list) seems to be telling me that there is an ambiguity between the two following types: Result boost::lambda::function_adaptor<Func>::apply<RET,A3>(Result (__thiscall shape::* )(Arg1),Object *,A1 &) Result boost::lambda::function_adaptor<Func>::apply<RET,A3>(Result (__thiscall shape::* )(Arg1),Object &,A1 &) I really could use a solution to this since it's stopping me from using lambda more often. Thanks in advance, Brian Martin

2 4

DLL names for special options
by John Dlugosz 09 Jun '10

09 Jun '10

I need to build 2 Boost libraries that are compatible with my project's debug build options. First, what do I name it? The section on "Library Naming" shows ...-gd... being included in the options, etc. But, how do I tag the name as having my needed #define options configured? In particular, _HAS_ITERATOR_DEBUGGING=0;_BIND_TO_CURRENT_VCLIBS_VERSION=1 causes incompatible code in standard includes and links to different runtimes. I also want to compile to be compatible with a so-called Debug build, but not disable optimizations within Boost code since I don't plan on stepping through it. On a related question, I don't see any marker in the name for platform, so how are x86 versions distinguished from x64 builds of the same code? Now, since I'm already trying to get every team and project in this large application suite to use a single "build_conventions" property sheet in Visual Studio to keep such things in one place and indeed matching for everyone, the simplest thing to do, since I only need two boost libraries so treated, is to just make a Visual Studio project with my "sheet" and all the source files, and let'er'rip. Assuming that's a good idea, I still need to know how to get the "auto link" feature to point to my correct version name. I see from auto_link.hpp that BOOST_LIB_RT_OPT is where the "gd" part is put in, and that might be where some of these are encoded. But I don't see anything for separate optimization vs. debug-target, platform type, etc. If I just throw in a 'z' to mean "my stuff", where is BOOST_LIN_RT_OPT supposed to be set? What would I change so that team members will automatically link against the custom versions? Thanks, --John (sorry about the footer) TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

1 0

[BGL] graphviz example
by Adam Spargo 09 Jun '10

09 Jun '10

Hi, I am trying to run Dijkstra example from the BGL book (chapter 5). I have built libbgl-viz.a and I am trying to link it like this: CPP = g++ CFLAGS = -O2 IFLAGS = -I ~/src/boost_1_42_0 LFLAGS = -L ~/src/boost_1_42_0/libs/graph/src ospf: ospf.o $(CPP) -o $@ $(LFLAGS) -l bgl-viz ospf.o ospf.o: ospf.cpp $(CPP) $(CFLAGS) $(IFLAGS) -c -o $@ ospf.cpp it seems to find the lib okay, but I get an error like this: ospf.cpp:(.text+0xa2): undefined reference to `boost::read_graphviz(std::basic_s ... Is there some obvious pitfall, or maybe someone could help me work it out? Thanks, Adam. -- Dr Adam Spargo High Performance Assembly Group email: aws(a)sanger.ac.uk Wellcome Trust Sanger Institute Tel: +44 (0)1223 834244 x7728 Hinxton, Cambridge CB10 1SA Fax: +44 (0)1223 494919 -- The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE.

2 1

boost::format help to format doubles.
by Mauricio Gomes 09 Jun '10

09 Jun '10

Hi I read the documentation and looked for some examples but I was not able to come up with a format string to do this: Format a double using desired grouping char and decimal separator. So 1234567.89 should print 1,234,567.89 First, how can I specify the usage of the grouping char ? sprintf uses the ' char but it gave me invalid format string every time I tried to use it. Also how can I specify the grouping char and decimal separator ? By setting the locale ? An example would be great. std::stringstream ss2; ss2 << boost::format ("%|.2f|") % src.get_valor_total (); String svalor_total = ss2.str (); Thank you in advance for your attention, Mau.

2 2

passing a shared pointer
by Lloyd 09 Jun '10

09 Jun '10

Hi, I have a shared pointer like this class a { }; boost::shared_ptr<a> Ptr(new a()); I need to pass this pointer as an argument to function (MyFunc) which access this as a reference void MyFunc(a& ref) { } Now I am calling the function like this MyFunc(*Ptr); But this causes the "Ptr" to destroy early (I am using this for an asynchronous-ASIO- operation) How can I do this right? Thanks a lot, Lloyd ______________________________________ Scanned and protected by Email scanner

3 4

contiguous Boost.Multimap?
by anony 09 Jun '10

09 Jun '10

Is it somehow possible to coax boost::multimap to store data contiguously, i.e. that all multimap elements are stored contiguously, without gaps between them, and all in the same memory block?

3 5

[Property Tree] Replace RapidXML with pugiXML?
by Brent Arias 09 Jun '10

09 Jun '10

How difficult is it to replace the RapidXML XML parser with another parser, such as pugiXML? Is this a change-case that has been planned and streamlined, and therefore a straight-forward effort, or is it a lengthy and detailed affair that requires intimate knowledge of PropertyTree internal workings? I ask because my organization predicts we likely will change back-and-forth between competing implementations (RapdidXML, Kranf xmlParser, pugiXML) as time goes by, which implies that we should have some kind of wrapper around the XML parser. I am investigating how well boost's "property tree" could server as that wrapper. And, in fact, we currently want pugiXML because of it's wide-char support<http://code.google.com/p/pugixml/source/browse/#svn/branches/unicode>, and because it has XPath support (granted PropertyTree assuredly won't "wrap" the XPath capability, but we'll deal with that separately). Brent Arias Software Architect Lumension Security(tm) Tel ~ (480) 444-1607 Fax ~ (480) 970-6323 brent.arias(a)lumension.com<mailto:brent.arias@lumension.com>

2 1

initialization of boost::thread::id object
by Boost lzw 09 Jun '10

09 Jun '10

Hi, I have a class (for illustrative purpose only): class X { public: X() : id_(0) // does not compile { } void f() { if (operation_failed) id_ = 0; // does not compile else id_ = boost::this_thread::get_id(); // get which thread is sucessfully executing f() } bool compare() { // compare current thread with the one that has sucessfully executed f() return id_ == boost::this_thread::get_id(); } private: boost::thread::id id_; }; My questions are: How can I set the id_ to an initial value (e.g., 0 here) for compare() function to get correct result, assuming multiple threads might be executing f() and compare() at the same time? Is there a way to do this: atomic<boost::thread::id> id_; to make them thread-safe? Thanks a lot in advance, Robert

2 3

Serialization of polymorphic types defined by plugins loaded at runtime
by Sam Spilsbury 09 Jun '10

09 Jun '10

Hi Everyone, I'm working on a project called "compiz" which is a plugin-based compositing window manager for X11 written in C++[1]. Recently, I've been adding serialization through the use of a wrapper class such that plugin class inherits this wrapper class, which automatically loads serialized data when instantiated and saves it when destroyed (by using a function that is called at the beginning of the plugin class destructor. My main problem at the moment is that I have a boost::ptr_vector of a class called "Element", which is a base for which other plugins derive and add their own "move" "start" "stop" functions as well as any additional properties they wish to attach to the "Element". So for example, this is where the ptr_vector is kept: class PrivateElementAnimation { public: PrivateElementAnimation (CompString type, int nElement, int size, int speed, int iter, bool rotate); PrivateElementAnimation () {}; template <class Archive> void serialize (Archive &ar, const unsigned int version) { ar & mElements; // Exception unregistered_type is raised here } boost::ptr_vector<Element> mElements; }; This is the definition of the class "Element" class Element { public: Element (); ElementAnimation *anim; float x, y, z; float dx, dy, dz; float rSpeed; int rDirection; int rAngle; float opacity; template <class Archive> void serialize (Archive &ar, const unsigned int version) { fprintf (stderr, "saving an element\n"); ar & x; ar & y; ar & z; ar & dx; ar & dy; ar & dz; ar & rSpeed; ar & rDirection; ar & rAngle; ar & opacity; } virtual bool init (); virtual void move (); virtual void fini (); private: void defaultInit (); void regenerateOffscreen (); friend class ElementAnimation; friend class PrivateElementScreen; }; And this is the definition of a derived class, FireflyElement, which is stored in a separate library loaded at runtime. (E.g, boost::serialization doesn't know this type exists at compile time of Element and PrivateElementAnimation. class FireflyElement : public Element { public: float lifespan; float age; float lifecycle; float dx[4], dy[4], dz[4]; bool init (); void move (); void fini (); template <class Archive> void serialize (Archive &ar, const unsigned int version) { boost::serialization::base_object <Element> (*this); ar & lifespan; ar & age; ar & lifecycle; ar & dx; ar & dy; ar & dz; } static Element * create (); ~FireflyElement (); FireflyElement (); friend class boost::serialization::access; private: }; BOOST_CLASS_EXPORT_GUID (FireflyElement, "FireflyElement"); // this class has a default constructor BOOST_SERIALIZATION_FACTORY_0 (FireflyElement); Of course, at this point, when I try to run the program, boost raises an exception unregistered_type, which as far as I know means that it doesn't know about the type relationship between FireflyElement and Element when trying to serialize Element *. I had a look at the documentation, and it appears that this is the way to register such type relationships (by using EXPORT_GUID, SERIALIZATION_FACTORY_0 and boost::serialization::base_object <Element> (*this); in the serialize function. I also had a look at the documentation for "plugins" and "dll's"[2] and couldn't find anything of use there. Does anybody know how to resolve this? Kind Regards, Sam [1] http://www.compiz.org [2] http://www.boost.org/doc/libs/1_38_0/libs/serialization/doc/special.html#de… -- Sam Spilsbury

1 0

Serialization Question
by Paul Dugas 08 Jun '10

08 Jun '10

I'm looking into recording some data collected by one of my programs into rotating files. Something like so: ofstream ofs; boost::archive::text_oarchive oa(ofs) size_t rotate = 0; for(;;) { foo_t foo = get_some_foo(); if (!rotate) { ofs.open(some_clever_new_filename); // attach oa again? rotate = 1000; } oa << foo; if (!--rotate) { ofs.close(); // detach oa? } } Do I need to do anything to detach or re-attach the oarchive? Paul

2 2

Boost documentation simple example fails
by hmiller＠hiwaay.net 08 Jun '10

08 Jun '10

I know I must be doing something wrong. I am trying to use the very simple example at: http://www.boost.org/doc/libs/1_43_0/libs/test/doc/html/tutorials/hello-the… #include <my_class.hpp> #define BOOST_TEST_MODULE MyTest #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_CASE( my_test ) { my_class test_object( "qwerty" ); BOOST_CHECK( test_object.is_valid() ); } I constructed and compiled a very simple class to use, but I get a compile error using the simple test above. I am using gcc version 4.1.2 on SUSE_linux. The compile error I get is: test_my_class.cpp:5: error: expected constructor, destructor, or type conversion before '(' token It is as if the compiler does not recognize BOOST_AUTO_TEST_CASE. Here is the my_class.hpp file: //==================== #include <string> using namespace std; class my_class { public: my_class( string str ); ~my_class(); private: bool valid; public: bool is_valid(); }; Here is my_class.cpp file: //==================== #include "my_class.hpp" my_class::my_class( string str ) { valid = false; } my_class::~my_class() { } bool my_class::is_valid() { return valid; } I may have made some typos while entering this code, but it does compile, and it does function properly when I instantiate it with another test program. What am I doing wrong? Thanks, Herb Miller

3 3

boost.mpi possible bug with mpi::scatter
by raj 08 Jun '10

08 Jun '10

Ok I'm new to mpi so maybe this is expected behaviour. But when using scatter, if number the of threads running is greater than number of elements in the array\vector sent to scatter, then there are the following issues 1) if the array contains a user-defined data then an out-of-bounds error will occur 2) otherwise (if is_mpi_datatype is true) the recieved data for threads that are "out-of-bound", will contain garbage and as such validity of the data cannot be tested. I assume this not correct behaviour? Raj

2 1

Build Questions
by John Dlugosz 08 Jun '10

08 Jun '10

I have two related questions. First, how can my source file assert that the correct version of Boost was found for an #include? Second, I need to build Microsoft debug DLLs to match incompatible compilation options, namely defines _HAS_ITERATOR_DEBUGGING=0;_BIND_TO_CURRENT_VCLIBS_VERSION=1. How should I decorate the resulting names to reflect this option? That is, besides ...-mt-dg...? I think this is referred to as the "ABI tag". --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 2

Problem with -Wall and Property Tree
by José Tomás Tocino García 08 Jun '10

08 Jun '10

Hi there, I've started to use boost::property_tree to parse some ini files, and I'm getting an error when I compile using -Wall on gcc 4.4.3. For this testcase: #include <boost/property_tree/ptree.hpp> #include <boost/property_tree/ini_parser.hpp> #include <iostream> using namespace std; class Reader{ boost::property_tree::ptree myTree; public: Reader(){ read_ini("config.ini", myTree); cout << myTree.get<float>("sonido.limite", 0.); } }; int main(int argc, char *argv[]) { Reader r; } If I compile using "g++ -o program testcase1.cpp -I." I get no errors, but if I add "-Wall", I get: ./boost/property_tree/detail/ptree_implementation.hpp: In function ‘int main(int, char**)’: ./boost/property_tree/detail/ptree_implementation.hpp:728: warning: dereferencing pointer ‘default_value’ does break strict-aliasing rules /usr/include/boost/optional/optional.hpp:422: note: initialized from here What's the problem? It compiles and works properly, but it would be great to have a warning-free compilation output. Thanks in advance. -- José Tomás Tocino García

2 2

[thread] Is Boost suitable for this application?
by Francisco Javier Porras Gálvez 08 Jun '10

08 Jun '10

Hi, all. I don't know anything about Boost, neither OpenMP, pthread nor Windows API thread library. So, before begin learning one of these, I'd like to explain to you how I want my application to work. It deals with running two parallel loops, one in LabView, and the other in C++. It doesn't matter if you don't know anything about LabView, this explanation is easy enough to understand. Attached you can see a picture which shows the desired execution behaviour. I want to know whether my application would be possible to achieve using Boost, or using another API. In another general C++ forum, I was suggested to use Boost. First, the execution begins in LabView. Here, I call a dll function written in C++. This is the "initialization" function. Its aim is to create another thread that will execute the main loop in C++. Then, this "initialization" function ends, because it has to return the control execution to LabView, but the new thread which has just been created must continue executing in its own loop. Now we have two loops in parallel: LabView and C++. In LabView's loop, we call periodically another dll function that has to send and retrieve data from the independent thread in C++. This function, like the other, has to end executing to pass control to LabView again, after it has done its purpose. This will be the behaviour until we call from LabView another dll function to stop the independent thread (not shown in the picture). After this, the LabView program ends. Is it possible to use Boost for this? If so, could you post a brief example about how you'd make it? Thanks in advance, Francisco

4 6

[GIL] Some questions about 2D iteration
by Brett Gmoser 08 Jun '10

08 Jun '10

Hello everybody, I've been toying with GIL a bit for the last few days in my spare time, and I've got a few questions that I can't seem to find answers to in the documentation. The following function which I've written works, but I'd like to get a better understanding of how to use iterators to accomplish the same task (the documentation says that view(x, y) = z is inefficient), as well as a few other things. The job of this function is to crop an image, discarding any white background. In other words, find the top-most, left-most, right-most, and bottom-most pixel that isn't the background color (white), and crop the image around those coordinates. template <typename ViewType, typename ImageType> void crop_image(ViewType& src, boost::scoped_ptr<ImageType>& dst) const { int top = 0; int right = 0; int bottom = 0; int left = 0; // Find the top, right, bottom, and left extents. First, find the top. for(int y=0; y < src.height(); ++y) { for(typename ViewType::x_iterator xpos = src.row_begin(y); xpos != src.row_end(y); ++xpos) { if(rgb8_pixel_t(255,255,255) != *xpos) { top = y; break; } } if(top) break; } // ... Do similar things for finding the right, bottom, and left extents. // We should now have the rectangle of the image, and it should be easy to copy everything over. dst.reset(new ImageType(1 + (right - left), 1 + (bottom - top))); for(int x = left; x <= right; ++x) for(int y = top; y <= bottom; ++y) view(*dst)((x - left), (y - top)) = src(x, y); } So, my first question is - is there a better way to do that first chore of finding each extent? Such as a way to use std::find or std::find_if, keeping in mind that I have to do two of the four sides in reverse (bottom-to-top to find bottom-most, right-to-left to find right-most)? The next question is about the final loop which copies from the source view to the newly created destination image. It seems that using iterators, I could do something like this, assuming that I had the above loops find iterator positions for all four extents (left, top, bottom, and right are iterators): // Start the copying process by creating a column iterator for the destination table beginning at column 0. // The top iterator starts at the top-most extent - and we cycle through the source image in a top-to-bottom, // left-to-right fashion. for(typename ImageType::view_t::y_iterator dst_y = view(*dst).col_begin(0); top <= bottom; ++dst_y,++top) { // The left iterator needs to be initialized with whatever row we're on - this appears to be impossible. There doesn't // seem to be any suitable way to do this, so consider it pseudocode. typename ViewType::x_iterator src_x = left.y(); // The destination X iterator also needs to know what Y position to begin on. You'd think you would be able to initialize // it with an iterator position, but it doesn't seem that you can. Again, this doesn't actually compile (it's the // view(*dst).row_begin(y) part). for(typename ImageType::view_t::x_iterator dst_x = view(*dst).row_begin(y); // Here, I need to be able to tell if src_x has hit the right extent. src_x <= right doesn't seem it will work, because // again right doesn't know it's Y position. src_x != right; // And increment both src and destination iterators. ++src_x,++dst_x ) { // Finally do the should-be-simple task of assigning the source pixel to the destination pixel. *dst_x = *src_x; } } If you read the comments, most of this doesn't seem to be possible. I've seen that there is a "transform_pixel_positions" algorithm that may do something close to what I want, but unfortunately I cannot find any documentation on it (other than mentioning that it exists, and a short example that really doesn't help - one of the few examples actually provided by the documentation). It seems to pass a unary argument to a functor, but that doesn't seem useful if I don't have both the X and Y coordinates, or an X iterator and Y iterator. Also, nowhere can I find where it says what the argument passed to the functor actually /is/. So after I figure these few things out, I'm hoping to be able to speed up my program a little bit. Right now everything is pretty slow using that view(x, y) method. I'd appreciate any help that anybody can give, thanks! Brett

4 9

get/set RTS on serial port
by Marcelo Adrian Lorenzati Sanz 08 Jun '10

08 Jun '10

Hi, As far as I've seen in the boost user mailing page, the topic about getting / setting the RTS and other control lines for the serial port have been already discussed, but my question is if it is planned to be added to boost asio serial port the control of those lines by the get_option / set_option. If not, how can I implement my own set_option to control the RTS line in a way to allow me to use it multiplatform, is there any example, hint, mock class, etx? The serial port is still very used in embedded designs and boost is the best way to go to connect with the pc. Best regards, Marcelo Lorenzati

2 2

Illegal instruction with Boost program options and Visual Studio 2008
by Basil Crow 07 Jun '10

07 Jun '10

Dear list, I am having some trouble using the Boost program options library with Visual Studio 2008 on Windows 7 (64-bit). My program is as follows: #include <boost/program_options.hpp> int wmain(int argc, wchar_t* argv[]) { boost::program_options::basic_command_line_parser<wchar_t> parser(argc, argv); return 1; } I compiled and linked the above as follows: > cl testcase.cpp /c /I "C:\dev\boost-1.42.0\include" /EHsc /Zi Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 15.00.21022.08 for 80x86 Copyright (C) Microsoft Corporation. All rights reserved. testcase.cpp > link testcase.obj /LIBPATH:"C:\dev\boost-1.42.0\lib" /DEBUG Microsoft (R) Incremental Linker Version 9.00.21022.58 Copyright (C) Microsoft Corporation. All rights reserved. When running the program, on line 504 of function_template.hpp I encounter an "Unhandled exception at 0x0018ffc4 in testcase.exe: 0xC000001D: Illegal Instruction." Please advise me as to how I can resolve this issue. Thank you, Basil

1 0

Bind questions
by John Dlugosz 07 Jun '10

07 Jun '10

Is the documentation for Boost::Bind matching the std::tr1::bind that got accepted? That is, can I read this to understand it since the docs are better than what comes with the compiler, or are there differences? Consider a member function C::foo. If I want to bind the object this parameter but leave the rest unchanged, would bind(&C::foo,x) automatically give me bind(&C::foo,x,_1,_2,_3)? What if I want to call x->foo(5,_3) and not use the first two parameters or the 4th, to make something that is passed to a callback function that wants 4 arguments? That is, the callback F is invoked F(a,b,c,d) and I want to make a function instance that can be assigned to F, and ends up calling x->foo(5,c). --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

4 6

[Range] Documentation for the combine function
by Sorin Fetche 07 Jun '10

07 Jun '10

Hi, While looking for a Boost.Range counterpart to the zip_iterator, I found the combine() function in boost/range/combine.hpp. But I didn't find any associated documentation neither in Boost 1.43 nor in trunk. In case I didn't overlook it, can I get a confirmation that this function really belongs to the public API, thus being safe to use? Thank you, Sorin

1 0

build problems on Mac OS with MacPorts gcc
by Rob Nikander 07 Jun '10

07 Jun '10

Hi, I'm on OS X, and I have a gcc 4.5 from macports (because I wanted C++0x features). I downloaded boost and compiled the regex library, but used the default boost build commands so it used apple's gcc, which led to weird bug when linked to a program created with gcc4.5. (Maybe that's not supposed to happen? I can start another thread about that.) Now I'm trying to build boost with the gcc 4.5, but that doesn't work... I followed these instructions to build bjam: http://www.boost.org/doc/libs/1_43_0/doc/html/jam/building.html I also edited a 'user-config.jam' to have: using gcc : 4.5 : g++-mp-4.5 ; Then went to BOOST_ROOT and ran: bjam toolset=gcc stage And it eventually barfs with this: gcc.link.dll bin.v2/libs/regex/build/gcc-4.5/release/threading-multi/libboost_regex.dylib ld: unknown option: -h collect2: ld returned 1 exit status Any ideas? thanks, Rob

2 2

[gil io_new] Problem writing a compressed jpeg
by Olivier Tournaire 07 Jun '10

07 Jun '10

Hi all, I am currently trying to write a compressed jpeg image without any success. Here is my code: #include <string> #include <boost/mpl/vector.hpp> #include <boost/gil/gil_all.hpp> #include <boost/gil/extension/io_new/tiff_read.hpp> #include <boost/gil/extension/io_new/tiff_write.hpp> #include <boost/gil/extension/io_new/jpeg_read.hpp> #include <boost/gil/extension/io_new/jpeg_write.hpp> using namespace std; using namespace boost; using namespace gil; int main(int argc, char** argv) { string base_image_path(argv[1]); string output_image_path(argv[2]); string output_compressed_image_path(argv[3]); typedef mpl::vector< rgb8_image_t, rgb16_image_t > rgb_image_types; typedef any_image<rgb_image_types> any_image_type; typedef any_image_type::view_t any_view_type; // Read tiff image image_read_info< tiff_tag > info = read_image_info( base_image_path , tiff_tag() ); any_image_type input_image; read_image(base_image_path.c_str(), input_image, tiff_tag()); write_view(output_image_path.c_str(), view(input_image), jpeg_tag()); image_write_info< jpeg_tag > compressed_jpeg_info; compressed_jpeg_info._quality = 30; //compressed_jpeg_info._dct_method = jpeg_dct_method::fastest; write_view(output_compressed_image_path.c_str(), view(input_image), compressed_jpeg_info); } Could you please help me ? Regards, Olivier

2 5

[utf] global fixture problem
by Olaf Peter 07 Jun '10

07 Jun '10

Hi, in the following snippet I get the compiler error: utf_global_fixture.cpp: In member function ‘void my_test1::test_method()’: utf_global_fixture.cpp:19: error: ‘check’ was not declared in this scope utf_global_fixture.cpp:20: error: expected ‘;’ before ‘do’ // Boost.Test #define BOOST_TEST_MODULE Unit_test_example_09 #include <boost/test/unit_test.hpp> // STL #include <iostream> struct MyConfig { MyConfig() { std::cout << "global setup part1\n"; } ~MyConfig() { std::cout << "global teardown part1\n"; } bool check() const { return true;} }; // structure MyConfig is used as a global fixture - it's invoked pre and post any testing is performed BOOST_GLOBAL_FIXTURE( MyConfig ) BOOST_AUTO_TEST_CASE( my_test1 ) { BOOST_REQUIRE( check() ) BOOST_CHECK( true ); } isn't the member access like the test case/module fixture (as I did read in the docs)? Thanks, Olaf

2 1

[BGL] Distributed Dijkstra vertex_iterator
by Bruno Santos 07 Jun '10

07 Jun '10

Hi all, I'm fairly new to boost so please bear my ignorance. I'm trying to implement the example of the distributed Dijkstra using the Parallel BGL and I'm having problems when iterating through the distance map. This is the code. Just updated the dijkstra_example.cpp with the Parallel BGL Dijkstra instructions. #include <boost/config.hpp> #include <iostream> #include <fstream> #include <stdio.h> #include <iomanip> #include <vector> #include <boost/graph/use_mpi.hpp> #include <boost/graph/distributed/mpi_process_group.hpp> #include <boost/graph/dijkstra_shortest_paths.hpp> #include <boost/graph/distributed/adjacency_list.hpp> using namespace boost; using boost::graph::distributed::mpi_process_group; int main(int argc, char* argv[]) { boost::mpi::environment env(argc,argv); typedef adjacency_list<listS, distributedS<mpi_process_group, vecS>, directedS, no_property, // Vertex properties property<edge_weight_t, int> // Edge properties > graph_t; typedef graph_traits < graph_t >::vertex_descriptor vertex_descriptor; typedef graph_traits < graph_t >::edge_descriptor edge_descriptor; typedef std::pair<int, int> Edge; const int num_nodes = 5; enum nodes { A, B, C, D, E }; char name[] = "ABCDE"; Edge edge_array[] = { Edge(A, C), Edge(B, B), Edge(B, D), Edge(B, E), Edge(C, B), Edge(C, D), Edge(D, E), Edge(E, A), Edge(E, B) }; int weights[] = { 1, 2, 1, 2, 7, 3, 1, 1, 1 }; int num_arcs = sizeof(edge_array) / sizeof(Edge); #if defined(BOOST_MSVC) && BOOST_MSVC <= 1300 graph_t g(num_nodes); property_map<graph_t, edge_weight_t>::type weightmap = get(edge_weight, g); for (std::size_t j = 0; j < num_arcs; ++j) { edge_descriptor e; bool inserted; tie(e, inserted) = add_edge(edge_array[j].first, edge_array[j].second, g); weightmap[e] = weights[j]; } #else graph_t g(edge_array, edge_array + num_arcs, weights, num_nodes); property_map<graph_t, edge_weight_t>::type weightmap = get(edge_weight, g); #endif std::vector<vertex_descriptor> p(num_vertices(g)); std::vector<int> d(num_vertices(g)); vertex_descriptor s = vertex(A, g); #if defined(BOOST_MSVC) && BOOST_MSVC <= 1300 // VC++ has trouble with the named parameters mechanism property_map<graph_t, vertex_index_t>::type indexmap = get(vertex_index, g); dijkstra_shortest_paths(g, s, &p[0], &d[0], weightmap, indexmap, std::less<int>(), closed_plus<int>(), (std::numeric_limits<int>::max)(), 0, default_dijkstra_visitor()); #else dijkstra_shortest_paths (g, s, predecessor_map( make_iterator_property_map(p.begin(), get(vertex_index, g))). distance_map( make_iterator_property_map(d.begin(), get(vertex_index, g))) ); #endif std::cout << "distances and parents:" << std::endl; graph_traits < graph_t >::vertex_iterator vi, vend; for (tie(vi, vend) = vertices(g); vi != vend; ++vi) { std::cout << "distance(" << name[*vi] << ") = " << d[*vi] << ", "; std::cout << "parent(" << name[*vi] << ") = " << name[p[*vi]] << std:: endl; } std::cout << std::endl; return 0; } The compiler line I'm using is: mpic++ -I/usr/include/boost/graph/distributed/ -I/usr/include/boost/ -L/usr/lib/ /usr/lib/libboost_mpi-mt.so /usr/lib/libboost_serialization-mt.so /usr/lib/libboost_graph_parallel-mt.so /usr/lib/libboost_system-mt.so dij.cpp -o dijkstr and the error is: In file included from /usr/include/c++/4.4/backward/hash_set:59, from /usr/include/boost/graph/adjacency_list.hpp:25, from /usr/include/boost/graph/distributed/adjacency_list.hpp:18, from dij.cpp:11: /usr/include/c++/4.4/backward/backward_warning.h:28:2: warning: #warning This file includes at least one deprecated or antiquated header which may be removed without further notice at a future date. Please use a non-deprecated interface with equivalent functionality instead. For a listing of replacement headers and interfaces, consult the file backward_warning.h. To disable this warning use -Wno-deprecated. dij.cpp: In function ‘int main(int, char**)’: dij.cpp:74: error: no match for ‘operator[]’ in ‘name[((boost::iterator_facade<boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, boost::detail::parallel::global_descriptor<long unsigned int>, boost::random_access_traversal_tag, boost::detail::parallel::global_descriptor<long unsigned int>, long int>*)(& vi))->boost::iterator_facade<I, V, TC, R, D>::operator* [with Derived = boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, Value = boost::detail::parallel::global_descriptor<long unsigned int>, CategoryOrTraversal = boost::random_access_traversal_tag, Reference = boost::detail::parallel::global_descriptor<long unsigned int>, Difference = long int]()]’ dij.cpp:74: error: no match for ‘operator[]’ in ‘d[((boost::iterator_facade<boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, boost::detail::parallel::global_descriptor<long unsigned int>, boost::random_access_traversal_tag, boost::detail::parallel::global_descriptor<long unsigned int>, long int>*)(& vi))->boost::iterator_facade<I, V, TC, R, D>::operator* [with Derived = boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, Value = boost::detail::parallel::global_descriptor<long unsigned int>, CategoryOrTraversal = boost::random_access_traversal_tag, Reference = boost::detail::parallel::global_descriptor<long unsigned int>, Difference = long int]()]’ /usr/include/c++/4.4/bits/stl_vector.h:610: note: candidates are: typename std::_Vector_base<_Tp, _Alloc>::_Tp_alloc_type::reference std::vector<_Tp, _Alloc>::operator[](size_t) [with _Tp = int, _Alloc = std::allocator<int>] /usr/include/c++/4.4/bits/stl_vector.h:625: note: typename std::_Vector_base<_Tp, _Alloc>::_Tp_alloc_type::const_reference std::vector<_Tp, _Alloc>::operator[](size_t) const [with _Tp = int, _Alloc = std::allocator<int>] dij.cpp:75: error: no match for ‘operator[]’ in ‘name[((boost::iterator_facade<boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, boost::detail::parallel::global_descriptor<long unsigned int>, boost::random_access_traversal_tag, boost::detail::parallel::global_descriptor<long unsigned int>, long int>*)(& vi))->boost::iterator_facade<I, V, TC, R, D>::operator* [with Derived = boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, Value = boost::detail::parallel::global_descriptor<long unsigned int>, CategoryOrTraversal = boost::random_access_traversal_tag, Reference = boost::detail::parallel::global_descriptor<long unsigned int>, Difference = long int]()]’ dij.cpp:75: error: no match for ‘operator[]’ in ‘p[((boost::iterator_facade<boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, boost::detail::parallel::global_descriptor<long unsigned int>, boost::random_access_traversal_tag, boost::detail::parallel::global_descriptor<long unsigned int>, long int>*)(& vi))->boost::iterator_facade<I, V, TC, R, D>::operator* [with Derived = boost::transform_iterator<boost::detail::parallel::global_descriptor<long unsigned int>::generator, boost::counting_iterator<long unsigned int, boost::use_default, boost::use_default>, boost::use_default, boost::use_default>, Value = boost::detail::parallel::global_descriptor<long unsigned int>, CategoryOrTraversal = boost::random_access_traversal_tag, Reference = boost::detail::parallel::global_descriptor<long unsigned int>, Difference = long int]()]’ /usr/include/c++/4.4/bits/stl_vector.h:610: note: candidates are: typename std::_Vector_base<_Tp, _Alloc>::_Tp_alloc_type::reference std::vector<_Tp, _Alloc>::operator[](size_t) [with _Tp = boost::detail::parallel::global_descriptor<long unsigned int>, _Alloc = std::allocator<boost::detail::parallel::global_descriptor<long unsigned int> >] /usr/include/c++/4.4/bits/stl_vector.h:625: note: typename std::_Vector_base<_Tp, _Alloc>::_Tp_alloc_type::const_reference std::vector<_Tp, _Alloc>::operator[](size_t) const [with _Tp = boost::detail::parallel::global_descriptor<long unsigned int>, _Alloc = std::allocator<boost::detail::parallel::global_descriptor<long unsigned int> >] The boost version I'm using is 1.40.0 in a Ubuntu Machine. If anyone could help me I would be grateful :) Thanks, Bruno Santos

2 2

[utf] handling 'gold' files
by Olaf Peter 07 Jun '10

07 Jun '10

Hi, what is the best way to give a test executable (BOOST_AUTO_TEST_CASE, Test suite with automated registration) a path to 'gold' files and use them? Thanks, Olaf

4 5

Fwd: boost::thread and corresponding stack array
by Aditya Arora 07 Jun '10

07 Jun '10

I'd like to use boost threads, if possible, in an upcoming project, just for the sake of portability. The problem is that boost threads seem a bit unwieldy when it comes to quickly creating multiple threads. Since the thread object constructor is what actually launches the thread, you can't simply create an array of thread objects on the stack, since each thread is only launched when the constructor is passed a functor. Therefore, if you want an array of threads, you need to allocate each thread object individually on the heap, like: boost::thread* thr[SIZE]; for (int i = 0; i < SIZE; ++i) thr[i] = new boost::thread(functor); Then of course, you need to iterate through the array later and call delete on each thread object. It's much simpler to do this with plain old pthreads, since pthreads need to be explicitly started by calling pthread_create, so you can create an array of pthread structs on the stack, and then call pthread_create on each one, without having to allocate/free objects on the heap. It would be really nice if boost threads simply had a start() member function, or something, so that you could do something like: boost::thread thr[SIZE]; for (int i =0 ; i < SIZE: ++i) thr[i].start(functor); Does anyone know the design considerations that went into creating boost threads, and why the designers chose to have the constructor the only way to launch a new thread? Aditya

3 2

Thread can not run concurrently with bind parameter
by coofucoo zhang 07 Jun '10

07 Jun '10

Hi All: It may be a simple question, but I have no idea now. So I want to post my question with you. Maybe you can help, thank you. It is about boost thread. I want to create a thread and pass a parameter to it. I found there are 2 ways in boost thread, they are: using bind, like *boost::thread thrd(boost::bind(&helloworld, "Bob"));* *thrd.join();* or construct a structure and do the following: *boost::thread thrd(helloworld("Bob"));* *thrd.join();* It works, really, like the demo code shown. But After I run 2 threads, I found 2 thread can not run concurrently. The second thread will blocked by the first thread. If the first thread never exit, the second thread never run. It is a very funny problem. But I really do not know what happen. Does anyone know about this topic? Thank you very much. -- With regards. Coofucoo Zhang

2 2

[intrusive] join lists
by Tim Blechmann 06 Jun '10

06 Jun '10

hi all, is there a way to join two instances of boost.intrusive in constant time? thanks, tim

2 2

[variant] newbie Q: cannot visit variant variable: "no matching function for call to apply_visitor ..."
by Riccardo Murri 06 Jun '10

06 Jun '10

Hello, I'm getting the following error when I try to compile the attached C++ source (repetitions of "boost::detail::variant::void_" omitted for clarity):: $ g++ variant_no_matching_function_call.cc variant_no_matching_function_call.cc: In member function ‘size_t Waterfall::Processor::step()::_row_size::operator()(boost::variant<Waterfall::Processor::SparseRow*, Waterfall::Processor::DenseRow*, boost::detail::variant::void_, ..., boost::detail::variant::void_>&)’: variant_no_matching_function_call.cc:47: error: no matching function for call to ‘apply_visitor(Waterfall::Processor::step()::__row_size, boost::variant<Waterfall::Processor::SparseRow*, Waterfall::Processor::DenseRow*, boost::detail::variant::void_, ..., boost::detail::variant::void_>&)’ I'm aware that the attached sample could be better handled by simple polymorphism, but it is the shortest excerpt from my real code that reproduces the problem. I'm getting the same error each time I try to "apply_visitor", so there must be something wrong in the way I'm using it... Thank you very much for any suggestions! I'm using g++ 4.4 and Boost.Variant 1.40 from the native Ubuntu 10.04 packages. Best regards, Riccardo -- Riccardo Murri, Hadlaubstr. 150, 8006 Zürich (CH)

2 2

[phoenix][lambda] lambda to phoenix translation: get sig (signature) from actor
by alfC 05 Jun '10

05 Jun '10

hi, what is the translation of boost.lambda typename LambdaExp::template sig<tuple<InputType> >::type in Boost.Phoenix? my best try was: actor<LambdaExp>::result< vector<InputType> >::type but it doesn't compile. Thank you, Alfredo

4 18

[Program Options] Requiring command-line options
by Jeremy Abramson 05 Jun '10

05 Jun '10

Heya! I just thinking about switching from simpleOpt to BoostPO for command line handling, mainly because of its support of configuration files. This may be a stupid question, but I haven't seen anything about making a command line parameter *required*. You'd think this is a pretty common thing, but I didn't see any mention of it in the tutorials or anything. I guess I could just make certain parameters default -1/NULL or something, and then test at the end if they're still equal to that, but that seems a bit ugly. Is there built in support (a flag in the "add_options" function or something) for making certain parameters required? Thanks a lot!

5 5

Interval cos fails on 64 bits platforms
by Christophe TROESTLER 05 Jun '10

05 Jun '10

Hi, I would like to use boost interval library in computer assisted proofs. However my machines are mostly 64 bits and the attached program fails -- returning [nan, nan] -- on them. On 32 bits, it works fine. Since this bug seems to be known [1], may somebody tell me how much progress has been done on solving it and whether there is there a workaround available? Thanks. Best, Christophe [1] https://svn.boost.org/trac/boost/ticket/2769 https://svn.boost.org/trac/boost/ticket/4011 #include <boost/numeric/interval.hpp> #include <fstream> #include <iostream> using namespace boost; using namespace numeric; using namespace interval_lib; typedef interval<double, policies<save_state<rounded_transc_opp<double> >, checking_base<double> > > I; template<class T, class Policies> std::ostream &operator<<(std::ostream &os, const boost::numeric::interval<T, Policies> &x) { os << "[" << x.lower() << ", " << x.upper() << "]"; return os; } int main() { std::cout << cos(I(-1, -0.6)) << "\n"; }

1 0

guessing Y-values
by Andrej van der Zee 05 Jun '10

05 Jun '10

Hi, In my program I have many points in a two-dimensional space, let's say on the X-axis I have time and on the Y-axis I have values ranging from -100 to +100. But unfortunately not for every value on the X-axis I have a Y-value. Though, I have to be able to "guess" the Y-values for every X-value. The (x,y)-points do correlate but I do not now how; sometimes it is a straight line, other times it is parabolic. My question is if boost has any support for this kind of problem. Thank you, Andrej

1 0

[Test] Tests execution order and depends_on method.
by Maria Kozinska 05 Jun '10

05 Jun '10

Hello, I have couple of question to Boost.Test framework. I'd like to ask why test_unit::depends_on(test_unit*) method is not described in User's guide. Is it only for internal use by framework or I can use it freely? I assume, that it can be used to determine precedence of one test before another. What happens if there is a cycle in dependence graph? If a test case fails, are dependant on it tests executed? In what order are executed automatically registered test suites and test cases? And in what order when they are manually registered (without any depends_on)? How this order is affected by depends_on method? Thanks for answer. Greetings, Maria

2 1

Boost MPI Serialization : Sending serialized array of user defined classes
by Noman Javed 04 Jun '10

04 Jun '10

I want to send my Array class using boost Mpi template<class T> class Array { private: int size; T* data; public: // constructors + other stuff }; Here T can be any built in type or user defined type. Suppose I have a class complex struct complex { std::vector<double> real_imag; // contain two elements }; How can I send Array<complex> using Boost::Mpi + serialization. Thanks in anticipation Regards Noman

2 3

[GIL] How to create (and read) a 4 pixels per sample image type?
by Olivier Tournaire 04 Jun '10

04 Jun '10

Hi all, I am currently trying to create and read an image which has 4 pixels per sample. Here is the output of tiffinfo on such an image : TIFF Directory at offset 0x8 (8) Image Width: 2273 Image Length: 2256 Resolution: 72, 72 pixels/inch Bits/Sample: 16 Sample Format: unsigned integer Compression Scheme: None Photometric Interpretation: RGB color Extra Samples: 1<unspecified> Samples/Pixel: 4 Rows/Strip: 2256 Planar Configuration: single image plane How can I achieve this gola with GIL? Best regards, Olivier

2 1

splitting non intrusive serialization into save/load
by Babineau, Denis 04 Jun '10

04 Jun '10

Hi I'm trying to serialize a class using non intrusive templates since I can't modify the class. The class has proper set/gets but doesn't expose the data directly so I would need to use the save/load method which requires a macro to be defined in the class but, again, I can't modify the class. Looking at the code, I tried the following to have the save/load implemented outside but get the following error with MSVC 7.1 (boost version 1.39). Is there any way to accomplish what I'm trying to do here? Thanks Denis error C2270: 'save' : modifiers not allowed on nonmember functions sample implementation: class MyClass { public : int GetValue() const; public : void SetValue(int v); private: int m_Value; }; ... namespace boost { namespace serialization { template <class Archive> void save(Archive & ar, MyClass & c, const unsigned int version) const { int v = c.GetValue(); ar & v; } template <class Archive> void load(Archive & ar, MyClass & c, const unsigned int version) { int v; ar & v; c.SetValue(v); } template <class Archive> void serialize(Archive & ar, MyClass & c, const unsigned int version) { boost::serialization::split_member(ar, c, version); } }} CONFIDENTIALITY NOTICE: The contents of this email are confidential and for the exclusive use of the intended recipient. If you receive this email in error, please delete it from your system immediately and notify us either by email, telephone or fax. You should not copy, forward, or otherwise disclose the content of the email.

2 2

filesystem and range::copy
by Olaf Peter 04 Jun '10

04 Jun '10

Hi, does the boost filesystem support boost.range (or vice versa?) I've got a compile error by fs::ifstream in( ... ); std::string storage; boost::copy( in, std::back_inserter( storage ) ); using trunk boost/mpl/eval_if.hpp: In instantiation of ‘boost::mpl::eval_if_c<true, boost::range_const_iterator<boost::filesystem::basic_ifstream<char, std::char_traits<char> > >, boost::range_mutable_iterator<const boost::filesystem::basic_ifstream<char, std::char_traits<char> > > >’: boost/range/iterator.hpp:63: instantiated from ‘boost::range_iterator<const boost::filesystem::basic_ifstream<char, std::char_traits<char> > >’ boost/range/concepts.hpp:221: instantiated from ‘boost::SinglePassRangeConcept<boost::filesystem::basic_ifstream<char, std::char_traits<char> > >’ boost/concept/detail/has_constraints.hpp:42: instantiated from ‘const bool boost::concepts::not_satisfied<boost::SinglePassRangeConcept<boost::filesystem::basic_ifstream<char, std::char_traits<char> > > >::value’ boost/concept/detail/has_constraints.hpp:45: instantiated from ‘boost::concepts::not_satisfied<boost::SinglePassRangeConcept<boost::filesystem::basic_ifstream<char, std::char_traits<char> > > >’ Thanks, Olaf

2 1

[BGL] Usage of stateful allocator
by Geoff Hilton 04 Jun '10

04 Jun '10

Is a static instance of a stateful allocator the only way to make the BGL graph interface accept it with boost::container_gen, or is the use of a stateful allocator with the BGL possible? The reason this is necessary is that I'd like to wrap an array class (which isn't STL compatible) provided by a third party for use with the BGL so that I can navigate the array of vertices and edges as modeled by the graph it contains. This array class however is useless unless it receives a reference to an instance of the stateful allocator upon its' construction. Currently I have to hold a BGL graph in memory which contains vertex and edge array indices reflecting matching positions in instances of the third party array class (one for the vertices, one for all the edges). (Unrelatedly but possibly useful to know) Once this graph is complete, I need access to the arrays themselves to pass them on to other objects that are part of the third party library. Thanks, Geoff

2 1

[shared_ptr] boost::make_shared & std::tr1::shared_ptr
by Geoff Hilton 04 Jun '10

04 Jun '10

It's not currently possible to mix boost::make_shared with the shared_ptr implementation provided by <boost/tr1/memory.hpp>. Could this be made possible or would mixing the two be a design flaw on my part? Thanks, Geoff

2 2

[Interprocess] Determine if named_mutex was opened or created
by Sönke Schau 03 Jun '10

03 Jun '10

Hello! In the constructor of named_mutex it is possible to specify that the mutex should be opened or otherwise created. Is there a way to determine which took place: creation or open? Best regards Sönke

2 1

configuring boost to use microsoft compute cluster pack on a windows xp pro box
by Sean Farrow 03 Jun '10

03 Jun '10

Hi: I'm trying to configure boost to use compute cluster pack on a non-server (windows xp) machine. Given that automatic mpi detection failed how do I go about doing this? Cluster pack installed to c:\program files\microsoft compute cluster pack and only client tools are installed. Regards Sean.

1 0

boost:any/boost::variant for integral/floating type
by Mathieu Malaterre 03 Jun '10

03 Jun '10

Hi, I am looking at boost::any and boost::variant which seems to be very usefull additions to C++ for building heterogenous containers. However in my case I need to build something that would be: - type safe - can only contains integral/floating type - allow operations such as max() / min() My typical operation on this 'generic_value_type' is simply `+= 1`. boost::any seems difficult to use, since you need to call any_cast any time you want to operate on the value. I am thinking I could just wrap boost::variant inside a custom class, and add my missing functionalities, something like: struct AttributeValue { private: struct add_one_visitor : public boost::static_visitor<> { template <typename T> void operator()(T & i) const { i++; } }; struct get_min_visitor : public boost::static_visitor<> { template <typename T> T operator()(T & i) const { return std::numeric_limits<T>::min(); } }; public: AttributeValue& operator++() { boost::apply_visitor( add_one_visitor(), Value ); return *this; } AttributeValue min() { return boost::apply_visitor( get_min_visitor(), Value ); } boost::variant<double, char, int> Value; }; Did I miss anything in boost ? Is boost::variant my best option (when only using fundamental types) ? Thanks -- Mathieu

3 5

boost and nvcc on msvc 9.0
by Prevrhal, Sven 03 Jun '10

03 Jun '10

Something to share - I had difficulties compiling a .cu file with #include <boost/random.hpp> with cuda 3.0 and msvc 9.0 The nvcc compiler first complained about not having stdint.h (true, it's not part of cuda or msvc). I #undef'd BOOST_HAS_STDINT_H in the config/compiler/nvcc.hpp header but then it complained about wrong int limits in base_traits. I figured out that setting #define BOOST_MSVC 1500 makes that error go away. I assume this breaks portability since nvcc.hpp is now locked to msvc but there doesn't seem to be a mechanism for telling boost about the nvcc host compiler. So ugly for now, but it seems to work at least for boost/random There's another problem with my code but it's probably not boost related, that is the mvsc debugger pointer jumps around wildly at some point in the code and then it crashes. In release mode, the program runs fine until the end when it too crashes. If you think you have an idea, I'll post the code. Sven ________________________________ The information contained in this message may be confidential and legally protected under applicable law. The message is intended solely for the addressee(s). If you are not the intended recipient, you are hereby notified that any use, forwarding, dissemination, or reproduction of this message is strictly prohibited and may be unlawful. If you are not the intended recipient, please contact the sender by return e-mail and destroy all copies of the original message.

1 0

[BGL] Graph coloring program
by Mike Douglass 03 Jun '10

03 Jun '10

I have uploaded the graph coloring program to http://groups.google.com/group/graph_coloring_program/ go to the files section and download the files and read README. I'm willing to license my code under the Boost license. Opening a track ticket ok.

2 2

Boost MPI
by Miguel Ángel Vázquez 03 Jun '10

03 Jun '10

Dear all, I already installed Boost MPI library and I tested like you suggest in the tutorial (executing bjam in /libs/mpi/test). Unfortunately, it doesn't pass all the tests and I don't know whether they are important or not or what to do to try to fix them. I enclose you the output, hope someone can help me somehow. Cheers, Miguel

1 0

Generalize boost::optional as boost::range to support boost::range::adaptors
by bradphelan＠xtargets.com 03 Jun '10

03 Jun '10

Hi all, I've implemented a small adaptor to make boost::optional behave like a range. This extends the functionality of boost::optional to be more like the scala Option class and the haskell Maybe monad. A trivial motivating example is int plus(const int & i){ return i + 1; } void foo(boost::optional<int> opt_v){ boost::copy ( opt_v | boost::adaptors::transformed(plus) , std::ostream_iterator<int>(std::cout) ); } foo(10) outputs 11 The full implementation and details are at http://xtargets.heroku.com/2010/06/03/using-boostoptional-as-a-range/ Does anybody think this is a useful extension to boost::optional or would be willing to provide feedback or suggestions on the implementation. Regards Brad

1 0

error compiling an accumulator sample
by Conoscenza Silente 03 Jun '10

03 Jun '10

Hi All I cannot build the following sample using accumulators. Do you know why? I used times ago an example and it worked without any problem. Can you help me with that? I have the following error error: ‘covariate1’ was not declared in this scope #include <iostream> #include <boost/accumulators/accumulators.hpp> #include <boost/accumulators/statistics/stats.hpp> #include <boost/accumulators/statistics/mean.hpp> #include <boost/accumulators/statistics/moment.hpp> #include <boost/accumulators/statistics/variance.hpp> #include <boost/accumulators/statistics/covariance.hpp> using namespace boost::accumulators; int main() { accumulator_set<double, stats<tag::mean, tag::variance, tag::covariance<double, tag::covariate1> > > x_acc; x_acc( 1.1, covariate1 = 2.2 ) ; return 0; }

1 1

Error:<no tag> cannot be initialized in a constructor.
by Conoscenza Silente 03 Jun '10

03 Jun '10

HI All I am using boost accumulators. These 2 lines use to work fine with current version of boost in LInux. accumulator_set< double, stats< tag::covariance<double, tag::covariate1> > > acc_cov; accumulator_set< double, stats< tag::variance > > acc_var; When I moved to a Sun machine where it is installed boost v1.40 I have this building error "/opt/boost/boost/accumulators/framework/depends_on.hpp", line 276: Error:<no tag> cannot be initialized in a constructor. "/opt/boost/boost/fusion/container/list/cons.hpp", line 85: Where: While instantiating "boost::accumulators::detail::accumulator_wrapper<int, int>::accumulator_wrapper(const boost::accumulators::detail::accumulator_wrapper<int, int>&)". "/opt/boost/boost/fusion/container/list/cons.hpp", line 85: Where: Instantiated from non-template code. 1 Error(s) Do you know how can I fix those errors and why I have this issue? Thanks

1 0

[lambda][phoenix] print expression tree
by alfC 03 Jun '10

03 Jun '10

As you can see from other posts. It's been quite painful to me to debug programs with expression templates. In order to debug some code I implemented some functions to print the expression trees, this would be a nice addition to Boost.Lambda and also to Boost.Phoenix (although Boost.Phoenix will probably have it through Boost.Proto which apparently already can print its expression trees). Here it is a *minimal* code for printing some Boost.Lambda expression (involving only [binary +,-,/,*],bind,constants), (let me know if I am reinventing the wheel) namespace boost{ namespace lambda{ using namespace boost::tuples; template<int N> //print placeholder std::string to_string(const lambda_functor<boost::lambda::placeholder<N> >&){ return "_"+boost::lexical_cast<std::string>(N); } //template<class T> //std::string to_string(const T& t){ // return boost::lexical_cast<std::string>(t); //} //binary operators #define BOOST_LAMBDA_TO_STRING_GEN(action_name, symbol_string)\ template<class LExp1, class LExp2> \ std::string to_string(const \ lambda_functor<lambda_functor_base<arithmetic_action<action_name>, tuple< \ LExp1, \ LExp2 \ > > >& action_name##_expr){ \ return to_string(get<0>(action_name##_expr.args)) + #symbol_string +to_string(get<1>(action_name##_expr.args)); \ } BOOST_LAMBDA_TO_STRING_GEN(plus_action, +); BOOST_LAMBDA_TO_STRING_GEN(divide_action, /); BOOST_LAMBDA_TO_STRING_GEN(minus_action, -); BOOST_LAMBDA_TO_STRING_GEN(multiply_action, *); //bind template<class FunctionType, class LExpr> std::string to_string(const lambda_functor<lambda_functor_base<action<2, function_action<2> >, tuple< FunctionType, //double (* const)(double), LExpr > > >& bind_expr){ return std::string("bind(") +boost::units::detail::demangle(typeid(FunctionType).name()) +", "+to_string(get<1>(bind_expr.args)) + ")"; } }} Cheers, Alfredo

1 0

Re: [Boost-users] issue when building boost with msvc2010
by Christian Henning 03 Jun '10

03 Jun '10

Hi there, I've been building boost with MSVC10 Express just fine. I usually use the Visual Studio Command Prompt window to get things done. Here you can double check if your environment variable "Include" is correct. Inside the command prompt just type "set include". My windows.h is in C:\Program Files (x86)\Microsoft SDKs\Windows\v7.0A\Include which is set inside include path. Regards, Christian On Wed, Jun 2, 2010 at 5:12 PM, Sean Farrow <sean.farrow(a)seanfarrow.co.uk> wrote: > Hi: > > When building boost with msvc 2010, I get a c1083 unable to open file with > windows.h. > > My command line is: > > Bjam –toolset=msvc –stage –build-type=complete > > Even though as the output asks I have called vcvarsall.bat. > > A portion of the log follows: > > compile-c-c++ > bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\runtime-link-static\debug.obj > > debug.cpp > > C:\Program Files\boost\boost/test/impl/debug.ipp(32) : fatal error C1083: > Cannot open include file: 'windows.h': No such file or directory > > > > call "C:\Program Files\Microsoft Visual Studio 10.0\VC\vcvarsall.bat" > x86 >nul > > cl /Zm800 -nologo > @"bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\runtime-link-static\debug.obj.rsp" > > > > ...failed compile-c-c++ > bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\runtime-link-static\debug.obj... > > Anyone got any ideas, do I need to set up some environment variables? > > Vs2008 builds fine with the latest windows sdk, the latest windows sdk > configuration tool doesn’t allow integration with vc2010. > > Help appreciated as always. > > Regards > > Sean. > > Regards > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

1 0

[spirit] what's wrong with my parser ?
by Kamil Zubair 03 Jun '10

03 Jun '10

I try to make parser for the following string: "value = x" or "value = [x1, x2]" I also want to retrieve the value of x or x1 and x2. My minimum code that will produce error is like this: typedef pair<float, boost::optional<float> > range_val; rule<string::const_iterator, range_val(), locals<float, boost::optional<float> > > range_value; rule<string::const_iterator> valid_line; range_value = (float_[_a = _1] | ("[" >> *space >> float_[_a = _1] >> ',' >> float_[_b = _1] >> *space >> "]"))[_val = boost::phoenix::bind(make_pair<float, boost::optional<float> >, _a, _b)]; valid_line = "value = " << range_value; I'm using vs 2008 with boost 1.43, the above code will produce this error: 1>f:\library\boost_1_41_0\boost\spirit\home\qi\nonterminal\rule.hpp(176) : error C2664: 'boost::mpl::assertion_failed' : cannot convert parameter 1 from 'boost::mpl::failed ************(__thiscall boost::spirit::qi::rule<Iterator>::=::error_invalid_expression::* ***********)(Expr)' to 'boost::mpl::assert<false>::type' 1> with 1> [ 1> Iterator=std::_String_const_iterator<char,std::char_traits<char>,std::allocator<char>>, 1> Expr=boost::proto::exprns_::expr<boost::proto::tag::shift_left,boost::proto::argsns_::list2<boost::proto::exprns_::expr<boost::proto::tag::terminal,boost::proto::argsns_::term<const char (&)[9]>,0>,boost::spirit::qi::rule<std::_String_const_iterator<char,std::char_traits<char>,std::allocator<char>>,range_val (void),boost::spirit::locals<float,boost::optional<float>>> &>,2> 1> ] 1> No constructor could take the source type, or constructor overload resolution was ambiguous 1> f:\project\experiment\spirit\spirit.cpp(82) : see reference to function template instantiation 'boost::spirit::qi::rule<Iterator> &boost::spirit::qi::rule<Iterator>::operator =<boost::proto::exprns_::expr<Tag,Args,Arity>>(const Expr &)' being compiled 1> with 1> [ 1> Iterator=std::_String_const_iterator<char,std::char_traits<char>,std::allocator<char>>, 1> Tag=boost::proto::tag::shift_left, 1> Args=boost::proto::argsns_::list2<boost::proto::exprns_::expr<boost::proto::tag::terminal,boost::proto::argsns_::term<const char (&)[9]>,0>,boost::spirit::qi::rule<std::_String_const_iterator<char,std::char_traits<char>,std::allocator<char>>,range_val (void),boost::spirit::locals<float,boost::optional<float>>> &>, 1> Arity=2, 1> Expr=boost::proto::exprns_::expr<boost::proto::tag::shift_left,boost::proto::argsns_::list2<boost::proto::exprns_::expr<boost::proto::tag::terminal,boost::proto::argsns_::term<const char (&)[9]>,0>,boost::spirit::qi::rule<std::_String_const_iterator<char,std::char_traits<char>,std::allocator<char>>,range_val (void),boost::spirit::locals<float,boost::optional<float>>> &>,2> 1> ] can someone point out what's wrong with my parser ?

3 3

Buffer overrun error when calling std::exception::what() on boost::asio exception
by Brodie Thiesfield 03 Jun '10

03 Jun '10

Hi, OS: Windows Compiler: MSVC 2008 SP1 (9.0) Boost: 1.43.0 I am using boost asio, with an implementation of an SSL acceptor. I am setting up the context for the SSL, calling "use_private_key_file" on boost::asio::ssl::context. The file that I am specifying is invalid or doesn't exist, which causes asio to throw an error via: boost/asio/ssl/basic_context.hpp:289 boost::asio::detail::throw_error(ec); I catch the exception in a catch block like: catch (const std::exception& e) { fprintf(stderr, "exception: %s\n", e.what()); return false; } When I call e.what() everything blows up with a buffer overrun error. The stack looks like the following. The m_cat->message() call is causing the error (but I can't step into it for some reason). boost/system/error_code.hpp:350 std::string message() const { return m_cat->message(value()); } boost/system/system_error.hpp:68 m_what += m_error_code.message(); All libraries are built as static libs with the same compiler (OpenSSL, Boost). I've tried compiling with BOOST_EXCEPTION_DISABLE defined but it makes no difference. I've tried removing the const from the catch clause, but no difference. Has anyone seen an error like this before? I am frustrated by this error and don't know how to look into it further. Any suggestions? Regards, Brodie

1 1

issue when building boost with msvc2010
by Sean Farrow 03 Jun '10

03 Jun '10

Hi: When building boost with msvc 2010, I get a c1083 unable to open file with windows.h. My command line is: Bjam -toolset=msvc -stage -build-type=complete Even though as the output asks I have called vcvarsall.bat. A portion of the log follows: compile-c-c++ bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\ru ntime-link-static\debug.obj debug.cpp C:\Program Files\boost\boost/test/impl/debug.ipp(32) : fatal error C1083: Cannot open include file: 'windows.h': No such file or directory call "C:\Program Files\Microsoft Visual Studio 10.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\ runtime-link-static\debug.obj.rsp" ...failed compile-c-c++ bin.v2\libs\test\build\msvc-10.0\debug\asynch-exceptions-on\link-static\ru ntime-link-static\debug.obj... Anyone got any ideas, do I need to set up some environment variables? Vs2008 builds fine with the latest windows sdk, the latest windows sdk configuration tool doesn't allow integration with vc2010. Help appreciated as always. Regards Sean. Regards

1 0

[lamda] smart self binding of a function object
by alfC 02 Jun '10

02 Jun '10

Hi, I usually have function-classes that take two arguments. class f{ double operator()(double x, double y) const { x*y+y; } }; called as: f myf; cout << myf(4.,3.); for some purposes I need a function of only one variable (parametrized in the first or second argument. So I end up doing: using boost::function; using namespace boost::lambda; function<double(double)> myf_at5 = bind(&f::operator(), myf, 5., _1); // myf_at5 is myf(5.,*) So, now I can pass myf_at5 to something that accepts a function of one variable. root_finder( myf_at5, ...); since I have access to f I would like to make f smart enough to handle its own binding. Eg. class f{ ... // same as before function<double(double)> operator()(double x_value, YYYY y_arg){ return bind( &f::operator(), //maybe needs cast to double(*) (double,double) *this, x_value, y_arg ); } function<double(double)> operator()(XXXX x_arg, double y_value){ return bind( &f::operator(), //maybe needs cast *this, x_arg, y_value ); } }; The question is: What should be the type of XXXX and YYYY? in such a way that I can just call root_finder( myf(5., _1) , ... ); or more complicated things like root_finder( myf(5., _1 * 2.), ...); in other words, myf(5.,_1) is the same as the old bind(&f::operator(), myf, 5., _1); Thank you, Alfredo answers using other libraries are also accepted :)

5 11

Is an iterator the answer?
by Eric Fowler 02 Jun '10

02 Jun '10

I have a bit of a problem here, and would like some input. I am working on an optimization algorithm that makes much use of shortest-paths and near-neighbor concepts. To get a reasonable near-neighbor search on points on the Euclidean plane, I have implemented (mostly) a Delaunay tree as a Boost graph. The points on the graph represent points in space, and are unique. I make use of angles and cosines and such to find neighbors and so on. The Delaunay graph is a basis for the optimization and shortest-paths stuff, which encompasses concepts other than just points-in-space, like type-of-point and so on. The basic idea is that when I am traversing my graph looking for optimal paths and such, I use the Delaunay graph to find near neighbors. Now, here is the problem: my graph really needs for vertices to share points in space. Each vertex has a location and other properties, so there needs to be several vertices at one point. To have a Delaunay graph with coincident points is not really meaningful; the geometry gets really weird (e.g. what is the angle between three coincident points?). So I had figured I could build the Delaunay graph, and put a thin layer on it that handled reference counts for points in space, and stored vertices with all their attendant extra information. So I would make a class that had a map<> of points-in-space and their refcounts, and as I added or deleted points I would bump the counter. As far as the Dr. Delaunay knew, all points would be unique; every coincident point would share it's neighbors with it's brothers. Problem is, how to do a (for instance) breadth-first search over this graph, in which the graph algorithm needs to "see" each point individually? My thinking right now is that I can have my structure as described and build a custom iterator, then pass the iterator to the various search algorithms and visitors and what not. The iterator would have the intelligence to sort out the non-unique points. So the simple question is: am I on the right track? Is this kind of thing feasible or supported by the library? If it is, how do I specify my iterator to the various algorithms? Alternatively, does anyone know if it is possible to build a Delaunay graph with coincident points that has geometric meaning, i.e. preserves the properties we associate with Delaunay graphs? This is really a math question ... Sorry if this is incoherent, it is late and I need to get it off my chest. Eric

2 1

New algorithm for four coloring planar graphs
by Mike Douglass 02 Jun '10

02 Jun '10

I have uploaded the output of my new algorithm (program) for four coloring planar graphs to http://groups.google.com/group/graph_coloring_B1 Go to the files section, download the files, and read README. The program was written in the BGL of course. The algorithm is a very simple alternating chain algorithm with some new ordering features. It works better than you would think. I think those interested in the four coloring of planar graphs will like it. I will upload three other examples probably tomorrow. Thanks

2 1