- Boost-users - lists.boost.org

Using boost library on MacOSx
by Richard L. Aurbach 24 Jul '09

24 Jul '09

On Jul 24, 2009, at 10:30 AM, boost-users-request(a)lists.boost.org wrote: > I have install boost library using this: > $ sudo port install boost > It seems download and compile fine on my local machine. > > But then i tried to compile a simple echo client example from > http://www.boost.org/doc/libs/1_35_0/doc/html/boost_asio/examples.html > > I get a bunch of compiler errors. I am using "g++ -o block_client > blocking_udp_echo_client.cpp " > > Here are some of the errors: > $ g++ -o block_client blocking_udp_echo_client.cpp > blocking_udp_echo_client.cpp:68:10: error: #include expects "FILENAME" > or <FILENAME> > blocking_udp_echo_client.cpp:69:10: error: #include expects "FILENAME" > or <FILENAME> > blocking_udp_echo_client.cpp:70:10: error: #include expects "FILENAME" > or <FILENAME> > blocking_udp_echo_client.cpp:71:10: error: #include expects "FILENAME" > or <FILENAME> > blocking_udp_echo_client.cpp:83: error: stray ‘\’ in program > blocking_udp_echo_client.cpp:108: error: stray ‘\’ in program > blocking_udp_echo_client.cpp:112: error: stray ‘\’ in program > > Thank you for any help. From my experience building software on MacOSX, I suspect that the problem is that you have not correctly defined an environment variable or something like that. If you write a line like "#include <boost/ blah>" and the compiler either cannot correctly resolve 'boost' or if the symbol resolves in such a way that an invalid file specification is created, then you might get the kind of error message that you are seeing. Another possibility -- if you have code left over from Windoze, look for backslashes in file specs. MacOSX is a Unix-like system that requires SLASHES, not BACKSLASHES. So lines line "#include <boost \header.h> will NOT work. Cheers, Rick Aurbach, Ph.D. President and Chief Engineer Aurbach & Associates, Inc. 8233 Tulane Avenue, Suite B St. Louis, MO 63132 www: http://www.aurbach.com/ eMail: support(a)aurbach.com [business] rick(a)aurbach.com [personal] Fax: 314/678-0869 Phone: 800/774-7239 314/726-1321 "Cowardice asks the question, 'Is it safe?' Expediency asks the question, 'Is it politic?' Vanity asks the question, 'Is it popular?' But, conscience asks the question, 'Is it right?' And there comes a time when one must take a position that is neither safe, nor politic, nor popular but one must take it because one's conscience tells one that it is right." -Martin Luther King, Jr. "If in the last few years you haven't discarded a major opinion or acquired a new one, check your pulse. You may be dead." - Gelett Burgess (1866-1951) "A life spent making mistakes is not only more honorable, but more useful than a life spent doing nothing." - George Bernard Shaw (1856-1950)

1 0

Re: [Boost-users] boost 1.39 serialization via base pointer ok on gcc, but crashes on AIX V10.1
by Avi Bahra 24 Jul '09

24 Jul '09

Ok first I hadn't realised my attachment was removed. So I include the source code below. This is small and self contained. o The compile error on AIX was removed because I saving via a non const variable. ( Note however gcc 4.2.1. on Linux it did not complain.) o To recap this compiles and runs on Linux with gcc 4.2.1 but crashes/exception on AIX V10.1. Because it reckons the derived/base relationship has not been registered. ... oserializer.hpp // convert pointer to more derived type. if this is thrown // it means that the base/derived relationship hasn't be registered vp = serialization::void_downcast( *true_type, *this_type, static_cast<const void *>(&t) ); if(NULL == vp){ boost::serialization::throw_exception( >>>> archive_exception(archive_exception::unregistered_cast) ); } Note if a save via the derived ptr, then it works. o If I explicitly register this relationship, (see code below) Then I get past the first exception only to get an assert. // since true_type is valid, and this only gets made if the // pointer oserializer object has been created, this should never // fail const basic_pointer_oserializer * bpos = archive_pointer_oserializer<Archive>::find(* true_type); >>> assert(NULL != bpos); if(NULL == bpos) boost::serialization::throw_exception( archive_exception(archive_exception::unregistered_class) ); ar.save_pointer(vp, bpos); o Initially I thought it was compiled without rtti on AIX. However this was not the case. Here is compile line. xlC -c -DDEBUG -qcpluscmt -qNOOPTimize -qnoinline -g -qfullpath -qfuncsect -qeh -qrtti -qtemplatedepth=700 -I"../SCRATCH/src" -I"/s1a/emos_esuite/emos_data/sms/boost_1_39_0" -o "bin/vacpp/debug/src/TestSerialisation.o" "src/TestSerialisation.cpp" This is show stopper bug that I must get fixed. Are there any more things I can try to debug this ? Any help would be much appreciated. //=============== the example ======================== #include <fstream> #include <assert.h> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/string.hpp> #include <boost/serialization/export.hpp> class RepeatBase { public: RepeatBase() {} virtual ~RepeatBase() {} virtual const std::string& name() const = 0; private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) {} }; class RepeatDerived : public RepeatBase { public: RepeatDerived( const std::string& name) : name_(name) {} RepeatDerived() {} virtual const std::string& name() const { return name_;} private: std::string name_; friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { // save/load base class information ar & boost::serialization::base_object<RepeatBase>(*this); ar & name_; } }; class Defs { public: Defs(const std::string& f) : fileName_(f), repeat_(new RepeatDerived(f)) {} Defs() : repeat_(NULL) {} ~Defs() { delete repeat_;} bool operator==(const Defs& rhs) const { return fileName_ == rhs.fileName_;} const std::string& fileName() const { return fileName_;} private: std::string fileName_; RepeatBase* repeat_; friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int /*version*/) { ar & fileName_; ar & repeat_; } }; BOOST_CLASS_EXPORT(RepeatBase); BOOST_CLASS_EXPORT(RepeatDerived); BOOST_SERIALIZATION_ASSUME_ABSTRACT(RepeatBase) int main() { // With this explicit register, we get a bit further, but still crashes on AIX // boost::serialization::void_cast_register<RepeatDerived, RepeatBase>(0,0); const Defs saveDefs("saveFile.txt"); { // Save the defs file std::ofstream ofs( saveDefs.fileName().c_str() ); boost::archive::text_oarchive oa( ofs ); oa << saveDefs; // BOMBS out here on AIX V10.1 } { // Restore the defs file Defs loadDefs; std::ifstream ifs(saveDefs.fileName().c_str()); boost::archive::text_iarchive ia(ifs); ia >> loadDefs; assert( saveDefs == loadDefs ); } return 0; } //=============================================== Best regards, Ta, Avi

2 1

Problem with boost::interprocess:basic_string : Error address out of bounds
by samuel.gallard 24 Jul '09

24 Jul '09

Hello, I first present to you my code and I will explain later : ShmString.h : #include #include #include namespace ip=boost::interprocess; typedef ip::allocator CharAllocator; class ShmString : public ip::basic_string, CharAllocator > { public: ShmString (); virtual ~ShmString (); private: /** * The segment to build the string */ static ip::managed_shared_memory* st_segment; } This class served to have some string in memory. The following Element.h used it : #include "ShmString.h" class Element { public: Element (const Element& elem); Element (const std::string dir, const std::string file, const std::string type, const std::string newpath, const std::string newfile, ip::managed_shared_memory * segment); virtual ~Element (); private: ShmString dir; ShmString file; ShmString target; ShmString type; ShmString newpath; ShmString newfile; }; I have four attributes that declare it. When I tried to define all the attributes in the constructor like this in Element.cpp : Element::Element (const string dir, const string file, const string type, const string newpath, const string newfile, ip::managed_shared_memory *segment){ try{ this->segment = segment; //test the type if(type == Request::st_rename){ this->type.assign(type.c_str()); } else{ string res = "Class Element in constructor -> The type should be : rename "; throw EventException(res.c_str(), __LINE__ ); } this->dir.assign(dir.c_str()); this->file.assign(file.c_str()); this->target.assign(""); this->newfile.assign(newfile.c_str()); this->newpath.assign(newpath.c_str()); } catch(ip::interprocess_exception &e){ ostringstream oss; const std::string what(e.what()); oss << "Class Element in constructor 3 -> "; oss << what; throw EventException(oss.str().c_str(), __LINE__ ); } } If for example "const string dir " has the size more than 10 (the capacity by default for a ShmString object), I obtained an error like this : 0x0807439f in boost::interprocess::basic_string, boost::interprocess::allocator, 0u>, boost::interprocess::iset_index> > >::assign (this=0xb6c4e274, s=0xb8af42d0 ) at /usr/local/include/boost/interprocess/containers/string.hpp:988 #3 0x08071d26 in Element (this=0xb6c4e26c, elem=@0xb7fa224c) at ..Element.cpp:22 I tried to used some other constructors be default for basic_string but it remains the same. This code only works when in the constructors the string size are below 10. Could you someone explain me why I obtain this? Do I have some faulty code? Could you provide some examples on how store string with interprocess::basic_string in share memory for a size for more than 10 characters? I will be very grateful because I am lost... Thanks you in advance Regards Samuel -- View this message in context: http://www.nabble.com/Problem-with-boost%3A%3Ainterprocess%3Abasic_string-%… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

Alternative to bjam threading=multi
by Avi Bahra 24 Jul '09

24 Jul '09

On AIX I want to compile with multi-threading, then means using xlC_r. Using bjam threading=multi Will do this. However is there a way achieving the same effect without putting "threading=multi" on the command line ? i.e putting the threading model in site-config.jam In my site-config.jam , have tried the following, all to no avail *using* xlc : : xlC_r ; using vacpp : : <threading>multi ; using vacpp : xlC_r ; Best regards, Ta, Avi

2 1

[Asio] How can you detect port collsions?
by Robert Jones 24 Jul '09

24 Jul '09

Hi All Thanks to those of you who have answered my previous asio queries - I'm gradually gaining on the problems! I am allocating port numbers on tcp connections dynamically, and must allocate and reserve the port number without blocking. I also have multiple instances of the process which needs to do this, and I'm plagued by port collisions. How do I detect, and avoid, port number collsions? Thanks, Rob.

2 1

[boost-users][bind] type of boost::bind
by Sakharuk, Vlad (GMI Development) 24 Jul '09

24 Jul '09

Here is example: struct a{ int foo(bool){ return 0; } }; boost::function1<int, void> func = bind(&a::foo,_1,false); a somea; func(somea); Is it possible to get a bind type? Something like: boost::mpl::identity<boost::bind<int,void,boost::arg<1>,a> >::type abind; Thanks Vlad. -------------------------------------------------------------------------- This message w/attachments (message) may be privileged, confidential or proprietary, and if you are not an intended recipient, please notify the sender, do not use or share it and delete it. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Merrill Lynch. Subject to applicable law, Merrill Lynch may monitor, review and retain e-communications (EC) traveling through its networks/systems. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or error-free. References to "Merrill Lynch" are references to any company in the Merrill Lynch & Co., Inc. group of companies, which are wholly-owned by Bank of America Corporation. Securities and Insurance Products: * Are Not FDIC Insured * Are Not Bank Guaranteed * May Lose Value * Are Not a Bank Deposit * Are Not a Condition to Any Banking Service or Activity * Are Not Insured by Any Federal Government Agency. Attachments that are part of this E-communication may have additional important disclosures and disclaimers, which you should read. This message is subject to terms available at the following link: http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you consent to the foregoing. --------------------------------------------------------------------------

2 1

[MPI] Serializing a class with a string data member
by Mathieu Larose 23 Jul '09

23 Jul '09

Hello, I'm trying to register a class (named A), because it will be a derived class. Calling BOOST_CLASS_EXPORT(A) will generate a lot of errors that look similar. Here is one of them: /usr/local/include/boost/mpi/datatype.hpp:184: error: no matching function for call to ‘assertion_failed(mpl_::failed************ boost::mpi::is_mpi_datatype<std::basic_string<char, std::char_traits<char>, std::allocator<char> > >::************)’ Here is the code: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - #include <iostream> #include <boost/mpi.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/mpi/packed_oarchive.hpp> #include <boost/mpi/packed_iarchive.hpp> #include <boost/serialization/export.hpp> #include <boost/serialization/string.hpp> class A { friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & _text; } }; BOOST_CLASS_EXPORT(A); // I need to export A, because I will use // abstract base pointer. A is not a derived // class for simplicity here. int main(int argc, char * argv[]) { return 0; } - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - I've found that someone had the same problem ( http://groups.google.com/group/boost-list/msg/c9a0c133bddd783d ), but no solution was found. What am I doing wrong? Thanks, Mathieu Larose

3 3

[BGL] Converting between adjacency_list with listS and vecS
by Ethan Kim 23 Jul '09

23 Jul '09

Hi all, The long standing question I've had for BGL is how to easily convert a graph (using adjacency_list) from listS as vertex list to vecS. (And vice versa) I understand why there needs to be 2 different implementations depending on different uses of algorithms. But is there an easy way to convert them back and forth? Any tip would be much appreciated. Thanks,

4 4

[Thread] Interruption Feature
by Timothy 23 Jul '09

23 Jul '09

Hi, I want to write a signal handler for SIGINT that notifies the main() thread to break out of a loop and exit. I was thinking of using Boost.Thread's interruption feature. Does it support this usage, or do I need to use some other way to do this? Thanks in advance. Tim

2 1

Other tcp stream examples? suggestion about asio examples... Re: usage of boost::asio::async_write and boost::asio::buffer
by Ozgur Ozturk 23 Jul '09

23 Jul '09

Hi All, Thanks to Igor's pointing to streambuf I realized availability of stream operations. I think for many newbies, stream examples would be easier to understand/reuse. I found the very simple daytime_client and daytime_server examples useful. (Examples could be reordered to start with simpler ones.) Are there any other asio examples using streams? If the contributors could add stream versions of chat server and client examples that would be great. I guess using streams is somewhat less efficient, is that correct? (since you have more control on the buffer without streams.) Best regards, Ozgur (Oscar) Ozturk www.DrOzturk.com Phone: +1 (908) DROZGUR i.e, +1 (908) 376-9487 On Tue, Jul 21, 2009 at 4:28 PM, Igor R <boost.lists(a)gmail.com> wrote: > > I tried to use the same string as buffer for both async_read_some and > async_write > > Who will allocate the buffer for the read operation? socket's > async_read_some() doesn't know to do this alone. > > > I guess the string can not be treated as a mutable buffer, so maybe I > should > > keep the original char * buffer for async_read_some and use the string > only > > for outputting with async_write. Is that right? > > (or any other recommendation?) > > Why wouldn't you use asio::streambuf? You've a set of functions that > async.read/write from/to it, and it grows automatically. You can see > relevant examples in asio docs. > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

2 1