March 2008 - Boost-users - lists.preview.boost.org

link problem: boost + stlport + msvc8
by CHABO Gabriel 05 Mar '08

05 Mar '08

Hi, This is my first post to the mailing list, and I hope I'm not bringing up a silly issue but I've been over this a million times already and I still can't figure it out, so any help is greatly appreciated. This is my setup: Windows XP SP2 Visual Studio 2005 STLport-5.1.5 boost_1_34_1 I'm trying to write a tiny program, just to get stlport and boost to play nice together and I still have "unresolved external symbol" errors. I downloaded stlport-5.1.5 and built it according to the instructions: * configure.bat -c msvc8 * nmake /f msvc.mak * nmake /f msvc.mak install I also added stlport\bin to my path Next, I downloaded the boost_1_34_1 windows installer and checked all the options to get all the libraries and added it to my path. Next, I configured visual studio 2005 as follows: Tools -> Options -> Project and Solutions -> VC++ Directories -> Show directories for: Include Files: Add the path to stlport headers (D:\Program Files\STLport-5.1.5\stlport) Library Files: Add the path to stlport libs (D:\Program Files\STLport-5.1.5\lib) Next, on the project level, I Added __STL_DEBUG to the list of preprocessors. I'm trying to build a Multi-threaded Debug DLL (/MDd). I also set the (/Zc:wchar_t-) option. When I tried to link, I got an error saying that libboost_program_options-vc80-mt-gdp-1_34_1.lib could not be found, so I had to build it. I edited the user-config.jam file to add the following line: using stlport : : D:/Program Files/STLport-5.1.5 D:/Program Files/STLport-5.1.5/lib ; This is the command I used to build the new lib bjam --toolset=msvc-8.0 link=static stdlib=stlport threading=multi cxxflags="-Zc:wchar_t-" --with-program_options -a -d2 I built the lib and moved it from bin.v2 to lib. As you can see I took every precaution with wchar_t so that all the libs are compatible, and still every time I try to link the simple program I get this error: error LNK2019: unresolved external symbol "public: __thiscall boost::program_options::options_description::options_description(class stlpd_std::basic_string<char,class stlpd_std::char_traits<char>,class stlpd_std::allocator<char> > const &,unsigned int)" (??0options_description@program_options@boost@@QAE@ABV?$basic_string@DV? $char_traits@D@stlpd_std@@V?$allocator@D@2@@stlpd_std@@I@Z) referenced in function _main This is the simple program in question #include <iostream> #include <boost/program_options.hpp> int main(int argc, char* argv[]) { std::cout<<"boost stlport test\n"; boost::program_options::options_description desc("Allowed options"); } Any help is greatly appreciated Thx Gabriel ******************************* This e-mail contains information for the intended recipient only. It may contain proprietary material or confidential information. If you are not the intended recipient you are not authorised to distribute, copy or use this e-mail or any attachment to it. Murex cannot guarantee that it is virus free and accepts no responsibility for any loss or damage arising from its use. If you have received this e-mail in error please notify immediately the sender and delete the original email received, any attachments and all copies from your system.

2 2

multi_array : boost::extents< is not a type
by Klett, Stefan 05 Mar '08

05 Mar '08

Hi List Members, While still trying putting boost::multi_array to work I ran into the next obstacle - my current code is fairly like in the multi_array "short example" That is : #include "boost/multi_array.hpp" class NwLayer { typedef boost::multi_array<double,3> neural_op; typedef neural_op::index index ; neural_op NeuralOperator(boost::extents[10][10][10]); } unfortunately I still get the compile time error In file included from ../NwLayer.cpp:1: ../NwLayer.h:12: Fehler: >boost::extents< is not a type on the last line of the above code snippet Is there anyone around who could explain to me what I can do to get rid of the error message ? Many thanks in advance Best regards Stefan -------------------------------------------------------- Stefan Klett (Dipl. Informatiker) | E-mail: stefan.klett(a)kit.edu Forschungszentrum Karlsruhe / KIT | Web: http://www.fzk.de SCC Steinbuch Centre for Computing| Phone: +49 (0)7247 828649 Hermann-von-Helmholtz-Platz 1 | Fax: +49 (0)7247 824972 D-76344 Eggenstein-Leopoldshafen | Mobile: +49 (0)176 20317250 -------------------------------------------------------------------

2 1

Boost.MPI - compilation problems
by Tom Smith 05 Mar '08

05 Mar '08

Hi, I seem to be having similar problems to this post: http://lists.boost.org/boost-users/2007/10/31648.php My MPI implementation is MPICH2; the first example in the docs at http://www.osl.iu.edu/~dgregor/boost.mpi/doc/mpi/tutorial.html works fine (this just creates a communicator and prints each process' rank). The second example in the tutorial (with recv and send): #include <boost/mpi.hpp> #include <iostream> #include <boost/serialization/string.hpp> namespace mpi = boost::mpi; int main(int argc, char* argv[]) { mpi::environment env(argc, argv); mpi::communicator world; if (world.rank() == 0) { world.send(1, 0, std::string("Hello")); std::string msg; world.recv(1, 1, msg); std::cout << msg << "!" << std::endl; } else { std::string msg; world.recv(0, 0, msg); std::cout << msg << ", "; std::cout.flush(); world.send(0, 1, std::string("world")); } return 0; } fails to link: output on g++ 4.0.1 on OSX 10.5 is Undefined symbols: "boost::mpi::communicator::operator int() const", referenced from: boost::mpi::status boost::mpi::communicator::recv_impl<std::basic_string<char, std::char_traits<char>, std::allocator<char> > >(int, int, std::basic_string<char, std::char_traits<char>, std::allocator<char> >&, mpl_::bool_<false>) constin ccxddhyu.o void boost::mpi::communicator::send_impl<std::basic_string<char, std::char_traits<char>, std::allocator<char> > >(int, int, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, mpl_::bool_<false>) constin ccxddhyu.o ld: symbol(s) not found My compiled invocation (which as mentioned compiles the first tutorial example) is just mpic++ -lboost_mpi-mt -lboost_serialization-mt main.cpp The problem (as for the person in the post I linked at the top) seems to be that mpi::communicator defines a conversion operator to MPI_Comm, which is just a typedef for int in MPICH2, but somehow the correct operator isn't found in the compiled lib.. Unlike the person in the linked post, I haven't had any other MPI libraries installed before, so there should be no inconsistencies. This problem occurs with 32 and 64-bit compiled libs built from subversion repos. code today. Thanks Tom

1 0

[regex] How to include named charact class [[:xdigit:]] in user defined character class
by Chandan Nilange 05 Mar '08

05 Mar '08

Hello, I want to include named character class [[:xdigit:]] as a part of another user defined character class. I am not getting the desired results. Pl. find my test example attached. Please let me know how to do this. Regards, Chandan ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs

2 1

Re: [Boost-users] Extending Multi Array
by Lang Stefan 05 Mar '08

05 Mar '08

Sorry folks for resending, but I forgot to change the subject line as appropriate when I first replied to the digest. My bad... For the sake of the initial poster (and anyone else interested in this issue) to help him spot my response, below is a copy of that initial reply. I cut away the rest of the digest as well as the original request that I did quote originally, to keep redundancy to a minimum. > Hi Stefan, > > My first guess would be to use initializers, as in > > NetworkMultiArray::NetworkMultiArray(int length) > : boost::multi_array<double,3>( ... ) > { > ... > } > > You shouldn't need to explicitly use the this keyword unless you are > using an obscure compiler and certain syntactically obscure > constructs. > However, if you want to refer to the methods of a superclass that you > have overridden in your own class, use this class's name as a > qualifier, > e. g. if you have a method 'begin()' for your own class that overrides > the 'multi_array::begin()' method, you will need to write > > boost::multi_array<double, 3>::begin(); > > Usually, when you try to extend an existing class it is a good idea to > create a typedef for your superclass, so it's easier to refer to it's > methods. In your case define a typdef like this: > > typedef boost::multi_array<double, 3> super; > > If you do that, you can forward calls to method overrides of your own > class to the baseclass by simply calling super::{_methodname_(...)}. E. > g. the above call would be reduced to > > super::begin(); > > Note that some methods cannot be overriden, such as the copy > constructors or assignment operators, so you will need to provide those > for your own class and explicitely call the base class methods inside: > > NetworkMultiArray& NetworkMultiArray::operator=(const NetworkMultiArray& other) > { > super::operator=(other); // explicit use of operator= neccessary here! > // copy extended stuff here > // ... > return *this; > } > > > Hope this helps! > > Regards, > Stefan (another one)

1 0

[foreach][range] How would BOOST_REVERSE_FOREACH be implemented?
by Pete Bartlett 05 Mar '08

05 Mar '08

Boost.Foreach is such a great piece of kit that once you start using it you just can't stop. Indeed I like working directly with values rather than iterators so much that is feels normal for me to write something like BOOST_FOREACH( value_type x , make_iterator_range( rbegin( seq ) , rend( seq ) ) whenever I wish to reverse iterate over a sequence. My co-workers are not big fans of typing though, so they want a REVERSE_FOREACH. At first naïve glance it looks easy to implement in terms of the original FOREACH: template<typename C> inline iterator_range</* something */> make_reverse_iterator_range(C& c) { return make_iterator_range( rbegin(c) , rend(c) ); } #define REVERSE_FOREACH( x , container ) BOOST_FOREACH( x , make_reverse_iterator_range( container ) ) This works fine when container is an lvalue. But then co-workers, used to the wonders of the original FOREACH, inevitably start to use the new macro with rvalues and get a crash due to dangling references (- the iterator_range's members point to departed quantities.....) and if you could somehow work around that, there is the issue hidden inside the "/* something */" that you have to cope with fact that MSVC has no proper const rvalue detection. Finally the #define MACRO(x,y) ANOTHER_MACRO( f(x) , g(y) ) style is not recommended because they are brittle in the face of x and/or y themselves being macros. ... but all these issues are exactly the ones that BOOST_FOREACH itself solves! Thus surely it is technically possible to write a REVERSE_FOREACH that works with rvalues as well as lvalues, e.g. by essentially copying and pasting the FOREACH implementation and replacing "begin" and "end" by "rbegin" and "rend" in the vital places. However one feels there may be a better way... has anyone had a go at creating a "fully-functional" REVERSE_FOREACH and could offer me any advice? If indeed the task is fairly intricate and requires re-use of foreach internals, would the library author consider including something in the library itself? Many thanks for reading, Pete Bartlett

4 6

Re: [Boost-users] Boost-users Digest, Vol 1561, Issue 2
by Lang Stefan 05 Mar '08

05 Mar '08

Hi Stefan, My first guess would be to use initializers, as in NetworkMultiArray::NetworkMultiArray(int length) : boost::multi_array<double,3>( {enter appropriate constructor parameters here} ) { ... } You shouldn't need to explicitly use the this keyword unless you are using an obscure compiler and certain syntactically obscure constructs. However, if you want to refer to the methods of a superclass that you have overridden in your own class, use this class's name as a qualifier, e. g. if you have a method 'begin()' for your own class that overrides the 'multi_array::begin()' method, you will need to write boost::multi_array<double, 3>::begin(); Usually, when you try to extend an existing class it is a good idea to create a typedef for your superclass, so it's easier to refer to it's methods. In your case define a typdef like this: typedef boost::multi_array<double, 3> super; If you do that, you can forward calls to method overrides of your own class to the baseclass by simply calling super::{_methodname_(...)}. E. g. the above call would be reduced to super::begin(); Note that some methods cannot be overriden, such as the copy constructors or assignment operators, so you will need to provide those for your own class and explicitely call the base class methods inside: NetworkMultiArray& NetworkMultiArray::operator=(const NetworkMultiArray& other) { super::operator=(other); // explicit use of operator= neccessary here! // copy extended stuff here // ... return *this; } Hope this helps! Regards, Stefan (another one) > -----Original Message----- > > Message: 3 > Date: Tue, 4 Mar 2008 16:21:12 +0100 > From: "Klett, Stefan" <stefan.klett(a)iwr.fzk.de> > Subject: [Boost-users] Extending Multi Array > To: <boost-users(a)lists.boost.org> > Message-ID: > <53834FBAA22A7F43B76C1BCD2D5A09D3026B16A0(a)FZKMSX6.ka.fzk.de> > Content-Type: text/plain; charset="us-ascii" > > Hi list members, > > This is a more or less a general kind of " how to extend container > classes" question - > I'm trying to write a class which inherits from > boost::multi_array - the > to get a kind of matrix class with some additional service routines. > Maybe it is a lack of experience - but I for the time being I > don't know > how to solve the problem to access the constructor of > multi_array in my > own class to initialize my matrix class - my current guess was to use > the "this" pointer but it had not worked out like I supposed > it to do. > > class NetworkMultiArray: public boost::multi_array<double,3> > { > public: > NetworkMultiArray(int); > virtual ~NetworkMultiArray(); > }; > > Here is the constructor code: > > > NetworkMultiArray::NetworkMultiArray(int length) > { > this::iterator iter3; > typedef this::template subarray<2>::type::iterator iter2; > typedef this::template subarray<1>::type::iterator iter1; > > for( iter1 i = *this.begin(); iter1 != *this.end(); iter1++){ > for(iter2 ii = (*i).begin(); ii != (*i).end(); ii++){ > for(iter3 iii = (*ii).begin(); iii != > (*ii).end(); iii++){ > *iii = 0.0 ; > } > } > } > } > sure this does not work - and is the result from trying to adapt the > iterator example from the documentation. > > my aim is to handle the NetworkMultiArray like a > boost::multi_array and > initialize it to zero - further it would be nice if somebody > could tell > me how I could assign a length to the array and the subarrays. > The key-point is that I want to use my class > (NetworkMultiArray) like a > multi_array by inheritance of the multi_array features of > boost:multi_array. > > Thank you very much in advance. > Best regards > Stefan

1 0

boost::ptr_array to use or not?
by j.c. 05 Mar '08

05 Mar '08

Hi, I have an pointer array of objects as follows: bucket * m_buckets[BUCKET_SIZE]; Because this array is a set size does it make sense to use a boost::ptr_array? boost::ptr_array<bucket, BUCKET_SIZE> m_buckets; * Note: One thing I first noticed with boost::array is that I cannot perform operations such as inserting at a desired index: // Existing code. m_buckets[bit] = new bucket(k,v); My goal is to use as many smart pointers as I can and less heap allocation. Thanks in advance, jc

2 1

[bind] Template deduction
by Robert Dailey 05 Mar '08

05 Mar '08

Hi, Suppose I have the following code: template< typename T > void DoFoo( int number, T object ) { } void main() { std::list<std::string> myStrings; // Assume myStrings is full of something... std::for_each( myStrings.begin(), myStrings.end(), boost::bind( &DoFoo, 5, _1 ) ); } On MSVC 9, this will not compile because DoFoo is a template, and at some point, boost::bind can't take a template as the function to call, or it can't deduce the function's template arguments. Is there a way I can make this work? Thanks.

2 1

boost posix time
by Aljaz 04 Mar '08

04 Mar '08

Hello I'm developing server and client applications and I need ping-pong mechanism to get ping time betwen them (time needed for data to arrive).. I've been thinking the best way would be to send Code: boost::posix_time::microsec_clock::universal_time()from server to client, which would then just reply with its own universal_time() .. The first problem is I dont know how to store universal_time() as an int/long.. I know I can store it as a string but this would take extra cpu to parse the data each time (with boost posix library).. All the data I send is serialized - maybe I could just serialize the Code: boost::posix_time::ptimeobject and send over? The next thing is what if server's time zone is gmt+1 and client's gmt+3? Will that affect the the time if I do the subtraction to get the ping time? If so, what can I do about it? Maybe anyone has better solution in mind? Thanks a lot for help

2 1