January 2006 - Boost-users - lists.preview.boost.org

newby problem: boost build problems under debian
by Grégoire Weber 23 Jan '06

23 Jan '06

Hi! I try to build boost from source (boost-1.33.1 from sf.net) under debian stable on x86. In the first run I followed the "Getting Started" document but run into problems. In a second round I tried out the following additionaly: * I downloaded and extracted the sources (tar) to my BOOST_ROOT * I built bjam according the "Building Boost.Jam" instructions: http://www.boost.org/tools/build/jam_src/index.html#building_bjam --> bjam seems ok to me * In Step 4 before carrying out the install uncommented the ``using gcc ;`` line in BOOST_ROOT/tools/build/v2/user-config.jam * Step 5: In the BOOST_ROOT I tried to trigger install``bjam --v2 install`` Now I'm stuck here, googling around didn't help. Any hints? Thanks. Gregoire P.S.: Everything I do I do as ordinary user. Do I need root priviledges? ----------------------------------------- bjam output ----------------------------------------- weber@stein:~/sz/nosvn/boost-1.33.1$ bjam --v2 install Building Boost.Regex with the optional Unicode/ICU support disabled. Please refer to the Boost.Regex documentation for more information (and if you don't know what ICU is then you probably don't need it). warning: Python location is not configured warning: the Boost.Python library won't be built /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/build/virtual-target.jam:868: in virtual-target.register-actual-name from module virtual-target error: Duplicate name of actual target: <p/usr/local/lib>libboost_date_time-gcc-d-1_33.so.1.33.1 error: previous virtual target { common%common.copy-libboost_date_time-gcc-d-1_33.so.1.33.1.SHARED_LIB { gcc%gcc.link.dll-libboost_date_time-gcc-d-1_33.so.1.33.1.SHARED_LIB { gcc%gcc.compile.c++-greg_month.o.OBJ { gregorian/greg_month.cpp.CPP } } { gcc%gcc.compile.c++-greg_weekday.o.OBJ { gregorian/greg_weekday.cpp.CPP } } { gcc%gcc.compile.c++-date_generators.o.OBJ { gregorian/date_generators.cpp.CPP } } } } error: created from ./install-lib error: another virtual target { common%common.copy-libboost_date_time-gcc-d-1_33.so.1.33.1.SHARED_LIB { common%common.copy-libboost_date_time-gcc-d-1_33.so.1.33.1.SHARED_LIB { gcc%gcc.link.dll-libboost_date_time-gcc-d-1_33.so.1.33.1.SHARED_LIB { gcc%gcc.compile.c++-greg_month.o.OBJ { gregorian/greg_month.cpp.CPP } } { gcc%gcc.compile.c++-greg_weekday.o.OBJ { gregorian/greg_weekday.cpp.CPP } } { gcc%gcc.compile.c++-date_generators.o.OBJ { gregorian/date_generators.cpp.CPP } } } } } error: created from ./install-lib error: added properties: none error: removed properties: none /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/build/virtual-target.jam:175: in actualize-no-scanner from module object(file-target)@291 /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/build/virtual-target.jam:110: in object(file-target)(a)291.actualize from module object(file-target)@291 /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/build-system.jam:285: in load from module build-system /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/kernel/modules.jam:259: in import from module modules /home/weber/sz/nosvn/boost-1.33.1/tools/build/v2/kernel/bootstrap.jam:153: in boost-buildfrom module /home/weber/sz/nosvn/boost-1.33.1/boost-build.jam:12: in module scope from module

1 0

boost::serialization stream exception
by Adrian Grigore 23 Jan '06

23 Jan '06

Hi, I encountered a mysterious stream exception when trying to deserialize an std::map<std::string,long> container. The exception seems to occur depending on the data stored in the container. Unfortunately none of the hints regarding stream exceptions from the manual seem to help. The source code is as follows: >>>> std::map<RHANDLE,long> ArchiveEntries; ... void ResArchive::SaveToFile() { std::ofstream ofs (INDEX_FILE_NAME); boost::archive::binary_oarchive OArchive(ofs); OArchive <<ArchiveEntries; } void ResArchive::LoadFromFile() { std::ifstream ifs(INDEX_FILE_NAME); boost::archive::binary_iarchive IArchive(ifs); IArchive >>ArchiveEntries; //STREAM EXCEPTION!!! } <<<< I do not open any file stream to the index file in any other part of my source code, so multiple simultaneous streams can't be the problem. I can consistently reproduce the error for 2 given sets of test data, but I also have some test data ( just 2 entries instead of a few thousand) where the problem does not occur. I have not been able to isolate the exact data that is needed to cause such a stream exception though. Any hints on how to proceed would be greatly appreciated. Thanks in advance for your time! :-) Best Regards, -- Adrian Grigore -- Adrian Grigore adrian(a)lobstersoft.com Lobstersoft - Fun Brain-Bending Games For All Ages http://www.lobstersoft.com Member of the Association of Shareware Professionals

2 1

Re: [Boost-users] STL-PORT and VC 7.1
by Sliwa, Przemyslaw (London) 23 Jan '06

23 Jan '06

This what I did. I had to create the projects from scratch. Fairly simple. Most of the test compile and link fine. Some of them fail. Thanks for help to anyone, Pshemek Kind regards Pshemek Przemyslaw Sliwa Global Foreign Exchange FX Risk Engine Development Merrill Lynch ( +44 (0)20 7995 4182 ) +44 (0)79 6325 2319 * Email: przemyslaw_sliwa(a)ml.com -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Nat Goodspeed Sent: 20 January 2006 15:57 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] STL-PORT and VC 7.1 > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sliwa, Przemyslaw (London) > Sent: Friday, January 20, 2006 10:32 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] STL-PORT and VC 7.1 > > I have a quick question - I would like to build the static library and > not the DLL. > Could you tell me what should be changed in the project apart form the > "General->static lib"? [Nat] Um, no. But since .vcproj files are specially-formatted text files, I assume the worst case would involve building a new static-library project by manually setting things the same way in your new project file as in my old one. The IDE may let you change the root type of a project after the fact; I've never tried. > Do you know how to disable the auto-link of boost? [Nat] These are things we changed in boost/boost/config/user.hpp: // BOOST_WHATEVER_DYN_LINK: Forces library "whatever" to be linked as a dll // rather than a static library on Microsoft Windows: replace the WHATEVER // part of the macro name with the name of the library that you want to // dynamically link to, for example use BOOST_DATE_TIME_DYN_LINK or // BOOST_REGEX_DYN_LINK etc (this macro is used to turn on __declspec(dllimport) // modifiers, so that the compiler knows which symbols to look for in a dll // rather than in a static library). // Note that there may be some libraries that can only be statically linked // (Boost.Test for example) and others which may only be dynamically linked // (Boost.Threads for example), in these cases this macro is unsupported. // #define BOOST_WHATEVER_DYN_LINK #define BOOST_FILESYSTEM_DYN_LINK #define BOOST_SIGNALS_DYN_LINK #define BOOST_PYTHON_DYN_LINK // BOOST_ALL_NO_LIB: Tells the config system not to automatically select // which libraries to link against. // Normally if a compiler supports #pragma lib, then the correct library // build variant will be automatically selected and linked against, // simply by the act of including one of that library's headers. // This macro turns that feature off. #define BOOST_ALL_NO_LIB _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users -------------------------------------------------------- If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/ --------------------------------------------------------

1 0

[iostreams] Error-handling with filters
by Paul Giaccone 23 Jan '06

23 Jan '06

How can you deal with a corrupt file when using filters? I have a chain of filters that includes a bzip2 and multichar filter have written myself. If the read() function fails to find what it expects in the file, it returns 0 (meaning "no characters read"). Otherwise it returns the number of characters read (which should be equal to the number of characters requested in the third function parameter) or -1 if the end of the file is reached. This works fine if the file is as is required, but if it is corrupted in any way, the program either hangs or crashes at if (s->strm != strm) return BZ_PARAM_ERROR; in bzlib.c, with s as a pointer that cannot be read (set to 0xcdcdcdcd). I'm working in Microsoft Visual C++ and gcc 3.2.3. How should this be handled? I haven't seen anything about it in the documentation. I imagine I should throw an exception instead of returning 0 (perhaps an iostreams exception?). Given that the crash is the bzip2 (bzip2 being the first operation done when deserialising the file), where would I put the associated catch block? Thanks, Paul

1 3

[Boost.Build v2] Incorrectly handled header dependencies
by Jaroslav Gresula 23 Jan '06

23 Jan '06

I'm using the CVS version of Boost.Build on WinXP. Consider the following project: top/ | +-- include/ | | | +-- dir/ | | | +--header.h +-- Jamroot +-- main.cxx // -- main.cxx #include <dir/header.h> int main() { return 0; } # -- Jamroot path-constant TOP : . ; exe app : main.cxx : <include>$(TOP)/include ; Modification of header.h does not invoke the app target. I'm not familiar with bjam at all, but there is one thing I noticed and which might be relevant. When bjam is run with -d7 the include directory appears in the bjam's output with leading '/' and 'header.h' itself is labeled as missing: set SEARCH = /c:/Temp/bb2/include set HDRSCAN = #[ t]*include[ ]*(<(.*)>|"(.*)") set SCANNER = object(c-scanner)@45 set HDRRULE = scanner.hdrrule set HDRGRIST = object(c-scanner)@45 build file: root = '/c:/Temp/bb2/include' dir = 'dir' base = 'header' search <object(c-scanner)@45>dir/header.h: /c:/Temp/bb2/include\dir/header.h build file: dir = '\c:\temp\bb2\include\dir' build file: dir = 'dir' base = 'header' search <object(c-scanner)@45>dir/header.h: dir/header.h build file: dir = 'dir' set SEARCH = set HDRSCAN = set SCANNER = set HDRRULE = set HDRGRIST = bind -- <object(c-scanner)@45>dir/header.h: dir/header.h time -- <object(c-scanner)@45>dir/header.h: missing made stable <object(c-scanner)@45>dir/header.h made* newer <p.-object(c-scanner)@45>main.cxx made+ update <pbin\msvc\debug>main.obj made+ update <pbin\msvc\debug>app.exe made update all -- Jarda

2 1

Hello: boost::filesystem with cygwin - windows style paths
by Simon Lam 22 Jan '06

22 Jan '06

Hello all, I've been fiddling around with the gnu gcc programming enviroment on cygwin, trying to make boost::filesystem work. I've hit a snag where the boost libraries consider my filesystem to be single rooted, and thus path won't take "c:\" as an argument. I've compiled the simple_ls.cpp program included in the documentation. It uses boost::filesystem::path::native_file_string(), and this returns a unix style path. I've tried running "simple_ls c:" but it complains that /home/<path to current dir>/c: is unavailable. I want to be able to use this in a program thats for windows, but independent of cygwin. Any ideas how to work around this? >Thanks >Simon (ps: yah! my first post!)

2 1

[Boost.Build v2] how to disable specific warnings on gcc
by Jaroslav Gresula 22 Jan '06

22 Jan '06

If I want to disable a specific warning on gcc (3.4.2, mingw) and use -Wall at the same time then on the gcc commnad line the -Wall option must precede the option disabling the specific warning. E.g. gcc -Wno-non-virtual-dtor -Wall main.cxx ignores the '-Wno-non-virtual-dtor' option, whereas gcc -Wall -Wno-non-virtual-dtor main.cxx works as expected. My problem is that I can't achieve the correct option order with Boost.Build v2 as the -Wall is used by default and the disabling option is always put before it. I tried to specify it in the toolset configuration: using gcc : : : <cxxflags>-Wno-non-virtual-dtor ; and also putting <toolset>gcc:<cxxflags>-Wno-non-virtual-dtor ; into project's requirements but neither works. Please, could anyone help me out with this? -- Jarda

1 0

glibc detected malloc(): memory corruption:
by Daniel Weiss 22 Jan '06

22 Jan '06

Hi I am working with the matrix libraries (#include "boost/numeric/ublas/matrix.hpp" etc...) and my program is running fine during several loops but suddenly stops, showing the message: *** glibc detected *** malloc(): memory corruption: 0x0865e638 *** /bin/sh: line 1: 8311 Aborted I found that the error occurs on the line where I initialize some matrices: matrix<float_type> Stiff_matrix(6,6), Qq_matrix(3,3); What I don't understand is, that the program runs through this line the first 3 times and only the 4th time the error occurs.... How can that be?? Greets, Dan. -- 10 GB Mailbox, 100 FreeSMS/Monat http://www.gmx.net/de/go/topmail +++ GMX - die erste Adresse für Mail, Message, More +++

1 0

[Win32 VC70 Linking] Hard coding library name for the final linking ('libboost_filesystem-vc7-mt-s-1_33_1.lib' ...)?
by Allen Zhao 22 Jan '06

22 Jan '06

Hi, All, I have a weired problem in linking libtorrent dll with my application. The libtorrent depends on 3 boost projects. This is what I got: test fatal error LNK1104: cannot open file 'libboost_filesystem-vc7-mt-s-1_33_1.lib' It looks like I have to build libboost_filesystem-vc7-mt-s-1_33_1.lib, and add this library explicitly in the final linking, no mater what my project setting is. The library filename must be named above. I also built a VC 7.0 from scratch (which contains all the needed boost source (not the prebuilt library)), it all ends the same linker error. Mystereous as it is, I am a little suspecious now. Is it possible that this library name is actually coded somewhere by some algorithm? Best regards, Allen Zhao

3 3

Re: [Boost-users] [tuple] no easy access to get function
by JOAQUIN LOPEZ MU?Z 22 Jan '06

22 Jan '06

----- Mensaje original ----- De: Frank Birbacher <bloodymir.crap(a)gmx.net> Fecha: Sábado, Enero 21, 2006 2:29 am Asunto: [Boost-users] [tuple] no easy access to get function > Hi! > > Today I wanted to std::transform a range of tuples into a sequence of > one element. I thought I could just take the address of some > tuple::get<N> function. Maybe also specify the Tuple type as template > parameter [...] > I'd rather want to take the address of boost::tuple::get<N,Tuple>, but > the template parameters won't match. > Maybe you forgot the final s in "tuples" when writing boost::tuples::get<N,Tuple> ?? Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 3

statechart: interaction with environment
by Oliver.Kowalke＠infineon.com 21 Jan '06

21 Jan '06

Hello, I try to understand how boost::statechart work - so I have some questions. If the FSM implements a communication protocol how does the FSM interact with the environment? 1. Problem: How to passing callback-functions to the FSM? * The callback should be invoked if the FSM is in a certain state and some conditions are met. * Do I have to pass the callback(s) with an event which has to be injected to the FSM from outside and store them in a given state? 2. Problem: How to pass initialization parameter into the FSM? * Some states need to be initialized with parameters (for instance host and port in order to create a socket). * Do I Have to pass the parameters via injected events? 3. Problem: How to pass data between several states in an FSM? * Same as Problem 2? With best regards, Oliver

2 1

statechart: howto connection two different FSMs
by Oliver.Kowalke＠infineon.com 21 Jan '06

21 Jan '06

Hello, The tutorial of boost::statechart contains an incomplete example of connecting two FSMs of the same type. For me it is not clear how the events are propagated between this two FSMs (concret C++-code) and if its possible to connect two FSMs of different type! Could you give me a simple (complete) example, please? Regards, Oliver

2 1

statechart: unable to compile examples - additional compiler flags needed?
by Oliver.Kowalke＠infineon.com 21 Jan '06

21 Jan '06

Hello, I'm using statechart on LINUX with gcc-3.4.2. bjam was able to compile the examples but if I copy the StopWatch-example into a Kdevelop-project I get errors?! I assume I've to put some additional compiler-flags which are not mentioned in the tutorial. Could you help please! In file included from /opt/boost/include/boost-1_33_1/boost/statechart/simple_state.hpp:13, from /opt/boost/include/boost-1_33_1/boost/intrusive_ptr.hpp:35, from /opt/boost/include/boost-1_33_1/boost/statechart/detail/state_base.hpp:1 6, from /opt/boost/include/boost-1_33_1/boost/statechart/state_machine.hpp:16, from /opt/boost/include/boost-1_33_1/boost/intrusive_ptr.hpp:34, from /opt/boost/include/boost-1_33_1/boost/statechart/event_base.hpp:15, from /opt/boost/include/boost-1_33_1/boost/statechart/event.hpp:11, from /home/kowalke/Projects/tcpproto/src/tcpproto.cpp:33: /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp:2 6: Fehler: Fehler beim Parsen before `<' token /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp:3 6: Fehler: destructors must be member functions /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp: In function `void boost::statechart::detail::leaf_state()': /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp:3 6: Fehler: ` void boost::statechart::detail::leaf_state()' redeclared as different kind of symbol /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp:2 6: Fehler: previous declaration of `template<class Allocator, class RttiPolicy> class boost::statechart::detail::leaf_state' /opt/boost/include/boost-1_33_1/boost/statechart/detail/leaf_state.hpp:2 6: Fehler: previous non-function declaration `template<class Allocator, class RttiPolicy> class boost::statechart::detail::leaf_state' ... with best regards, Oliver

2 1

gcc3.2 preprocessor concatenation or not on negative numbers
by Andy Little 21 Jan '06

21 Jan '06

Hi all, Using gcc3.2 the concatenation of 1##Num where Num is negative leaves a space eg 1e#-1 --> 1e -1 as can be seen by the preprocessed output below. The only way round it that I can see is to simply not use a macro for this compiler Is there a preprocessor workaround for this problem? Is this even the correct behaviour? Any help appreciated. #define POW_EVAL(Num, Float_type)\ template <>\ struct pow_eval< \ Float_type BOOST_PP_COMMA() Num BOOST_PP_COMMA()\ 1 BOOST_PP_COMMA() false\ >{\ enum{ required = true};\ typedef Float_type result_type;\ result_type operator()()const\ { result_type result\ = static_cast< Float_type > (1e ## Num);\ return result;\ }\ }; POW_EVAL(-60, double) POW_EVAL(-59, double) POW_EVAL(58, double) POW_EVAL(57, double) // preprocessor output template <> struct pow_eval< double , -60 , 1 , false >{ enum{ required = true}; typedef double result_type; result_type operator()()const { result_type result = static_cast< double > (1e- 60); return result; } }; template <> struct pow_eval< double , -59 , 1 , false >{ enum{ required = true}; typedef double result_type; result_type operator()()const { result_type result = static_cast< double > (1e- 59); return result; } }; template <> struct pow_eval< double , 58 , 1 , false >{ enum{ required = true}; typedef double result_type; result_type operator()()const { result_type result = static_cast< double > (1e58); return result; } }; template <> struct pow_eval< double , 57 , 1 , false >{ enum{ required = true}; typedef double result_type; result_type operator()()const { result_type result = static_cast< double > (1e57); return result; } }; regards Andy Little

2 5

regex format string extension for more than nine captures
by Evan Burkitt 21 Jan '06

21 Jan '06

I need to be able to reformat match_results that contain more than nine captures using format strings. I'm looking for advice as to the easiest way to accomplish this. Is there some existing work I can build on or borrow from? I need syntax similar to the boost extended syntax, possibly using two digits for the capture number, e.g., '$01' It's important to me to be able to use the extended syntax's conditional expression '?Ntrue-expression:false-expression', again probably with two digits 'NN' in place of the single 'N'. Any help would be greatly appreciated. -eb-

2 1

[tuple] no easy access to get function
by Frank Birbacher 21 Jan '06

21 Jan '06

Hi! Today I wanted to std::transform a range of tuples into a sequence of one element. I thought I could just take the address of some tuple::get<N> function. Maybe also specify the Tuple type as template parameter. But from the source I figured the get function is not designed in a way that would allow me easy access. I had to write myself an additional function for this task (see below). Am I trying the wrong way? Or would you fix the implementation? template<size_t N, typename Tuple> typename boost::tuples::element<1,Tuple>::type const& getTupleElement(const Tuple& tuple) { return boost::get<N>(tuple); } to do: transform(..., ..., ..., &::getTupleElement<1,Tuple>); or //using boost::lambda for_each(..., ..., var(i) += bind(&::getTupleElement<1,Tuple>, _1)); I'd rather want to take the address of boost::tuple::get<N,Tuple>, but the template parameters won't match. Frank

1 0

Parse errors when building against Insure++ by ParaSoft.
by Sven Delmas 20 Jan '06

20 Jan '06

I am trying to do some memory allocation debugging using Insure++ on my Windows machine. They use their own parser, and inject all kinds of fun stuff. However, their parser appears to be more restrictive (or just plain different) than the usual parsers. I end up with two parser errors: One in boost/details/indirect_traits.hpp (line 254 in boost 1.33.1): BOOST_STATIC_CONSTANT( bool, value = sizeof(detail::is_function_ref_tester(t,0)) == sizeof(::boost::type_traits::yes_type)); The error is: namespace "boost::detail" has no member is_function_ref_tester I can resolve that by taking the detail:: away, which seems to make sense if it indeed not part of the namespace. Can anyone confirm whether that is a valid solution/issue? The second issue is in boost/pool/detail/ct_gcd_lcm.hpp (line 68 in boost 1.33.1): BOOST_STATIC_CONSTANT(unsigned, value = ct_gcd2<B>::helper<A % B>::value); The error is: the global scope has no value I can resolve this by switching to the other branch with respect to the macro BOOST_NO_TEMPLATE_PARTIAL_SPECIALIZATION. That compiles fine for VC 7.0, VC 7.1 and VC 8.0, so my question is whether it will produce the correct result, and if so, whether the visualc.hpp should be adjusted for those compiler. Any help would be appreciated. Thanks, Sven

1 0

Serialization "stream error" when using binary_iarchive/binary_oarchive & make_binary_object, but xml_iarchive & text_iarchive suceeds
by Terence Wilson 20 Jan '06

20 Jan '06

I've been scratching my head trying to figure this one out for a few days. My code serializes xml_iarchive & text_iarchive with no problems, however, when I substitute the binary_iarchive I get a "stream error" serialization exception thrown during the make_binary_object call when reading a binary file: template<class Archive> void serialize(Archive & ar, const unsigned int) { ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(CImageDataBase2D); if (Archive::is_loading::value) { Alloc(); } //ar & boost::serialization::make_nvp("ImageData", boost::serialization::make_binary_object(m_pData, sizeof(T)*m_x*m_y)); const int blobSize=sizeof(T)*m_x*m_y; _ASSERTE(blobSize); ar & boost::serialization::make_nvp("ImageData", boost::serialization::make_binary_object(m_pData, blobSize)); } I've tried to excise the relevant code to a console project, however, it has tentacles and will take a few more days. In the mean time I would appreciate help from anyone that might have a suggestion. I've looked at the code and can find nothing wrong, as I said, ASCII and XML serialization works flawlessly I'm using Visual Studio 8 with Boost 1.33.1.

2 2

Getting Environment Variables
by Merrill Cornish 20 Jan '06

20 Jan '06

Is there a Boost way to get environment variables? I _thought_ I had seen a reference to that in the program options libarary, but I can't find it now. Merrill

5 7

Re: [Boost-users] gcc3.2 preprocessor concatenation or not on negative numbers
by Merrill Cornish 20 Jan '06

20 Jan '06

Andy, Thinking about it some more, I think I was wrong. Merrill

1 0

Re: [Boost-users] gcc3.2 preprocessor concatenation or not on negative numbers
by Merrill Cornish 20 Jan '06

20 Jan '06

Andy, I _think_ I may know why concatenation is working like it is. I believe the purpose of the ## concatentation operator is to cause one symbol (i.e., letters, numbers, underscore) to be created out of two symbols. Since the result of the concatenation that you specified is _not_ a legal symbol, the preprocessor appears to have made a best guess at what should be done rather than flag an error. Merrill

2 1

Compiling with egcs-2.91.66
by Emre Birol 20 Jan '06

20 Jan '06

Hi all, My application uses Graph library files of Boost 1.33.1. It is a very very simple shortest path implementation. When I try to compile it with egcs-2.91.66, Red Hat 6.2 (don't ask why), I get this weird error: boost/mpl/aux_/integral_wrapper.hpp:72: sorry, not implemented: conversion of template_type_parm as template parameter In file included from boost/mpl/integral_c.hpp:32, .... ... It uses following headers from boost: #include <boost/config.hpp> #include <boost/graph/graph_traits.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/dijkstra_shortest_paths.hpp> Any idea? How can I fix it? thanks, Emre

2 2

Re: [Boost-users] STL-PORT and VC 7.1
by Nat Goodspeed 20 Jan '06

20 Jan '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sliwa, Przemyslaw (London) > Sent: Friday, January 20, 2006 10:32 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] STL-PORT and VC 7.1 > > I have a quick question - I would like to build the static library and > not the DLL. > Could you tell me what should be changed in the project apart form the > "General->static lib"? [Nat] Um, no. But since .vcproj files are specially-formatted text files, I assume the worst case would involve building a new static-library project by manually setting things the same way in your new project file as in my old one. The IDE may let you change the root type of a project after the fact; I've never tried. > Do you know how to disable the auto-link of boost? [Nat] These are things we changed in boost/boost/config/user.hpp: // BOOST_WHATEVER_DYN_LINK: Forces library "whatever" to be linked as a dll // rather than a static library on Microsoft Windows: replace the WHATEVER // part of the macro name with the name of the library that you want to // dynamically link to, for example use BOOST_DATE_TIME_DYN_LINK or // BOOST_REGEX_DYN_LINK etc (this macro is used to turn on __declspec(dllimport) // modifiers, so that the compiler knows which symbols to look for in a dll // rather than in a static library). // Note that there may be some libraries that can only be statically linked // (Boost.Test for example) and others which may only be dynamically linked // (Boost.Threads for example), in these cases this macro is unsupported. // #define BOOST_WHATEVER_DYN_LINK #define BOOST_FILESYSTEM_DYN_LINK #define BOOST_SIGNALS_DYN_LINK #define BOOST_PYTHON_DYN_LINK // BOOST_ALL_NO_LIB: Tells the config system not to automatically select // which libraries to link against. // Normally if a compiler supports #pragma lib, then the correct library // build variant will be automatically selected and linked against, // simply by the act of including one of that library's headers. // This macro turns that feature off. #define BOOST_ALL_NO_LIB

1 0

Re: [Boost-users] STL-PORT and VC 7.1
by Sliwa, Przemyslaw (London) 20 Jan '06

20 Jan '06

Thanks a lot for this. I have a quick question - I would like to build the static library and not the DLL. Could you tell me what should be changed in the project apart form the "General->static lib"? Do you know how to disable the auto-link of boost? Many thanks for help, Pshemek Kind regards Pshemek Przemyslaw Sliwa Global Foreign Exchange FX Risk Engine Development Merrill Lynch ( +44 (0)20 7995 4182 ) +44 (0)79 6325 2319 * Email: przemyslaw_sliwa(a)ml.com -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Nat Goodspeed Sent: 20 January 2006 14:35 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] STL-PORT and VC 7.1 > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sliwa, Przemyslaw (London) > Sent: Friday, January 20, 2006 8:18 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] STL-PORT and VC 7.1 > > I did not understand your statement properly -> did you create your own > VC 7.1 projects for the three libraries? [Nat] Yes, using the IDE, we made new VC 7.1 projects, which I've attached for you to review & modify to suit your own needs. (There's already a libs/python/build/VisualStudio/boost_python.dsp file, which we loaded into VC 7.1, implicitly converting it to .vcproj.) We created them in libs/filesystem/build libs/signals/build libs/python/build/VisualStudio under the Boost main directory. (It's important to know where we put the .vcproj files because we use .. in the projects' include paths. We do that because the project file is checked into our version control system, and we want people to be able to check out multiple copies in different locations on their own machine and build each of them independently.) The include paths reference $(STLPORTDIR), which you must set in the environment. boost_python.vcproj additionally references $(PYTHONDIR), which you must likewise set. > Is this a standard procedure what people do when something is not > working in boost? [Nat] It's a standard procedure for when bjam produces build errors and you don't even WANT to learn enough about its idiosyncrasies to try to coax it. :-) -------------------------------------------------------- If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/ --------------------------------------------------------

1 0

Re: [Boost-users] STL-PORT and VC 7.1
by Nat Goodspeed 20 Jan '06

20 Jan '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sliwa, Przemyslaw (London) > Sent: Friday, January 20, 2006 8:18 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] STL-PORT and VC 7.1 > > I did not understand your statement properly -> did you create your own > VC 7.1 projects for the three libraries? [Nat] Yes, using the IDE, we made new VC 7.1 projects, which I've attached for you to review & modify to suit your own needs. (There's already a libs/python/build/VisualStudio/boost_python.dsp file, which we loaded into VC 7.1, implicitly converting it to .vcproj.) We created them in libs/filesystem/build libs/signals/build libs/python/build/VisualStudio under the Boost main directory. (It's important to know where we put the .vcproj files because we use .. in the projects' include paths. We do that because the project file is checked into our version control system, and we want people to be able to check out multiple copies in different locations on their own machine and build each of them independently.) The include paths reference $(STLPORTDIR), which you must set in the environment. boost_python.vcproj additionally references $(PYTHONDIR), which you must likewise set. > Is this a standard procedure what people do when something is not > working in boost? [Nat] It's a standard procedure for when bjam produces build errors and you don't even WANT to learn enough about its idiosyncrasies to try to coax it. :-)

1 0