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

uint_t<bits>::value where bits is a template parameter fails to compile
by Phil Endecott 07 Jul '06

07 Jul '06

Dear All, I am having trouble using the integer library. The following fails to compile: #include <boost/integer.hpp> template <int bits> void fn(void) { //typedef boost::uint_t<8>::least t; typedef boost::uint_t<bits>::least t; t var; } int main(int argc, char* argv) { fn<8>(); } With g++ 4.0.0 on Linux and Boost 1.32, I get: boost_int_test.cc: In function â€˜void fn()â€™: boost_int_test.cc:7: error: expected initializer before â€˜tâ€™ boost_int_test.cc:8: error: â€˜tâ€™ was not declared in this scope boost_int_test.cc:8: error: expected `;' before â€˜varâ€™ The problem seems to occur when I use a template parameter - "bits" in this case - as the template parameter to uint_t. If I use a constant (as in the commented-out line) it works. Is this a C++ feature, or a bug somewhere? Can anyone suggest a workaround? Thanks, Phil.

2 1

[spirit] working with end_p
by james.jones＠firstinvestors.com 07 Jul '06

07 Jul '06

From: Neil Hunt <boost(a)thehunts.id.au> > Hello all, > > Given a simple spirit grammar (simplified from my real grammer and > untested) such as: > > file = *line; > line = sensible_line >> ending >> *comment; > sensible_line = confix_p( '\"', *c_escape_ch_p, '\"' ); > comment = !( ch_p(';') >> ( anychar_p - eol_p ) ) >> ending; > ending = *blank_p >> ( eol_p | end_p ); > > > Note: any given line can be legitimately terminated by the end of the > file or end of line > > And there can be arbitrary numbers of comment lines between sensible lines. > > However since 'comment' is terminated with 'ending', the parser ends up > in an infinite loop matching the kleene star on end_p because it returns > zero length (doesn't advance the scanner). At least I think this is the > problem. Indeed. You can fix it by adjusting your grammar thus: line = sensible_line >> (ending >> comment)* >> ending; sensible_line ... as before comment = !( ch_p(';') >> ( anychar_p - eol_p ) ); ending ... as before - James Jones Administrative Data Mgmt. Webmaster 375 Raritan Center Pkwy, Suite A Data Architect Edison, NJ 08837

1 0

[serialization] Facets for writing and reading infinity and nan to text streams
by Johan Råde 07 Jul '06

07 Jul '06

Following a suggestion of Robert Ramey's, I have put together facets for writing and reading non-finite floating point numbers, such as infinity and nan, to text streams, in a consistent and portable way. This is useful when serializing to text archives, but should have other applications as well. I have attached a zip file. It contains three files: 1. the actual code: non_finite_num_facets.hpp 2. documentation: non_finite_num_facets_doc.txt 3. tests: non_finite_num_facets_test.cpp The code handles char and wchar_t. It handles float, double and long double. The code has been tested with VC++ 7.1 but should work with other compilers as well. Any feedback is wellcome. --Johan Råde (rade(a)maths.lth.se)

2 3

std::find_if, boost::mem_fn and boost::not1
by Pete Hodgson 06 Jul '06

06 Jul '06

Hi All, I'm just getting into using boost and stl algorithms, and I'm getting some indecipherable compiler errors that I'd like some help understanding. Here's some sample code: <------------------------SNIP------------------------> #include <boost/shared_ptr.hpp> #include <boost/functional.hpp> #include <boost/bind.hpp> class CEgg { public: bool IsHatched() const; }; typedef std::vector< boost::shared_ptr<CEgg> > tEggsVector; //this function compiles with no problems bool SomeEggsHaveHatched() { tEggsVector basketOfEggs; tEggsVector::const_iterator it = std::find_if( basketOfEggs.begin(), basketOfEggs.end(), boost::mem_fn( &CEgg::IsHatched ) ); bool bSomeEggsHaveHatched = ( it != basketOfEggs.end() ); return bSomeEggsHaveHatched; } //this function doesn't compile, //presumably because of the boost::not1 usage bool AllEggsHaveHatched() { tEggsVector basketOfEggs; tEggsVector::iterator it = std::find_if( basketOfEggs.begin(), basketOfEggs.end(), boost::not1( boost::mem_fn( &CEgg::IsHatched ) ) ); bool bSomeEggsHaventHatched = ( it != basketOfEggs.end() ); return !bSomeEggsHaventHatched; } <------------------------SNIP------------------------> When I try and compile this in Visual C++ 2003 I get the following compile errors for AllEggsHaveHatched(): c:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\include\algorithm(64) : error C2664: 'bool boost::unary_negate<Predicate>::operator ()(const boost::call_traits<T>::param_type) const' : cannot convert parameter 1 from 'std::allocator<_Ty>::value_type' to 'const boost::call_traits<T>::param_type' with [ Predicate=boost::_mfi::cmf0<bool,CEgg>, T=boost::unary_traits<boost::_mfi::cmf0<bool,CEgg>>::argument_type ] and [ _Ty=boost::shared_ptr<CEgg> ] and [ T=boost::unary_traits<boost::_mfi::cmf0<bool,CEgg>>::argument_type ] No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called c:\WORK\4DCapture_Soln\VideoCapture\PixelinkCaptureFacade.h(46) : see reference to function template instantiation '_InIt std::find_if<std::vector<_Ty>::iterator,boost::unary_negate<Predicate>>(_InIt,_InIt,_Pr)' being compiled with [ _InIt=std::vector<boost::shared_ptr<CEgg>>::iterator, _Ty=boost::shared_ptr<CEgg>, Predicate=boost::_mfi::cmf0<bool,CEgg>, _Pr=boost::unary_negate<boost::_mfi::cmf0<bool,CEgg>> ] Any idea why boost::not1 seems to be causing problems? Should I be using std::not1? or boost::mem_fun instead of boost:mem_fn? Cheers and TIA, Pete

3 6

Boost compile fails on Visual C++ 6, warnings on VS8
by Sagnes, Frederic 06 Jul '06

06 Jul '06

Hello, I'm using the Boost libraries for my project (archive, threads, program_options, date_time) and these are really helpful, saved me a lot of time and the great designs make them easy to use. I'm compiling my project with Visual Studio 2005 (8.0) and gcc 4.0.1. Both work quite well, I only have warnings on VS8 because it seems to render integer operation in 64 bit and no cast is being done (a static cast would fix this), some deprecation warning which are normal, and the GCC build works well. But the VC++ 6 compilation is a real pain, here is the command line for BJam: bjam "-sTOOLS=msvc" "-sMSVC_ROOT=C:\Programme\Microsoft Visual Studio\VC98" "-sVISUALC=C:\Programme\Microsoft Visual Studio\VC98" --prefix=D:\ALM\Boost\msvc install I get a lot of compilation errors, including internal compiler errors which are quite alarming. I thought Boost libraries were meant to be portable and compile on VS6. The archive library for instance fails to build. Is there a workaround to build Boost libraries on VS6? Did I miss a switch or setting? Is VS6 that bad that it can't render complex but standard C++ code? Where can I check with which compiler Boost libraries are compatible? Best regards _______________________________________________ Frédéric Sagnes, A&D AS RD DH K3 Siemens AG Karlsruhe

2 1

boost::bind question
by Surya Kiran Gullapalli 06 Jul '06

06 Jul '06

Hi, I've a std::map with the following structure. std::map < int, tuple<std::string, std::string> > mapTuple ; typedef map < int, tuple<string, string> > mapTuple ; Now i've say 10 elements in this container. Now i'd like to iterate thru this container and find a specific element which matches a certain criteria. Right now I'm doing this. MapIntTuple::iterator iter ; for (iter = mymap.begin(); iter != mymap.end(); ++iter) { if (iter->second.get<0>() == current_field) break ; } I'd like to iterate thru the container and find the element whose first string in the tuple matches with my search string. How can i do this using boost::lamdba::bind and std::find_if. May be that will help me in understanding boost::bind better. iter = find_if (mymap.begin(), mymap.end(), ???) ; Thanks in advance, Surya

1 0

[date_time] unused parameter warnings with gcc
by Brian Neal 05 Jul '06

05 Jul '06

Just a suggestion: We like to compile with lots of warning turned on and try to strive for warning free builds. Under gcc, with the -Wall -W options, we get a lot of "warning: unused parameter" warnings with the date_time library. For example, /usr/include/boost-1_33/boost/date_time/time_facet.hpp:202: warning: unused parameter 'a_ref' It looks like a lot of these could be fixed by /* commenting out */ the unused parameter names; e.g. explicit time_facet(::size_t /* a_ref */ = 0) ... Thanks.

1 0

Re: [Boost-users] [boost] [Announce] Boost Installer for Visual C++ 7.1/8.0
by Trent Nelson 05 Jul '06

05 Jul '06

> Boost Consulting is pleased to announce the availability of a Windows > installer for the current Boost release (1.33.1). See: > > http://www.boost-consulting.com/download.html Any plans to make the .nsi file you're using public? There may be a few of us out there that have experience with Nullsoft and could contribute to the installer in some way, shape or form. Trent.

4 4

vectors with multiindex
by Adán Cosgaya 05 Jul '06

05 Jul '06

hi, i am new using the multiindex library, and I need to do the following. I have a set of d-dimensional points (e.g (3,5,8,9,5) is a 5-dimensional point), and I need to sort the points on each of the coordinates. I want to store the points in a vector, and use multiindex to have all the points (vectors in this case) sorted in each of the coordinates. E.g. sorted by x: x y z w ----------- 0 0 0 0 1 2 7 4 2 3 1 1 3 1 3 3 4 5 2 9 sorted by y: 0 0 0 0 3 1 3 3 1 2 7 4 2 3 1 1 4 5 2 9 etc... would you mind pointing me in the right direction with an example of how to do it with multiindex library? thanks

3 3

[mpl] a few questions about set and map
by Yuval Ronen 05 Jul '06

05 Jul '06

I spent some time reading the mpl docs, and I have a few questions about mpl::set and mpl::map. 1. The docs assert there is a 'front' algorithm for set and map. This sounds weird to me, as the order of the types is them unspecified. But if I'm wrong, and 'front' is in place, then where is 'back'? 2. Where's multiset and multimap? In the boost/mpl directory I see something about multiset, but no real boost/mpl/multiset.hpp header, and there's absolutely nothing about multimap. 3. The docs say that the size of joint_view is the arithmetic sum of the sizes of the sequences of which it's composed. Is that true also in the case of set/map? For an mpl newbie it sounds reasonable that it will removed duplicates and might end up with a size that is smaller than the sum of its components. If it's not, then it sounds quite useless to use it for set/map. Thanks, Yuval

4 20

How to find ancestor nodes in a graph, thanks
by sean yang 05 Jul '06

05 Jul '06

Hi folks, I want to find all the nodes that can reach a specific node 'target_n'. Let's say, an example graph looks like this: G= (V, E), where V={A, B, C, target_1, target_2, D, E, F}, and E= { (A-->B), (B-->target_1), (C-->target_2), (A-->target_2), (D-->E), (E-->F), (F-->D), } In this example, I want to find out(say print them out) 'A', 'B', 'C'. Now, I have a graph representation like the following one . ------------------------------ typedef adjacency_list<vecS, vecS, bidirectionalS, VertexNameProperty > Graph; Graph travel; --------------------------------- And I have already read in the above information to the Graph travel. One way I am thinking of to achieve the above goal is: reverse the direction of edges in G, i.e., getting a G', first; and then do BFS from target_n nodes. My question is: (1) is there an easy (some API, sorry I am not familar with the library) way to reverse edge direction? (2) is there an easy way to add a 'printf()' during BFS? Thanks, Sean _________________________________________________________________ FREE pop-up blocking with the new MSN Toolbar get it now! http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/

2 1

[serialization] infinite floating point values
by Johan Råde 05 Jul '06

05 Jul '06

I just replaced the binary archives by text archives in an app, to make sure that the archives will be portable between 32 and 64 bit platforms. But when I use text archives, the app crashes. The problem is: Text archives do not handle infinite values of type float and double. --- Example ---------------------------------------------------------- #include <iostream> #include <limits> #include <sstream> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/format.hpp> int main() { std::stringstream ss; { boost::archive::text_oarchive oa(ss); double inf1 = std::numeric_limits<double>::infinity(); std::cout << boost::format("Save: %1%\n") % inf1; oa << inf1; } { boost::archive::text_iarchive ia(ss); double inf2; ia >> inf2; std::cout << boost::format("Load: %1%\n") % inf2; } } --- Output ---------------------------------------------------------- Save: 1.#INF Load: 1 --------------------------------------------------------------------- There are two infinite values of type double, std::numeric_limits<double>::infinity() and -std::numeric_limits<double>::infinity(). These could be represented as "inf" and "-inf" in the archive. I would be very useful for me if this functionality were added to the serialization library. (Otherwise I will have to do a lot of code rewriting, or stick binary archives.) One could also consider handling nan (not a number), but that is maybe a lower priority. --Johan Råde rade(a)maths.lth.se

2 4

[spirit] working with end_p
by Neil Hunt 05 Jul '06

05 Jul '06

Hello all, Given a simple spirit grammar (simplified from my real grammer and untested) such as: file = *line; line = sensible_line >> ending >> *comment; sensible_line = confix_p( '\"', *c_escape_ch_p, '\"' ); comment = !( ch_p(';') >> ( anychar_p - eol_p ) ) >> ending; ending = *blank_p >> ( eol_p | end_p ); Note: any given line can be legitimately terminated by the end of the file or end of line And there can be arbitrary numbers of comment lines between sensible lines. However since 'comment' is terminated with 'ending', the parser ends up in an infinite loop matching the kleene star on end_p because it returns zero length (doesn't advance the scanner). At least I think this is the problem. By using: ending = *blank_p >> eol_p ; the parser runs to completion, but any given line must be terminated by an end of line sequence. Any suggestions on how to fix this problem? Thanks for any help offered, Neil Hunt

2 1

error using lambdas: conversion from 'x' to non-scalar type 'y'
by klaus triendl 05 Jul '06

05 Jul '06

Hi, I'm rather new to lambdas. I wanted to use them to stream objects into a special stream (not a stl stream). the following code outlines the situation: ---- #include <iostream> #include <list> #include <algorithm> #include <boost/ref.hpp> #include <boost/shared_ptr.hpp> #include <boost/lambda/lambda.hpp> using namespace std; using namespace boost; using boost::lambda::_1; class MyStream { public: MyStream() {} }; class MySubStream: public MyStream { public: MySubStream(MyStream*) {} }; MyStream& operator <<(MyStream& os, const int&) { return os; } int main() { list<shared_ptr<int> > cont; MyStream osTop; MySubStream osSub(&osTop); for_each(cont.begin(), cont.end(), ref(osSub) << *_1); return 0; } ---- g++ outputs error: '/usr/include/boost/lambda/detail/operator_lambda_func_base.hpp:212: error: conversion from 'MyStream' to non-scalar type 'MySubStream' requested'. MS VC 2003.net says: boost\lambda\detail\operator_lambda_func_base.hpp(212) : error C2664: 'MySubStream::MySubStream(MyStream *)' : cannot convert parameter 1 from 'MyStream' to 'MyStream *'. The question is whether my code is correct at all and how I could use lambdas the right way. Thanks! Klaus Triendl

2 4

Re: [Boost-users] shared_ptr + const_cast woes
by Evan Drumwright 04 Jul '06

04 Jul '06

Hi guys, Thanks for your help, in particular Stefan's reworked code fragment. I actually did read the documentation for shared_ptr (in particular, the section on const_pointer_cast), but got confused by the counterexample (I thought that it was saying not to use a temporary). Thanks again, Evan >Message: 5 >Date: Wed, 28 Jun 2006 10:33:19 -0400 >From: me22 <me22.ca(a)gmail.com> >Subject: Re: [Boost-users] shared ptr + const_cast woes >To: boost-users(a)lists.boost.org >Message-ID: <fa28b9250606280733u5125fa01p92da87e1e8f9ad0f(a)mail.gmail.com> >Content-Type: text/plain; charset=UTF-8; format=flowed >Take a closer look at what this is doing: >> boost::shared_ptr<const int> y(const_cast<const >int*>(x.front().get())); >To be more explicit, it's the same as > int *ip = x.front().get(); > int const *cip = ip; > boost::shared_ptr<const int> y( cip ); >At y you're creasing another shared_ptr to the same address that >doesn't share ownership with the one in x.front(). > >What you really want is const_pointer_cast : >http://boost.org/libs/smart_ptr/shared_ptr.htm#const_pointer_cast > >From the documentation for the above-mentioned const_pointer_cast: >Notes: the seemingly equivalent expression >shared_ptr<T>(const_cast<T*>(r.get())) >will eventually result in undefined behavior, attempting to delete the same >object twice. >Which is exactly your problem. > >~ Scott McMurray ------------------------------ >Message: 6 >Date: Wed, 28 Jun 2006 11:59:12 -0230 >From: Stefan Tarrant <stefan.tarrant(a)c-core.ca> >Subject: Re: [Boost-users] shared ptr + const_cast woes >To: boost-users(a)lists.boost.org >Message-ID: <44A29238.9040800(a)c-core.ca> >Content-Type: text/plain; charset=ISO-8859-1; format=flowed >When you use get() all bets are off. You are basically taking >responsibility for the ownership of the returned pointer yourself, which is >generally a bad idea. Use a const_pointer_cast<T> instead. >Here is my re-worked example: > >#include <vector> >#include <iostream> >#include <boost/shared_ptr.hpp> > >int main() >{ > std::vector<boost::shared_ptr<int> > x; > x.push_back(boost::shared_ptr<int>(new int)); > *x.front() = 4; > std::cout << "use count: " << x.front().use_count() << std::endl; > boost::shared_ptr<const int> y = x.front(); > std::cout << "use count: " << x.front().use_count() << std::endl; > std::cout << "use count: " << y.use_count() << std::endl; > boost::shared_ptr<int> z = boost::const_pointer_cast<int>(y); > std::cout << "use count: " << z.use_count() << std::endl; > std::cout << "use count: " << y.use_count() << std::endl; > x.pop_back(); > std::cout << *y << std::endl; > std::cout << "use count: " << z.use_count() << std::endl; > std::cout << "use count: " << y.use_count() << std::endl; > > y.reset(); > std::cout << "use count: " << z.use_count() << std::endl; > std::cout << "use count: " << y.use_count() << std::endl; >} > >And the output: > >use count: 1 >use count: 2 >use count: 2 >use count: 3 >use count: 3 >4 >use count: 2 >use count: 2 >use count: 1 >use count: 0 >Press any key to continue . . . > >-- >Stefan Tarrant >Senior Project Engineer >C-CORE >Captain Robert A. Bartlett Building >Morrissey Road >St. John's, NL >Canada A1B 3X5

1 0

boost pool
by michelle Tchicaya 04 Jul '06

04 Jul '06

Hello, I am very new to C++. And I want to use Boost thread and Boost Pool to create a multithreaded application which will handle real time messages. I need ONE Boost Pool. But I dont know how to combine it with thread Thank a lot for your help mimi

1 0

Re: [Boost-users] [boost] Convert function pointer type to function type
by Peter Soetens 04 Jul '06

04 Jul '06

Quoting Peter Dimov: > Peter Soetens wrote: >> Hi, >> >> I'm desperately seeking for doing this conversion using the boost >> libraries. User code often provides a function pointer, while library >> code often works with function types. [...] > > One obvious way is > > template<class F> boost::function<F> function_wrapper( F * pf ) > { > return boost::function<F>( pf ); > } > > but a more interesting question is why do you need a function_wrapper at > all. Everything that you could do with the returned function<F> (call it) is > also possible with pf itself. Looking at the simplicity of that answer, this was clearly a boost-users question :-]. Sorry. This was a minimalistic example, removing the context. The big picture uses boost::function<F> to store partially bound objects, since this is a way to store the result of boost::bind for later use in a not templated class. I got confused by the boost::function_traits<T>, which only takes a "function type" and not a "function pointer type." I assumed one was not convertible to the other. I tried using boost::remove_pointer<F>, which didn't work and confirmed back then my assumption (confusingly, is_pointer<F> did work, but I could not remove it using type_traits...) My books write about function pointer types but do not mention 'function types'. These concepts appear to me underdocumented in standard C++ works. *Thanks a lot*. Peter -- www.fmtc.be ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

1 0

Re: [Boost-users] [BGL] All Possible Matchings (Hossein Haeri)
by Grégoire Dooms 04 Jul '06

04 Jul '06

boost-users-request(a)lists.boost.org wrote: Message: 9 Date: Mon, 3 Jul 2006 16:46:55 +0100 (BST) From: Hossein Haeri <powerprogman(a)yahoo.com> Subject: [Boost-users] [BGL] All Possible Matchings To: boost-users(a)lists.boost.org Message-ID: <20060703154655.35182.qmail(a)web53405.mail.yahoo.com> Content-Type: text/plain; charset=iso-8859-1 > Dear all, > > I have a bipartite graph, and want to generate all its > possible matchings. Any suggestions? > You should have a look at Constraint Programming and the Alldifferent constraint. In CP, the task of filtering for a constraint over some integer variables amounts to partitioning, for each of its variables, the potential values into three categories: the values participating in all solutions (obviously at most one per variable), the values participating in some solutions, and the values participating in no solutions. In the case of the Alldifferent constraint, it can be reinterpreted in graph matching terms: Build a bipartite variable-value graph. A solution to the Alldifferent constraint is a maximum matching covering all variables in this graph. Hence filtering can be done by identifying edges which belong to some/all/no max matching. In a nutshell, the following property of matchings is used [Peterson] [Berge] : M1 (+) M2 = P M1 (+) P = M2 where P is an alternating path (for instance, look at the LEDA book for a good exposition of this property and matching algorithms). This property is used in matching algorithms with P an augmenting alternating path. If you consider "maintaining" paths (eg. non-augmenting and non-decreasing -- even length, starting at a free node and ending at a covered node, or an alternating cycle) and apply them on a max matching you get another max matching. Here are a few references: Ref. to the original paper : http://citeseer.ist.psu.edu/context/210815/0 A survey on this subject: http://citeseer.ist.psu.edu/579391.html (or http://www.cs.cornell.edu/~vanhoeve/papers/alldiff.pdf for a more recent/in progress version) Slides by JC Regin: http://www.constraint-programming.com/people/regin/papers/GraphAndCP.pdf Finally, an open-source implementation: http://www.gecode.org/gecode-doc-latest/distinct_2dom_8icc-source.html Hope this helps, -- Grégoire Dooms.

1 0

When to use thread.join()
by Simon Reye 04 Jul '06

04 Jul '06

My apologies if this has been covered. I am new to boost and have been working with the boost thread class. I assume that the join() function waits for the thread to finish execution. But what if the thread has already finished executing? Can I still call join() and have it not wait for infinity? Simon

2 2

[Spirit] puts namespace std into the global scope
by Johan Råde 04 Jul '06

04 Jul '06

It seems that the header boost/spirit.hpp puts the entire namespace std into the global scope. That is not very nice! --- Example 1 -------- #include <string> #include <sstream> #include <fstream> #include <vector> #include <boost/spirit.hpp> void f() { string s; istringstream iss; ofstream ofs; vector<int> v; } ---------------------- Example 1 should not compile (undefined symbols), but does compile. If the line #include <boost/spirit.hpp> is removed, then Example 1 does not compile. --- Example 2 -------- #include <boost/filesystem/path.hpp> #include <boost/filesystem/fstream.hpp> #include <boost/spirit.hpp> using boost::filesystem::path; using boost::filesystem::ifstream; void g() { path p; ifstream ifs(p); } ---------------------- Example 2 should compile but does not. The compiler complains that the symbol ifstream is ambigous. If the line #include <boost/spirit.hpp> is removed, then Example 2 does compile. I´m using Boost 1.33.1 and MSVC 7.1 on a Windows XP computer. --Johan Råde rade(a)maths.lth.se

3 3

auto-diff expression templates-> MPL
by Russell L. Carter 03 Jul '06

03 Jul '06

Greetings, This is a beginner question about the MPL. I'm building a general purpose simulation program which will have an optimization module as well as an internal ODE solver which requires derivatives of the implemented algorithm in various forms. Around 1998-2002 there was a lot of enthusiasm for generating the derivatives automatically using expression templates. I'm not quite through with an exhaustive review of the literature but it appears the various efforts ran to ground on the compilation memory and cpu requirements that evaluating expression templates in their original form required. So 1st question, has anybody done something like this with mpl/fusion sequences and algorithms? Has there been an improvement in compile time performance? Do you have any pointers to trivial examples? I've read Alexandrescu, and Abrahams & Gurtovoy and it's not clear to me how to map the expression templates as described in the earlier efforts onto mpl sequences and algorithms. That's probably enough for now. Thanks for any help. Russell

1 0

[shmem] and kernel 2.4 / linuxthreads
by Sascha Lumma 03 Jul '06

03 Jul '06

Hi all, I noticed that kernel 2.6 is a requirement for shmem on linux. Is it possible to get shmem to run - maybe with constraints - together with kernel 2.4 and the older pthread version (linuxthreads)? A start on a kernel 2.4 box produces a segmentation fault in shared_memory.hpp. To reach most users as possible, it would be nice to be compatible or if not possible to get as a minimum a normal termination. Sascha Lumma

2 2

[BGL] python bindings build graph from edge list
by John Reid 03 Jul '06

03 Jul '06

Hi, I was trying to build a graph from an edge list and ran into some problems. The following code throws an exception on write_graphviz(). I think it is because there's no property map called node_id in the graph. import boost.graph as bgl edges = [ ( 'a', 'b' ), ( 'b', 'c' ), ( 'c', 'a' ) ] g = bgl.Graph( edges ) g.write_graphviz( 'bind-copy.dot' ) I get the following: Traceback (most recent call last): File "<stdin>", line 1, in ? KeyError: 'node_id' I couldn't manage to generate the node_id property map automatically. So after you have built the graph with the edge list constructor how do you know which vertex is which? I ended up using the following code to create the graph: def build( edges ): # Create a graph to build g = bgl.Graph() # Construct some property maps node_ids = g.vertex_properties[ 'node_id' ] = g.vertex_property_map( 'string' ) color = g.vertex_properties[ 'color' ] = g.vertex_property_map( 'string' ) centrality = g.vertex_properties[ 'centrality' ] = g.vertex_property_map( 'float' ) weight = g.edge_properties[ 'weight' ] = g.edge_property_map( 'float' ) length = g.edge_properties[ 'len' ] = g.edge_property_map( 'float' ) # create the edges and vertices name_2_vertex = { } for ( name1, name2 ) in edges: if not name_2_vertex.has_key( name1 ): v1 = name_2_vertex[ name1 ] = g.add_vertex() node_ids[ v1 ] = name1 else: v1 = name_2_vertex[ name1 ] if not name_2_vertex.has_key( name2 ): v2 = name_2_vertex[ name2 ] = g.add_vertex() node_ids[ v2 ] = name2 else: v2 = name_2_vertex[ name2 ] e = g.add_edge( v1, v2 ) weight[ e ] = 0.1 length[ e ] = 4 # Color articulation points red for v in bgl.biconnected_components( g ): color[ v ] = 'red' # Assign centrality values to property map bgl.brandes_betweenness_centrality( g, centrality ) return g This code seems to works fine. So I thought I'd post this just to see if I was missing something obvious and/or in the hope someone else might find it useful. Best, John.

3 6

[BGL] All Possible Matchings
by Hossein Haeri 03 Jul '06

03 Jul '06

Dear all, I have a bipartite graph, and want to generate all its possible matchings. Any suggestions? TIA, --Hossein ___________________________________________________________ Yahoo! Messenger - with free PC-PC calling and photo sharing. http://uk.messenger.yahoo.com

1 0

[BGL] Bipartite Graphs
by Hossein Haeri 03 Jul '06

03 Jul '06

Dear all, Any special facility there is in BGL for bipartite graphs? TIA, --Hossein ___________________________________________________________ All New Yahoo! Mail Tired of Vi@gr@! come-ons? Let our SpamGuard protect you. http://uk.docs.yahoo.com/nowyoucan.html

2 2

[BGL]Customising the Adjacency List Storage
by Louis Lavery 03 Jul '06

03 Jul '06

According to the book [BGL p232], in order to customise the EdgeList I need to define a container generator and parallel edge traits class, and over load the push() and erase() functions for the custom container [BGL p234]. My custom container is user::PushBackList<T>... namespace user { template<typename ValueType> struct PushBackList { //... }; struct PushBackListS { }; } ...and I've put the (specialised) container_gen and parallel_edge_traits classes in namespace boost... namespace boost { template<typename ValueType> struct container_gen<user::PushBackListS,ValueType> { typedef user::PushBackList<ValueType> type; }; template<> struct parallel_edge_traits<user::PushBackListS> { typedef allow_parallel_edge_tag type; }; } ...now all I need do is overload the push() and erase() functions. The problem is they are in nested namespace graph_detail and are called with full qualification (i.e. boost::graph_detail::push(...)) from adjacency_list class. That means I need to put my overloads in boost::graph_detail. Is that right? Thanks, Louis.

2 1

[BGL] Budled Properties
by Hossein Haeri 03 Jul '06

03 Jul '06

Dear BGL-ers, I'm so new to BGL, and a little mixed up about the properties. Is the "Six Degrees of Kevin Bacon" example in the documentation of 1.33.0 using Bundled Properties? Or, is it using that old style "using_adjacency_list.html" strongly recommended me to leave it? TIA, --Hossein ___________________________________________________________ All New Yahoo! Mail Tired of Vi@gr@! come-ons? Let our SpamGuard protect you. http://uk.docs.yahoo.com/nowyoucan.html

2 1

Re: [Boost-users] Retrieve multindex value type?
by JOAQUIN LOPEZ MU?Z 03 Jul '06

03 Jul '06

----- Mensaje original ----- De: Francesco Biscani <bluescarni(a)gmail.com> Fecha: Domingo, Julio 2, 2006 6:19 pm Asunto: [Boost-users] Retrieve multindex value type? > Hello, > > is there a way to retrieve from a multiIndex container the type of > its content? > > I need to do something like this: > > template <class T> > class my_class > { > private: > T set; > > public: > void insert(const T::value_type &); <------ ? > }; > > Here T is a multiIndex container, and the insert function should > add a term to the set. Hence "insert" must accept as input the > kind of values that are stored in the multiIndex container. Is > there a way to do so or should I just add another parameter to > the template and use that as argument for insert? > Thanks, Hello Francesco, Indeed, multi_index_container provides a nested value_type typedef that you can use exactly in the way you describe. Have you encountered any problem in doing so? Thank you for using Boost.MultiIndex, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 2

boost 64bit support on windows x64 on amd64 hardware
by Chu-Ming Ng 03 Jul '06

03 Jul '06

Hi, I am porting some code that uses boost from 32-bit to the 64-bit platform and would like to know if the boost library supports this platforms. thanks! Chu-Ming

7 7

Retrieve multindex value type?
by Francesco Biscani 02 Jul '06

02 Jul '06

Hello, is there a way to retrieve from a multiIndex container the type of its content? I need to do something like this: template <class T> class my_class { private: T set; public: void insert(const T::value_type &); <------ ? }; Here T is a multiIndex container, and the insert function should add a term to the set. Hence "insert" must accept as input the kind of values that are stored in the multiIndex container. Is there a way to do so or should I just add another parameter to the template and use that as argument for insert? Thanks, Francesco

1 0

Locking smart pointer
by Mattias Brändström 02 Jul '06

02 Jul '06

Hello! I have a small class to show you and perhaps get some comments on: <code> #include <boost/shared_ptr.hpp> #include <boost/thread.hpp> template <typename T> class LockPointer { public: class LockProxy { public: LockProxy(boost::shared_ptr<T> p, boost::shared_ptr<boost::recursive_mutex> mutex) : p_(p), lock_(new boost::recursive_mutex::scoped_lock(*mutex)) { } boost::shared_ptr<T> operator->() { return p_; } private: boost::shared_ptr<T> p_; boost::shared_ptr<boost::recursive_mutex::scoped_lock> lock_; }; LockPointer(boost::shared_ptr<T> o) : p_(o), mutex_(new boost::recursive_mutex()) { } LockProxy operator->() { return LockProxy(p_, mutex_); } private: boost::shared_ptr<T> p_; boost::shared_ptr<boost::recursive_mutex> mutex_; }; </code> The purpose of this class is to lock a mutex before each method call on an object of class T, and unlock the mutex when that call returns. It seems that this would be something that one would like to have in some cases. Maybe so many cases that it might be part of boost somewhere? I haven't found any library in boost that does this. Is there one? I haven't done anything like this before so it would be nice if anyone here could make some comments on the code. Will it do what I think it does or will it blow up in my face? :-) Regards, Mattias

3 2

[serialization] Out of memory with Borland C++ Builder 60
by Federico Feller 02 Jul '06

02 Jul '06

Hi everyone. Im having problems compiling some code with Borland C++ Builder 6 that includes libboost serialization. In particular Im having troubles registrating a new class in order to be serialized. The new class, say A3, is a subclass of antother that is declared abstract, say A. A has already some subclases (A1, A2) that are serialized correctly. As I did for other subclasses on the same hierarchy, I added the instruction ar.register_type(static_cast<A3 *>(NULL)); in the load/save method of another class that serializes a pointer to the base class of the hierarchy. After doing that, I tried to compile the code and got the following error from the compiler: [C++ Fatal Error] if.hpp(72): F1008 Out of memory I really have no idea whats going on. Any ideas? Thanks Federico __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

4 7

Re: [Boost-users] Boost.Variant: char[], a bool or std::string ?
by Nindi Singh 01 Jul '06

01 Jul '06

This has the desired effects BUT ... I am not a software engineer .. so I would like to know is it would the proffessionals would consider ugly code ? #include<string> #include<boost/variant.hpp> typedef boost::variant< bool,std::string> BoostVariant; struct MyVariant:public BoostVariant { template<typename type> MyVariant(const type &theType):BoostVariant(theType){} MyVariant(const char *theCString):BoostVariant(std::string(theCString)){} }; int main () { MyVariant v("A String"); std::string &Sref(boost::get<std::string&>(v)); // NOw this does NOT throw bool &Bref(boost::get<bool&>(v)); // but this does return 0; }

2 1

Re: [Boost-users] Boost.Variant: Is a char * really a bool ?
by Nindi Singh 01 Jul '06

01 Jul '06

Hi Yuval sorry .. i will ensure I put a RE: on my furtue replies ... but i have posted a reply with subject 'Boost.Variant: char[], a bool or std::string ? ' N

1 0

Boost.Variant: char[], a bool or std::string ?
by Nindi Singh 01 Jul '06

01 Jul '06

I sort of guessed that this implicit conversion was taking place ... but unfortunatley it wil make my variant very error prone .. and on the other hand I do want the capability to construct from a C string appropriatley. I think one possible soultion would be to derive from boost::variant<bool,std::string> and put in an explicit constructor for const char *, this will also let me get around other similar problems. However deriving from boost::variant .... ??? not sure how healthy that is. >To answer the question in the subject line: char* isn't bool, but it's >implicitly convertible to bool, which is probably (haven't really >checked, just an educated guess) what's causing you trouble. > >Nindi wrote: >> When I have a Boost.Variant templated on a bool and a std::string, I get >> the following behaviour .. >> >> #include<string> >> >> #include<boost/variant.hpp> .> >> typedef boost::variant< bool,std::string> BoostVariant; >> >> int main () { >> >> BoostVariant v("A String"); >> >> bool &Bref(boost::get<bool&>(v)); // this is ok >> >> std::string &Sref(boost::get<std::string&>(v)); // But this throws >> >> return 0; >> >> } > >Instead of > BoostVariant v("A String"); >try > BoostVariant v(std::string("A String")); > >I think it will solve your problems.

1 0

Boost.Variant: Is a char * really a bool ?
by Nindi 01 Jul '06

01 Jul '06

When I have a Boost.Variant templated on a bool and a std::string, I get the following behaviour .. #include<string> #include<boost/variant.hpp> typedef boost::variant< bool,std::string> BoostVariant; int main () { BoostVariant v("A String"); bool &Bref(boost::get<bool&>(v)); // this is ok std::string &Sref(boost::get<std::string&>(v)); // But this throws return 0; }

2 1

Boost.Variant: char[], a bool or std::string ?
by Nindi Singh 01 Jul '06

01 Jul '06

... .. but this works fine .. #include<string> #include<boost/variant.hpp> typedef boost::variant<std::string> BoostVariant; int main () { BoostVariant v("A String"); std::string &Sref(boost::get<std::string&>(v)); // NOW this does NOT throw return 0; }

1 0

Boost.Variant: char[], a bool or std::string ?
by Nindi Singh 01 Jul '06

01 Jul '06

If I instantiate a Boost.Variant templated on a bool and a std::string , then I get the following behaviour ... not sure why #include <string> #include <Boost/Variant.hpp> typedef boost::variant< bool,std::string> BoostVariant; int main () { BoostVariant v("A String"); bool &Bref ( boost::get<bool &>(v)) ; // this is ok std::string &Sref ( boost::get<std::string> &(v) ) ; // but this throws return 0; } Is this what is supposed to happen ? Ninds

1 0

[serialization] Error serializing std::wstring to text_oarchive on VC++
by William Kemper 01 Jul '06

01 Jul '06

The fill method on mb_from_wchar calls std::wctomb, which is backed by 'WideCharToMultiByte'. This function uses the codepage defined in the current locale, however, the utf-8 codepage does not function properly on VC++ 7 or 8. Would it be difficult to have the conversion use the code_cvt facet defined for the stream underlying the archive? It also seems that doing so would ensure serialization / deserialization would be more consistent and self-sufficient, since neither process would depend on a (mutable) global locale setting. At any rate, the serialization library works with certain window's codepages, but *not* with the utf-8 codepage (65001). It would be nice to have the library use its own utf8_code_cvt facet to translate wide characters to multibyte. If there is some work around I'm not aware of, I would appreciate any advice. Thanks.

2 1

Compile-time constant CRCs
by Michael Marcin 01 Jul '06

01 Jul '06

I didn't tag this because I'm not sure if it should belong to [preprocessor], [mpl] or [crc]. Given the black-box (macro, function, class, whatever) I'll call ConstCRC32 that is assignable to a unsigned int is there a way to write: unsigned int hash = ConstCRC32( "HELLO" ); and have an optimized build yield the equivalent of: unsigned int hash = 0x58a10fc0; I had a solution (although obviously non-portable) in msvc7.1 but after moving to msvc8 and stepping through the code I noticed that msvc8's "improved" optimizing compiler cannot optimize away the calculation like msvc7.1's could. I also doubt that the code would work if I ever had to port it to another compiler. I tried using simple TMP techniques I've used in the past but couldn't find a way to access the characters of the string. So looking around I came upon boost.crc which is geared towards computing crcs of blocks of data which isn't really what I care about right now. I also found boost.mpl which is very cool but I have a hard time grasping still. The problems seems solvable with mpl if I could transform the data into boost::mpl::list_c<char,'H','E','L','L','O'> but I still coulnd't find a way to access the characters of the string and I think others at my company would rather revert to the old #define CRC( str, crc ) crc unsigned int hash = CRC( "HELLO", 0x58a10fc0 ); than move to that syntax directly. So looking further I found boost.preprocessor which I haven't explored enough to see if there is away to have PP calculate the CRC directly or have it transform the string into the boost::mpl::list_c form. Can anyone offer some advice? Thanks Mike P.S. The old version was actually fairly simple it just used a precomputed static const unsigned int lookup table and did a strlen and a for loop with a lookup into the table for each character (in a __forceinline function that took a const char * const string).

1 0

Replacing C functions with boost::function for callback functions in OpenGL
by Stephen Torri 01 Jul '06

01 Jul '06

I have a main file from my OpenGL program which contains global variables and the functions that are used by OpenGL as callback functions. For example: int a; float b; void display () { .... display code calls ... } int main (int argc, char **argv) { glutDisplayFunc( display ); } Now I thought I would put all the variables and functions into a class and use that in the main. Here is where I am getting stuck on the use of boost::function. I changed the code to look something like the following: class MyMain { int a; float b; void display() { ... display code calls ... } } int main (int argc, char **argv) { MyMain m_ref; boost::function<void()> disp_func = boost::bind ( &MyMain::display, &m_ref ); glutDisplayFunc ( disp_func ); } What I am getting back from the g++ compiler is the following error message: error: cannot convert `boost::function<void ()(), std::allocator<void> >' to `void (*)()' for argument `1' to `void glutDisplayFunc(void (*)())' Any ideas where I am going wrong? Stephen

4 6

[regex] Support for named captures?
by Olenhouse, Jason 30 Jun '06

30 Jun '06

I've read the boost regex docs and I don't see anything about named captures. I see information about captures using the number notation, $1, $2, etc. I'm assuming that the regex lib doesn't support named captures, correct? I looked at the spirit docs and didn't see any mention of it in the regex parser portions. I'm guessing that spirit doesn't support it either, correct? And by not supporting it I mean an out-of-the-box solution with the regex parser. I'm sure something more complicated than I could understand is possible with spirit. I can post to that mailing list if no one here knows. I'm trying to regex syslog messages for specific pieces of data. Depending on what systems send me syslog messages the order of the pieces of data may be different. I'd like a solution where I can ask for the specific piece of data explicitly, instead of having to remember that on system A dataItem1 is $1, but on system B it's $2. I will always have the same fields that I look for and that won't change. I'm trying to not complicate my database schema or the make the client's use any more difficult since they're the ones that write the regex statement. If named captures with the regex lib or with spirit aren't supported, any ideas on how this might be done would be appreciated. I'm stuck on boost 1.32, but have been looking for a good excuse to force our development dept. to a newer release. -Jason

2 1

Re: [Boost-users] Replacing C functions with boost::function for callback functions in OpenGL
by Baranowski, Daniel 30 Jun '06

30 Jun '06

[snip] > Now I thought I would put all the variables and functions into a class > and use that in the main. Here is where I am getting stuck on the use of > boost::function. I changed the code to look something like the > following: > > class MyMain { > int a; > float b; > > void display() > { > ... display code calls ... > } > } > > int main (int argc, char **argv) > { > MyMain m_ref; > boost::function<void()> disp_func = boost::bind ( > &MyMain::display, &m_ref ); > glutDisplayFunc ( disp_func ); > } [snip] Maybe you could leave the display function as it was before, have OpenGL call that function, and then have Display call disp_func(). Something like this (though I haven't actually tried it): typedef boost::function<void()> DisplayFuncType; DisplayFuncType CurrentDisplayObject; class MyMain { int a; float b; void display() { ... display code calls ... } } void display () { if(!CurrentDisplayObject.empty()) CurrentDisplayObject(); } int main (int argc, char **argv) { MyMain m_ref; CurrentDisplayObject = boost::bind ( &MyMain::display, &m_ref ); glutDisplayFunc (display); } Let me know if that works out for you! I'd be interested to know. I think that's a good idea, BTW. Daniel ********************************************************************************************** Disclaimer - This email and any files transmitted with it are proprietary and may contain privileged or copyright information. You must not present this message to another party without gaining permission from the sender. If you are not the intended recipient you must not copy, distribute or use this email or the information contained in it for any purpose other than to notify us. If you have received this message in error, please notify the sender immediately, and delete this email from your system. We do not guarantee that this material is free from viruses or any other defects although due care has been taken to minimize the risk. eSafe scanned this email for viruses, vandals and malicious contentAny views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of LSI. **********************************************************************************************

1 0

[asio] has anybody enabled SSL in a large project?
by Scott 30 Jun '06

30 Jun '06

We've been using ASIO for a while now in a large project and it's been great. I've recently been tasked to add optional SSL support between our client and server. Since there are some SSL classes in ASIO I started with the ASIO SSL example code and got that working. I then pretty much copied the relevant pieces from the sample code over to our project to see if it would work there. Using the same certificates and key files from the sample code, I added the handshaking and got our client and server to communicate. It all seemed great at first, because it seemed to seemlessly work. But eventually the communications break down. I either get asserts in my code from unexpected data in the stream, or the client hangs waiting for data that never appears or any of several different stream problems. It's rarely the exact same thing. But it eventually always fails. Especially on long sustained transfers. I have no idea what I might be doing wrong. There's really no documentation at all on ASIO and SSL. For all I know, it might be a problem with the SSL implementation in ASIO. Which brings me to my question - has anybody successfully used the SSL classes in ASIO on complex projects? If so, that would point to a problem in my code somewhere. Perhaps I need to do more than just add the initial SSL handshake (aside from using the SSL stream beyond that point)? Thanks, Scott

1 0

Re: [Boost-users] multi_array: creating a view to a single element
by Baranowski, Daniel 30 Jun '06

30 Jun '06

[snip] > [snip] > > Basically, I have a 2 Dimensional array and I > > would like to be able to access the center of that array > > using a variable that maps to some index (for example, > > element[3][3] in a 7x7 array). > [snip] > > Would T & center = element[3][3] be satisfactory? > > I don't think multiarray supports any type of reallocation after > construction, so it should be ok. > > Sohail Thanks Sohail. While you can resize a multi_array, I won't be, so it's not an issue. What you've got is pretty obvious, I guess I was just had gone too far down the "view" path to see the simple answer! Thanks also to Cromwell for chiming in too as that is definitely a good, more generic approach, but probably overkill for my particular instance. But, still good to know! Daniel ********************************************************************************************** Disclaimer - This email and any files transmitted with it are proprietary and may contain privileged or copyright information. You must not present this message to another party without gaining permission from the sender. If you are not the intended recipient you must not copy, distribute or use this email or the information contained in it for any purpose other than to notify us. If you have received this message in error, please notify the sender immediately, and delete this email from your system. We do not guarantee that this material is free from viruses or any other defects although due care has been taken to minimize the risk. eSafe scanned this email for viruses, vandals and malicious contentAny views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of LSI. **********************************************************************************************

1 0

[serialization] customize xml archive
by Jabe 30 Jun '06

30 Jun '06

hello, i try to create a xml archive that fit our needs. the default xml serializer handles not the format i would like to implement. some differences are: - primitive types (int, float, ...) should be written as xml attributes - array types should not provide any parent xml element, all array elements will be serialized as childs from the element containing the array - tagnames should always contain the classname - parent classes should create a new xml child element - and some more, but so far as an example the default xml code looks like this: <foo class_name="Class1" class_id="1"> <name>example</name> <container class_id="2"> <count>2</count> <item class_id_reference="2"> <a>123</a> </item> <item class_id_reference="3" class_name="Class3"> <b>456</b> </item> </foo> i would like this output: <Class1 role="foo" name="example"> <Class2 role="container" a="123" /> <Class3 role="container" b="456" /> </Class1> is that possible with the current boost::serialization and boost::archive libraries? i tried to derive all baseclasses of xml_oarchive and reimplemented it, without success. maybe a first question: why don't i get a classname for each serialized object in common_oarchive::save_override(const class_name_type & t, int)? sometimes i get a class_id, sometimes a class_reference_id and so on. classnames are offered rarely. regards, jabe

4 12

Re: [Boost-users] multi_array: creating a view to a single element
by Sohail Somani 29 Jun '06

29 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Cromwell Enage > > > Would T & center = element[3][3] be satisfactory? > > Try: > > typename boost::multi_array<T,2>::index > index1 = (element.index_bases()[0] > + element.strides()[0]) / 2; > typename boost::multi_array<T,2>::index > index2 = element.index_bases()[1]; > + element.strides()[1]) / 2; > T& center = element[index1][index2]; To make it general, yes. > > I don't think multiarray supports any type of > > reallocation after construction, so it should > > be ok. > > boost::multi_array has a resize() function, if that's > what you're afraid of. See > <http://www.boost.org/libs/multi_array/doc/reference.html> In this case, obviously center would not be center anymore, so the OP should beware.

1 0

[date_time] Get age in years
by Jason Dolan 29 Jun '06

29 Jun '06

I'm trying to get the age of someone in Years and/or months and/or days. I've tried: date_period dp(day_clock::local_day(), birthdate); cout << "You are " << dp.length() << " old" << endl; But that only calculates days. Is there a static function available that will turn days into months. and days into years? Thanks -Jason

4 4

Re: [Boost-users] multi_array: creating a view to a single element
by Sohail Somani 29 Jun '06

29 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Baranowski, Daniel > Sent: Wednesday, June 28, 2006 7:55 PM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] multi_array: creating a view to a > single element > [snip] > Basically, I have a 2 Dimensional array and I > would like to be able to access the center of that array > using a variable that maps to some index (for example, > element[3][3] in a 7x7 array). [snip] Would T & center = element[3][3] be satisfactory? I don't think multiarray supports any type of reallocation after construction, so it should be ok. Sohail

2 1

Embedded Python issue
by Trevor M Cickovski 29 Jun '06

29 Jun '06

Hello, I am trying to run some embedded Python code, following the boost.Python tutorial. What I was hoping to do was import some external modules. So basically in my .cpp file I have: Py_Initialize(); object main_module(( handle<>(borrowed(PyImport_AddModule("__main__"))))); object main_namespace = main_module.attr("__dict__"); try{ handle<> ignored((PyRun_String( "import Numeric\n" // Other Python commands , Py_file_input , main_namespace.ptr() , main_namespace.ptr()) )); } catch(error_already_set) { PyErr_Print(); } // Finish the Python Interpreter Py_Finalize(); However when I 'import Numeric', I get an undefined symbol: Traceback (most recent call last): File "<string>", line 4, in ? File "/afs/nd.edu/user25/tcickovs/Research/PYTHON/lib/python2.4/site-packages/Numeric/Numeric.py", line 91, in ? import multiarray ImportError: /afs/nd.edu/user25/tcickovs/Research/PYTHON/lib/python2.4/site-packages/Numeric/multiarray.so: undefined symbol: PyExc_ValueError Now if I just run python from the command line and call 'import Numeric' everything is fine. Does anyone have any idea what I need to do differently if I use Boost? Thanks for the help, Trevor -- ------------------------------------------------------- This is your time, this is your dance Live every moment, leave nothing to chance. Swim in the sea, drink of the deep Embrace the mystery of all you can be. - Michael W. Smith, "This Is Your Time"

1 0

Re: [Boost-users] boost 64bit support on windows x64 on amd64 hardware
by Sohail Somani 29 Jun '06

29 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of ChuMing > > > > On Tuesday, June 27, 2006 at 08:42:30 (+0000) Chu-Ming Ng writes: > > > > > >Hi, > > > > > >I am porting some code that uses boost from 32-bit to the > 64-bit platform > and > > >would like to know if the boost library supports this platforms. > But is there anyone I can ask to for official > confirmation? You can confirm it for yourself by looking at the regressions linked from the www.boost.org page. Or run them yourself.

1 0

how to compile first.cpp example
by francis 29 Jun '06

29 Jun '06

Hello, I try to compile first.cpp with bjam, but I don't know ho to specify the program_options library. I have a binary release of the boost library (Linux Kubuntu). I'm looking for a good Jamfile. Thanks for your help.

2 2

Statechart: Comple error with MinGW and not MSVC-7.1
by Eoin 29 Jun '06

29 Jun '06

Hello, My name is Eoin O'Callaghan I've run into a difficulty with the Boost Statechart library as I was porting my code from MSVC-7.1 to work with MinGW also. MinGW gives an error regarding the ctx parameter used with non simple states when the state is a template class; MinGWFail.cpp:24: error: expected `)' before "ctx" MinGWFail.cpp:29: error: expected constructor, destructor, or type conversion before '(' token I've attached a small file which replicates the problem. Note that if the content state is a simple_state then both MSVC-7.1 and MinGW compile without errors. Hope someone can make some sense of this, thanks the the author for making such an excellent library free for everyone. (Indeed thanks to all those involved in Boost :) ) Kind regards, Eoin. #include <boost/statechart/event.hpp> #include <boost/statechart/state_machine.hpp> #include <boost/statechart/state.hpp> #include <boost/statechart/simple_state.hpp> #include <boost/statechart/transition.hpp> #include <boost/statechart/custom_reaction.hpp> template<class P> class content; class parser : public boost::statechart::state_machine<parser, content<parser> > { public: parser(); virtual ~parser() {} }; template<class P> struct content : boost::statechart::state<content<P>, P> { public: content(my_context ctx); virtual ~content() {} }; template<class P> content<P>::content(my_context ctx) : my_base(ctx) {}

2 3

multi_array: creating a view to a single element
by Baranowski, Daniel 29 Jun '06

29 Jun '06

Hey All! I'm sure there's got to be a way, and unfortunately, the boost users archive is "down for maintenance" so I can't search there. Basically, I have a 2 Dimensional array and I would like to be able to access the center of that array using a variable that maps to some index (for example, element[3][3] in a 7x7 array). Can someone tell me how to do this? I tried to create a view to a 0 Dimensional array, but that wouldn't work (not surprisingly). FWIW, here's what I tried: array_type::array_view<0>::type center(myarray[ boost::indices[3][3] ]); I can do this: Array_type::array_view<1>::type center(myarray[ boost::indices[range(3,3)][3] ]); But then I access the center by doing center[0]... I guess that'll work, but I wouldn't consider that ideal. Anyone? Anyone? Bueller? Bueller? Thanks a bunch! Daniel ********************************************************************************************** Disclaimer - This email and any files transmitted with it are proprietary and may contain privileged or copyright information. You must not present this message to another party without gaining permission from the sender. If you are not the intended recipient you must not copy, distribute or use this email or the information contained in it for any purpose other than to notify us. If you have received this message in error, please notify the sender immediately, and delete this email from your system. We do not guarantee that this material is free from viruses or any other defects although due care has been taken to minimize the risk. eSafe scanned this email for viruses, vandals and malicious contentAny views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of LSI. **********************************************************************************************

1 0

Re: [Boost-users] shared_ptr in a boost::any [config] [hpux]
by Thomas Matelich 28 Jun '06

28 Jun '06

On 6/27/06, David Abrahams <dave(a)boost-consulting.com> wrote: > "Thomas Matelich" <matelich(a)gmail.com> writes: > > > I use boost::any for sending data to my thread's command queue, and I > > need to send a shared_ptr in. I'm looking for any one to tell me this > > is fine or bad. Based on my understanding of boost::any I think it > > will be fine, but I'm slightly concerned I'll screw up the reference > > count somehow. > > > > Can anyone confirm or allay my fears? > > Allay :) Thanks. Unfortunately, I just ran into boost/config/platforms/hpux.hpp and its belief that gcc and threads don't mix on HP-UX. I'm not sure why this is stated. Perhaps its left over from HP 10.x and DCE threads? I've been using gcc and threads for a long time, even back on 10.20 using ACE. I currently use both ACE thread wrappers and Qt's threads wrappers on 11i with gcc 3.2.3. I would just go ahead and comment out that assumption, but I'm concerned someone knows something I don't know. It looks like Jens Maurer added this assumption 2 years ago ( http://tinyurl.com/ku9wm ). Has anyone experimented with removing it? I'd really like mutexes for my shared pointers. Short-term, I'm currently working on a patch release and wouldn't really want to mess with every shared pointer in my whole system. I guess my workaround would be to wrap the shared_ptr in a mutexed_shared_ptr and lock every option that messes with the shared count?

1 0

Re: [Boost-users] [date_time] extract date from string using a list of patterns
by Brian Neal 28 Jun '06

28 Jun '06

> Thats why I was looking into it this way, I was hoping not to have to > create my own parser. I'm actually kinda of surprised that there isn't > a open source natural language date string parser out there already. GNU date is such a library. In PHP, there is the strtotime() function, which I believe uses GNU date C code to parse stuff like what you list. It accepts a bewildering variety of strings like "now", "last Monday", "10 September 2004", "10 Sep 2004", "+1 week 2 days 3 seconds", etc. I can't find a GNU link for you though. But here is the man page for PHP's strtotime() for a starting point: http://us3.php.net/manual/en/function.strtotime.php Good luck

3 3

Boost threads built from source causes app freeze, using prebuilt works fine
by Zachary Mark 28 Jun '06

28 Jun '06

Hello, We are using Boost threading as a part of our applicaiton. On multiple Linux platforms, when we compile Boost's thread library from source and use the compiled shared libraries (we are aware of the static linking issues with the thread library) with our application, we get a deadlock at different places in the application depending on which platform we are using. When using a pre-built binary (for example, from Debian's apt repository) for the thread library, the application does not deadlock. There is otherwise no difference between the application builds. We would like for us to be able to build Boost libs from source on every target platform, so I would like to figure out exactly what our build script is doing wrong with the thread library. For reference, here is the relevant portion of the Boost build bash script (this is a part of a larger script): # run configure cd libs/config || fail-configure chmod +x configure || fail "error: could not change permissions on ${PACKAGE} configure script" ./configure || fail-configure cp user.hpp ../../boost/config/ cd ../.. # build and install libraries BOOST_LIBS="date_time filesystem program_options regex thread" for LIB in ${BOOST_LIBS}; do echo "Building library ${LIB}..." bjam -sTOOLS=gcc -sBUILD="release" --with-${LIB} \ --layout="system" stage 2>&1 || \ fail "error: could not build boost library ${LIB}" bjam -sTOOLS=gcc -sBUILD="release" --with-${LIB} --layout="system" \ --prefix=`pwd`/../../${INSTALL_DIR} install 2>&1 || \ fail "error: could not install boost library ${LIB}" done Is there something that this script is missing or doing wrong that might cause some of these deadlock issues? Thanks in advance for your help. -- Zach

1 0

Re: [Boost-users] Interesting question, may be potential feature
by Ovanes Markarian 28 Jun '06

28 Jun '06

Nat, thanks for the quick reply! Unfortunately, people ideally would like to work with this data types without any functions and my job is to give them type safety ;) The scenario should be: struct PossibleValues { accessor<0> value_name_0(my_union); accessor<1> value_name_1(my_union); accessor<N> value_name_n(my_union); private: discriminated_union<mpl_vector_over_pairs> my_union; }; usage: void ValueHandler(PossibleValues& ps) { if(ps.value_name_0) ps.value_name_0->netsted_value = new_value; .... } Accessor makes havy usage of mpl::if_, enable_if, etc. so that -> operator works will regular types, pointer types, types where the -> operator is defined etc... Accessor analyses the type associated with the discrminator and overloads a correct set of functions/operators for that type. For example, if a stored type is pointer -> returns the pointer itself, if the stored type is normal object the address of it is returned, but if the type contains the -> operator then the object is returned by value (in case of shared_ptr, because you can not apply the -> operator twice ->->). Thanks, Ovanes

1 0

Re: [Boost-users] Interesting question, may be potential feature
by Nat Goodspeed 28 Jun '06

28 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Ovanes Markarian > Sent: Wednesday, June 28, 2006 12:39 PM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] Interesting question, may be potential feature > > But as some of you might have seen I have to pass to all accessors the > reference or pointer to the > accessed discrminated union type. So every instance requires on 32bit OS > at least 4 bytes. The > banal thing is that a group of accessors accesses the same type. The > question/idea here would be > to use or to make these objects share the reference, so that all related > accessor references > require 4 bytes. What do you think of this? Is it possible at all? [Nat] Um, make accessors be methods of a small object that contains the reference? But in that case, why wouldn't the convenience methods be attached to the referenced union object itself?

1 0

Interesting question, may be potential feature
by Ovanes Markarian 28 Jun '06

28 Jun '06

Dear Users and Developers, I am currently working on a kind of discriminated union type, e.g. XDR specific union type implementation (please refer to http://www.faqs.org/rfcs/rfc1832.html section) which maps enumeration values to types. In this case one wants to provide users as much comfort as possible for accesing this union. I have written a template class, which is defined by mpl::vector pairs of discriminant and related type. This class contains a boost::variant definition from this sequence. The problem is, that accessing the values etc. of this variant is a little bit clumsy: discriminated_union_inst.get<some const int for discrminant>(); discriminated_union_inst.set<some const int for discrminant>(new value); So I defined some accessor template. If initialized the accessor is responsible to access only one particular discriminant related type for example if discriminated_union is defined by sequence of: mpl_derived_pair<0, int>, mpl_derived_pair<1, std::string> ... mpl_derived_pair<N, MyClass> I define an accessor object: accessor<0> accessor_for_0(union_instance); later on this accessor allows the access if valid data is contained in the union. if union contains second type (mpl_derived_pair<1, string>) accessor_for_0->some_value //throws exception or if(accessor_for_0) // is true: means unions contains the correct type //do some access if true Now I must say: It was a pleasure to implement this using many boost libraries! But as some of you might have seen I have to pass to all accessors the reference or pointer to the accessed discrminated union type. So every instance requires on 32bit OS at least 4 bytes. The banal thing is that a group of accessors accesses the same type. The question/idea here would be to use or to make these objects share the reference, so that all related accessor references require 4 bytes. What do you think of this? Is it possible at all? With Kind Regards, Ovanes Markarian

1 0

shared ptr + const_cast woes
by Evan Drumwright 28 Jun '06

28 Jun '06

Hi all, I am having trouble with pointer ownership as a result of the const_cast operator. In particular, for the code below, when a const_cast is used to create a const int pointer, the use count for x does not increase. The result of this is that when x.pop_back() is called, y is no longer valid (though the use count for y remains 1). The documentation for shared_ptr does not help here. Any suggestions? Thanks, Evan #include <vector> #include <iostream> #include <boost/shared_ptr.hpp> int main() { std::vector<boost::shared_ptr<int> > x; x.push_back(boost::shared_ptr<int>(new int)); *x.front() = 4; std::cout << "use count: " << x.front().use_count() << std::endl; boost::shared_ptr<const int> y(const_cast<const int*>(x.front().get())); std::cout << "use count: " << x.front().use_count() << std::endl; std::cout << "use count: " << y.use_count() << std::endl; boost::shared_ptr<int> z(const_cast<int*>(y.get())); std::cout << "use count: " << z.use_count() << std::endl; std::cout << "use count: " << y.use_count() << std::endl; x.pop_back(); std::cout << *y << std::endl; } Output: use count: 1 use count: 1 use count: 1 use count: 1 use count: 1 0

3 2

[BGL] Two Grpah Matching Algorithms
by Hossein Haeri 28 Jun '06

28 Jun '06

Dear all, (I don't know whether I'm supposed to post BGL stuff to a more specialised list. Am I?) I was wondering if anybody here has implemented the maximum weight graph matching algorithm for bipartite graphs. Is that already a part of BGL? And, what about the stable marriage algorithm? To stop people from lynching me, this is NOT an assignement; We [1] are running out of time for a paper submission, and, we still need a few more evaluations. That's why I'm in a hurry... So, what I need for the moment is an implementation of the above two algorithms. Any help in this respct would be highly appriciated. TIA, --Hossein [1] Semantic Web Research Laboratory, Computer Engineering Department, Sharif University of Technology (SUT), Tehran, IRAN. http://sw.ce.sharif.edu ___________________________________________________________ Try the all-new Yahoo! Mail. "The New Version is radically easier to use" The Wall Street Journal http://uk.docs.yahoo.com/nowyoucan.html

2 3

[preprocessor] CAT and '-' token
by Johan Nilsson 28 Jun '06

28 Jun '06

Hi, I guess this is more of a regular preprocessor question than a Boost.Preprocessor specific one, but: Shouldn't it be possible to concat tokens starting with (or solely consisting of) a '-'? The example at the end of this posting yields the errors: error: pasting "-" and "debug" does not give a valid preprocessing token error: pasting "foo" and "-" does not give a valid preprocessing token --- #include <boost/preprocessor/cat.hpp> #include <boost/preprocessor/stringize.hpp> #include <iostream> #define MODULE foo #define VARIANT debug #define NAME BOOST_PP_CAT(MODULE, BOOST_PP_CAT(-, VARIANT)) void foo() { std::cout << "The name is: " << BOOST_PP_STRINGIZE(NAME) << '\n'; } --- Any workaround? I'm using Boost from cvs at RC_1_34_0 and g++ 4.0.2 under OpenSuse 10. / Johan

2 2

Comments for Filesystem Library Proposal for TR2 (N1975)
by Daniel Krügler 28 Jun '06

28 Jun '06

Hello! One week ago I wrote a list of comments to the above mentioned proposal provided by Beman Dawes to his email address <mangled_address>bdawes at acm dot org</mangled_address> given in the public paper http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2006/n1975.html Since I got no answer, I assume that the mail got lost due to filter barriers or something else. Is someone aware whether the above mentioned address should be still valid? I apologize for the noise in this group, but currently I'm not able to use either comp.std.c++ or comp.lang.c++.moderated due to news group rearrangements in our company to address this issue there and have hopes that Beman Dawes might read this group ;-) Thanks for your patience, Daniel Krügler

2 2

shared_ptr in a boost::any
by Thomas Matelich 28 Jun '06

28 Jun '06

I use boost::any for sending data to my thread's command queue, and I need to send a shared_ptr in. I'm looking for any one to tell me this is fine or bad. Based on my understanding of boost::any I think it will be fine, but I'm slightly concerned I'll screw up the reference count somehow. Can anyone confirm or allay my fears? Thanks, Tom

2 1

Re: [Boost-users] [date_time] extract date from string using a list of patterns
by james.jones＠firstinvestors.com 27 Jun '06

27 Jun '06

From: Jason Dolan <jason(a)pcc.com> > Thats why I was looking into it this way, I was hoping not to have to > create my own parser. I'm actually kinda of surprised that there isn't > a open source natural language date string parser out there already. Maybe take a look at the source from the GNU "at" command? It would be in C, but might yield some useful ideas. HTH, - James Jones Administrative Data Mgmt. Webmaster 375 Raritan Center Pkwy, Suite A Data Architect Edison, NJ 08837

1 0

still problems with has_xxx and shared_ptr...
by Ovanes Markarian 27 Jun '06

27 Jun '06

Hello all! Sorry to disturb again... I have the following code: namespace detail { BOOST_MPL_HAS_XXX_TRAIT_DEF(element_type) BOOST_MPL_HAS_XXX_TRAIT_DEF(value_type) BOOST_MPL_HAS_XXX_TRAIT_DEF(type) BOOST_MPL_HAS_XXX_TRAIT_DEF(type_t) } struct B1 {}; int main() { typedef boost::shared_ptr<B1> ptr_t; typedef mpl::if_c < detail::has_type<ptr_t>::value , mpl::int_<0> , mpl::if_c < detail::has_element_type<ptr_t>::value , mpl::int_<1> , mpl::if_c < detail::has_value_type<ptr_t>::value , mpl::int_<2> , mpl::if_c < detail::has_type_t<ptr_t>::value , mpl::int_<3> , mpl::int_<-1> > > > > type_id; type_id just_for_debug_an_instance; } MSVC 8.0 debugger shows for just_for_debug_an_instance the following type: boost::mpl::if_c<1,boost::mpl::int_<0>,boost::mpl::if_c<1,boost::mpl::int_<1>,boost::mpl::if_c<1,boost::mpl::int_<2>,boost::mpl::if_c<0,boost::mpl::int_<3>,boost::mpl::int_<-1> > > > > How is it possible, if I really can find in the shared_ptr template class the following typedefs: public: typedef T element_type; typedef T value_type; typedef T * pointer; typedef typename detail::shared_ptr_traits<T>::reference reference; For me it is really unclear why does: detail::has_type<ptr_t>::value has the value 1 in the first static if statement? I looked through the shared_ptr template class and found nowhere the 'type' typedef, function, or anything else, what could be evaluated by has_value. Any help would be really apreciated. Many Thanks, Ovanes Markarian

1 1

Re: [Boost-users] boost 64bit support on windows x64 on amd64 hardware
by Sohail Somani 27 Jun '06

27 Jun '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Chu-Ming Ng > Sent: Tuesday, June 27, 2006 1:42 AM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] boost 64bit support on windows x64 on > amd64 hardware > > > Hi, > > I am porting some code that uses boost from 32-bit to the > 64-bit platform and > would like to know if the boost library supports this platforms. I don't know about officially, but I haven't had any problems on 64 bit.

1 0

inspecting templates and boost::is_pointer
by Ovanes Markarian 27 Jun '06

27 Jun '06

Hello all, I am trying to specialize boost::is_pointer so, that it can determine if a type is a pointer wrapper (shared_ptr, auto_ptr, weak_ptr). Does anybody has ideas how this can be done? May be it would be possible to handle if e.g. one can determine if a class/struct has some typedef inside: template<...> class shared_ptr { typedef ... element_type; }; class some_other_class { typedef ... type; }; class some_other_class2 { //no typedef... }; Somehow I am not certain if I am on the right path. My problem is that I have a variant container type with different types. One could access a value using -> operator. The operator returns the pointer to the value. But if this value is a pointer (that can be determined by boost::is_pointer type trait) it should return the pointer and not pointer to pointer value. Now what happens if I store shared_ptr or any other type which overload -> operator? This will result in compiler error, because the compiler tries to access member of the pointer to pointer and not of the pointer instance. Is there may be a way to determine if the member access operator is overloaded? What if -> operator is defined in the base? And what if child class inherits base as protected or private? The other solution would be to force every user to write a specialized class is_pointer for every type stored in variant, but this would be not so nice. With Kind Regards, Ovanes Markarian

3 4

Re: [Boost-users] function traits
by Gottlob Frege 27 Jun '06

27 Jun '06

> > From: David Abrahams <dave(a)boost-consulting.com> "Ovanes Markarian" <om_boost(a)keywallet.com> writes: > > > Does anyone know if it possible to determine if a class > > overloads a certain function _where_ the return type is unknown. > > > > An example is the -> operator. It can return any type. My question > > is: is it possible to determine at compile time if some type > > overloads the -> operator, without knowing the return type when this > > operator is overloaded? > > That would only be possible if -> could be a free function, but IIRC > it is required to be a member function... so the answer is no. > > You can do it for any function that can be implemented as a non-member > (see e.g. boost/detail/is_incrementable.hpp). > > Also, if your end goal is still something like is_pointer_like<T>, instead of detecting operator->, you might be able to use the same techniques as is_incrementable to detect operator * (unary) - ie implement is_dereferenceable<T>. Would it be OK to assume is_pointer_like<T> '==' is_dereferenceable<T> ? Tony

1 0

[date_time] extract date from string using a list of patterns
by Jason Dolan 27 Jun '06

27 Jun '06

I'm looking to take a string and convert it to a date. The only problem is the string can be one of many patterns. i.e. ("%Y%m%d", "%Y-%m-%d", "%d/%m/%Y", etc...). It is also possible that the given string will fail all pattern matches, and thus return false. Can anyone give me an example of how to do this? I know you need to use the date_input_facet, but I am VERY new to boost and can't get it to work. Thanks! -Jason

3 5

[serializtion] Problem with multiple archives and Intel compiler
by Bill Lear 27 Jun '06

27 Jun '06

I am having great difficulty in architecting my boost serialization solution with the Intel compiler. The problem is this: I need to support text, xml, and binary archive formats. I allow the user to select which format to save in, by default using the portable text format. If the user wants greater speed, they can choose binary, or if they want to visually inspect the data, they can choose xml. On load, I dynamically detect which archive is used and call the appropriate load routine (text, xml, binary). However, if I include all three archives in the same compilation unit, the default (text) serialization is slowed down dramatically, with the runtimes increasing approximately 50% (from 200 to 300 seconds, and this on a fast 64-bit Opteron box). I tried breaking things apart, putting my archive includes and BOOST_CLASS_EXPORT calls into separate compilation units. I have the following code structure: class AbstractBase { virtual void method() = 0; }; BOOST_IS_ABSTRACT(AbstractBase) class Derived: public AbstractBase { virtual void method() {} }; class A { Def* def; template <class Archvie> void serialize(Archive& ar, const unsigned int version) { ar & def; } }; I break my serialization code into three different files, one for each type of archive, each of which includes a common file that has a templated method for loading a serialized file, along with BOOST_CLASS_EXPORT for the derived class. Here is the text version: #include <boost/archive/text_oarchive.hpp> #include "Serialization.hh" A* BoostTextLoader::load() { return Serialization<text_iarchive>::load(fileName); } #include <boost/serialization/export.hpp> BOOST_CLASS_EXPORT(Derived) Here is the binary: #include <boost/archive/binary_oarchive.hpp> #include "Serialization.hh" A* BoostTextLoader::load() { return Serialization<binary_iarchive>::load(fileName); } #include <boost/serialization/export.hpp> BOOST_CLASS_EXPORT(Derived) and a similar one for XML. The Serialization.hh is roughly this: template<typename Archive> struct FanSerialization { static A* load(const std::string& filename) { A* a = 0; stream<mapped_file_source> ifs(filename); Archive ia(ifs); ia >> BOOST_SERIALIZATION_NVP(a); return a; } } When I compile and try to link a test program, I get errors saying "multiple definition of `boost::archive::detail::guid_initializer<Derived>::instance'" and so on. So, I try removing the BOOST_CLASS export from the binary and XML code modules, leaving it in the text version. Compile and link are now good. The text archive performance is back down to 200 seconds. When I try to serialize binary (or xml) archives, I get. /boost-1_33_1/boost/archive/detail/oserializer.hpp:418: static void boost::archive::detail::save_pointer_type<Archive, TPtr>::polymorphic<T>::save(Archive &, const T &, const boost::archive::detail::basic_pointer_oserializer *) [with T = AbstractBase, Archive = boost::archive::xml_oarchive, TPtr = AbstractBase*]: Assertion `0L != bpos_ptr' failed. The code that is failing is from the assert below (oserializer.hpp): // sice true_type is valid, and this only gets made if the // pointer oserializer object has been created, this should never // fail bpos_ptr = archive_pointer_oserializer<Archive>::find(* true_type); assert(NULL != bpos_ptr); if(NULL == bpos_ptr) boost::throw_exception( archive_exception(archive_exception::unregistered_class) ); ar.save_pointer(vp, bpos_ptr); So, needless to say, I'm a bit confused by what is going on. Is there any way at all to fix this? I have run a profiler on the Intel code and it simply seems to be "doing more" than the non-Intel version (more comparison operators, chiefly). Could there be any way to manually instantiate things, or substitute the containers boost uses for look-ups at run-time with different, separate ones? Thanks. Bill

2 3

[BGL][compressed_sparse_row_graph] add_vertices
by Johan Oudinet 27 Jun '06

27 Jun '06

Hi, I try using the compressed_sparse row_graph graph type, and I want add lots vertices. When I read the documentation (http://www.boost-consulting.com/boost/libs/graph/doc/compressed_sparse_row.…) I saw the appropriate add_vertices function described as bellow: vertex_descriptor add_vertices(vertices_size_type count, compressed_sparse_row_graph& g) But, in compressed_sparse_row_graph.hpp:346: <--8----------------------------------------------------- template<BOOST_CSR_GRAPH_TEMPLATE_PARMS> inline Vertex add_vertices(Vertex count, BOOST_CSR_GRAPH_TYPE& g) { Vertex old_num_verts_plus_one = g.m_rowstart.size(); g.m_rowstart.resize(old_num_verts_plus_one + count, EdgeIndex(0)); return old_num_verts_plus_one - 1; } <--8----------------------------------------------------- I don't understand why `count' is a `Vertex' type! Why not `vertices_size_type' type ? Regards, -- Johan

2 1

[BGL] Betweenness Centrality Clustering
by Rui Carvalho 27 Jun '06

27 Jun '06

Hi, Is the BGL betweenness centrality clustering algorithm an implementation of: Community structure in social and biological networks M. Girvan and M. E. J. Newman Proc. Natl. Acad. Sci. USA 99, 7821-7826 (2002). http://aps.arxiv.org/PS_cache/cond-mat/pdf/0112/0112110.pdf thanks, Rui

2 1

1.33.0 documentation
by Joe McCay 27 Jun '06

27 Jun '06

I do I generate readable documentation for 1.33.0 out of a docbook file or a .fo file? jmccay

2 1

progress display for parallel environment
by Levent Yilmaz 27 Jun '06

27 Jun '06

Hi All, This little utility (progress_display in progress.hpp) is so useful that I couldn't help but send out a feature request (hoping that this list is the right place to do that). For programs using multiple threads or multiple distributed processes (as in MPI) it is useful to output progress indication only from some root node. However, these programs are usually SPMD (single program multiple data) and it is not convenient/possible to simply not create a progress_display object. One solution would be: // create a progress_display that does nothing for non root nodes int total_run = 100; boost::progress_display pg( i_am_root ? total_run : pg::DUMMY_DISPLAY ); // Then operator++() will output only from some root process. for (int i=0; i!=total_run; pg++); So my feature request is, can we have some facilities for creating trivial and does-nothing-useful type of progress_displays? best regards, - Levent.

2 1

Time-dependent graphs
by Thomas Hanne 27 Jun '06

27 Jun '06

Hi, I am currently using an adjacency_list for calculating shortest paths using BGL. Now I would like to make the edge weights time-dependent by calling some own function during the algorithm. Unfortunately, I did not really get things working as described in the book "The Boost Graph Library" with respect to self-defined property maps. (Or should I use a visitor for that purpose?) Maybe you could helb me with some code example on using own code for the edge weights to be used during the Dijkstra algorithm? Many thanks and best regards, Thomas Hanne My current code for calculating shortst paths is like ... typedef adjacency_list<listS, vecS, directedS, no_property, property<edge_weight_t, double> > tGraph; ... int anzKnoten = num_vertices(m_Wegezeiten); vector<vertex_descriptor> vorg(anzKnoten); vector<double> wege(anzKnoten); property_map<tGraph, edge_weight_t>::type gewichte = newget(edge_weight, m_Wegezeiten); property_map<tGraph, vertex_index_t>::type indizes = get(vertex_index, m_Wegezeiten); vertex_descriptor start = vertex(nrVon, m_Wegezeiten); dijkstra_shortest_paths(m_Wegezeiten, start, &vorg.at(0), &wege.at(0), gewichte, indizes, less<double>(), closed_plus<double>(), numeric_limits<int>::max(), 0, default_dijkstra_visitor() );

2 1

[BGL] Automatic cleanning in the graph.
by Seweryn Habdank-Wojewódzki 27 Jun '06

27 Jun '06

Hi everyone Is there a possibility to build a graph which nodes are automatically deleted when in the users code there is no reference to them? Maybe something that combines smart pointers and a graphs? Regards. -- |\/\/| Seweryn Habdank-Wojewódzki `\/\/

2 1

[PATCH] [Parameter] typo in index.rst
by Roman Neuhauser 27 Jun '06

27 Jun '06

Index: libs/parameter/doc/index.rst =================================================================== RCS file: /cvsroot/boost/boost/libs/parameter/doc/index.rst,v retrieving revision 1.30 diff -u -r1.30 index.rst --- libs/parameter/doc/index.rst 10 Apr 2006 20:41:06 -0000 1.30 +++ libs/parameter/doc/index.rst 27 Jun 2006 09:29:46 -0000 @@ -273,7 +273,7 @@ We're going to define our interface in namespace ``graphs``. Since users need access to the keyword objects, but not the tag types, we'll define the keyword objects so they're acceessible through -``graphs``, and we'll hide the tag types away in a tested +``graphs``, and we'll hide the tag types away in a nested namespace, ``graphs::tag``. The library provides a convenient macro for that purpose (MSVC6.x users see this note__):: -- How many Vietnam vets does it take to screw in a light bulb? You don't know, man. You don't KNOW. Cause you weren't THERE. http://bash.org/?255991

1 0

Re: [Boost-users] Boost Variant OutputStreamable
by Ovanes Markarian 27 Jun '06

27 Jun '06

The operator << must be defined in the boost namespace. See the error message of VC:see reference to function template instantiation 'std::basic_ostream<_Elem,_Traits> &boost::operator <<<char,std::char_traits<char>,std::vector<_Ty>,int,std::string,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_>(std::basic_ostream<_Elem,_Traits> &,const boost::variant<T0_,T1,T2> &)' being compiled VC looks for boost::operator<< function and not ::operator<< function. But this should be documented in OutputStreamable. This would compile: #include<iostream> #include<vector> #include<string> #include<boost/variant.hpp> namespace boost { template <typename T> std::ostream& operator<<(std::ostream& os, const std::vector<T>& v) { for(typename std::vector<T>::size_type i = 0; i < v.size(); ++i) os << v[i] << std::endl; return os; } } struct TpTemplateVariant { typedef boost::variant<std::vector<unsigned long>,int, std::string> BoostVariant; }; typedef TpTemplateVariant::BoostVariant Tpvariant; int main () { Tpvariant p(5); std::cout << p; return 0; } Good Luck, Ovanes Markarian On Mon, June 26, 2006 12:45, CLAIRE, Narinder, Group Risk Mgmt wrote: > I apologise for emailing you directly .. but i canno access the mailing list > from work.. > However in response to your response. > > > The following doesn't compile either .. I get the error > > error C2679: binary '<<' : no operator found which takes a right-hand > operand of type 'const T0' (or there is no acceptable conversion) > > On VC8 > > #include<iostream> > #include<vector> > #include<string> > > #include<boost/variant.hpp> > > > > std::basic_ostream<char> &operator<<(std::basic_ostream<char> > &theostream,const std::vector<unsigned long> &theVector) { > for(unsigned long i(0);i<theVector.size();i++) > theostream << theVector[i] << "\n" ; > return theostream; > } > > > struct TpTemplateVariant { > typedef boost::variant< std::vector<unsigned > long>, > > int, > > std::string > >> BoostVariant; > }; > > typedef TpTemplateVariant::BoostVariant Tpvariant; > > int main () { > > Tpvariant p(5); > std::cout << p; > > return 0; > > > > > *********************************************************************************** > The Royal Bank of Scotland plc. Registered in Scotland No 90312. Registered Office: 36 St Andrew > Square, Edinburgh EH2 2YB. > Authorized and regulated by the Financial Services Authority > > This e-mail message is confidential and for use by the > addressee only. If the message is received by anyone other > than the addressee, please return the message to the sender > by replying to it and then delete the message from your > computer. Internet e-mails are not necessarily secure. The > Royal Bank of Scotland plc does not accept responsibility for > changes made to this message after it was sent. > > Whilst all reasonable care has been taken to avoid the > transmission of viruses, it is the responsibility of the recipient to > ensure that the onward transmission, opening or use of this > message and any attachments will not adversely affect its > systems or data. No responsibility is accepted by The Royal > Bank of Scotland plc in this regard and the recipient should carry > out such virus and other checks as it considers appropriate. > Visit our websites at: > http://www.rbos.com > http://www.rbsmarkets.com > *********************************************************************************** > Download this as a file

2 4

[signals] unresolved overloaded function type with gcc 4.1
by Lars Monecke 26 Jun '06

26 Jun '06

Hi, i'm stucked and google does not lead me to the right pages :-( The follwing code example does not compile on gcc 4.1.1 an boost 1.33.1 . It works on msvc.net 7.1. ---code starts here #include <boost/bind.hpp> #include <boost/signal.hpp> #include <iostream> struct Printer { void operator()(int iValue) { std::cout << iValue << "\n"; } }; int main(int /* argc */, char* /* argv[] */) { typedef boost::signal<void (int)> tSignal; tSignal stSignal; Printer stPrinter; stSignal.connect(stPrinter); typedef std::vector<int> tConInts; tConInts cInts; cInts.push_back(1); cInts.push_back(2); cInts.push_back(3); for_each(cInts.begin(), cInts.end(), boost::bind(&tSignal::operator(), boost::ref(stSignal), _1)); return 0; } ---code ends here ---gcc output starts here main.cpp: In function 'int main(int, char*)': main.cpp:37: error: call of overloaded 'bind(<unresolved overloaded function type>, const boost::reference_wrapper<boost::signal<void ()(int), boost::last_value<void>, int, std::less<int>, boost::function<void ()(int), std::allocator<void> > > >, boost::arg<1>&)' is ambiguous /usr/local/gcc-4.1.1/boost-1.33.1/boost/bind/bind_mf_cc.hpp:43: note: candidates are: boost::_bi::bind_t<R, boost::_mfi::mf1<R, T, A1>, typename boost::_bi::list_av_2<A1, A2>::type> boost::bind(R (T::*)(B1), A1, A2) [with R = boost::last_value<void>::unusable, T = boost::signal1<void, int, boost::last_value<void>, int, std::less<int>, boost::function<void ()(int), std::allocator<void> > >, B1 = int, A1 = boost::reference_wrapper<boost::signal<void ()(int), boost::last_value<void>, int, std::less<int>, boost::function<void ()(int), std::allocator<void> > > >, A2 = boost::arg<1>] /usr/local/gcc-4.1.1/boost-1.33.1/boost/bind/bind_mf_cc.hpp:54: note: boost::_bi::bind_t<R, boost::_mfi::cmf1<R, T, A1>, typename boost::_bi::list_av_2<A1, A2>::type> boost::bind(R (T::*)(B1)const, A1, A2) [with R = boost::last_value<void>::unusable, T = boost::signal1<void, int, boost::last_value<void>, int, std::less<int>, boost::function<void ()(int), std::allocator<void> > >, B1 = int, A1 = boost::reference_wrapper<boost::signal<void ()(int), boost::last_value<void>, int, std::less<int>, boost::function<void ()(int), std::allocator<void> > > >, A2 = boost::arg<1>] make: *** [main.o] Error 1 ---gcc output ends here I think something with static_cast is the solultion, but i don't know how? Any hints? Thanks in advance, Lars

3 4

[optional] Why is operator-> disabled
by Jiri Palecek 26 Jun '06

26 Jun '06

Hello, I have been using boost::optional with a reference parameter recently, and I was puzzled by the fact that it is not possible to use operator-> to get the content. I could understand that get_pointer is disabled, but there is always a pointer such as *p==optional.get() and you can't actually get the pointer from operator-> Regards Jiri Palecek

1 0

Re: [Boost-users] Nindi ... Boost Variant
by Ovanes Markarian 26 Jun '06

26 Jun '06

The operator << must be defined in the boost namespace. See the error message of VC:see reference to function template instantiation 'std::basic_ostream<_Elem,_Traits> &boost::operator <<<char,std::char_traits<char>,std::vector<_Ty>,int,std::string,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost::detail::variant::void_>(std::basic_ostream<_Elem,_Traits> &,const boost::variant<T0_,T1,T2> &)' being compiled VC looks for boost::operator<< function and not ::operator<< function. But this should be documented in OutputStreamable. This would compile: #include<iostream> #include<vector> #include<string> #include<boost/variant.hpp> namespace boost { template <typename T> std::ostream& operator<<(std::ostream& os, const std::vector<T>& v) { for(typename std::vector<T>::size_type i = 0; i < v.size(); ++i) os << v[i] << std::endl; return os; } } struct TpTemplateVariant { typedef boost::variant<std::vector<unsigned long>,int, std::string> BoostVariant; }; typedef TpTemplateVariant::BoostVariant Tpvariant; int main () { Tpvariant p(5); std::cout << p; return 0; } Good Luck, Ovanes Markarian On Mon, June 26, 2006 12:45, CLAIRE, Narinder, Group Risk Mgmt wrote: > I apologise for emailing you directly .. but i canno access the mailing list > from work.. > However in response to your response. > > > The following doesn't compile either .. I get the error > > error C2679: binary '<<' : no operator found which takes a right-hand > operand of type 'const T0' (or there is no acceptable conversion) > > On VC8 > > #include<iostream> > #include<vector> > #include<string> > > #include<boost/variant.hpp> > > > > std::basic_ostream<char> &operator<<(std::basic_ostream<char> > &theostream,const std::vector<unsigned long> &theVector) { > for(unsigned long i(0);i<theVector.size();i++) > theostream << theVector[i] << "\n" ; > return theostream; > } > > > struct TpTemplateVariant { > typedef boost::variant< std::vector<unsigned > long>, > > int, > > std::string > >> BoostVariant; > }; > > typedef TpTemplateVariant::BoostVariant Tpvariant; > > int main () { > > Tpvariant p(5); > std::cout << p; > > return 0; > > > > > *********************************************************************************** > The Royal Bank of Scotland plc. Registered in Scotland No 90312. Registered Office: 36 St Andrew > Square, Edinburgh EH2 2YB. > Authorized and regulated by the Financial Services Authority > > This e-mail message is confidential and for use by the > addressee only. If the message is received by anyone other > than the addressee, please return the message to the sender > by replying to it and then delete the message from your > computer. Internet e-mails are not necessarily secure. The > Royal Bank of Scotland plc does not accept responsibility for > changes made to this message after it was sent. > > Whilst all reasonable care has been taken to avoid the > transmission of viruses, it is the responsibility of the recipient to > ensure that the onward transmission, opening or use of this > message and any attachments will not adversely affect its > systems or data. No responsibility is accepted by The Royal > Bank of Scotland plc in this regard and the recipient should carry > out such virus and other checks as it considers appropriate. > Visit our websites at: > http://www.rbos.com > http://www.rbsmarkets.com > *********************************************************************************** >

1 0

Issue with program_options with STLPort using long descriptions
by Tim 26 Jun '06

26 Jun '06

Dear All, I have an issue when using program_options with long description strings when using STLport in debug mode. This occurs within format_paragraph in options_description.cpp and is being found by the _Incrementable check within the STLPort iterator checking. This is using vc71 STLPort 5.0 and Boost 1.33.1 Any help on this issue would be gratefully received. Apologies for asking a question that has been asked previously but there was no resolution posted to the group.

2 2

Strange compiler behaviour with mpl::vector
by Ovanes Markarian 26 Jun '06

26 Jun '06

Hello all! I posted the following message http://lists.boost.org/boost-users/2006/06/20271.php a few days ago, but did not get any replies. I know there is an error in the const_ref class. But nevertheless the problem which occures is that my class accepts a vector of supported types as template parameter. It has two constructors (no default ctor): template<class Seq_> class my_class { public: typedef my_class<Seq_> this_type; explicit inline my_class(const this_type& copy_from) :... {...} template<class TypeFromVector> explicit inline my_class(typename const_ref<TypeFromVector>::type inst) :... {...} }; Now if I try to initialize my_class with type declared in vector I get an error: error C2664: 'discriminated_union<Seq_>::discriminated_union(const discriminated_union<Seq_> &)' : cannot convert parameter 1 from 't1' to 'const discriminated_union<Seq_> &' with [ Seq_=types_vector ] Reason: cannot convert from 't1' to 'const discriminated_union<Seq_>' with [ Seq_=types_vector ] No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called Ok, if I define the ctor param without typename const_ref<...>::type everything goes fine, if I use const_ref it fails. const_ref is defined as follows: template<class Ty> struct const_ref : boost::add_reference < typename boost::add_const < typename boost::remove_reference<Ty>::type >::type > {}; I can understand that the compiler is unable to make a decision which ctor to use, but I do not get an ambiguaty error. It seems like the compiler simply quits expanding templates and uses the last most successful ctor matched before the error. This happens only when I use mpl::vector templated type in conjunction with templated ctor. Other types like below work fine: template<int D_, class Ty_> struct my_pair : mpl::pair<mpl::int_<D_>, Ty_> { typedef my_pair<D_, Ty_> my_type; inline my_pair() {} explicit inline my_pair( typename const_ref<Ty_>::type ty) : second_(ty) {} explicit inline my_pair(const my_type& other) : second_(other.second_) {} inline my_type& operator =(const my_type& other) { second_=other.second_; return *this; } second second_; }; May be some of you have any suggestions. Thanks a lot, Ovanes Markarian

2 1

Serialization of an object with multiple formats
by Vidal Roca 26 Jun '06

26 Jun '06

It is possible to define more than one serialization format for a data structure with the serialization library? If not, has someone some good purposes to make it to? Thanks, Vidal

4 9

Re: [Boost-users] Boost Variant OutputStreamable
by Ovanes Markarian 26 Jun '06

26 Jun '06

Hi! Can you please to compile this code either with string in the variant or a vector. Not sure if it helps but the variant documentation at http://www.boost.org/doc/html/variant/reference.html states: namespace boost { template<typename T1, typename T2 = unspecified, ..., typename TN = unspecified> class variant; template<typename Sequence> class make_variant_over; template<typename T1, typename T2, ..., typename TN> void swap(variant<T1, T2, ..., TN> &, variant<T1, T2, ..., TN> &); template<typename ElemType, typename Traits, typename T1, typename T2, ..., typename TN> std::basic_ostream<ElemType,Traits> & operator<<(std::basic_ostream<ElemType,Traits> &, const variant<T1, T2, ..., TN> &); } Which means that the types are not zero-index based. So your T5 would be a string and not a vector. Make some tests like with simple integers, strings and finally a vector and see what happens. Best Regards, Ovanes -----Original Message----- From: Nindi [mailto:Nindi73@yahoo.co.uk] Sent: Sunday, June 25, 2006 15:36 To: boost-users(a)lists.boost.org Subject: [Boost-users] Boost Variant OutputStreamable I have started using Boost recently, and though I should switch my own variant to the Boost Variant . However I cannot compile the following code #include<iostream> #include<vector> #include<string> #include<boost/variant.hpp> std::basic_ostream<char> &operator<<(std::basic_ostream<char> &theostream,const std::vector<unsigned long> &theVector) { for(unsigned long i(0);i<theVector.size();i++) theostream << theVector[i] << "\n" ; return theostream; } struct TpTemplateVariant { typedef boost::variant< int, unsigned long, double, bool, std::string, std::vector<unsigned long> > BoostVariant; }; typedef TpTemplateVariant::BoostVariant Tpvariant; int main () { Tpvariant p(5); std::cout << p; return 0; } The error I get is error C2679: binary '<<' : no operator found which takes a right-hand operand of type 'const T5'

1 0

[BGL] total newbie question
by pixelpusher＠cox.net 25 Jun '06

25 Jun '06

Sorry for the newbieishness of the question, but I'm just getting started with C++ templates in general and Boost.Graph in particular. It seems to me that, while there is plenty of it, the documentation assumes a level of understanding of templates and other OO principles that I simply don't have at this point. Normally, I'd start digging into the example programs and look for common points of reference with things I already know and use the comments in the code for guidance. However, the example programs for BGL are astonishingly devoid of comments. They seem to be written for folks who already know how BGL works, not for folks who are trying to learn it. Most of what I see in the examples leaves me scratching my head and wondering WTF am I getting myself into here. Anyway, what I was hoping to find was more of a tutorial in how to get started writing a path-finding program using BGL and (probably) the Dijkstra-shortest paths template. - I have a set of edges that I will read from a text file. I don't know in advance how many edges will be read, so I plan to use an std::vector to store each edge's id, nodes, and weight. (All of the examples construct an array of known size and use that to build the graph. None use variable sized structures to build the graph.) Unless, of course, I can load the graph without pre-constructing a storage scheme. - The edge weights have to be modified based on the user's preferences, which will change with every run. (None of the examples show how to change edge weights.) - I will have known starting and ending nodes. I don't need to know the paths from my source to every possible destination. I only need to extract the best one that leads to my destination. I'd be happy enough for now just to know how to get started on the first bullet. Can anyone provide any guidance here? I'd really appreciate it. Thanks, Thom -- ------------------------------------------------------- Man will occasionally stumble over the truth, but most of the time he will pick himself up and continue. -- Winston Churchill

3 2

Re: [Boost-users] Strange compiler behaviour with mpl::vector
by Ovanes Markarian 25 Jun '06

25 Jun '06

Sorry, in the error message you will read the type name discriminated_union. This is the real type name, think of it as my_class. Thanks and sorry again, Ovanes _____ From: Ovanes Markarian [mailto:om_boost@keywallet.com] Sent: Sunday, June 25, 2006 23:18 To: 'boost-users(a)lists.boost.org' Subject: Strange compiler behaviour with mpl::vector Hello all! I posted the following message http://lists.boost.org/boost-users/2006/06/20271.php a few days ago, but did not get any replies. I know there is an error in the const_ref class. But nevertheless the problem which occures is that my class accepts a vector of supported types as template parameter. It has two constructors (no default ctor): template<class Seq_> class my_class { public: typedef my_class<Seq_> this_type; explicit inline my_class(const this_type& copy_from) :... {...} template<class TypeFromVector> explicit inline my_class(typename const_ref<TypeFromVector>::type inst) :... {...} }; Now if I try to initialize my_class with type declared in vector I get an error: error C2664: 'discriminated_union<Seq_>::discriminated_union(const discriminated_union<Seq_> &)' : cannot convert parameter 1 from 't1' to 'const discriminated_union<Seq_> &' with [ Seq_=types_vector ] Reason: cannot convert from 't1' to 'const discriminated_union<Seq_>' with [ Seq_=types_vector ] No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called Ok, if I define the ctor param without typename const_ref<...>::type everything goes fine, if I use const_ref it fails. const_ref is defined as follows: template<class Ty> struct const_ref : boost::add_reference < typename boost::add_const < typename boost::remove_reference<Ty>::type >::type > {}; I can understand that the compiler is unable to make a decision which ctor to use, but I do not get an ambiguaty error. It seems like the compiler simply quits expanding templates and uses the last most successful ctor matched before the error. This happens only when I use mpl::vector templated type in conjunction with templated ctor. Other types like below work fine: template<int D_, class Ty_> struct my_pair : mpl::pair<mpl::int_<D_>, Ty_> { typedef my_pair<D_, Ty_> my_type; inline my_pair() {} explicit inline my_pair( typename const_ref<Ty_>::type ty) : second_(ty) {} explicit inline my_pair(const my_type& other) : second_(other.second_) {} inline my_type& operator =(const my_type& other) { second_=other.second_; return *this; } second second_; }; May be some of you have any suggestions. Thanks a lot, Ovanes Markarian

1 0

Boost Variant OutputStreamable
by Nindi 25 Jun '06

25 Jun '06

I have started using Boost recently, and though I should switch my own variant to the Boost Variant . However I cannot compile the following code #include<iostream> #include<vector> #include<string> #include<boost/variant.hpp> std::basic_ostream<char> &operator<<(std::basic_ostream<char> &theostream,const std::vector<unsigned long> &theVector) { for(unsigned long i(0);i<theVector.size();i++) theostream << theVector[i] << "\n" ; return theostream; } struct TpTemplateVariant { typedef boost::variant< int, unsigned long, double, bool, std::string, std::vector<unsigned long> > BoostVariant; }; typedef TpTemplateVariant::BoostVariant Tpvariant; int main () { Tpvariant p(5); std::cout << p; return 0; } The error I get is error C2679: binary '<<' : no operator found which takes a right-hand operand of type 'const T5'

1 0

Re: [Boost-users] inspecting templates and boost::is_pointer
by Michael Nicolella 24 Jun '06

24 Jun '06

You passed has_xxx to the macro - It would make sense to only pass xxx (looks like an error in the example): BOOST_MPL_HAS_XXX_TRAIT_DEF(xxx) Also, you'll need to instantiate the has_xxx template: typedef has_xxx<test2> my_test; Mike -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Ovanes Markarian Sent: Saturday, June 24, 2006 12:22 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] inspecting templates and boost::is_pointer Dave, thanks for your answer. I think this should be definitely a part of FAQ, otherwise someone could spent endless hours on matters why some of boost classes do not work as expected. I have one more question. I tried the following code with VC 8.0: #include <boost/mpl/has_xxx.hpp> struct test2 { typedef int xxx; }; BOOST_MPL_HAS_XXX_TRAIT_DEF(has_xxx) typedef has_xxx my_test; And got a compiler error: main.cpp(22) : error C2146: syntax error : missing ';' before identifier 'my_test' Somehow the macro was not expanded. I replaced xxx with some other name before but end up in compiler errors, so that I decided to check if the standard example is compilable at all by using the typedef. Do you have any suggestions? Thanks, Ovanes -----Original Message----- From: David Abrahams [mailto:dave@boost-consulting.com] Sent: Saturday, June 24, 2006 04:51 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] inspecting templates and boost::is_pointer "Ovanes Markarian" <om_boost(a)keywallet.com> writes: > Hello all, > > I am trying to specialize boost::is_pointer so, that it can determine > if a type is a pointer wrapper (shared_ptr, auto_ptr, weak_ptr). First of all, don't do that. It's a really bad idea, and it will probably break some (Boost) library code. is_pointer is supposed to detect real pointers, and if you start changing what it does with other types, you'll break the expectations of every bit of code that currently uses is_pointer. This has come up before, so maybe it should be a FAQ (John?) > Does anybody has ideas how this can be done? May be it would be > possible to handle if e.g. one can determine if a class/struct has > some typedef inside: http://www.boost.org/libs/mpl/doc/refmanual/has-xxx-trait-def.html -- Dave Abrahams Boost Consulting www.boost-consulting.com _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

[iostreams] '.gnu.linkonce' errors when linking mapped_file code
by Bill Lear 24 Jun '06

24 Jun '06

When I use the Intel 9.0 compiler on 64-bit Linux, I get strange link errors that do not appear with gcc: `.gnu.linkonce.t._ZN5boost9iostreams6detail16direct_streambufINS0_18mapped_file_sourceESt11char_traitsIcEE7seekoffElSt12_Ios_SeekdirSt13_Ios_Openmode' referenced in section `.gnu.linkonce.d...1..TPK T.845_0.0.844.343222' of ./libnet_x64icco2a.a(BoostBinaryFanSerialization.o): defined in discarded section `.gnu.linkonce.t._ZN5boost9iostreams6detail16direct_streambufINS0_18mapped_file_sourceE St11char_traitsIcEE7seekoffElSt12_Ios_SeekdirSt13_Ios_Openmode' of ./libnet_x64icco2a.a(BoostBinaryFanSerialization.o) I've been poring over linker docs, compiler output, preprocessor output, google searches, etc., all morning. I've installed the newest version of the GNU binutils package, to no avail. If anyone can shed light on this, I'd really appreciate it. Bill

1 0

function traits
by Ovanes Markarian 24 Jun '06

24 Jun '06

Hello! I do not really want to bother you all, but I have some issue, which I am not certain is possible to solve and this drives me crazy. I posted some similar question at this location: http://lists.boost.org/boost-users/2006/06/20309.php and got an interesting answer, but not really one which satisfies me. Does anyone know if it possible to determine if a class overloads a certain function _where_ the return type is unknown. An example is the -> operator. It can return any type. My question is: is it possible to determine at compile time if some type overloads the -> operator, without knowing the return type when this operator is overloaded? Many thanks, Ovanes

2 1

Re: [Boost-users] date_time microsec_clock::univeral_time() implementation
by Brian Neal 23 Jun '06

23 Jun '06

> Jeff Garland <jeff(a)crystalclearsoftware.com> wrote: > > Brian Neal wrote: > > We have an application that was using our own custom functions for > > getting the OS time. We were basically representing time as the number > > of milliseconds since the UNIX epoch in an unsigned long, or as the > > number of nanoseconds since the epoch in a 64-bit type. > > > > We later discovered boost, and cut everything over to > > boost::date_time. This has worked out very, very well and has > > uncovered several time related bugs in our old code. > > Glad it was helpful :-) > > > However, we do > > have a small percentage of code that is time critical, and we have > > noticed that calling microsec_clock::univeral_time() is 2.6 times > > slower than our old function, and for this one bit of code that is > > significant. > > Ok. > > > Looking at the implementation for UNIX like systems, I see that > > create_time() in microsec_time_clock.hpp is calling gettimeofday() > > followed by a call to gmtime() or localtime() as appropriate. I think > > it's this second call that is the difference between our old code and > > boost. > > Ok. > > > In our time critical chunk of code we are just time tagging events, > > and we aren't concerned about timezone. I was thinking about adding a > > function to our code that called clock_gettime() (which is what our > > old code did), but then constructing a time_type out of that using the > > UNIX epoch for the date part and the results of clock_gettime() for > > the time_duration part, i.e, in pseudo code: > > > > time_type our_get_time() { > > clock_gettime(CLOCK_REALTIME, &timespec); > > time_duration td = massage(timespec); > > return time_type(unix_epoch, td); > > } > > Looks like a solid approach. Of course you don't have to modify > Boost.date_time, you can simply extend it with your own clock implementation. I replied to myself about this to the list, but my messages seem to have gotten out of order, so I'll repeat my findings and amplify a bit below: It turned out, much to my surprise, that the above approach was much, much slower than the current implementation of universal_time(). I suppose that is because it's got an enormous amount of seconds to add onto a date that is 36 years ago, and thus you got all your leap year calculations and stuff. I dunno, I didn't pursue it further. But it was much worse performance. I was doing something like this: const date unixEpoch(1970, Jan, 1); ptime getTime() { struct ::timespec ts; ::clock_gettime(CLOCK_REALTIME, &ts); // error checking omitted const time_duration td(seconds(ts.tv_sec) + nanoseconds(ts.tv_nsec)); return ptime(unixEpoch, td); } So I went back and rethought about how I was using boost::date_time. In our pre-boost code, we were just getting these 32-bit numbers, essentially tick counts, from the OS. We were just using these tick counts for comparisons and deltas, timestamping and stuff. We certainly didn't need full dates, as our events last, at most, 10 seconds. If you think about it, these tick counts are just a duration from some OS specific epoch. Aha, time_duration! So in our time critical code, we replaced all of our ptimes with time_durations, and replaced all calls to universal_time() with a function of our own making (getTimeStamp()), which calls clock_gettime() and forms up a time_duration with the results. This worked perfectly and brought back our performance to the pre-boost levels. For our other, non-critical time code, we just left the ptimes in. > > I'm also curious why gettimeofday() was chosen over clock_gettime()? > > Is one more available than the other? > > Honestly, I don't remember why at this point. There is the issue with > timezone adjustment for the other cases. My guess is portability was a factor > since gettime_clock is only supported in the POSIX real-time spec, although > it's probably pretty widespread now. That said, I'd be willing to improve the > implementation of microsecond_clock with this change where it is possible -- > looks like we can tell by checking CLOCK_REALTIME. I would guess, also, that portability might have been a driver. I do not believe the clock_gettime() is available on cygwin, for example, but gettimeofday() is. > > If you send me a working implementation I can drop it in and try it out ;-) > Well, as I said, the current universal_time() is much faster than treating the results of clock_gettime() as a time_duration from the UNIX epoch. But again, after rethinking our porting to boost, we brought the overkill on ourselves for replacing our tick counts with ptimes. time_durations were a better fit for what we were doing for that one specific case. Thanks. Boost::date_time is a very useful and powerful library!

2 1

Re: [Boost-users] inspecting templates and boost::is_pointer
by Gottlob Frege 23 Jun '06

23 Jun '06

> From: "Ovanes Markarian" <om_boost(a)keywallet.com> > > Hello all, > > I am trying to specialize boost::is_pointer so, that it can determine if a > type is a pointer > wrapper (shared_ptr, auto_ptr, weak_ptr). > > Does anybody has ideas how this can be done? May be it would be possible > to handle if e.g. one can > determine if a class/struct has some typedef inside: > > > You can make a template to detect typedefs - I wrote a has_value_type<T> for value_type. In my case I needed to know if the underlying type was char or wchar_t, so I ended up with a recursive thing that would even resolve things like a 'list of vectors of strings', etc. > > Somehow I am not certain if I am on the right path. My problem is that I > have a variant container > type with different types. One could access a value using -> operator. The > operator returns the > pointer to the value. But if this value is a pointer (that can be > determined by boost::is_pointer > type trait) it should return the pointer and not pointer to pointer value. > Now what happens if I > store shared_ptr or any other type which overload -> operator? This will > result in compiler error, > because the compiler tries to access member of the pointer to pointer and > not of the pointer > instance. You can also write a has_specific_member_function<T>, and thus I suspect you could also do it for operator->(). I think it depends on whether you can take a member function pointer to an operator function. google for has_foo, or try http://groups.google.com/group/comp.lang.c++.moderated/browse_frm/thread/a6… Is there may be a way to determine if the member access operator is > overloaded? What if -> > operator is defined in the base? And what if child class inherits base as > protected or private? Haven't tried any variations on public/protected/private - I suspect that you need to have a template that checks publicly, then another that checks for protected (I suspect that version of has_arrow_op<T> would probably need to derive from T) and I doubt you could tell the difference between a class that doesn't have ->, and a class that has a private ->. But I doubt you would want to know - if it doesn't have a callable ->, then you should probably act like it doesn't have one at all - ie it doesn't want you to think of it as a pointer. The other solution would be to force every user to write a specialized class > is_pointer for every > type stored in variant, but this would be not so nice. > > > With Kind Regards, > > Ovanes Markarian Tony

2 1

Re: [Boost-users] 'standard' free functions for operators, etc
by Gottlob Frege 23 Jun '06

23 Jun '06

> > In other words, if you have an operation F, and two type aliases X and Y > that both resolve to 'int' and you want F to have different semantics when > applied to X and Y, you are effectively not using the C++ type system. You > have imposed your own "logical type system" that is based on names (X and > Y) > and not types. I understand, and to "effectively not using the C++ type system" I would add a subtle twist: "not effectively using the C++ type system" either :-). Yet there is lots of code out there that does this. So, with regards to intrusive_ptr, I guess the answer is that that kind of code is not worth considering when designing intrusive_ptr. Is it worth considering for the_next_boost_thing<T> ? Maybe it depends on the thing, but much of boost is generic, so I think it does make sense to talk about the_next_boost_thing, even out of context. Note also that things like map<> allow the predicate to be specified, instead of using a free map_less(T, T) and assuming map_less is always the same, for a given type. Now I understand that map is a different situation than intrusive_ptr - I'm definitely stretching things a bit. But I'm looking for reasons to use one technique over the other (and also instead of, say, virtual functions / runtime polymorphism, but I think I understand most of the differences there). Obviously passing in a predicate/policy is the way to go when it is reasonable to expect different policies/predicate used *on the same type*, and intrusive_ptr is probably a case where different policies don't make sense. Is that enough to make it better? ie: - a policy based add_ref for intrusive_ptr would allow it to be used in those rare cases where there is a 'type-overlap' or a 'lack of strong typing' HOWEVER, - a policy based approach would *also* allow different add_refs for a single type (ie 'logical' or 'strong' type), which would be senseless and *disasterous* when the type really only has one way of doing add_ref, THUS intrusive_ptr chooses free functions, over a policy-based approach ie we thought about a policy-based approach, but specifically wanted to *enforce* one-add_ref-per-type, or we didn't even think about it (at least not consciously), it was just the natural and obvious thing to do Anyhow, looking at it that way maybe teaches me something.... Tony

2 1

gcc + stlport + boost 1_33_0
by William Deegan 22 Jun '06

22 Jun '06

Greetings, We're using gcc 3.2.3 stlport 4.61 and boost 1_33_0. When compiling some code which uses scoped_ptr, it was necessary to remove the explicit call to std::auto_ptr and change it to use just plain auto_ptr. I've attached the diffs below. Is there a boost configuration or other change I would need to make to allow this combination to compile without the below changes. So std::auto_ptr<> would be resolvable? Thanks, Bill These are the files I had to modify: M boost/intrusive_ptr.hpp diff -r1.2 intrusive_ptr.hpp 210c210 < return std::less<T *>()(a.get(), b.get()); --- > return less<T *>()(a.get(), b.get()); 257c257,258 < template<class E, class T, class Y> std::basic_ostream<E, T> & operator<< (std::basic_ostream<E, T> & os, intrusive_ptr<Y> const & p) --- > using _STL::basic_ostream; > template<class E, class T, class Y> basic_ostream<E, T> & operator<< (basic_ostream<E, T> & os, intrusive_ptr<Y> const & p) M boost/scoped_ptr.hpp diff -r1.2 scoped_ptr.hpp 63c63 < explicit scoped_ptr(std::auto_ptr<T> p): ptr(p.release()) // never throws --- > explicit scoped_ptr(_STL::auto_ptr<T> p): ptr(p.release()) // never throws M boost/shared_array.hpp diff -r1.2 shared_array.hpp 160a161 > using _STL::less; 163c164 < return std::less<T*>()(a.get(), b.get()); --- > return less<T*>()(a.get(), b.get()); M boost/shared_ptr.hpp diff -r1.2 shared_ptr.hpp 197c197 < explicit shared_ptr(std::auto_ptr<Y> & r): px(r.get()), pn() --- > explicit shared_ptr(_STL::auto_ptr<Y> & r): px(r.get()), pn() 221c221 < shared_ptr & operator=(std::auto_ptr<Y> & r) --- > shared_ptr & operator=(_STL::auto_ptr<Y> & r) 437c437,438 < template<class E, class T, class Y> std::basic_ostream<E, T> & operator<< (std::basic_ostream<E, T> & os, shared_ptr<Y> const & p) --- > using _STL::basic_ostream; > template<class E, class T, class Y> basic_ostream<E, T> & operator<< (basic_ostream<E, T> & os, shared_ptr<Y> const & p) M boost/detail/shared_count.hpp diff -r1.2 shared_count.hpp 138c138 < explicit shared_count( std::auto_ptr<Y> & r ): pi_( new sp_counted_impl_p<Y>( r.get() ) ) --- > explicit shared_count( _STL::auto_ptr<Y> & r ): pi_( new sp_counted_impl_p<Y>( r.get() ) ) 213c213 < return std::less<sp_counted_base *>()( a.pi_, b.pi_ ); --- > return _STL::less<sp_counted_base *>()( a.pi_, b.pi_ ); 307c307 < return std::less<sp_counted_base *>()(a.pi_, b.pi_); --- > return _STL::less<sp_counted_base *>()(a.pi_, b.pi_);

1 0

[Review results][PQS]
by fred bertsch 22 Jun '06

22 Jun '06

The Physical Quantities System review has ended, and, unfortunately, we are going to have to turn it down at this time. However, it is clear from the reviews that there is an enormous amount of interest in the library. I'm really hoping that that Andy Little will continue his work on it and resubmit it for review. Andy has been following the large discussion of PQS, so I will not enumerate the more minor criticisms included there. Some of the major criticisms that influenced my difficult decision include the following: * Many of the reviews liked the documentation, but improvements there would help out for a large number of reviewers. * If the library were more flexible about the type system used, several reviewers would have been happier. However, I would suggest that as long as a plan is in place for improving the flexibility, this should not prevent an initial version of the library from being accepted. * Since the library includes a large number of compile-time type manipulations, it would be great if the library included standard metafunctions to query these types. In addition, following other template metaprogramming conventions established within Boost such as documenting the concepts used would be a great help. I'd like to thank Andy Little for the work done so far on PQS and also for responding rapidly to the prolific discussion on the developer's list. -Fred Bertsch

4 5

Re: [Boost-users] 'standard' free functions for operators, etc
by Gottlob Frege 22 Jun '06

22 Jun '06

> From: "Peter Dimov" <pdimov(a)mmltd.net> > intrusive_ptr_add_ref is passed a pointer to a class with an embedded > reference count. It doesn't matter how many typedefs resolve to X*, X > still > has its reference count that needs to be incremented. > > You only run into the problem when "assign" is underspecified and > overloaded > to mean different things in different contexts. I understand your point, and maybe it is not a problem for intrusive_ptr in all pratical cases, but intrusive_ptr_add_ref is passed a T *. T may or may NOT be a class. ie T could be an int or an HWND (which is often just void *, depending on #define STRICT) or some other handle, or whatever. Maybe all those examples are a bit ugly - ie the int would have to be something like an index into a global array, or something like that. But it is not necessarily a distinct 'class' that holds its refcount intrinsicly. Or maybe it is necessarily that, but only because of how intrusive_ptr_add_ref was done, not for any other reason. So intrusive_ptr is ever so slightly more restrictive than it need be. Probably not a big deal. Note also, that for those few strange cases, you could also just wrap the HWND or whatever into a class, but then of course, intrusive_ptr could have just required the class to have a specific add_ref member function (and then you might need to wrap all classes). So maybe intrusive_ptr is just an example and not necessarily the best one, but I still wonder 1. if policy based could have been the same for 95% of the cases, and better for the last 5%, 2. there are any other techniques to consider. Either for intrusive_ptr or boost iterators or other places where I want to use the same techniques. I think that's the real question - if I am going to borrow some idioms for boost or wherever, I'd like to borrow the right ones for the right situations. Tony

2 1

linking
by Tamas Hegedus 22 Jun '06

22 Jun '06

Hi, I am newbie to c/cpp and I have linking and/or loading shared library problem. I can not link with dynamic libraries. Thanks for your help in advance, Tamas *** 1 *** [violin]$ ldconfig -p | grep libboost_program_options libboost_program_options.so.1 (libc6) => /usr/lib/libboost_program_options.so.1 *** 2 *** As I do not have the file /usr/lib/libboost_program_options.so.1 I created a link: libboost_program_options.so.1 -> libboost_program_options-gcc-1_33_1.so.1.33.1 *** 3 *** Compiling my src file [violin]$ g++ -Wall -I /usr/local/include/boost-1_33_1 -c test.cpp *** 4A *** [violin]$ g++ -Wall -o test test.o /usr/local/lib/libboost_program_options-gcc-1_33_1.so.1.33.1 [violin]$ ./test ./test: error while loading shared libraries: libboost_program_options-gcc-1_33_1.so.1.33.1: cannot open shared object file: No such file or directory *** 4B *** [violin]$ g++ -Wall -o test test.o /usr/local/lib/libboost_program_options.so.1 [violin]$ ./test ./test: error while loading shared libraries: libboost_program_options-gcc-1_33_1.so.1.33.1: cannot open shared object file: No such file or directory *** 4C *** [violin]$ g++ -Wall -o test test.o /usr/local/lib/libboost_program_options-gcc-1_33_1.a [violin]$ ./test Compression level was not set. -- Tamas Hegedus, PhD | phone: (1) 919-966 0329 UNC - Biochem & Biophys | fax: (1) 919-966 5178 6107 Thurston-Bowles Bldg | mailto:hegedus@med.unc.edu Chapel Hill, NC, 27599-7248 | http://biohegedus.org

3 2