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

Combining static and dynamic policies
by Jens Müller 23 Nov '06

23 Nov '06

Hi, I have the following task: There is an algorithm where you can set different behaviors for creating a BFS tree. Currently this is done by setting a policy object (dynamic), a function of which is called to perform the search. I'd like to use static policies here (as is done in generic Boost algorithms), but still keep the possibility of setting the behavior manually. The reason for this is: You can set the behavior using a command line parameter, and it would be impractical to instantiate all templates and then use a huge switch (think of combinations of several policies) to choose the appropriate instantiation. Here a wrapper (dynamic polymorphism) would be needed anyway, or any code using the algorithm would need to be duplicated as well. So my idea is as follows: There is a template parameter for a policy class. Static policy exist for each possible behavior, as well for a dynamic behavior. The "dynamic" policy class has virtual member functions and there exists an inheriting dynamic class for each behavior.[1] The algorithm then has a named param for the policy class. Default value if to default-construct an object of the policy class. (If useful, the template might have a default value for the policy template parameter as well, so the user needn't do anything). Has someone already taken this or a similiar approach? If so, could you point me to a reference? What do you think of the approach? Greeting, Jens [1] Hmmm ... Could this inheriting class be used as template parameter without overhead? Other solution would be: class dyn_behavior{ public: virtual void perform_search(something ...) = 0; } class specific_dyn_behavior<class T> : dyn_behavior { T behavior; public: void perform_search(something ...){ behavior.perform_search(something ...); } }

1 0

shared_ptr. circular references
by bringiton bringiton 23 Nov '06

23 Nov '06

Can anyone offer advice on detecting circular references with shared_ptrs. I have a large app, 50+ classes, and most objects are managed with shared_ptrs. No consideration has gone into avoid circular references, so I'm sure there will be a few in there. I thinking that looking though all the classes and their attributes will take a while. Any suggestions/advice in finding circular references (so i can use weak_ptrs)? I was thinking of maybe getting a tool that draws the class diagram. Or having a base class for all my classes that reference count to a global variable to detect the number of leaks. It seems the shared_ptr is a good idea for memory management, but it opens up the original problem of memory leaks...

4 3

[serialization] linking
by Deane Yang 23 Nov '06

23 Nov '06

I am using bjam with Boost Build v2 to compile and link an executable that calls the boost serialization library using MSVC++ 8.0. For some reason, it tries to link to a library named "libboost..." but the actual library is called "boost...." (no "lib" prefix). Why is this happening and how do I fix it? It doesn't happen when I compile the same executable in Linux.

5 11

[BGL] working file_dependencies.cpp
by Alexander Holler 22 Nov '06

22 Nov '06

Hello, I've already though some time that something does not work as expected using the way the various file_dependencies.cpp examples have done for calculating what files could be compiled in parallel. Because I've now got to a point where I really need something working, I've just done it myself. ;) My solution is pretty simple, and it basically works like it is described in the documentation for the example, walk through the tree and calculate the maximum distance. To achieve that, I'm just using the result from topological_sort and calculating the maximum through a walk through the in_edges of the vertices. Because a diff is almost as large as the resulting example, I've just attached the modified example. It might not look as nice as the original example, but it works as expected. ;) Kind regards, Alexander Holler //======================================================================= // Copyright 1997, 1998, 1999, 2000 University of Notre Dame. // Authors: Andrew Lumsdaine, Lie-Quan Lee, Jeremy G. Siek // // Distributed under the Boost Software License, Version 1.0. (See // accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) //======================================================================= // Some small modifications are done by Alexander Holler /* Paul Moore's request: As an example of a practical problem which is not restricted to graph "experts", consider file dependencies. It's basically graph construction, plus topological sort, but it might make a nice "tutorial" example. Build a dependency graph of files, then use the algorithms to do things like 1. Produce a full recompilation order (topological sort, by modified date) 2. Produce a "parallel" recompilation order (same as above, but group files which can be built in parallel) 3. Change analysis (if I change file x, which others need recompiling) 4. Dependency changes (if I add a dependency between file x and file y, what are the effects) */ #include <boost/config.hpp> // put this first to suppress some VC++ warnings #include <iostream> #include <iterator> #include <algorithm> #include <time.h> #include <boost/utility.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/topological_sort.hpp> #include <boost/graph/depth_first_search.hpp> #include <boost/graph/dijkstra_shortest_paths.hpp> #include <boost/graph/visitors.hpp> using namespace std; using namespace boost; enum files_e { dax_h, yow_h, boz_h, zow_h, foo_cpp, foo_o, bar_cpp, bar_o, libfoobar_a, zig_cpp, zig_o, zag_cpp, zag_o, libzigzag_a, killerapp, N }; const char* name[] = { "dax.h", "yow.h", "boz.h", "zow.h", "foo.cpp", "foo.o", "bar.cpp", "bar.o", "libfoobar.a", "zig.cpp", "zig.o", "zag.cpp", "zag.o", "libzigzag.a", "killerapp" }; struct print_visitor : public bfs_visitor<> { template <class Vertex, class Graph> void discover_vertex(Vertex v, Graph&) { cout << name[v] << " "; } }; struct cycle_detector : public dfs_visitor<> { cycle_detector(bool& has_cycle) : m_has_cycle(has_cycle) { } template <class Edge, class Graph> void back_edge(Edge, Graph&) { m_has_cycle = true; } protected: bool& m_has_cycle; }; int main(int,char*[]) { typedef pair<int,int> Edge; Edge used_by[] = { Edge(dax_h, foo_cpp), Edge(dax_h, bar_cpp), Edge(dax_h, yow_h), Edge(yow_h, bar_cpp), Edge(yow_h, zag_cpp), Edge(boz_h, bar_cpp), Edge(boz_h, zig_cpp), Edge(boz_h, zag_cpp), Edge(zow_h, foo_cpp), Edge(foo_cpp, foo_o), Edge(foo_o, libfoobar_a), Edge(bar_cpp, bar_o), Edge(bar_o, libfoobar_a), Edge(libfoobar_a, libzigzag_a), Edge(zig_cpp, zig_o), Edge(zig_o, libzigzag_a), Edge(zag_cpp, zag_o), Edge(zag_o, libzigzag_a), Edge(libzigzag_a, killerapp) }; const std::size_t nedges = sizeof(used_by)/sizeof(Edge); typedef adjacency_list<vecS, vecS, bidirectionalS, property<vertex_distance_t, int> > Graph; #if defined(BOOST_MSVC) && BOOST_MSVC <= 1300 // VC++ can't handle the iterator constructor Graph g(N); for (std::size_t j = 0; j < nedges; ++j) { graph_traits<Graph>::edge_descriptor e; bool inserted; tie(e, inserted) = add_edge(used_by[j].first, used_by[j].second, g); } #else Graph g(used_by, used_by + nedges, N); #endif typedef graph_traits<Graph>::vertex_descriptor Vertex; // Determine ordering for a full recompilation // and the order with files that can be compiled in parallel { typedef list<Vertex> MakeOrder; MakeOrder make_order; property_map<Graph, vertex_distance_t>::type distance = get(vertex_distance, g); topological_sort(g, std::front_inserter(make_order)); cout << "make ordering: "; for (MakeOrder::iterator i = make_order.begin(); i != make_order.end(); ++i) { cout << name[*i] << " "; // Walk through the in_edges an calculate the maximum distance. Graph::in_edge_iterator j, j_end; tie(j, j_end) = in_edges(*i, g); if(j==j_end) distance[*i]=0; else { int maxdist=0; // Through the order from topological sort, we are sure that every // distance we are using here is already initialized. for (; j != j_end; ++j) maxdist=std::max(distance[source(*j, g)], maxdist); distance[*i]=maxdist+1; } } cout << endl << endl; cout << "parallel make ordering, " << endl << "vertices with same group number can be made in parallel" << endl; graph_traits<Graph>::vertex_iterator i, iend; for (tie(i,iend) = vertices(g); i != iend; ++i) cout << "time_slot[" << name[*i] << "] = " << distance[*i] << endl; } cout << endl; // if I change yow.h what files need to be re-made? { cout << "A change to yow.h will cause what to be re-made?" << endl; print_visitor vis; breadth_first_search(g, vertex(yow_h, g), visitor(vis)); cout << endl; } cout << endl; // are there any cycles in the graph? { bool has_cycle = false; cycle_detector vis(has_cycle); depth_first_search(g, visitor(vis)); cout << "The graph has a cycle? " << has_cycle << endl; } cout << endl; // add a dependency going from bar.cpp to dax.h { cout << "adding edge bar_cpp -> dax_h" << endl; add_edge(bar_cpp, dax_h, g); } cout << endl; // are there any cycles in the graph? { bool has_cycle = false; cycle_detector vis(has_cycle); depth_first_search(g, visitor(vis)); cout << "The graph has a cycle now? " << has_cycle << endl; } return 0; }

2 2

[enable_if] Compiler workaround for constructor overloads
by Michael Marcin 22 Nov '06

22 Nov '06

Hello I'm working with a Codewarrior compiler (Nokia Codewarrior C/C++ Compiler for Windows/x86 version 3.2.3 build 451). The following test code finds the proper overload for MSVC 2005 and doesn't for the Codewarrior compiler but only with the = syntax. Does anyone have any idea how to make "c" find its overload as well? It would make the client code a lot cleaner to read/write. #include <boost/type_traits/is_integral.hpp> #include <boost/utility/enable_if.hpp> struct tester { int value; tester( float f ) : value((int)f) { } tester( double d ) : value((int)d) { } template< typename T > tester( T val, typename boost::enable_if< boost::is_integral<T> >::type* = 0 ) : value(val) { zzz } }; void main() { tester a = 1.f; // works (chooses float overload) tester b(1); // works (i.e. error on zzz) tester c = 1; // works on vc2005 (i.e. error on zzz) // fails on cw (ambiguous overload looking at float and double) }

1 0

wave - can line markers be suppressed?
by Grant, Joan 22 Nov '06

22 Nov '06

Using the wave library is there any way to suppress line markers (#line...) in a manner similar to the '-P' option of the gcc compiler's preprocessor? Thanks, Joan The information contained in or attached to this email may be subject to the Export Administration Regulations (EAR), administered by the U.S. Department of Commerce, or the International Traffic in Arms Regulations (ITAR), administered by the U.S. Department of State, and may require an export license from the Commerce or State Department prior to its export. An export can include a release or disclosure to a foreign national inside or outside the United States. Include this notice with any reproduced portion of this information.

2 2

Another generic program that won't compile with pointee, bind, and array
by Terry G 22 Nov '06

22 Nov '06

Here's another (simplified) program I can't get to work. I'm trying to write a generic routine that finds the largest element of a sequence. It is assumed that element->size() is defined, as it is for boost::array and std::string. For some reason, the Largest template function works okay for std::strings, but not for boost::arrays. Can you tell me what is wrong? Will you please offer some improvements for error detection and debugging techniques that would have helped here? Perhaps I need a "typename" or "template" somewhere. Take it as given that I have a String class that needs to be a fixed-size array. Here's the code: #include <boost/operators.hpp> #include <boost/static_assert.hpp> #include <boost/function.hpp> #include <boost/tuple/tuple.hpp> #include <boost/iterator/iterator_traits.hpp> #include <boost/bind.hpp> #include <boost/array.hpp> #include <boost/pointee.hpp> #include <cstddef> #include <iterator> #include <algorithm> #include <functional> #include <string> #include <vector> #include <iostream> #include <iomanip> template <class InIter> InIter Largest(InIter first, InIter last) { typedef typename boost::pointee<InIter>::type ObjType; InIter largest = std::max_element(first, last, boost::bind(std::less<std::size_t>(), boost::bind(&ObjType::size, _1), boost::bind(&ObjType::size, _2)) ); return largest; } // Largest template <std::size_t N> class String: public boost::array<char, N> { public: std::string str() const { if (this->back() == '\0') return std::string(this->data()); return std::string(this->data(), this->size()); } // str String() { this->assign('\0'); } String(const std::string& str_) { if (str_.size() > this->size()) throw std::runtime_error("String::ctor: str_ too big"); int copied = str_.copy(this->c_array(), this->size()); std::fill_n(this->begin() + copied, this->size() - copied, '\0'); } // ctor operator std::string() const { return str(); } }; // String using namespace std; int main(int argc, const char* argv[]) { // typedef String<64> ArgStr; // <-- This won't compile. typedef string ArgStr; // <-- This compiles/runs ok. vector<ArgStr> args; while (--argc) args.push_back(string(*++argv)); if (args.empty()) { cout << "no args!" << endl; return EXIT_FAILURE; } cout << "args="; copy(args.begin(), args.end(), ostream_iterator<string>(cout, " ")); cout << endl; vector<ArgStr>::iterator iter = Largest(args.begin(), args.end()); cout << "Largest arg is '" << string(*iter) << "'." << endl; return EXIT_SUCCESS; } // main

2 1

'configure' does not recognise '--without-python'
by MikeW 22 Nov '06

22 Nov '06

During Boost build, a line pops up saying "to avoid seeing this message again, use option --without-python". However, if you specify this as a "./configure" option, configure halts saying --without-python is unrecognised. I can live without it/work around it, but it should presumably be fixed ... Regards, Mike

2 4

[BGL] how to actually search with DFS/BFS?
by martin f krafft 22 Nov '06

22 Nov '06

This may seem like a silly question, but how does one actually search with DFS or BFS? Say I want to find the first node for which a predicate returns true in a BFS traversal starting at node N... The function signatures don't have a parameter for "criteria" or "predicate", so I guess it's up for a Visitor to implement such a check. What is not clear to me is how the visitor functions can break out of the search once a condition is true. Thanks, -- martin; (greetings from the heart of the sun.) \____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck invalid/expired pgp subkeys? use subkeys.pgp.net as keyserver! spamtraps: madduck.bogus(a)madduck.net "information superhighway" is just an anagram for "i'm on a huge wispy rhino fart".

9 12

[BGL] Constantness bug in kruskal_minimum_spanning_tree
by Fernando Cacciola 22 Nov '06

22 Nov '06

Hi I have a graph-like data structure which supports both const and non-const vertex and edge iterators and handles, and I want to interface it to the BGL. FWIW. the iterators and handles are class types, and the const versions are different classes than the non-const ones, so removing const qualification is not the way to bind a const type to its non-const version. AFAICT, the proper way to address both types of iterator/handles is by specializing both graph_traits<MyGraph> and graph_traits<MyGraph const>. It all works fine with my own BGL-like algorithms, but now I want to call kruskal_minimum_spanning_tree and I can't. The problem is that the kruskal function takes a "const Graph&" argument but it declares the internal types like this: typedef typename graph_traits<Graph>::edge_descriptor Edge ; (and so for the vertex_descriptor, etc) but then, lines like this: tie(vb,ve) = vertices(G) or this source(e, G) fails to compile with my graph because it attempts to bind a const iterator/handle to a non-const iterator/handle, which my data structure intentionally forbids. Unless I am missing something obvious (and I hope I am), this is a bug in the kruskal function: it should declare the working types like this: typedef typename graph_traits<Graph const>::edge_descriptor Edge ; Am I right? If yes, what can I do?? Is there any "constant correct" BGL algorithm? I'm calling kruskal just as an example showing how the data structure can be interfaced to the BGL. TIA Fernando Cacciola

2 1

Re: [Boost-users] Boost configure options for crosscompilation ?
by Hughes, James 22 Nov '06

22 Nov '06

This is the command line I used to build boost using Bjam for an arm platform, not sure if that helps with what you need, but here it is anyway!! bjam "-sGCC_ROOT_DIRECTORY=/opt/montavista/mobilinux/devkit/arm/iwmmxt_le" "-sGCC=iwmmxt_le-gcc" "-sGXX=iwmmxt_le-g++" stage Stage is the folder in which all the libs are placed, other parameters should be self explanatory. James > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of MikeW > Sent: 20 November 2006 15:59 > To: boost-users(a)lists.boost.org > Subject: [Boost-users] Boost configure options for crosscompilation ? > > I would like to crosscompile Boost under Buildroot. > > I created what appeared to be a successful package file, but > found that the native compiler and tools were used rather > that Buildroot's crosscompiler toolchain. > > However, I could not see any other 'configure' options to > switch to using a crosscompiler (EPREFIX looked promising at > first but seems to be a path prefix, rather than a tool > prefix such as "arm-linux-"). > > I note an earlier post (unable to post a 'follow-up' to it) > seemed to successfully do a cross build. > > Could you tell me how Boost is 'configure'd to use the > crosstools rather than the native tools ? > > Thanks, > Mike > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > This message (including any attachments) contains confidential and/or proprietary information intended only for the addressee. Any unauthorized disclosure, copying, distribution or reliance on the contents of this information is strictly prohibited and may constitute a violation of law. If you are not the intended recipient, please notify the sender immediately by responding to this e-mail, and delete the message from your system. If you have any questions about this e-mail please notify the sender immediately.

3 5

[regex] Conditional Replacement?
by Scott Meyers 22 Nov '06

22 Nov '06

If I want to replace all the regex matches in a string, no problem, regex_replace is just what I'm looking for. But suppose I want to replace each match only if some predicate is satisfied, e.g., to implement something like a string replacement in a text editor where the user is allowed to approve or reject each proposed replacement. Is there any direct support for something like this? The only way I can think of to do this is to iterate over the string to be searched, finding each match and then consulting the predicate to see if the replacement should be performed for that match. If so, I can call regex_replace on the found substring, but then I'll get back a new string, and I'll have to manually handle the process of knitting together the final revised string from the original string and the results of each substitution. This is essentially what regex_replace already does, except it unconditionally does a substitution for each match. What it seems I want is something like regex_replace_if. Is there such a thing in Boost or TR1? If not, is it because there is little demand for such functionality? That would surprise me, because pretty much every text editor allows one to iteratively search for a match and query the user to see if a replacement should be performed at that point. Thanks, Scott

2 1

thread: not joinable?
by Robert Marion 21 Nov '06

21 Nov '06

Hi, I've written my first boost::thread application, and I get this assertion: /boost_1_33_1/libs/thread/src/thread.cpp:223: void boost::thread::join(): Assertion 'm_joinable' failed. I call the thread as: void _pollMode( Controller& ctrl ) { // do polling ... for my testing, left blank } void SomeClass::startPolling() { m_pCtrl = new Controller; boost::function<void (Controller&)> funcPoll; funcPoll = &_pollMode; m_thr = new boost:thread(boost::bind( funcPoll, *m_pCtrl)); // never get past this line (unless copy ctor modified) } The Controller class has a copy constructor defined, which gets called several times (by the thread class?) Controller( const Controller& ctrl ) { boost::mutex::scoped_lock lock( m_mutex ); m_thr = ctrl.m_thr; // bad ?? BUT ... if I do m_thr = NULL; then everything is ok. Why??? What's going on here??? } Any help/explanation would be appreciated. Rob ____________________________________________________________________________________ Sponsored Link Mortgage rates near 39yr lows. $420k for $1,399/mo. Calculate new payment! www.LowerMyBills.com/lre

2 1

Statechart tutorial problem and possible solution
by Michael 21 Nov '06

21 Nov '06

I am going through the tutorial for Statechart. Going through the section "adding reactions", while debug build links fine, in a release mode linker issues following complaints: 1>FSM.obj : error LNK2001: unresolved external symbol "public: void __thiscall boost::statechart::detail::no_context::no_function<struct EvReset>(struct EvReset const &)" (??$no_function@UEvReset@@@no_context@detail@statechart@boost@@QAEXABUEvReset@@@Z) 1>FSM.obj : error LNK2001: unresolved external symbol "public: void __thiscall boost::statechart::detail::no_context::no_function<struct EvStartStop>(struct EvStartStop const &)" (??$no_function@UEvStartStop@@@no_context@detail@statechart@boost@@QAEXABUEvStartStop@@@Z) I ended up adding following line to my cpp file: template< class Event > void sc::detail::no_context::no_function( const Event & ) {}; and it cured the linker complaint. Michael

2 3

Problem with sub_range
by abir basak 21 Nov '06

21 Nov '06

Hi, I am not sure if it is the proper newsgroup to get answers related to boost usage. I am facing a problem with boost::sub_range for a const container which is not allowing me to access the indexed operator. Below a short program to demonstrate the problem. Necessary headers are included. Boost version 1.33.1 typedef vector<int> VI; VI v(10); int data[] = {0,1,2,3,4,5,6,7,8,9}; copy(data,data+10,v.begin());///inserted a few int to std::vector<int> ///prints 0 1 2 3 4 5 6 7 8 9 copy(v.begin(),v.end(),ostream_iterator<int>(cout," ")); cout<<endl; const VI& cv = v; /// a const reference to the container. VI::iterator it = v.begin(); ///got an random access iterator it[1] = 100; /// index operator allows to change it copy(v.begin(),v.end(),ostream_iterator<int>(cout," ")); cout<<endl; ///prints 0 100 2 3 4 ... VI::const_iterator it1 = v.begin(); ///got a const_iterator cout<<it1[1]<<endl; /// returns const_reference, assignment is not allowed, access is allowed. prints 100 boost::sub_range<VI> r(v.begin()+2,v.begin()+8); ///got a sub_range ///prints 2 3 4 5 6 7 copy(r.begin(),r.end(),ostream_iterator<int>(cout," ")); cout<<endl; r[2] = 400; cout<<r[2]<<endl;/// prints 400 ///prints 2 3 400 5 6 7 copy(r.begin(),r.end(),ostream_iterator<int>(cout," ")); ///so far everything is fine. Here problem starts /// a const sub_range ? boost::sub_range<const VI> r1(cv.begin()+2,cv.begin()+8); ///prints 2 3 400 5 6 7 copy(r1.begin(),r1.end(),ostream_iterator<int>(cout," "));cout<<endl; ///next statement is don't compile cout<<r1[2]<<endl; ///=> not allowed gives error under Visual Studio 7.1 cannot convert from 'const std::allocator<_Ty>::value_type' to 'boost::iterator_range<IteratorT>::value_type &' Why it tried to convert it to value_type& instead of const value_type& ? cout<<typeid(r1.begin()).name()<<endl; says r1.begin() is a const_iteartor. So the statement cout<<r1[2]<<endl; should return a const_reference. Am I mis interpreting the const version of sub_range ? If so, what is a const counterpart of sub_range, which behaves just like a pair of const_iterator ? Any help is appreciated. Thanks abir

3 8

[regex] compile errors on Mac OS X
by Richard Dingwall 21 Nov '06

21 Nov '06

Ran into a strange problem with boost::regex today. I get this just from including boost/regex.hpp: powerbook:~/Desktop/bre_test richard$ make g++ -pedantic -Wall -g -I/usr/local/include/boost-1_33_1/ -c -o bre_test.o bre_test.cpp /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function `bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_greedy_single_repeat(bool)': /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:991: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:992: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function `bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_slow_dot_repeat(bool)': /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1041: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1042: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function `bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_char_repeat(bool)': /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1159: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1160: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function `bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_short_set_repeat(bool)': /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1223: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1224: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function `bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_long_set_repeat(bool)': /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1288: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' /usr/local/include/boost-1_33_1/boost/regex/v4/perl_matcher_non_recursive.hpp:1289: error: invalid conversion from 'boost::re_detail::re_syntax_base*' to 'long int' make: *** [bre_test.o] Error 1 powerbook:~/Desktop/bre_test richard$ A test case is attached. OS X version: Darwin powerbook.local 8.8.0 Darwin Kernel Version 8.8.0: Fri Sep 8 17:18:57 PDT 2006; root:xnu-792.12.6.obj~1/RELEASE_PPC Power Macintosh powerpc

2 4

[program_options] replace environment variables in config file
by Chris Weed 21 Nov '06

21 Nov '06

Hi, I would like to be able to replace environment variables specified in config file values. For example I would like to use the env. var. (BOOST_PATH=c:\Boost) to expand the following file1=${BOOST_PATH}\include into this: file1=c:\Boost\include So I can then parse the value (c:\Boost\include) as a boost::filesystem::path. It seems like I want something similar to the extra parser for command-line options, but it doesn't seem to be possible for the config file parser. Do you have any ideas how to accomplish this? Also, is there a parser in program_options or property_tree that can parse "Java Property"-style property files, such as # My comment here cache=1kb file=c:/autoexec.bat Thanks, Chris

1 0

How to avoid specifying the boost .libs in my MSVS 2003 projects?
by Joshua Stewart 21 Nov '06

21 Nov '06

I have a static lib that uses boost/regex.hpp. (boost version 1.33.1) However when I want to use my static lib and link into a separate executable project, I have to specify the libboost_regex-vc71-mt-gd-1_33_1.lib as an input library for the executable project, so that it can link correctly with the static lib. Otherwise I get linking errors. One solution might be to make my static lib a dll... but this is not on the cards at the moment. The boost headers on some systems will make use of a #pragma to do this linking for you. I thought that if I added #include <boost/regex.hpp> To a header in my executable project, that boost would automatically identify and link against the correct boost .lib and my static lib would link without errors also. I am trying to find a nice way that I can use boost in a static lib project in visual studio 2003, and then link that in a separate project without needing to specify to specific boost .lib that the static lib requires. Any help or suggestions are appreciated. Regards, Josh This message contains confidential information and is intended only for the individual(s) named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please immediately delete it and all copies of it from your system, destroy any hard copies of it, and notify the sender. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. To the maximum extent permitted by law, Immersive Technologies Pty. Ltd. does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission.

1 0

Bug report in date_time, 1.33.0, not fixed in CVS
by Eric Teutsch 20 Nov '06

20 Nov '06

Boost::date_time uses FILETIME internally to get subsecond time resolution timestamps. When using date_time in a managed C++ environment, the compiler (VC7.1) can't differentiate between the windows global FILETIME, and the .NET System::Runtime::InteropServices::FILETIME, and therefore throws an error. If the boost header file microsec_time_clock.hpp (in boost/date_time) changes the types FILETIME to the global namespace ::FILETIME, then the naming collision goes away. This has only been tested in a limited arena, but I believe it should not affect other areas). Thank you Eric T. ________________________________________ I let my mind wander and it didn't come back.

2 2

Re: [Boost-users] Beginner's shared_ptr question...
by Hughes, James 20 Nov '06

20 Nov '06

> void MyClass::init() > { > // either > spOther = shared_ptr<someOtherClass>(new someOtherClass(this)); or > spOther.reset(new someOtherClass(this)); } > > Dave. > -- > Dave Slutzkin > Melbourne, Australia > daveslutzkin(a)fastmail.fm Another beginners question... Are the two ways of doing this, as shown above, exactly the same? i.e. execute the same code and are therefore at the same cost, or is one more efficient than the other? James > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > This message (including any attachments) contains confidential and/or proprietary information intended only for the addressee. Any unauthorized disclosure, copying, distribution or reliance on the contents of this information is strictly prohibited and may constitute a violation of law. If you are not the intended recipient, please notify the sender immediately by responding to this e-mail, and delete the message from your system. If you have any questions about this e-mail please notify the sender immediately.

2 1

Boost configure options for crosscompilation ?
by MikeW 20 Nov '06

20 Nov '06

I would like to crosscompile Boost under Buildroot. I created what appeared to be a successful package file, but found that the native compiler and tools were used rather that Buildroot's crosscompiler toolchain. However, I could not see any other 'configure' options to switch to using a crosscompiler (EPREFIX looked promising at first but seems to be a path prefix, rather than a tool prefix such as "arm-linux-"). I note an earlier post (unable to post a 'follow-up' to it) seemed to successfully do a cross build. Could you tell me how Boost is 'configure'd to use the crosstools rather than the native tools ? Thanks, Mike

1 0

Beginner's shared_ptr question...
by Martin Waller 20 Nov '06

20 Nov '06

(g++ 3.3.5, boost 1.32) I have a program as follows: Header: //multiple include guard omitted for clarity #include <boots/shared_ptr> class someOtherClass; class MyClass { public: MyClass(); void init(); . . . private: boost::shared_ptr <someOtherClass> spOther; . . . }; Implementation #1: #include <MyClass.hpp> #include <someOtherClass.hpp> MyClass::MyClass() { init(); } void MyClass::init() { spOther(new someOtherClass(this)); } Implementation #2: #include <MyClass.hpp> #include <someOtherClass.hpp> MyClass::MyClass():spOther(new someOtherClass(this)) { } Question: Why does implementation #2 fail to compile with the error: error: no match for call to '(boost::shared_ptr<someOtherClass>)(someOtherClass*)' ? yet implementation #2 compiles fine? Please help me understand what going on here...is it possible to get implementation #2 to compile or is it 'bad code'? Thanks

3 2

How to use bind with explicit constructor conversion?
by Terry G 20 Nov '06

20 Nov '06

This doesn't compile. How best to accomplish this using generic STL or Boost algorithms/adapters? terry #include <iostream> #include <iomanip> #include <algorithm> #include <vector> #include <iterator> #include <cstdlib> #include <boost/bind.hpp> #include <boost/mem_fn.hpp> using namespace std; using namespace boost; class MyInt { int _value; public: explicit MyInt(int x=0) : _value(x) { } int value() const { return _value; } }; // MyInt int main() { vector<MyInt> v; // Do this: // for (int i=0; i!=30; ++i) v.push_back(MyInt(rand())); generate_n(back_inserter(v), 30, bind(MyInt, bind(rand))); // Now print v. transform(v.begin(), v.end(), ostream_iterator<int>(cout, "\n"), mem_fn(&MyInt::value)); return 0; } // main

2 1

boost::numeric::ublas::matrix adaptors for C style arrays
by Edward Grace 20 Nov '06

20 Nov '06

Dear All, Is it possible to adapt e.g. a pointer to double in to a ublas::matrix<double>, or construct a matrix from a pointer to double? I have used the Blitz++ libraries, with those it is possible to use storage allocated elsewhere. This is to allow the painless interface of my code with legacy C code. I would like to do something like the following: using namespace boost::numeric::ublas; double *p; p = new double[100]; .... Fill p[*] with data // Construct a 10x10 matrix using externally allocated storage. matrix<double> A(10,10,p); There does appear to be a hint that I can use my own allocator, but it is far from clear how I go about doing this. For example coming up with an Allocator function that doesn't allocate but instead uses existing memory. class MyAllocator : public GeneralAllocator { double *TheMemory; public: MyAllocator(double *p) { TheMemory = p; } allocate() {} }; matrix<double,MyAllocator> A; Failing that, I'd like an adaptor that allows me to effectively do something like: matrix<double> A; double *p; // Allocate p and fill with stuff! A = adaptor<matrix<double> >(p,nCol,nRows); A is now a 10x10 matrix<double> using the memory pointed to by p. Obviously there are issues regarding who is now responsible for this memory, something handled by the memory policy in Blitz++ (neverDelete, etc). Help!! Is there a way to do this? -ed

2 3

[Boost.General] Improvement proposals for boost data access
by Markus Werle 20 Nov '06

20 Nov '06

Hi Boosters! While trying to trace the development of some boost libraries I see opportunities for improvement concerning the boost data distribution system. I am a little bit tired of looking up 3 or 4 different places in order to find out how some code snippet or library part under development is distributed, whether it is commited to the sandbox cvs, the vault, or stored hidden below boost-consulting.com (like SummerOfCode stuff with only a mailing list post pointing to it) or whether the developer has deliberately choosen to set up his own cvs/svn tree or other distribution method e.g. via sourceforge (which might be perfectly OK, Joel!). Yet I am missing the single entry point to all these GoodThings. Things are suboptimal: Example 1: Try to find out via boost web pages where one can download the latest and greatest version of boost::fusion. You can find a version at http://spirit.sourceforge.net/, but I am unsure if it is the latest bleeding edge version ... Example 2: Function types is under review. The files are accessible via tinyurl, so boost ML content is not self-contained. If I look at it in 6 years (when <your favorite .com monster here> bought tinyurl and crashed the database), I will not be able to find any hint to where to find the files. Also if I'd like to see how that lib looked like at the time of review in order to understand the review discussion on the mailing list I am out of luck. Example 3: files in the vault are rather volatile, no file history available. Still waiting to obtain some old files mentioned in some post ... I'd like to make proposals for improvement: 1. Rule: Libraries under review _must_ be commited to cvs/svn tree of boost or boost-sandbox. In order to help people behind firewalls and resistent IT administrators, a script creates daily snapshots of cvs/svn tree which are accessible via ftp/http from boost.org similar to the way gcc is distributed. If a developer has no access via cvs/svn he/she has to find someone who can take this task (or boost has a server which provides svn on port 80 ...) Any announcement concerning reviews points to the ftp/http access point and contains the cvs/svn command which has to be used in order to obtain the files. Rationale: KISS for users. Blindly trust to find the code at one single point. 2. Rule: Posts to boost ML which link to any code snippet or fileset uploaded to the vault for discussion use a http link to the true URL at boost::vault. Using the full link is declared as netiquette rule for posting to ML. Rationale: Internet data is not persistent. A full link gives you the chance to find the file either by contacting the admin of the mentioned http server or by filename via standard search engines. 3. Technical measure: The vault is under some automatic version control (http://subversion.tigris.org/faq.html#filesystem http://noedler.de/projekte/wdfs/ http://svnbook.red-bean.com/en/1.1/apcs03.html) and versions can be retrieved via http on a per date basis, which allows to follow older discussions easily. Rationale: I recently read a discussion between Jeremy S., Joel d. G. and Dave A. which was hard to follow anyway due to the subject and due to the skill level of those who were talking. It would be much easier to get at least some minimal insight into what they were talking about if the files were still available (so I hope). Sometimes it really helps to read early stage code that is not yet polluted by all this CIRCUMVENT_BROKEN_COMPILER macros. [sidenote: looking forward to the new book: "The design and evolution of boost::mpl, boost::fusion, boost::graph and 12 other libraries" including all the chapters Dave and Alex left out in their first book ;-P] What do you think? Markus

2 4