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

[Serialization] crash in boost::serialization::extended_type_info::key_unregister ()
by Avi Bahra 15 Feb '10

15 Feb '10

I am using boost 1.39 on suse linux with gcc 4.2.1 and using a debug build. In order to rationalize the long compile times, I was attempting to split a file into multiple translation units. When I ran my tests, they all passed, but then crashed with a memory fault. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0xb7cd26d0 (LWP 21897)] 0xb7d40a08 in strcmp () from /lib/libc.so.6 (gdb) bt #0 0xb7d40a08 in strcmp () from /lib/libc.so.6 #1 0x08e19924 in boost::serialization::extended_type_info::key_unregister () #2 0x08e19ace in boost::serialization::extended_type_info::~extended_type_info () #3 0x08c41afc in ~extended_type_info_typeid (this=0x92179cc) at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/extended_type_info_typeid.hpp:80 #4 0x08c41bbe in ~singleton_wrapper (this=0x92179cc) at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/singleton.hpp:104 #5 0x08c209e0 in __tcf_25 () at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/singleton.hpp:121 #6 0xb7d00599 in exit () from /lib/libc.so.6 #7 0xb7ce9fe8 in __libc_start_main () from /lib/libc.so.6 #8 0x08a833f1 in _start () It appears that boost serialisation type registration has problems when it is split into multiple files. Also it appears that most of serialisation test all appear in a single file, hence these kinds of issues can easily be missed. Is there any work around to this problem ? Best regards, Ta, Avi

2 1

[Boost Graph] Two-graphs common spanning tree problem: generic algorithm (with concept checking, of course)
by Michele Caini 15 Feb '10

15 Feb '10

Hi all, my name is Michele Caini, I write from Italy, Florence, and this is my first mail on this list. I'd like to propose a new generic implementation of the two-graphs common three algorithm, proposed first by Grimbleby and then modified by Schach, that made a revolution on the first idea basing his algorithm on the solution for the single spanning three problem proposed and developed by Mint, Read and Tarjan. This algorithm is mainly used in symbolic circuit analysis, and I develop a generic solution for my symbolic solutor: QSapecNG, Qt-based frontend for symbolic circuit solution. If Boost Graph library mantainers are interested on publishing this algorithm with the next release of library, I'm happy to submit it, and to work on to optimize the code. Thanks to all, Michele ps: sorry for my English. :-)

2 2

Re: [Boost-users] [Serialization] crash in boost::serialization::extended_type_info::key_unregister ()
by dada＠lamef.bordeaux.ensam.fr 15 Feb '10

15 Feb '10

Hi, I had the same problem... This problem was fixed in the next release (1.40) Regards, Damien. Le lundi 15 février 2010 à 10:30 +0000, Avi Bahra a écrit : > I am using boost 1.39 on suse linux with gcc 4.2.1 and using a debug > build. > In order to rationalize the long compile times, I was attempting to > split a file > into multiple translation units. When I ran my tests, they all passed, > but then > crashed with a memory fault. > > Program received signal SIGSEGV, Segmentation fault. > [Switching to Thread 0xb7cd26d0 (LWP 21897)] > 0xb7d40a08 in strcmp () from /lib/libc.so.6 > (gdb) bt > #0 0xb7d40a08 in strcmp () from /lib/libc.so.6 > #1 0x08e19924 in > boost::serialization::extended_type_info::key_unregister () > #2 0x08e19ace in > boost::serialization::extended_type_info::~extended_type_info () > #3 0x08c41afc in ~extended_type_info_typeid (this=0x92179cc) > at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/extended_type_info_typeid.hpp:80 > #4 0x08c41bbe in ~singleton_wrapper (this=0x92179cc) > at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/singleton.hpp:104 > #5 0x08c209e0 in __tcf_25 () > at /var/tmp/ma0/boost/boost_1_39_0/boost/serialization/singleton.hpp:121 > #6 0xb7d00599 in exit () from /lib/libc.so.6 > #7 0xb7ce9fe8 in __libc_start_main () from /lib/libc.so.6 > #8 0x08a833f1 in _start () > > It appears that boost serialisation type registration has problems > when it is split into multiple files. Also it appears > that most of serialisation test all appear in a single file, hence > these kinds of issues can easily be missed. > Is there any work around to this problem ? > > Best regards, > Ta, > Avi > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

[Python]
by Oliver Rudolph 15 Feb '10

15 Feb '10

Hi folks! When trying to run Python-wrapped C++ code, I receive the following message (this was a simple call to one of the c'tors of class TextFile): Traceback (most recent call last): File "./FileTest.py", line 5, in <module> t_out = TextFile("sample.txt", "This is some sample text. It says:\n\nHello world!") Boost.Python.ArgumentError: Python argument types in TextFile.__init__(TextFile, str, str) did not match C++ signature: __init__(_object*, std::string filename, dretchy::io::DeviceDataType type, std::string content, dretchy::io::CompressionType zipper) __init__(_object*, std::string filename, dretchy::io::DeviceDataType type, dretchy::io::CompressionType zipper) __init__(_object*, std::string filename, dretchy::io::DeviceDataType type, std::string content) __init__(_object*, std::string filename, dretchy::io::DeviceDataType type) __init__(_object*, std::string filename, std::string content, dretchy::io::CompressionType zipper) __init__(_object*, std::string filename, dretchy::io::CompressionType zipper) __init__(_object*, std::string filename, std::string content) __init__(_object*, std::string filename) Strangely enough, this only happens when doing so on my openSuSE 11.0 machine. If I try the same thing on my Ubuntu 9.10 notebook, it works perfectly. What could be the source of this problem? The code is a bit lengthy, so I attached the non-compilable headers only. Personally, I presume the error is rooted in the C-string / C++ std::string conversion. The Python wrapping itself was done via Py++, the boost libs in use are of version 1.41.0. Thanks a lot and best regards, Oliver

1 0

[accumulators] question
by Neal Becker 15 Feb '10

15 Feb '10

If I have: typedef accumulator_set<el_t,stats<tag::mean> > accum_t; Is there a way to get the count of the number of samples?

2 3

installing boost 1.34.0 in snow leopard
by Murali Krishna Pusala 15 Feb '10

15 Feb '10

Hi every one, I am trying to install boost 1.34.0 on snow leopard, it is able to build .a files but it is not building .dylib libraries. It giving an error ld: unknown option: -h krishna

5 6

Asio exception handling
by Matthias Vallentin 15 Feb '10

15 Feb '10

Hi, Consider the example where an io_service objects is used as scheduler, with tasks handed to io_service::post(). According to the manual [1], exceptions should propagate to the caller of io_service::run. However, this does not seem to be the case with the attached example. Essentially, I am doing the following: while (true) { try { io_service.post(task(&work)); io_service.post(task(&work)); io_service.post(task(&work)); io_service.run(); break; } catch (...) { // Never reached! std::cout << "catching exception" << std::endl; } } I attached a minimal example that demonstrates this issue. Any help would be appreciated. Matthias [1] http://www.boost.org/doc/libs/1_42_0/doc/html/boost_asio/reference/io_servi… -- Matthias Vallentin vallentin(a)icsi.berkeley.edu http://www.icir.org/matthias

2 2

[filesystem] Mac OS default codecvt facet
by Beman Dawes 14 Feb '10

14 Feb '10

On Mac OS, Boost.Filesystem currently uses std::locale() as the default, since Apple doesn't see fit to support std::locale(""). There has been a request (ticket #3928) to change the codecvt facet to UTF-8 for the Boost.Filesystem default on Mac OS. This would at least potentially be a breaking change, so I wanted to ask other's opinions before making it. Questions: * Is UTF-8 OK with Mac OS users as the Boost.Filesystem default? * Would Linux users also prefer UTF-8 as the Boost.Filesystem default? Thanks, --Beman

1 0

generate structs with all combinations of members from a given struct
by Hicham Mouline 14 Feb '10

14 Feb '10

Hello, struct params { double field1; double field2; double field3; double fiedl4; }; I would like to generate all the types that hold any combination of the members of params. struct params_field1 { double field1; }; struct params_field2 { double field2; }; struct params_field3 { double field3; }; struct params_field4 { double field4; }; struct params_field1_field2 { double field1; double field2; }; struct params_field1_field3 { double field1; double field3; }; struct params_field1_field4 { double field1; double field4; }; struct params_field2_field3 { double field2; double field3; }; struct params_field2_field4 { double field2; double field4; }; struct params_field3_field4 { double field3; double field4; }; struct params_field1_field2_field3 { double field1; double field2; double field3; }; struct params_field1_field2_field4 { double field1; double field2; double field4; }; struct params_field1_field3_field4 { double field1; double field3; double field4; }; struct params_field2_field3_field4 { double field2; double field3; double field4; }; struct params_field1_field2_field3_field4 { double field1; double field2; double field3; double field4; }; There are 15 of these, which is 2^4 -1, where 4 is the number of members in params. Assuming params is adapted as a fusion sequence, this looks doable with PP? Should I be looking at BOOST_PP_LIST_FOLD_xxx macros? regards,

5 28

[Function] Takes functor by value not reference?
by Alexander Lamaison 14 Feb '10

14 Feb '10

I may be misunderstanding as the boost::function code is tricky to follow but it seems to me that the function contructor takes the functor it is passed by value rather than by reference. This requires the functor to be copyable. In my case I'm trying to pass a functor as its abstract supertype which prevents copying. Is it not possible to take the functor my reference, thereby removing the need for copying? The constructor in question appears in function_template.hpp at line 1056 in version 1.40.0: Summarised: template<typename R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_PARMS> class function<BOOST_FUNCTION_PARTIAL_SPEC> : public BOOST_FUNCTION_FUNCTION<R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_ARGS> { typedef BOOST_FUNCTION_FUNCTION<R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_ARGS> base_type; template<typename Functor> function(Functor f) : base_type(f) { } } Many thanks. Alex Lamaison -- http://swish.sourceforge.net

2 2

Re: [Boost-users] [random] number generator as static class member
by Moritz Beber 14 Feb '10

14 Feb '10

(I'm sorry for creating a new message in reply but I don't get my own messages.) The problem was of course the uniform_01 distribution which uses the underlying generator by value rather than reference. I now tie in a uniform_real in the same interval and it works just fine. Enjoy the rest of the weekend, Moritz

1 0

[Function] Overloaded functor as a boost::function
by Alexander Lamaison 14 Feb '10

14 Feb '10

I'm trying to pass a function with an overloaded operator() to a function expecting a boost::function parameter. This fails to compile as it appears to pick the wrong overload. Here is the functor interface with the two ways it can be called: class Command { public: virtual void operator()( const com_ptr<IShellItemArray>& items, const com_ptr<IBindCtx>& bind_ctx); virtual void operator()( const com_ptr<IDataObject>& data_object, const com_ptr<IBindCtx>& bind_ctx); }; When I try to pass an instance of this functor to a method expecting a boost::function of this type: boost::function<void( const com_ptr<IShellItemArray>& items, const com_ptr<IBindCtx>& bind_ctx)> I get a compiler error C2440: '=' : cannot convert from 'IShellItemArray *' to 'IDataObject *'. This seems to indicate that the boost::function is trying to pass its parameters to the second overload rather than the first. Any ideas how I can make boost::function pick the correct overload? I'm using Boost 1.40.0. Many thanks. Alex Lamaison -- http://swish.sourceforge.net

1 1

[random] number generator as static class member
by Moritz Beber 14 Feb '10

14 Feb '10

Dear list users, I have attached a reduced case of what I want to do. Basically, I thought it would be a good idea to have one general number generator that I attach to specific distributions in my member functions according to my needs (uniform_01 in this example, but I also want to use others). So for example I do the following in my real code, where 'rng' is my static member object: typedef boost::lagged_fibonacci607 rand_gen; boost::bernoulli_distribution<double> b_dist(p); boost::variate_generator<rand_gen&, boost::bernoulli_distribution<double> > bool_gen(MyClass::rng, b_dist); boost::uniform_smallint<short unsigned> in_dist(0, this->nodes_in - 1); boost::variate_generator<rand_gen&, boost::uniform_smallint<short unsigned> > in_gen(MyClass::rng, in_dist); Unfortunately, as the code I attached demonstrates, every time I do that the original state of the 'rng' seems to be used. Do you have suggestions for fixing this? Thank you in advance, Moritz

1 0

[boost.asio] Getting/setting RTS and CTS/DSR/DCD with serial_port
by David Faure 13 Feb '10

13 Feb '10

Hello, I'm porting old platform-specific serial port code to use asio::serial_port, and it's working nicely overall, but I'm missing some functionality. How can I get/set the status of the DTR, RTS and DCD pins? This seems to be needed for some devices. I'm no expert, but the current documentation says: // RTS on if Data in Transmit queue or while BREAK transmitted // RTS off iff no data in transmit queue and // Output Handshaking via CTS/DSR/DCD // data is transmitted only when all lines which are enabled for // output handshake are up. (BREAKs can be transmitted) There is also this one, can this be implemented in asio? // setErrorChar ... enable/disable Error handling and lets you define // the error replacement char. // Specifies whether bytes received with parity errors are replaced // with the character specified by the ErrorChar member. void setErrorChar(bool on, char errorChar); If it helps, the Windows implementation was: dcbI.fbFlowReplace |= MODE_ERROR_CHAR; dcbI.bErrorReplacementChar = cError; but I suppose I don't have access to the DCBINFO used inside boost, and anyway the whole purpose is to make this cross-platform :) -- David Faure | dfaure(a)kdab.com | KDE/Qt Senior Software Engineer Klarälvdalens Datakonsult AB, a KDAB Group company Tel. Sweden (HQ) +46-563-540090, USA +1-866-777-KDAB(5322) KDAB - Qt Experts - Platform-independent software solutions

2 2

metaprogramming -- does a specialization exist?
by Peter Foelsche 13 Feb '10

13 Feb '10

Is there some way to check at compile time whether a particular specialization exists? E.g. I have a template class with two template arguments which are commutative -- means they can be exchanged and the result is the same. Can I check whether type<A, B> exists and if not select type<B, A>? Peter

3 2

Sun C++ 5.10 and BGL adjacency_list
by Nicholas Brealey 13 Feb '10

13 Feb '10

I am trying to use the Boost Graph Library from Boost 1.42.0 with Sun Studio 12 update 1 but the adjacency_list.cpp example from the libs/graph/example directory does not compile. Does anyoone have a workaround? > $ CC -V > CC: Sun C++ 5.10 SunOS_i386 128229-04 2009/11/25 > Usage: CC [ options ] files. Use 'CC -flags' for details > $ CC -library=stlport4 -features=tmplife -features=tmplrefstatic -I/home/nick/boost_1_42_0 -o adjacency_list adjacency_list.cpp > "/home/nick/boost_1_42_0/boost/graph/detail/adjacency_list.hpp", line 1722: Error: Could not find a match for boost::detail::get_dispatch<boost::detail::Config, boost::detail::Base, boost::detail::Property>(boost::adj_list_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config, boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config>>, unsigned VertexProperties::*, int) needed in boost::get<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undir ectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config, boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config>, unsigned VertexProperties::*>(unsigned VertexProperties::*, boost::adj_list_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config, boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config>>&). > "adjacency_list.cpp", line 63: Where: While instantiating "boost::get<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config, boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config>, unsigned VertexProperties::*>(unsigned VertexProperties::*, boost::adj_list_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, Vertex Properties, EdgeProperties, boost::no_property, boost::listS>::config, boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>::config>>&)". > "adjacency_list.cpp", line 63: Where: Instantiated from non-template code. > "adjacency_list.cpp", line 71: Warning (Anachronism): Assigning int to boost::detail::error_property_not_found. > "adjacency_list.cpp", line 71: Error: Cannot cast from int to boost::detail::error_property_not_found. > "adjacency_list.cpp", line 82: Error: The operation "std::ostream<<boost::detail::error_property_not_found" is illegal. > "adjacency_list.cpp", line 84: Error: The operation "boost::adj_list_vertex_property_map<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::detail::error_property_not_found, boost::detail::error_property_not_found&, std::string EdgeProperties::*>[boost::detail::edge_desc_impl<boost::undirected_tag, void*>]" is illegal. > "adjacency_list.cpp", line 93: Error: The operation "std::ostream<<boost::detail::error_property_not_found" is illegal. > "adjacency_list.cpp", line 98: Error: The operation "std::ostream<<boost::detail::error_property_not_found" is illegal. > "adjacency_list.cpp", line 100: Error: The operation "boost::adj_list_vertex_property_map<boost::adjacency_list<boost::vecS, boost::listS, boost::undirectedS, VertexProperties, EdgeProperties, boost::no_property, boost::listS>, boost::detail::error_property_not_found, boost::detail::error_property_not_found&, std::string EdgeProperties::*>[boost::detail::edge_desc_impl<boost::undirected_tag, void*>]" is illegal. > 7 Error(s) and 1 Warning(s) detected. I posted a similar question on the Sun C++ forum last year but no one had a solution: http://forums.sun.com/thread.jspa?threadID=5419942 g++ works fine: > $ g++ -I/home/nick/boost_1_42_0 -o adjacency_list adjacency_list.cpp > $ g++ --version > g++ (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) > Copyright (C) 2004 Free Software Foundation, Inc. > This is free software; see the source for copying conditions. There is NO > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. > Thanks Nick

2 3

Complete vs incomplete (fwd decl'd) types
by Dominique Devienne 12 Feb '10

12 Feb '10

In client code including the header for MyDerived, I'm using a MyPtr<MyDerived>, but the static assert below was failing. It took me a little while to figure out that in myderived.h, the MyDerived::nested_type typedef is using a forward-declared MyNestedDerived type declared in my_nested_derived.h, that the client code was not including, which was the reason is_base_of was failing with the error below (Boost 1.41, VC9). Adding a #include <my_nested_derived.h> fixed it. boost\type_traits\is_base_and_derived.hpp(228) : error C2139: 'MyNestedDerived' : an undefined class is not allowed as an argument to compiler intrinsic type trait '__is_base_of' But is there a way to get a more obvious error message stating that MyNestedDerived is an incomplete type or something? In retrospect the message is not that bad, but I'm hoping there's a better way. Thanks, --DD PS: Of course there's a reason the nested type is in a separate header, and it's an idiom repeated often in our code base. template <class E> class MyPtr { public: ... typedef typename E::nested_type nested_type; explicit MyPtr(const shared_ptr<nested_type>& n) : ... { BOOST_STATIC_ASSERT((is_base_of<MyNestedBase, nested_type>::value)); } ... };

2 1

PP: LOCAL_ITERATE: how to get rid of extra empty lines
by Hicham Mouline 12 Feb '10

12 Feb '10

Hello I get the following preprocessed output # 34 "my_header.hpp" # 1 "/usr/include/boost/preprocessor/iteration/detail/local.hpp" 1 3 4 # 34 "/usr/include/boost/preprocessor/iteration/detail/local.hpp" 3 4 //result of LOCAL_macro expansion //result of macro expansion //result of macro expansion //result of macro expansion How do I get rid of the blank lines in the middle? Regards,

2 1

Boost File Locks Do not work as advertised?
by Michael Dehmlow 12 Feb '10

12 Feb '10

I'm following the example at: http://www.boost.org/doc/libs/1_35_0/doc/html/interprocess/synchronization_… for locking and writing to a file, but when a flush is called either implicitly due to a full buffer, or an explicit flush no data is written and the fstream bad() call returns true. If I write a small enough amount of data and don't explicitly flush it works because I unlock the file then call close (where an implicit flush occurs). However, if a flush occurs at any time the file lock is locked no data is written and bad() return true. [START OF CODE] #include <fstream> #include <iostream> #include <boost/interprocess/sync/file_lock.hpp> #include <boost/interprocess/sync/scoped_lock.hpp> int main() { std::string tocFile("C:\\temp\\myfile.bin"); std::fstream file; file.open(tocFile.c_str(), std::ios::out | std::ios::binary | std::ios::trunc); if (!file.is_open() || file.bad()) { std::cout << "Open failed" << std::endl; exit(-1); } boost::interprocess::file_lock f_lock(tocFile.c_str()); { boost::interprocess::scoped_lock<boost::interprocess::file_lock> e_lock(f_lock); file.write(tocFile.c_str(),tocFile.length()); if (!file.is_open() || file.bad()) { std::cout << "Write failed" << std::endl; exit(-1); } file.flush(); if (!file.is_open() || file.bad()) { std::cout << "Flush failed" << std::endl; exit(-1); } } file.close(); return 0; } [END OF CODE] -- View this message in context: http://old.nabble.com/Boost-File-Locks-Do-not-work-as-advertised--tp2757007… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

PP: is macro expansion cached?
by Hicham Mouline 12 Feb '10

12 Feb '10

Hello, Say I define a macro #define MYMACRO(x,y) .. MYMACRO because of various reasons takes very long to expand (1mn) (uses PP_FOR, PP_WHILE, PP_SEQ_FOR_EACH .). If later in the translation unit, I have a "call" to MYMACRO MYMACRO(a,b) and then a couple of lines later MYMACRO(a,b) Will it be faster to define line 15: #define MYMACRO_ab MYMACRO(a,b) and then use line 25: MYMACRO_ab . line 35: MYMACRO_ab When does MYMACRO expansion happen? Is it at line 15 once, or at lines 25 and 35? Is the answer compiler-dependent? Regards,

1 0

PP: are there known incompatibilities with vs2008?
by Hicham Mouline 12 Feb '10

12 Feb '10

Hi, I have some PP macros that expand correctly under g++4.4 but when I came to try them on vs2008sp1, it generated errors. I will dig inside the code now, but if anyone is aware of msvc-specific problems... Regards,

1 0

abandoned named_mutex
by Carla Strembicke 12 Feb '10

12 Feb '10

Hello, My question is with regard to named_mutexes that are abandoned by a process that dies abnormally. Essentially I have shared memory( utilizing boost shared memory libraries ) and am incorporating boost named mutexes to limit access to one process at a time. There is no concept of a server/client or parent/child process....the process have equal opportunity. This is Microsoft centric code that I am required to support cross-platforms( Unix-like). Microsoft's mutexes have the concept of a mutex object being abandoned if the thread terminates without releasing. Is there any support in the boost mutex to determine if the mutex has been abandoned? Option: What I am considering is perhaps before the mutex is locked, it check a file that has been created with the mutex name and the pid of the process that has it locked. If the process does not exisit the file the mutex is unlocked and the file entry is removed and the new process id is entered. I am not sure if there is a chance of a race condition I would have to scope it out. FYI - I am using 1.41. is there any changes in 1.42 ( i will be checking ). Any other suggestions? Carla Strembicke Software Developer [cid:imagea03ff0.jpg@eeae924c.570740bf] SUBNET Solutions Inc. www.SUBNET.com<http://www.subnet.com/> Making Substations More Intelligent T: 403.270.8885 F: 403.270.9631 #100, 4639 Manhattan Road SE, Calgary, Alberta, Canada; T2G 4B3 Follow SUBNET: Twitter<http://twitter.com/subnetsolutions> | Facebook<http://www.facebook.com/subnetsolutions> | Linkedln<http://www.linkedin.com/groups?gid=2733031&trk=hb_side_g> Visit the all new SUBNET website at http://www.SUBNET.com<http://www.subnet.com/> CONFIDENTIAL INFORMATION NOTICE: The information contained in this e-mail is privileged, confidential and intended solely for the use of the addressee named above. If the reader of this e-mail is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail is strictly prohibited. If you received this in error, please contact the sender and destroy all copies of this e-mail. Thank you.

2 2

Asio sharedptr heap corruption
by Lloyd 12 Feb '10

12 Feb '10

Hi, My program uses Asio to connect to a remote server. The connection class make use of sharedptr() as shown below... class Connect:public boost::enable_shared_from_this<Connect> { .... } All the async calls use sharedptr() And the object is created like this... boost::shared_ptr<Connect> Conn=boost::shared_ptr<Connect>(new Connect(IoService)); When the destructor of Conn is called the program crashes with the error... First-chance exception at 0x7c81eb33 in MYPROGRAM.exe: 0x000006D9: There are no more endpoints available from the endpoint mapper. HEAP[MYPROGRAM.exe]: Heap block at 00369710 modified at 00369918 past requested size of 200 Windows has triggered a breakpoint in MYPROGRAM.exe. What could be the reason? (If I avoid using sharedptr() and let it leak the memory, the program does not crash) Please give some hints to come out of this problem... Thanks a lot, Lloyd ______________________________________ Scanned and protected by Email scanner

2 1

[type_traits] remove_pointer from member function types?
by Rutger ter Borg 12 Feb '10

12 Feb '10

Hello, although mentioned as a tip on the function_traits documentation page, it seems that remove_pointer doesn't work for member functions. It does work for normal function pointers. See attached an example. Is this the described/correct behaviour? If so, how do I easily transform a member function pointer type to a function type? Thanks, Rutger

1 0

[proto] using proto with nvcc
by Eric Jourdanneau 12 Feb '10

12 Feb '10

Hi everybody, Nice to meet you :) I would like to test proto with nvcc, and when I compile the simple Hello world example of 1.42, I get the following error message: /usr/include/boost/proto/detail/decltype.hpp:314: error: invalid use of template-name ‘boost::type’ without an argument list. Of course it works with g++ I just don't know what to do, can somebody help me on that point? Thanks Eric

10 25