- Boost-users - lists.boost.org

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