December 2009 - Boost-users

program_options: main and other translation units
by Hicham Mouline 10 Dec '09

10 Dec '09

Hello, I have a main.cpp translation unit with the main(int argc, char* argv[]) function. My application runs in 1 of 4 modes. The mode is selected by the command line options, something like --mode1 --mode2 --mode3 --mode4 In main(), I have defined a mode_options instance of options_description, added these modes to it and defined the variables_map 'vm' that will be passed around to functions (1 for each mode) that reside in other translation units. I then get to a try/catch block, where: I store( command_line_parser(argc, argv).options(mode_options).run(), vm ); notify(vm); I have a sanity check that only 1 and at least 1 of the 4 modes is selected. I then call the function relevant to the mode selected, and pass to it 'vm'. In each of these other functions, I have other instances of options_description, reuse the same 'vm' and store again in vm the parsing ran on the appropriate options_description instance. Each mode option has specific suboptions. Question 1: I want the help option to print usage for all modes and all suboptions if no mode is selected. I want the help option to print usage for just 1 mode if a mode is selected. Question 2: I still want to program_options to throw if an unrecognized option is input, however, I throw my own exceptions if an option for mode_i is set with --mode_j for e.g. Question 3: I may eventually move to a GUI (probably c#-based) for passing options. Is there any particular advice to isolate the machinery of program_options from the functional part of my application. A big thanks to the library developer(s) and to the mailing list contributors for all the useful answers. Best regards, H

2 1

mpl: POD decoration with constructors and getters/setters
by Hicham Mouline 10 Dec '09

10 Dec '09

Hello, I posted a question to comp.lang.c++ first and now realize this maybe a mpl or PP question as well. Here is the original question: > Hicham Mouline wrote: >> I have functions that each take 1 parameter. This parameter is of type >> >> struct S { >> <arithmetic_or_bool_type_1> member_name_1; >> <arithmetic_or_bool_type_n> member_name_n; >> }; // n can be of order 10 or so >> >> class C { >> public: >> return_type apply( const S& ) const; >> }; >> >> Now I want to optimize the returned value. > > "Optimize"? > >> I would optimize by fixing any subset of the n members fixed, ie none at >> all, or just member 5, or members 4 and 8, or members 7,3 and n, or n-1 >> members of the {1....n} set. >> >> How can I do this? >> >> maybe >> >> class C{ >> public: >> pair<return_type, S> optimize(...) const; // returns the optimum S >> that gives the best return value > > What's "return_type" here? What does it mean for the return value to be > "the best"? > >> }; >> >> Some interface like this looks nice to me (as a user). >> >> The question is how to pass the fixed members to optimize. A response came as: > If your 'return_type' has to be the same *type*, there is no better > solution than to return an 'S', if the members that you don't want to > "return" are set to some "invalid" value. As an alternative you could > consider supplying another member to 'S', which would indicate the "valid" > or "active" members. You can have another (nested) struct in 'S' with bit > fields: > > struct S { > ... // as you have now > struct { > unsigned use_1:1; > unsigned use_2:1; > ... > unsigned use_n:1; > } which_members_to_use; > }; > My intial email not being clear, further clarifications are here: return_type is just double. C c; S s1; c.apply( s1 ); //// returns double , say 15.0 By optimize, I mean in most cases maximize, but sometimes minimize. So I'm trying to get the largest double by changing parts or all of s1. I was thinking of some metaprogramming magic that would generate automatically S constructors with arguments all the possible subsets of the member fields, and depending on the ctor selected by the user, would set those bits in the nested struct appropriately. This could generate also getters/setters that the optimize function could use. This could be a metafunction that returns a type SS based on the type S. I realize this is definitely a question for boost as well. The C::optimize function would then return the "optimal" S, along with the maximum/minimum double found. The details and meaning of the optimization would be specific to C::optimize. Best regards,

1 0

Boost Unit Testing Macros on Visual Studio 2005/Visual C++ 8.0
by Andrew Leer 10 Dec '09

10 Dec '09

I'm reading this article (http://www.ibm.com/developerworks/aix/library/au-ctools1_boost/?S_TACT=105A…) on the Boost Unit Testing Framework. However I'm having a bit of trouble with the first example, my guess is that they left something out (something that would be obvious to hardcore C++ coders) as IBM often does in their articles. Another possibility is that my Visual Studio 2005 C++ compiler is just too old for the example. #define _MYSTRING #define BOOST_TEST_MODULE stringtest #include <boost/test/unit_test.hpp> #include "stdafx.h" //#include "mystring.h" BOOST_AUTO_TEST_SUITE (stringtest) // name of the test suite is stringtest BOOST_AUTO_TEST_CASE (test1) { /* mystring s; BOOST_CHECK(s.size() == 0); */ BOOST_CHECK(0 == 0); } BOOST_AUTO_TEST_CASE(test2) { /* mystring s; s.setbuffer("hello world"); BOOST_REQUIRE_EQUAL('h', s[0]); // basic test */ BOOST_CHECK(0 == 0); } BOOST_AUTO_TEST_SUITE_END( ) To me the BOOST_AUTO_TEST_SUITE and BOOST_AUTO_TEST_CASE lines look a little suspect (especially since they don't have quotes around the arguments, and they are undeclared identifiers...but this probably means they are macros and I'm not certain I understand the concept or if that is available in VC++ 8.0)... Is there any reason why this code won't work? Bellow are the errors I'm receiving: 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(7) : error C2065: 'stringtest' : undeclared identifier 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(9) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(9) : error C2146: syntax error : missing ';' before identifier 'BOOST_AUTO_TEST_CASE' 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(9) : error C2065: 'test1' : undeclared identifier 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(10) : error C2448: 'BOOST_AUTO_TEST_CASE' : function-style initializer appears to be a function definition 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(18) : error C2065: 'test2' : undeclared identifier 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(19) : error C2448: 'BOOST_AUTO_TEST_CASE' : function-style initializer appears to be a function definition 1>c:\users\andy\documents\visual studio 2005\projects\unittesttests\unittesttests\unittesttests.cpp(29) : fatal error C1004: unexpected end-of-file found

5 13

socket type?
by pedro chaparro 10 Dec '09

10 Dec '09

hi, i wanna to pass the same socket decriptor among differents functions, that way all work over the same socket, but i'm no sure what kinf o data is a socket descriptor? what kind o data must retunr my function? thanks ? getsocket() { } -- Ing Pedro Alonso Chaparro Valero Ciudad Politecnica de la Innovación iTEAM - Mobile Communications Group Polytechnic University of Valencia C\ Camino de Vera S/N, Edificio 8G 46022 Valencia, Spain

2 1

[interprocess] mapped_region::flush()
by Stefan Strasser 10 Dec '09

10 Dec '09

Hi, I have a few questions regarding mapped_region. 1. first of all, is mapped_region::flush() safe to use? on linux it is implemented as msync(MS_SYNC) which seems fine, but on windows it uses FlushViewOfFile alone, which IIUC does not flush the changes to disk. MSDN: "The FlushViewOfFile function [...] does not wait to return until the changes are flushed from the underlying hardware disk cache and physically written to disk. To flush all the dirty pages plus the metadata for the file and ensure that they are physically written to disk, call FlushViewOfFile and then call the FlushFileBuffers function." I'm not sure if they really mean "hardware disk cache", since no sync or flush function can entirely make sure of that. but FlushFileBuffers is documented to flush operating system buffers. and the quote above says a call to FlushFileBuffers is necessary. 2. can the mapped memory be accessed while another thread is calling flush()? does that affect the flushing? it is still required that all data that was written to the mapped region at the point of the flush() call is written to disk when it returns. Interprocess' documentation doesn't say anything about that and I can't seem to find any information about that on the underlying functions mmap/FlushViewOfFile either. and 3, not entirely boost related: I get some strange performance results using mapped_region::flush() on linux (not tested on windows so far). I already reduced the flush() calls to a minimum but I still get better performance in many cases using my own filebuf implementation that performs paging without hardware support, using a std::filebuf. using std::filebuf shouldn't be faster than hardware file mapping. when I call flush() relatively frequently that in itself drives performance down, but when I call it less frequently it blocks for noticeable amounts of time. (thus, question 2) do the offset/size arguments to mapped_region::flush() affect performance? I could track what pages are changed, but I was under the assumption the OS already does that. Thanks

2 2

Runtime error with Boost's unit test framework
by Joey Mink 10 Dec '09

10 Dec '09

Hi all, I'm a newbie using the boost unit test framework. I'm getting a run-time exception that presents itself with the following message: Unhandled exception at 0x61c3b130 (msvcr80d.dll) in test.exe: 0xC0000005: Access violation reading location 0xcccccccc. If I step through the code, the exception is thrown the 3rd time this line from Boost is invoked: unit_test_suite_impl.hpp:111 callback0<> const& test_func() const { return m_test_func; } I can't seem to uncover what the problem is - has anyone else had this problem? Might anyone have any insight to my possible misuse of it? I'm linking with libboost_program_options-vc80-mt-gd-1_41.lib. Running Visual Studio 2005 (VC8) Thanks! Joey

4 3

boost::lambda questions
by Peter Foelsche 09 Dec '09

09 Dec '09

I would like to be able to: * return multiple results from a lambda function * use temporary variables inside a lambda function, declared using BOOST_AUTO * I would need the ability to deduce the type of the returned values via BOOST_AUTO to declare the matching variables. I think I could use boost::tuple for this: const BOOST_AUTO(sTupleVar, someThing); const BOOST_AUTO(sFirstResult, get<0>(sTupleVar)); const BOOST_AUTO(sSecond, get<1>(sTupleVar)); But how to declare a local variable inside a lambda function? Peter

2 1

Re: [Boost-users] Problem using Boost regex
by John Emmas 09 Dec '09

09 Dec '09

[Addendum...] Following my original post, firstly, apologies - I should have mentioned that I'm trying to build the regex example using Visual Studio 8 Express Edition (i.e. VC++2005). Also, I've got some more information.... Apart from those libs that I mentioned in my original post, I've found a whole bunch of libs beginning with "libboost-regex-vc80-" and ending with these suffixes:- mt-1_40.lib mt-gd-1_40.lib mt-s-1_40.lib mt-sgd-1_40.lib s-1_40.lib I guess these must be the static link libs. Anyway, I've discovered that if I link to any lib that ends in "mt-1_40.lib" I can build my app and launch it successfully as long as I build in Release mode. But in Debug mode, the app either doesn't launch - or if it does launch, it gives the unhandled exception that I described earlier. Hope that helps, John

4 19

ASIO async I/O with iostream?
by Jay Sprenkle 09 Dec '09

09 Dec '09

Good afternoon all, I'm interested in writing a program using asio. I'm new to both boost and asio. Is it possible to use async I/O with streams? A stream is a good method for what I'm interested in doing but I don't see how to make it work. What I've got so far: A class using event driven async I/O. The class parses the data into variable arbitrary sized messages. I read from the tcp socket into a buffer, then create a stream from the buffer, then deserialize the data from the stream. unsigned char buffer[1024]; // connect receive event socket->async_read_some( boost::asio::buffer(buffer), boost::bind( &TcpConnection::receive, this, boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred ) ); I'd like to be able to skip the buffer and just read directly to the stream with an event to notify me of the arrival of data. Thanks! -- Cause united breaks guitars http://www.youtube.com/watch?v=5YGc4zOqozo

2 3

Bug in mingw with gcc 4.3.0: boost_concept_check.h
by Eloi Du Bois 09 Dec '09

09 Dec '09

Hi all, I found a probable bug in mingw. Here is the minimal code that can not compile: #include <boost/ptr_container/ptr_list.hpp> > > #include <iostream> > > class A > { > public: > A(const int id):_id(id){} > virtual ~A()=0; > bool operator==(const A& other) const { return _id == other._id; } > int _id; > }; > > A::~A(){} > > class B : public A > { > public: > B(const int id):A(id){} > virtual ~B() {} > }; > > class C : public A > { > public: > C(const int id):A(id){} > virtual ~C() {} > }; > > int main() { > typedef boost::ptr_list<A> AList; > AList a,b; > a.push_back( new B(0) ); > a.push_back( new C(1) ); > b.push_back( new B(0) ); > b.push_back( new C(1) ); > > return a == b; > } > And the compilation error: gcc.compile.c++ > T:\Analyse1\dist\lib\bin\gcc-mingw-4.3.0\debug\threading-multi\src\main.o > c:\qt\mingw\bin\../lib/gcc/mingw32/4.3.0/include/c++/bits/boost_concept_check.h: > In instantiation of '__gnu_cxx::_EqualOpConcept<A, A>': > c:\qt\mingw\bin\../lib/gcc/mingw32/4.3.0/include/c++/bits/boost_concept_check.h:62: > instantiated from 'void __gnu_cxx::__function_requires() [with _Concept = > __gnu_cxx::_EqualOpCon > cept<A, A>]' > c:\qt\mingw\bin\../lib/gcc/mingw32/4.3.0/include/c++/bits/stl_algobase.h:943: > instantiated from 'bool std::equal(_II1, _II1, _II2) [with _II1 = > boost::void_ptr_iterator<std::_List_ > const_iterator<void*>, const A>, _II2 = > boost::void_ptr_iterator<std::_List_const_iterator<void*>, const A>]' > T:\Analyse1\3rdParty\boost/boost/ptr_container/detail/reversible_ptr_container.hpp:531: > instantiated from 'bool > boost::ptr_container_detail::reversible_ptr_container<Config, CloneA > llocator>::operator==(const > boost::ptr_container_detail::reversible_ptr_container<Config, > CloneAllocator>&) const [with Config = > boost::ptr_container_detail::sequence_config<A, std:: > list<void*, std::allocator<void*> > >, CloneAllocator = > boost::heap_clone_allocator]' > tests\src\main.cpp:38: instantiated from here > c:\qt\mingw\bin\../lib/gcc/mingw32/4.3.0/include/c++/bits/boost_concept_check.h:296: > error: cannot declare field '__gnu_cxx::_EqualOpConcept<A, A>::__a' to be of > abstract type 'A' > tests\src\main.cpp:6: note: because the following virtual functions are > pure within 'A': > tests\src\main.cpp:14: note: virtual A::~A() > c:\qt\mingw\bin\../lib/gcc/mingw32/4.3.0/include/c++/bits/boost_concept_check.h:296: > error: cannot declare field '__gnu_cxx::_EqualOpConcept<A, A>::__b' to be of > abstract type 'A' > tests\src\main.cpp:6: note: since type 'A' has pure virtual functions > > "g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -mthreads > -DBOOST_SYSTEM_DYN_LINK=1 -I"T:\Analyse1\3rdParty\boost" -I"tests\src" -c > -o "T:\Analyse1\dist\lib\bin\gcc-mingw- > 4.3.0\debug\threading-multi\src\main.o" "tests\src\main.cpp" > The problem might come from this: #define _GLIBCXX_DEFINE_BINARY_OPERATOR_CONSTRAINT(_OP,_NAME) \ template <class _Ret, class _First, class _Second> \ struct _NAME { \ void __constraints() { (void)__constraints_(); } \ _Ret __constraints_() { \ return __a _OP __b; \ } \ _First __a; \ _Second __b; \ } I hope it comes from me and not from boost... If anyone have an idea of what I can do to solve this, It would be great :) Regards, Eloi.

1 0