March 2011 - Boost-users - lists.preview.boost.org

[interprocess] set doesn't work when key_type is pointer.
by skywalker 26 Mar '11

26 Mar '11

Hello all, It seems that boost::interprocess::set won't compile if the key_type is pointer. However, bip::vector or std::set works fine. I'm using g++ 4.5.2 and boost 1.46.0 with linux 2.6.37. Here is my code: // compile command: g++ a.cpp -lboost_thread namespace bip = boost::interprocess; struct comp { template<class Pointer> bool operator()(Pointer a, Pointer b) const { return *a < *b; } bool operator()(int a, int b) const { return a < b; } }; typedef bip::managed_mapped_file::segment_manager segment_manager_t; #define fset(value_t) \ bip::set<value_t, comp, bip::allocator<value_t, segment_manager_t> > bip::managed_mapped_file managed_shm(bip::open_or_create, "/tmp/bip_test", 1024); void bip_example() { fset(int*) s(comp(), managed_shm.get_segment_manager()); s.insert(new int (7)); ///< error: key_type can't be pointer in bip. fset(int) s2(comp(), managed_shm.get_segment_manager()); s2.insert(int (7)); ///< ok: non-pointer is accepted. std::set<int*, comp> s3; s3.insert(new int (7)); ///< ok: int* works with STL, as well as shared_ptr, offset_ptr. } void more_example() { typedef bip::offset_ptr<int> optr; fset(optr) s3(comp(), managed_shm.get_segment_manager()); bip::allocator<int, segment_manager_t> alloc_int = managed_shm.get_segment_manager(); optr p = alloc_int.allocate(1); alloc_int.construct(p, 7); s3.insert(p); ///< error: bip doesn't work with offset_ptr too. typedef boost::shared_ptr<int> sptr; fset(sptr) s4(comp(), managed_shm.get_segment_manager()); s4.insert(sptr(new int(7))); ///< ok: shared_ptr works fine. } Won't that fail to compile, or am I missing something? Regards, Tianjiao

2 1

[Graph] Difference between DF search and visit / Graph Printing
by Marcus Riemer 26 Mar '11

26 Mar '11

Hi there! I am struggling a little understanding the difference between the DF search and visit. I thought that the DF visit would revisit "back" nodes while the DF search would not. But as I just found out that does seem to be the difference. What I want: I would like to print all following nodes for every node of a acylic graph. Lets say I have this very simple graph, I hope spaces are preserved: -> Long \ Begin / -> End \ -> -> Short / I would like to print that graph in the following Fashion: Begin .Long ..End .Short ..End I don't mind the exact order of "Long" and "Short", as long as they are both printed. So far I did this by implementing the following dfs_visitor<> [0]. I then apply the visitor with a call to boost::depth_first_visit() and my graph. However, the problem is, that the "End" node is only visited, and therefore printed, once. Of course I could roll out the DF visit myself, but as I will be demonstrating this at my unversity, I would prefer a clean solution. So what am I probably missing to print the graph in the fasion I want it? [0] http://pastebin.com/u43TtU7F

3 4

encryption with boost.iostreams
by Sean Farrow 26 Mar '11

26 Mar '11

Hi: I'm currently writing a serialization routine that needs to encrypt data. Are there any examples of using encryption with boost.iostreams. Also, given an xml file I need to determine that a file is encrypted and then decrypt the file. Anyone got any experience with this? Cheers Sean.

1 0

[units] unit conversion on construction in relation to argument passing
by alfC 25 Mar '11

25 Mar '11

Hi, Boost.Units offers automatic conversion (if conversion factor is defined) on construction of variables: quantity<cgs::lengh> A = 4.*cgs::centimeter; quantity<si::length> a(A); nice! But for some reason I believe this would work for argument functions double f(quantity<si::length> v){ return 1.;} ... quantity<cgs::lengh> A = 4.*cgs::centimeter f(A); // doesn't work! not matching function why is this? isn't the argument of the call a sort of construction argument for the function argument, or is more like a plain assignment. Is there a way to force the automatic conversion of the function call. Or I am forced to use this other long call? f(quantity<si::length>(A)); Thanks Alfredo Full code: #include<boost/units/systems/si.hpp> #include<boost/units/systems/cgs.hpp> double f(quantity<si::length> v){ return 1.; } int main(){ quantity<cgs::length> A(4.*cgs::centimeter); f(A); //not matching function return 0; }

4 4

Assertion `!pthread_mutex_destroy(&m)' failed.
by Panagiotis Foteinos 25 Mar '11

25 Mar '11

Hello users. I am new to boost:threads, so my apologies in advance. I am getting an assertion failure right before main() exits. Specifically, I get the following error: "*/usr/include/boost/thread/pthread/mutex.hpp:45: boost::mutex::~mutex(): Assertion `!pthread_mutex_destroy(&m)' failed.*" Below is the code snippet: *********************************************************************************************************************************** class Test { private: boost::mutex mtx; public: int counter; Test() { counter = 10; } void DoWork() { for(unsigned int i = 0; i < 1000000; i++) { mtx.try_lock(); counter++; mtx.unlock(); } } }; int main() { unsigned int N = 7; boost::thread* test_threads = new thread_type[N]; for(unsigned int i = 0; i < N; i++) { test_threads[i] = thread_type(boost::lambda::bind<void>(&Test::insert, &object)); } for(unsigned int i = 0; i < N; i++) { test_threads[i].join(); } return 1; } *********************************************************************************************************************************** Can somebody shed some light? Main() exits normally, if I replace mtx.try_lock() with mtx.lock(). Why this behavior? Best Regards, Panagiotis Foteinos

2 1

Fw: Boost.thread - infinite thread and expired thread object
by Emil Kaffeshop 25 Mar '11

25 Mar '11

--- On Fri, 3/25/11, Emil Kaffeshop <emildg(a)yahoo.com> wrote: > From: Emil Kaffeshop <emildg(a)yahoo.com> > Subject: Boost.thread - infinite thread and expired thread object > To: boost-users(a)lists.boost.org > Date: Friday, March 25, 2011, 7:16 PM > Hello all, > > I am new to the group and boost. > > I have a question. > > void f (void) > { > while (1) > { > //do thread job > } > } > > void start_thread () > { > boost::thread t(f); > } > > void main (void) > { > start_thread (); > pthread_exit (NULL); > } > > What are consequence for the thread "f" from disappearing > of thread object? > Does thread specific data available? > Does boost::this_thread functions available? > So is there something which f needs from "t" object? > > Thank you in advance! > > Best Regards > > > >

1 0

Connecting Windows Messages/Events with Boost callbacks
by asif saeed 25 Mar '11

25 Mar '11

Hi, I recall sometime back someone posted this question on this list and the response/answer was very good. Could someone please provide link to that earlier post? Or, alternatively, could you please post the solution again? Best regards / Asif

4 13

Re: [Boost-users] Collecting boost multi-index meta data
by Joaquin M Lopez Munoz 25 Mar '11

25 Mar '11

Gokulakannan Somasundaram <gokul007 <at> gmail.com> writes: > > > Hi, > Say i have a boost::tuple / MPL Sequence, on which i have > created a multi-index. Is it possible for me collect all the > attribute numbers on which i have declared the > IndexSpecifierList( the one passed as template parameter to > boost::multi_index::indexed_by ) ? Hi Gokul, I'm not sure I undersand your question. Have you taken a look at the index_type_list nested type? http://www.boost.org/libs/multi_index/doc/reference/multi_index_container.h… types >From this MPL sequence you can easily extract the list of key extractor types used for defining the multi_index_container, which hopefully is close to what you're after. Don't hesitate to elaborate a little more on your question and I'll see whether I can be more helpful. Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 4

Math/Statistical Distributions
by Andrej van der Zee 25 Mar '11

25 Mar '11

Hi, Statistics have been a while ago... I was reading the documentation, at least part of it, of the Boost library Math/Statistical Distributions. But I think I need a bit of help... I have many values Yi that are closely distributed around a mean value Y-mean. The collection of Yi is obtained by an algorithm that sometimes produces trash values that need to be discarded. Such false values are typically not "close" to the Y-mean. I was hoping somebody could tell what distribution I should use and how I should use this. If I recall correctly from my school-days, I need the Normal (Gaussian) Distribution. What would be the algorithm to find those trash values? Thank you, Andrej

2 2

Serialization: multiple definition of 'boost::archive::detail::init_guid...
by Marco Piacentini 25 Mar '11

25 Mar '11

Hi guys...I've found on google some people with the same problem, but all of them have a partial solution... In practice this error happens where I include a .h file which contain the directive #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> and the macro BOOST_CLASS_EXPORT() in more than one .cpp file... I have the necessity to serialize/deserialize in more than one .cpp file and at these points I have to include the .h which contains the polymorphic struct that I have to serialize/deserialize. So the partial solution proposed from others to put all the the include in a single file, is not praticable for me.. Do you know another solution?thank you very much... Bye..

2 1

Boost Threads are not parallel
by Panagiotis Foteinos 25 Mar '11

25 Mar '11

Hello users. I observed the boost threads behave strangely, depending on the way I call the function to be executed in parallel. In the following example (example #1), the threads launched are executed in parallel: **************************************************************************************** 1) class Test { int insert(){}; } object; boost::thread* threads = new boost::thread[N]; for(unsigned int i = 0; i < N; i++) { threads[i] = boost::thread(&Test::insert, &object); } **************************************************************************************** However, in the following example (example #2), the threads are executed sequentially. Only the line in the for-loop body differs from example#1. **************************************************************************************** 2) class Test { int insert(){}; } object; boost::thread* threads = new boost::thread[N]; int ret; for(unsigned int i = 0; i < N; i++) { threads[i] = boost::thread(boost::lambda::var(ret) = object.insert()); } **************************************************************************************** Can anybody explain why the threads in example#2 are executed sequentially (i.e., one at a time)? Best Regards, Panagiotis Foteinos

2 1

io_service::stop() call safe from within post()ed functor?
by Dom Driver 25 Mar '11

25 Mar '11

Hi guys, Picture a server class with the following methods (pseudocode): void Server::Work() { // Start the Async receive. mSocket.async_receive_from( *blah* ) ); // Start the Async IO service. mASIOService.run(); } void Server::Start() { // Launch a new thread to do the work. mThread = boost::thread( boost::bind( &Server::Work, this ) ); } void Server::Stop() { // Post the stop request to the io_service. mASIOService.post( boost::bind( &Server::DoStop, this ) ); // Join the comms thread. mThread.join(); } void Server::DoStop() { // Actually stop the io_service. mASIOService.stop(); } Assuming handling of packets is all correctly implemented, is it safe for the user to simply call the Stop() object method to stop the server? Basically, can a function that's been post()ed to an io_service call io_service.stop() for that io_service? TIA, Dom

1 0

[iostreams] Can I apply different char_type on a chain of filters?
by 甜瓜 25 Mar '11

25 Mar '11

Hi, all, I usually want to read a file, and group the bytes into some object (eg: struct MyData) for further use. boost::iostreams provides the filtering_stream, but it seems all filters along the chain MUST have the same char_type. Is there a way that transform a byte stream into MyData stream? eg: ifstream file("d.gz", ios_base::in | ios_base::binary); filtering_streambuf<input> in; in.push(MyDataExtractor()); in.push(gzip_decompressor()); in.push(file); class MyDataExtractor { public: typedef MyData char_type; struct category ...; template<typename Source> std::streamsize read(Source& src, char_type* s, std::streamsize n) {...} }; Thank you in advance! -- melon

1 0

[BOOST Interprocess] file lock error on MSVC 10
by Claude 24 Mar '11

24 Mar '11

Hi, I'm trying to compile simple program with MSVC++ 10 that use <boost/interprocess/sync/file_lock.hpp> include file (boost version 1.44) But I have a compile error in this file at row 68: C:\Users\Claudio\Desktop\boost_1_44_0\boost/interprocess/sync/file_lock.hpp(68): vedere il riferimento all'istanza del modello di funzione 'boost::interprocess::file_lock &&boost::interprocess::move<boost::interprocess::file_lock&>(T)' in corso di compilazione with [ T=boost::interprocess::file_lock & ] "Impossibile associare un elemento lvalue a un riferimento di rvalue" ("Can not associate a lvalue to a rvalue reference") This program compiles without error on GCC 4.5 (Linux Ubuntu) Why? -- View this message in context: http://boost.2283326.n4.nabble.com/BOOST-Interprocess-file-lock-error-on-MS… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Boost gregorian date time from preformated string
by Louis Hoefler 24 Mar '11

24 Mar '11

I use boost 1.46.0 and visual studio 2010. I try to convert a string into a date time object. I found some examples that use stringstreams. Is it possible to convert custom formatted date/time strings into gregorian date objects without using a stringstream? Thank you, Louis

1 0

Re: [Boost-users] Thread Interruption Best Practice
by Dom Driver 24 Mar '11

24 Mar '11

> if I understand your problem you want the thread blocking in select to return? Correct. I need this call to return when: 1) I've got some data waiting to be read. 2) The application is terminating and I want to stop my RX communications thread. The first one is straightforward. I'll investigate the use of pipes as a secondary option. As for 2), from other suggestions, I've gone back to the ASIO library. I read about slushies and brain-freeze, and thought it might be the way to go. However, I think I've managed to fall victim to brainfreeze myself. >From the ASIO example at: http://www.boost.org/doc/libs/1_46_1/doc/html/boost_asio/example/echo/async… This is an asynchronous server similar to what I'm trying to do here. I've got a couple of questions on this, 1) When the server object is created, the call to socket.async_read_from() is made immediately. If there is no data waiting, to be read, this call returns immediately, right? 2) When data does arrive, the handler 'handle_receive_from()' is called. Regardless of whether some bytes are read or not, we eventually call socket.async_read_from() again. This is probably quite fundamental to what I'm misunderstanding, but doesn't this mean I'll be effectively calling socket.async_read_from() from within the handler of a call to socket.async_read_from()? Will I be recursing indefinitely until the application terminates? That doesn't sound right, so I must be missing something very obvious here. Hmmm. Biblical-style ray-of-light guidance appreciated!

1 0

Project using Boost won't build with Xcode 4; complains about "global weak symbols"
by G S 23 Mar '11

23 Mar '11

Hi all. I'm using Boost in an iPhone app as a statically linked lib, and after updating to Xcode 4 the project won't build. It fails with this: *ld: bad codegen, pointer diff in boost::system::error_category:* *:error_category()to global weak symbol vtable for boost::system::error_**categoryfor architecture i386* Happens for both simulator and device, after the same project built fine under previous Xcodes. Anybody have some insight on this and a solution? Thanks! Gavin

2 3

[uBLAS] Matrix creation vs resize
by Ryan 23 Mar '11

23 Mar '11

I need to remove rows from a matrix if it fails a comparison. I have two possible options to remove the matrix rows. Option 1 adds a row to the new matrix for each row from the old matrix that passes the comparison. Option 2 keeps track of the rows from the old matrix that passed the comparison and then creates a new matrix all at once. I'm wondering which option is more efficient or is there a simpler way to achieve the same results? Ryan Option 1 void( matrix<double> const& H, matrix<double> const& P, matrix<double> const& R, vector<double> const& v) { double const rejection_criterion = 2; matrix<double> M = prod(H, matrix<double>(prod(P, trans(H))) + R; matrix<double> result; int matrix_size = 1; for (unsigned i = 0; i < v.size(); ++i) { if ( v[i] * v[i] < rejection_criterion * M(i,i)) { result.resize(matrix_size, H.size2(), true); matrix_row<matrix<double> > oldH (H, i); matrix_row<matrix<double> > newH (result, matrix_size - 1); newH = oldH; matrix_size++; } } } Option 2 void( matrix<double> const& H, matrix<double> const& P, matrix<double> const& R, vector<double> const& v) { double const rejection_criterion = 2; matrix<double> M = prod(H, matrix<double>(prod(P, trans(H))) + R; list<unsigned> goodrows; for (unsigned i = 0; i < v.size(); ++i) { if ( v[i] * v[i] < rejection_criterion * M(i,i)) { goodrows.push_back(i); } } matrix<double> result(goodrows.size(), H.size2()); for ( unsigned i = 0, list<unsigned>::iterator iter = goodrows.begin(); i < result.size1(); ++i, ++iter) { matrix_row<matrix<double> > oldH (H, *iter); matrix_row<matrix<double> > newH (result, i); newH = oldH; } }

1 0

String transformation using string_algo
by Remko Tronçon 23 Mar '11

23 Mar '11

Hi, I want to apply an (un-)escaping transformation on a string (e.g., map "<" to "<", map ">" to "<", ...). I was wondering whether boost string algorithms can provide a good way of doing this or not. I was corsidering find_format, but the problem there is that the finder has to check for occurrence of "<" (by matching a string), but then the formatter needs to check again which of the strings the finder matched ("<", ">") to emit the correct character, which sounds suboptimal. A workaround would be to make the finder always succeed, and let the formatter do the actual check and emit the string if it matches. Is there a better alternative? I also noticed there's a "transform_range_copy", but this isn't documented. thanks! Remko

1 0

any __cxa_demangle / UnDecorateSymbolName in boost ?
by Vial, Florent 23 Mar '11

23 Mar '11

Hi all, i want to retrieve the class name of some std::type_info in my application. is there any function to help me with this within boost ? I looked for it in the boost 1.46.1 code but did not find anything. My problem is that UnDecorateSymbolName is not thread safe under Windows. Thank you in advance, Florent ________________________________ Arnold & Richter Cine Technik GmbH & Co. Betriebs KG Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: HRA 57918 Persönlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH Sitz: München - Registergericht: Amtsgericht München - Handelsregisternummer: HRB 54477 Geschäftsführer: Franz Kraus; Dr. Martin Prillmann [http://www.arri.de/fileadmin/mail_images/CTBanner.gif] A new star is born – ALEXA the most versatile digital camera system. Get all the latest information from arridigital.com and YouTube.

1 0

Traversing a graph
by Vipin Varma 23 Mar '11

23 Mar '11

Hi, I have a quick question about graph traversal: what is the appropriate algorithm to be used if I want to find the shortest distance between 2 specified vertices {v_s, v_d} such that every edge on the graph is traversed at least once? I've already used the breadth_first_search() and dijkstras_shortest_paths() for other problems wherein v_s = v_d. Seemingly, these algorithms cannot be used when the source and destination vertices are dissimilar. Thanks, Vipin

2 2

serialization example
by Marco Piacentini 23 Mar '11

23 Mar '11

Hi guys..a very simple question... In the example of the serialization, before serialize, an header is linked to the data. It will be handle both in write and read operations. Is this header essential for the serialization process, or it can be omitted and I can serialize only the data? thank you...

2 2

issues when using boost.bind from boost.thread
by Sean Farrow 23 Mar '11

23 Mar '11

Hi: I'm trying to use boost.bind when starting a thread. The following code starts a thread and tries to pass in a boost.asio.io_service as well. boost::thread* t = new boost::thread(boost::bind(&boost::asio::io_service::run, &io)); I'm getting the following exception when running this code. Unhandled exception at 0x7c812afb in MDRLite.exe: Microsoft C++ exception: boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::system::system_error> > Removing the previous block of code removes the error. Having read the boost.bind, The boost.tread/boost.asio documentation I carn't seem to find what is wrong. This seems to be the final piece of the jigsaw that I carn't seem to crack! Any help appreciated. Cheers Sean.

2 1

[Statechart] Why does this state machine enter unstable state only in debug mode?
by Zhu Bicen 23 Mar '11

23 Mar '11

I am trying to use statechart in a player application. And found a weird issue, so I write following code to reproduce the issue and ask for help. 1. call the MciWndPlay win32API to play a music file within a state entry action 2. Windows msg , which indicate the player current player position, is coming 3. call process_event within Windows procedure 4. but assert failed: get_pointer(pOutermostUnstableState_) == 0 It seems the PlayingState entry action does not execute completely, The debug message after MciWndPlay call is not also printed out, but another event is coming, But This is a single thread application, the MciWndPlay is just Send Message to something and return immediately. I also print out the thread ID, all is the same. Another thing is this assert failure is only occurred in Debug Mode, in release mode all goes well. I am using Visual C++ 2008 express, and boost 1.46.1. Any comments and indication is welcome. I am confused by this issue several days. Thanks in advance. // BugFix.cpp : Defines the entry point for the application. // #include "stdafx.h" #include "BugFix.h" #include "Vfw.h" #include <iostream> #include <boost/statechart/event.hpp> #include <boost/statechart/state_machine.hpp> #include <boost/statechart/simple_state.hpp> #include <boost/statechart/state.hpp> #include <boost/statechart/custom_reaction.hpp> #include <boost/statechart/transition.hpp> #include <boost/statechart/result.hpp> #include <mmsystem.h> #pragma comment(lib, "vfw32.lib") namespace fsm = boost::statechart; #define MAX_LOADSTRING 100 // Global Variables: HINSTANCE hInst; TCHAR szTitle[MAX_LOADSTRING]; TCHAR szWindowClass[MAX_LOADSTRING]; std::wstring file_name = L"D:\\VOA\\5. Deep Six It Is Well Hidden.mp3"; // Forward declarations of functions included in this code module: ATOM MyRegisterClass(HINSTANCE hInstance); BOOL InitInstance(HINSTANCE, int); LRESULT CALLBACK WndProc(HWND, UINT, WPARAM, LPARAM); INT_PTR CALLBACK About(HWND, UINT, WPARAM, LPARAM); static std::wstring FILE_NAME = L""; struct OpenFileEvent: public fsm::event< OpenFileEvent >{ std::wstring file_name; OpenFileEvent(const std::wstring p_file_name):file_name(p_file_name){} }; struct UpdatePosEvent: public fsm::event< UpdatePosEvent >{ long pos; UpdatePosEvent(long position): pos(position){} }; class WaitingForFileState; class PlayingState; class PlayerFsm : public fsm::state_machine< PlayerFsm, WaitingForFileState > { public: friend class WaitingForFileState; friend class HandlingFileState; friend class PlayingState; }; class WaitingForFileState: public fsm::simple_state< WaitingForFileState, PlayerFsm > { public: typedef fsm::custom_reaction< OpenFileEvent > reactions; fsm::result react( const OpenFileEvent& evt) { std::cout << "OpenFileEvent is received " << std::endl; FILE_NAME = evt.file_name; return transit< HandlingFileState >(); } }; class HandlingFileState: public fsm::state< HandlingFileState, PlayerFsm, PlayingState> { public: HandlingFileState( my_context ctx ): my_base( ctx){ MCIWndOpen(_playerWnd, FILE_NAME.c_str(), NULL); } }; class PlayingState: public fsm::simple_state< PlayingState, HandlingFileState> { public: PlayingState() { std::cout << "Entering PlayingState" << std::endl; std::cout << "fsm threadId = " << GetCurrentThreadId() << std::endl; assert(MCIWndPlay(_playerWnd) == 0); // below message is not printed out in debug mode, // and another message UpdatePosEvent come // and StateChart assert failure std::cout << "After MCIWndPlay" << std::endl; } typedef fsm::custom_reaction< UpdatePosEvent > reactions; fsm::result react( const UpdatePosEvent& evt){ std::cout << "UpdatePosEvent is received in PlayingState" << std::endl; return fsm::detail::result_utility:: make_result( fsm::detail::consumed ); } private: long current_pos_; }; HWND _playerWnd; PlayerFsm _player_fsm; int APIENTRY _tWinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPTSTR lpCmdLine, int nCmdShow) { AllocConsole(); freopen("CONOUT$", "w+t", stdout); freopen("CONOUT$", "w+t", stderr); std::cout << "threadId = " << GetCurrentThreadId() << std::endl; UNREFERENCED_PARAMETER(hPrevInstance); UNREFERENCED_PARAMETER(lpCmdLine); // TODO: Place code here. MSG msg; HACCEL hAccelTable; // Initialize global strings LoadString(hInstance, IDS_APP_TITLE, szTitle, MAX_LOADSTRING); LoadString(hInstance, IDC_BUGFIX, szWindowClass, MAX_LOADSTRING); MyRegisterClass(hInstance); // Perform application initialization: if (!InitInstance (hInstance, nCmdShow)) { return FALSE; } hAccelTable = LoadAccelerators(hInstance, MAKEINTRESOURCE(IDC_BUGFIX)); // Main message loop: while (GetMessage(&msg, NULL, 0, 0)) { if (!TranslateAccelerator(msg.hwnd, hAccelTable, &msg)) { TranslateMessage(&msg); DispatchMessage(&msg); } } return (int) msg.wParam; } ATOM MyRegisterClass(HINSTANCE hInstance) { WNDCLASSEX wcex; wcex.cbSize = sizeof(WNDCLASSEX); wcex.style = CS_HREDRAW | CS_VREDRAW; wcex.lpfnWndProc = WndProc; wcex.cbClsExtra = 0; wcex.cbWndExtra = 0; wcex.hInstance = hInstance; wcex.hIcon = LoadIcon(hInstance, MAKEINTRESOURCE(IDI_BUGFIX)); wcex.hCursor = LoadCursor(NULL, IDC_ARROW); wcex.hbrBackground = (HBRUSH)(COLOR_WINDOW+1); wcex.lpszMenuName = MAKEINTRESOURCE(IDC_BUGFIX); wcex.lpszClassName = szWindowClass; wcex.hIconSm = LoadIcon(wcex.hInstance, MAKEINTRESOURCE(IDI_SMALL)); return RegisterClassEx(&wcex); } BOOL InitInstance(HINSTANCE hInstance, int nCmdShow) { HWND hWnd; hInst = hInstance; // Store instance handle in our global variable hWnd = CreateWindow(szWindowClass, szTitle, WS_OVERLAPPEDWINDOW, CW_USEDEFAULT, 0, CW_USEDEFAULT, 0, NULL, NULL, hInstance, NULL); if (!hWnd) { return FALSE; } ShowWindow(hWnd, nCmdShow); UpdateWindow(hWnd); return TRUE; } LRESULT CALLBACK WndProc(HWND hWnd, UINT message, WPARAM wParam, LPARAM lParam) { int wmId, wmEvent; PAINTSTRUCT ps; HDC hdc; switch (message) { case WM_COMMAND: wmId = LOWORD(wParam); wmEvent = HIWORD(wParam); // Parse the menu selections: switch (wmId) { case IDM_ABOUT: _player_fsm.process_event(OpenFileEvent(file_name)); //DialogBox(hInst, MAKEINTRESOURCE(IDD_ABOUTBOX), hWnd, About); break; case IDM_EXIT: DestroyWindow(hWnd); break; default: return DefWindowProc(hWnd, message, wParam, lParam); } break; case WM_CREATE: _player_fsm.initiate(); _playerWnd = MCIWndCreate( hWnd, hInst, MCIWNDF_NOTIFYPOS|MCIWNDF_NOMENU|MCIWNDF_NOTIFYMODE, NULL); break; case MCIWNDM_NOTIFYPOS: std::cout << "received MCIWNDM_NOTIFYPOS msg, threadId = " << GetCurrentThreadId() << std::endl; _player_fsm.process_event(UpdatePosEvent(lParam)); break; case WM_PAINT: hdc = BeginPaint(hWnd, &ps); EndPaint(hWnd, &ps); break; case WM_DESTROY: PostQuitMessage(0); break; default: return DefWindowProc(hWnd, message, wParam, lParam); } return 0; } -- Best Regards Bicen Zhu

2 3

Boost.Test: test suite with parameter
by Hendrik Fuß 23 Mar '11

23 Mar '11

Hi folks, I've got a suite of tests that test the functions of an abstraction layer. What I would like to do is to apply this suite to any number of concrete implementations of the abstract interface. That is, my test suite should have one parameter, say, a pointer to the abstract interface's implementation. I know I can register unary function based test cases, but how can I provide an argument to an entire test suite? I wasn't able to figure this out. Appreciate any help. cheers Hendrik

1 0