May 2010 - Boost-users - lists.boost.org

BCP question
by Robert Ramey 06 May '10

06 May '10

I'm investigating BCP and have at least one question. When I use --list with BCP to generate a list of dependencies for the serialization library, I get among otherthings: ... boost/iterator.hpp boost/iterator/detail/config_def.hpp boost/iterator/detail/config_undef.hpp boost/iterator/detail/enable_if.hpp boost/iterator/detail/facade_iterator_category.hpp boost/iterator/filter_iterator.hpp boost/iterator/interoperable.hpp boost/iterator/iterator_adaptor.hpp boost/iterator/iterator_categories.hpp boost/iterator/iterator_facade.hpp boost/iterator/iterator_traits.hpp boost/iterator/transform_iterator.hpp ... The first one - boost/iterator.hpp - what I wonder about. I searched the files in used in the serialization library and I could find on direct usage of the convenience header "boost/iterator.hpp". I'm guess that this is a secondary dependency introduced by something I've included directly. So a) Is this a good guess? b) Is there an easy way of determining which of the things I'm including is including boost/iterator.hpp? I've taken care while working on the serialization library to minimize gratuitous dependencies. It seems that other libraries might be including the "convenience headers" which subverts my intention to minimize dependencies. Would this not be a good argument for library writters to refrain from using these convenience headers? Robert Ramey BTW - I had never looked at BCP before. An incredible accomplishment to write such a thing - correctly. RR

3 2

[spirit] need help with the parser
by Andrzej Krzemienski 06 May '10

06 May '10

Hi, Could someone help me understand why my fairly simple spirit::qi parser does not work as I expect? I try to parse string "2 c" into the structure: struct Data { char val; }; std::map<int, Data> map; such that the map contains one element with an int 2 as key and char "c" wrapped into structure Data as the value. I am using Boost 1.42. The rules for the grammar w/o semantic actions are as follows. value = qi::char_; list = (qi::int_ >> value); // the start rule I know the grammar could be expressed more simply; my real life example is more complex and makes sense, but I tried to reduce it as much as possible. The full grammar definition reads: typedef std::string::const_iterator Iter; BOOST_FUSION_ADAPT_STRUCT(Data,(char, val)); struct TestGrammar : qi::grammar<Iter, std::map<int,Data>(), ascii::space_type> { qi::rule<Iter, std::map<int,Data>(), ascii::space_type> list; qi::rule<Iter, Data(), ascii::space_type> value; TestGrammar() : TestGrammar::base_type(list) { value = qi::char_[ at_c<0>(_val) = _1 ]; list = (qi::int_ >> value)[ _val[_1] = _2 ]; } }; The code that does the parsing reads int main( ) { std::string input = "2 c"; Iter iter = input.begin(), end = input.end(); TestGrammar grammar; std::map<int,Data> map; bool ans = phrase_parse( iter, end, grammar, space, map ); assert( ans ); assert( iter == end ); BOOST_AUTO( it, map.find(2) ); assert( it != map.end() ); assert( it->second.val == 'c' ); // FAILS! } I marked my expectations with assertions. The parsing does succeed, i.e. phrase_parse returns true and the entire input is consumed, the key of the map's element is right, but its value is somehow lost. Can anyone tell what I am missing? I enclose the entire program code in the attachment. Thanks in advance for help. Regards, &rzej

2 2

[BGL] std::pair<edge_descriptor, bool> edge()
by Hossein Haeri 05 May '10

05 May '10

Dear all, Am I right in believing that the bool returned in the function below is meant to signal to the user whether there is an edge between u and v or not? std::pair<edge_descriptor, bool> edge(vertex_descriptor u, vertex_descriptor v, const adjacency_list& g) The only explanation that the online documentation provides for this function is: "Returns an edge connecting vertex u to vertex v in graph g." Perhaps somebody should amend that explanation... Best, --Hossein

2 3

boost uuids
by Carla Strembicke 05 May '10

05 May '10

I am trying to convert a legacy GUID on mircosoft to use the boost uuid. I am utilizing the stringstream as mentioned in the boost uuid documentation. I have managed to get the GUID converted into a string but am unable to inject the string value Into the uuid variable. Below is just a snippet of a test program. I read it back into a buffer variable to make sure that value was properly set. ss gets the value of guid_str but does not set the value in uuid_test. WCHAR guid_buf[50]; Boost::uuids::uuid uuid_test; StringFromGUID2(guid_value, guid_buf, 50); char *guid_str = OLE2A(guid_buf); std::stringstream ss; char buf[50]; ss << guid_str; ss >> uuid_test; ss << uuid_test; ss >> buf; Obvioulsy I am using stringstream wrong. Does anyone know how to convert a GUID to a boost UUID? I can't seem to find the documentation or interface. 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.

1 0

[asio] Timer cancellation behavior
by BKnoth 05 May '10

05 May '10

Environment: Windows XP Boost 1.40 Problem: Canceling an asio::deadline_timer doesn't pass an operation_aborted value to the timer handler. --------------- From the async_wait doc: For each call to async_wait(), the supplied handler will be called exactly once. The handler will be called when: * The timer has expired. * The timer was cancelled, in which case the handler is passed the error code boost::asio::error::operation_aborted. --------------- When I call my Restart() function (which calls the cancel() function) my TimerHandler() receives an err value of 0. When the timer is allowed to expire, the TimerHandler doesn't get called at all. Thoughts? - Bruce Here's the relevant code: boost::asio::io_service mIOService; boost::asio::deadline_timer mTimer(mIOService, boost::posix_time::seconds(3)) inline void TimerHandler(const boost::system::error_code& err) { cout << err; if (err && err != boost::asio::error::operation_aborted) { cout << endl << "Watchdog elapsed" << endl; } } void Start() { mTimer.async_wait(TimerHandler); mIOService.run(); } // Cancels and restarts the watchdog timer - prevents a timeout void Restart() { mTimer.cancel(); mTimer.async_wait(TimerHandler); mIOService.reset(); mIOService.run(); }

2 2

[statechart] exception_translator < >
by Georg gast 05 May '10

05 May '10

Hi all! Currently i'm working on a project and i'm using boost::state_chart for implementing my asyncrone statecharts. Everything seems ok so far. But i have a small problem. My used client library, which is handled in the statemachine, throws sometime an exception. I said to me: "Not a problem, i will use an exception_translator." At home i made a simple project under Linux/g++ 4.4 (-Wall -pedantic) and MSVC2008 . No problems. At work we use MSVC2005 SP1 and boost 1.42.0. There i cant compile my simple statechart example. (currently i dont have the error messages here at my PC). The documentation states at http://www.boost.org/doc/libs/1_42_0/libs/statechart/doc/tutorial.html#Exce… <----------------------------------------> Discriminating Exceptions .... Unfortunately, this idiom (using throw; inside a try block nested inside a catch block) does not work on at least one very popular compiler. If you have to use one of these platforms, you can pass a customized exception translator class to the state_machine class template. This will allow you to generate different events depending on the type of the exception. <----------------------------------------> and under http://www.boost.org/development/tests/release/developer/statechart.html All tests "pass" for MSVC2005SP1 but i can't find a exception_translator test there (or i'm too blind) ;) So my question: Is MSVC2005SP1 "this very popular compiler" which cant handle exception_translators? @Andreas: Thanks for that very useful library! :-D Best Regards Georg Gast

2 2

Rv: [Boost MPI]Problems with instalation the boost in visual 2008
by Juliet Villanueva 05 May '10

05 May '10

--- El mié 5-may-10, Juliet Villanueva <wizard_pag(a)yahoo.com.mx> escribió: De: Juliet Villanueva <wizard_pag(a)yahoo.com.mx> Asunto: [Boost MPI]Problems with instalation the boost in visual 2008 A: boost-users(a)lists.boost.org Fecha: miércoles, 5 de mayo de 2010, 18:13 Hello: I'm trying to install the boost libraries MPI, but so far I have not been successful. So far the steps I performed are: 1) Run the file in the folder bjam.exe of boost, this procedure creates libs, some of the folders boost, but skips the folder of MPI, which does not generate the appropriate libs. 2) When the library is installed tells me that you must modify the user-config.jam, adding "usign MP" I, for installation. 3) After editing the file is placed bjam command-with-mpi install, and start the installation indicating that boost jump MPI library. Another way I try it from the compiler to use Visual stdio 2008, from the terminal set, follow these steps: 1) Set within the boost folder 2) Run the file bajm.exe, here starts the installation again skipping the boost folder MPI. 3) When the library is installed tells me that you must modify the user-config.jam, adding "usign MP" I, for installation. 4) After editing the file put the following command: C: \ boost_1_42_0 \ boost_1_42_0> bjam - build-type = complete - toolset = MSVC - build-dir = C: \ src \ boost_1_42_0 \ build \ 90 \ 32 - stagedir = C: \ src \ boost_1_36_0 \ stage \ 90-32 stage. This line starts installing again but skip the folder you need. The truth is no longer what happens, because so far I can already detect the libraries, all I need are generating. Lib, which tells me that the compiler can not locate the libs are missing me the following: • libboost_mpi-vc90-mt.lib • libboost_mpi-vc90-mt-1_36.lib • boost_mpi-vc90-mt.lib • boost_mpi-vc90-mt-1_36.dll • boost_mpi-vc90-mt-1_36.lib Someone can help me fix this, scheduled to begin in MPI boost, I would greatly appreciate. Julieth

1 0

[serialization][1.42.0] gcc(4.01)/xcode(3.1.2) failure to register type
by Jeff Flinn 05 May '10

05 May '10

I've followed the doc's 'Static Libraries and Serialization' approach. To move the serialize implementation from the header to the cpp file for my classes in my hierarchy which are serialized through base class pointer. Which runs on MSVC8, after linking with /OPT:NOREF. Unfortunately the BOOST_CLASS_EXPORTxxx facilities appear to be optimized away on gcc/xcode, even though all of the 'strip' settings are off. What approach are others taking to resolve this? Robert's reply to Emil Dotchevski on Sep 25, 2008 suggests that explicitly calling ar.register(static_cast<myType *>(0)); is the only portable way to ensure registration, is this still the case? Should explicit registration be done in combination with or en lieu of BOOST_CLASS_EXPORTxxx? Thanks, Jeff

2 6

[Serialization]Substituting std::istringstream for text_iarchive?
by Laszlo Orban 05 May '10

05 May '10

We have encountered a problem during deserialization via boost::archive::text_iarchive. It is somehow related to the gcc (4.4.1 via MinGW) Windows implementation of the std::stream, that is the parameter of the constructor of text_iarchive. The exact same code caused no problems on a Linux machine (with gcc 4.4.1 as well). The setup is a multi-threaded application (that causes the issue) with frequent deserializations on different threads. The problem is in short: 1: MessagePtr msg; // the structure to be deserialized below 2: std::istringstream archive_stream(buffer); // buffer contains the data, it is a simple char[] array 3: boost::archive::text_iarchive archive(archive_stream, 0); //constructor of the text_iarchive for deserialize 4: archive >> msg; // the deserialization The line causing the problem is the second and third one: in the MinGW implementation of istringstream handling there seems to be some static variables working in the background, and without proper protection of these lines, eventually the application is terminated improperly. Important, that the buffer above is thread-binded, thus it is surely not modified in the background. Using global boost::mutex-es solve the problem: boost::mutex istringstream_locker; //global , local mutexes doesn't work as the behaviour is bound to the std::stream implementation (...) istringstream_locker.lock(); std::istringstream archive_stream(buffer); boost::archive::text_iarchive archive(archive_stream, 0); istringstream_locker.unlock(); (...) However this being a frequently used code, we are wondering if there is any better (more time effective) solution or workaround. (Note: we tried using the boost/format/alt_sstream, but we couldn't get it to work.) Or maybe we can substitute text_iarchive with something else that doesnt require an std::stream? Thanks for reading this!

2 1

How to declare a structure containing vertex or edge properties?
by Eric Fowler 05 May '10

05 May '10

Sorry if this is a FAQ. I have done this in the past, but can't remember the exact syntax, and don't have the old code. It is possible to declare a struct and use it to hold vertex or edge properties, rather than using the property_map interface, e.g. struct tagFOO { int length; byte color; }FOO; etc. How can I do this? Where is the doc on it? Thanks Eric

2 1