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

Boost.Log: Set formatter for named_scope attr.
by David Puffer 14 Nov '13

14 Nov '13

Hi, Being new to Boost.Log (and Boost in general) I have spent the last few days trying to implement Boost.Log - which Is working fairly well by now. However - I encountered one problem that I've not been able to solve by now, although it seems to me quite trivial: I have successfully registered the "named_scope" attribute and I am populating its value by a call to BOOST_LOG_FUNCTION. This works well and I can see the function-name in my log-output. However - I also want the line-number and the file-name, and discovered that there's a possibility to set the output-format of the "named_scope" attribute (which already incorporates file-name and line-number info) like so: #include <boost/log/expressions/formatters/named_scope.hpp> sink->set_formatter(expr::stream << expr::format_named_scope("Scopes", "%n")); As documented here: http://www.boost.org/doc/libs/1_54_0/libs/log/doc/html/log/detailed/express… I just don't seem to find any way to get to my sink - which is configured in a settings-file and evaluated by a call to the init_from_stream() function. Is there any way to query for all the sinks - or get a sink by name? How is one supposed to set a sink formatter for a sink which was generated by the library and not the user? Or is there a way to specify the output-format for named attributes via a settings-file as well? As far as I can see it's only possible to format the output of a log record as a whole - but not single attributes. Also: Is it correct, that the above function will only set the format for the "Scopes" attribute - which means I could call this function repeatedly for lets say the "TimeStamp" attribute as well? Best regards, David

2 2

[Log] Error with operator<< in basic_formatting_ostream
by Víctor Yo 14 Nov '13

14 Nov '13

Hi ewveryone, I'm working lately with Boost.Log (I'm a newbie) library and I want to write a logger with some parameters and formatting options. I followed the documentation, I created a backend and a frontend and I tried to format the output in the frontend with the set_formatter() function. I tried with the Lambda style and with the Boost.Format style but I'm getting an error that I can not locate. Iwas working on it for the last two whole days but I couldn't find a solution. It is related with the basic_formatting_ostream, but I don't understand what that means. I defined the backend and attached to it two streams (to a file and to cout) typedef boost::log::sinks::text_ostream_backend our_backend; typedef boost::log::sinks::synchronous_sink<our_backend> our_frontend; //............................................................................................................................................ boost::shared_ptr<our_backend> backend = boost::make_shared<our_backend>(); //Stream definitions boost::shared_ptr<std::ostream> stream_out(&std::cout, boost::log::empty_deleter()); boost::shared_ptr<std::ostream> stream_file(new std::ofstream("file_logger.log")); backend -> add_stream(stream_out); backend -> add_stream(stream_file); Created the frontend (and wrapping the backend to it) //Wrap the backend into the frontend and register in the core. //The frontend provides synchronization for the backend. boost::shared_ptr<our_frontend> frontend(new our_frontend(backend)); And now come the offending code which is supposed to format the output (note that the first commented block was my first try). None of them works and the error is the same in both cases: */*frontend -> set_formatter* * (* * boost::log::expressions::format("[%1%] %2%, File: %3%, Line: %4% <%5%>: %6% ")* * % boost::log::attributes::counter< unsigned int >(1)* * % boost::log::attributes::local_clock()* * % boost::log::expressions::attr< severity_level >("Severity")* * % boost::log::expressions::message* * );*/* //We are going to define the formatter in a different style to try to get formatted output frontend -> set_formatter ( boost::log::expressions::stream << boost::log::attributes::counter< unsigned int >(1) <<" ["<< boost::log::attributes::local_clock() <<"] <"<< boost::log::expressions::attr< severity_level >("Severity") <<"> "<< boost::log::expressions::message ); The error I get is the next one: */home/victor/Programacion/workspace/opice/build-parallel/../libs/include/boost/log/utility/formatting_ostream.hpp:664:5: error: no match for ‘operator<<’ in ‘(& strm)->boost::log::v2_mt_posix::basic_formatting_ostream<CharT, TraitsT, AllocatorT>::stream [with CharT = char, TraitsT = std::char_traits<char>, AllocatorT = std::allocator<char>, boost::log::v2_mt_posix::basic_formatting_ostream<CharT, TraitsT, AllocatorT>::ostream_type = std::basic_ostream<char>]() << value’* And I really don't know what is happening. I also included a lot of files (I'm sure I included too much files, indeed) to be sure it was not caused by some undefined operator or something similar. However I still have the same problem. The includes are: #include <iostream> #include <fstream> #include <string> #include <boost/thread/shared_mutex.hpp> #include <boost/log/core.hpp> #include <boost/log/common.hpp> #include <boost/make_shared.hpp> #include <boost/log/expressions.hpp> #include <boost/log/utility/setup/file.hpp> #include <boost/log/utility/setup/console.hpp> //puede ser innecesario #include <boost/log/utility/setup/common_attributes.hpp> //puede ser innecesario #include <boost/log/utility/empty_deleter.hpp> #include<boost/thread/mutex.hpp> #include <boost/log/attributes/timer.hpp> #include <boost/log/attributes/mutable_constant.hpp> #include <boost/log/sources/severity_logger.hpp> #include <boost/log/utility/formatting_ostream.hpp> #include <boost/log/sinks/text_ostream_backend.hpp> #include <boost/log/sinks/sync_frontend.hpp> #include <boost/log/support/date_time.hpp> As you can see, they are probably redundant or innecessary but I could rip off that files when I could get the program compiled. Now, I'm out of ideas so any help you could bring me should be very appreciated. Thank you in advance. Victor Hevia

2 4

[1.55] PDF Documentation build available.
by John Maddock 14 Nov '13

14 Nov '13

The final PDF documentation build for Boost-1.55 is now available at https://sourceforge.net/projects/boost/files/boost-docs/1.55.0/ As ever this only contains documentation for the subset of Boost libraries that use quickbook/boostbook/docbook based documentation. Enjoy! John.

1 0

Boost 1.54.0 Header Warnings: Shadowed Variables
by Tom Browder 13 Nov '13

13 Nov '13

I am trying to use Boost 1.54.0 to replace a previous version (1.47) and am using g++ 4.7 with most warnings turned on. I am getting many -Wshadow errors and wonder why they exist. Surely the developers test for such. Here is one example from xpressive/detail/utility/hash_peek_bitset.hpp: Both 'icase' and 'count' are reported to shadow members of 'this'. Note that the bool arg is named 'icase' but the compiler reports 'this' has a member named 'icase' (as can be plainly seen below). In this example it's fairly clear what is intended, but then again maybe not (and it doesn't give the user much confidence in the code). It isn't good practice IMHO to do such so that is why I like to use some prefix or suffix to disambiguate the two variables. // Make sure all sub-expressions being merged have the same case-sensitivity bool test_icase_(bool icase) { std::size_t count = this->bset_.count(); if(256 == count) { return false; // all set already, nothing to do } else if(0 != count && this->icase_ != icase) { this->set_all(); // icase mismatch! set all and bail return false; } this->icase_ = icase; return true; } I could file many bugs but it begs the question of why the warnings are allowed in the first place. Suggestions? Thanks and best regards, -Tom

8 17

[Boost.Iterator] Problem compiling zip_iterator with clang and C++11
by Jonathan Jones 13 Nov '13

13 Nov '13

Hi folks, The following code fails to compile on OSX 10.8.5 with Xcode 5.0.1 and Boost 1.49.0. This is not real world code, but it demonstrates the problem I'm facing. #include <boost/iterator/zip_iterator.hpp> template<typename T> void foo() { boost::zip_iterator<boost::tuple<T*, T*> > iter; std::fill(iter, iter, std::make_pair(T(),T())); } void bar() { foo<int>(); } Here are the commands I used to compile (if I drop "-stdlib=libc++" from the compile line, the code compiles just fine): setenv DEVELOPER_DIR /Applications/Xcode5.0.1.app/Contents/Developer xcrun -sdk macosx10.8 clang++ -c zip_iterator.cpp -I<boost-path>/include -std=c++11 -stdlib=libc++ and here is a heavily formatted snippet of the key error: /Applications/Xcode5.0.1.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/c++/v1/algorithm:1824:1: note: candidate function [with _RandomAccessIterator = boost::zip_iterator<boost::tuples::tuple<int*, int*>>, _Tp = std::__1::pair<int, int>] not viable: no known conversion from typename iterator_traits<zip_iterator<tuple<int*, int*>>>::iterator_category (aka 'boost::detail::iterator_category_with_traversal<std::__1::input_iterator_tag, boost::random_access_traversal_tag>') to 'std::__1::random_access_iterator_tag' for 4th argument __fill(_RandomAccessIterator __first, _RandomAccessIterator __last, const _Tp& __value_, random_access_iterator_tag) Any help is appreciated. Thanks, Jon

1 0

[Math] BOOST_FPU_EXCEPTION_GUARD on Linux
by David Roberts 13 Nov '13

13 Nov '13

Hello, I have a program that makes considerable use of the gamma distribution from Boost.Math. Profiling with OProfile has shown that, on Linux, this program spends about 20% of its runtime calling the functions fesetexceptflag() and feclearexcept(). The code responsible for calling these is in the BOOST_FPU_EXCEPTION_GUARD macro which is defined in boost/math/tools/config.hpp. I have looked back through the source history to see when and why this macro was added, and it was in changeset 3186 from September 2006 with the comment "More linux fixes: finally got AMD64 machines to behave correctly" (see https://svn.boost.org/trac/boost/changeset/3186) In Boost 1.54 the conditions under which the guard macro is defined have been refined since it was originally added, but the guard class itself remains the same as in 2006. Given the performance penalty I am suffering as a result of constantly clearing and reseting the floating point exception flags, I have a couple of questions about this functionality: 1) Can anyone provide a specific test case that did not work without the BOOST_FPU_EXCEPTION_GUARD functionality but was fixed by its addition? It would be very useful to have such a test case because then I can check if this is still a problem on recent versions of Linux. 2) The commit comment for changeset 3186 refers to AMD64. Does this mean the original problem only occurred on AMD CPUs, or does this refer to the architecture, and hence it affected both Intel and AMD CPUs? Many thanks, David Roberts

2 1

[Graph] Static allocation of nodes and edges
by Sensei 13 Nov '13

13 Nov '13

Dear all, I need to optimize my code with regards to both memory and time. The problem is simple: pre-allocating the adjacency matrix. The graph may have a huge number of nodes N, but the number of edges K per node is limited. While N can be in the order of one hundred million, K is at most 10. The graph will be directed, so no symmetry is needed. Do you know if I can pre-allocate such an adjacency matrix? If not, I suppose I must use an adjacency list in order to minimize memory consumption, am I right? Thanks!

2 9

Help with lockfree queue
by TYSON Kieran 13 Nov '13

13 Nov '13

[@@ OPEN @@] Hello, I am using a global lockfree queue to transfer a message from different threads. The program compiles and runs without fault up until the program exits and the queues destructor is called. An unhandled exception is thrown on the destruction of the queue. Below is the code used to set up the queue. #define B1_OUTPUT_LENGTH_PER_CHANNEL (32000) #define B2_OUTPUT_LENGTH_PER_CHANNEL (32000) #define B3_OUTPUT_LENGTH_PER_CHANNEL (64000) #define NB_CHANNELS (8) typedef struct s_OutputMsg { float dataB1[(int)B1_OUTPUT_LENGTH_PER_CHANNEL*NB_CHANNELS]; float dataB2[(int)B2_OUTPUT_LENGTH_PER_CHANNEL*NB_CHANNELS]; float dataB3[(int)B3_OUTPUT_LENGTH_PER_CHANNEL*NB_CHANNELS]; int dataLength; } OutputMsg; boost::lockfree::queue<OutputMsg, boost::lockfree::fixed_sized<true>> dataMsgQueue(100); My first thoughts are that there are limitations on the size of the message because I have used simple integer queues in the past with no problem. Are there size limitations on the message of a lockfree queue? Thanks, Kieran This message contains OPEN information that is not sensitive and can be freely accessed by people both inside and outside of the Thales Group This email was classified by Tyson, Kieran on Wednesday, 13 November 2013 10:58:41 AM ------------------------------------------------------------------------- DISCLAIMER: This e-mail transmission and any documents, files and previous e-mail messages attached to it are private and confidential. They may contain proprietary or copyright material or information that is subject to legal professional privilege. They are for the use of the intended recipient only. Any unauthorised viewing, use, disclosure, copying, alteration, storage or distribution of, or reliance on, this message is strictly prohibited. No part may be reproduced, adapted or transmitted without the written permission of the owner. If you have received this transmission in error, or are not an authorised recipient, please immediately notify the sender by return email, delete this message and all copies from your e-mail system, and destroy any printed copies. Receipt by anyone other than the intended recipient should not be deemed a waiver of any privilege or protection. Thales Australia does not warrant or represent that this e-mail or any documents, files and previous e-mail messages attached are error or virus free. -------------------------------------------------------------------------

2 1

How to convert cmyk jpg file to rgb jpg file with BOOST.GIL?
by Guangchao Wang 13 Nov '13

13 Nov '13

I used following code: rgb8_image_t rgb8Img; jpeg_read_and_convert_image("CMYK.jpg", rgb8Img); //CMYK.jpg is CMYK image jpeg_write_view("RGB.jpg", const_view(rgb8Img)); to convert CMYK image to RGB image. The saved RGB.jpg is not correct. Could you give me some suggestion? Thanks.

2 2

[Locale] Getting correct system locale under Mac OS X
by Orhun Birsoy 11 Nov '13

11 Nov '13

I am using boost 1.55.0.rc2. I am running 10.8.5 and when I select "Norway (Norwegian Bokmal)" (nb_NO) as the region, LANG env. variable is undefined, making boost::locale::generator("") to generate a "C" locale. I can of-course get the locale identifier from NSLocale (or CFLocale) and initialize the boost::locale::generator with "nb_NO" which seems to be working, but with Mavericks things got even more interesting. It looks like, in 10.9 you can mix and match language with a country anyway you like i.e. I can select "Norway" as my region and "English" as my language and again LANG is empty but NSLocale returns "en_NO". But this time when I use boost::locale::generator("en_NO"), boost::locale::as::number uses "." (dot) as decimal separator, although NSLocale reports "," (comma). 1. Can boost::locale handle such combinations (en_NO, en_TR, etc)? (I am using ICU backend, others are explicitly turned off). a. for number formatting? b. for messages formatting (translations)? which mo will it pick for en_TR ? 2. How can I 'customize' locale generated by boost::locale to reflect the changes user might do in system locale settings (i.e. select en_US but change the decimal symbol)? Thanks. -- Orhun Birsoy

2 1

Thanks to everyone for all the work on 1.55.0
by Marshall Clow 11 Nov '13

11 Nov '13

I just wanted to take off my "release manager" hat and say, personally, thank you to all the people who tried out the RCs/Betas/etc, reported bugs, and suggested fixes. You were all a big help in making the release. -- Marshall Marshall Clow Idio Software <mailto:mclow.lists@gmail.com> A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait). -- Yu Suzuki

1 0

Boost 1.55.0 released!
by Marshall Clow 11 Nov '13

11 Nov '13

Release 1.55.0 of the Boost C++ Libraries is now available. These open-source libraries work well with the C++ Standard Library, and are usable across a broad spectrum of applications. The Boost license encourages both commercial and non-commercial use. This release contains one new library and numerous enhancements and bug fixes for existing libraries. === New Libraries === • Predef: This library defines a set of compiler, architecture, operating system, library, and other version numbers from the information it can gather of C, C++, Objective C, and Objective C++ predefined macros or those defined in generally available headers, from Rene Rivera. === Known bugs with Visual Studio 2013/Visual C++ 12 === Visual Studio 2013 was released quite late in the release process, so there exist several unresolved issues. These include: • Serialization can't compile because of a missing include. • Using has_member_function_callable_with from Boost.Container's allocator_traits causes a compile error (#9332). • For several libraries overloads with intializer lists are incorrectly reported as ambiguous. This is a Visual C++ bug and it isn't clear if there's a good workaround. • In libraries such as Unordered and MultiIndex, calling overloaded functions with initializer lists can result in a compile error, with Visual C++ claiming that the overloads are ambiguous. This is a Visual C++ bug and it isn't clear if there's a good workaround. This won't affect code that doesn't use initializer lists, or uses an initializer list that doesn't require an implicit conversion (i.e. an initializer list of the container's exact value type). • Thread: ex_scoped_thread compile fails (#9333). For details, including download links, see http://www.boost.org/users/news/version_1.55.0 You can also download directly from SourceForge: http://sourceforge.net/projects/boost/files/boost/1.55.0/ To install this release on your system, see http://www.boost.org/doc/libs/release/more/getting_started/index.html Thanks, --The Boost release team Vladimir Prus Rene Rivera Marshall Clow Eric Niebler Daniel James Beman Dawes

2 1

[Polygon] attaching data to polygon boundary segments
by Andreas Buhr 10 Nov '13

10 Nov '13

Dear Boost community, I would like to process 2D geometry for a finite element solver using boost.polygon. I programmed it, it's working fine. I use the polygon_data and polygon_set_data types from boost.polygon. Now I would like to attach information to the boundary segments. Then I'd like to do boolean operations. At the end I would like to have the boundary information merged. Is this possible with boost.polygon? A attached a little drawing of what behavior I would like to achieve, colors code the boundary markers. thanks a lot in advance, Best regards, Andreas

1 1

[BGL] how to use write_graphviz_dp with graph bundle properties
by Pablo Madoery 10 Nov '13

10 Nov '13

Hello. I know how to use write_graphviz_dp with vertices and edges bundle properties but i can't figure out how to print the graph_bundle property of a graph. i have the following code where the line with the error is commented: #include <boost/graph/properties.hpp> #include <boost/graph/graphviz.hpp> #include <boost/graph/adjacency_list.hpp> using namespace boost; using namespace std; struct VertexInfo { int id; }; struct EdgeInfo { int weight; }; struct GraphInfo { int duration; }; typedef adjacency_list<vecS, vecS, undirectedS, VertexInfo, EdgeInfo, GraphInfo> Graph; void printGraph(Graph graph, string file) { ofstream ofs(file.c_str()); dynamic_properties dp; dp.property("node_id", get(vertex_index, graph)); dp.property("label", get(&EdgeInfo::weight, graph)); ///dp.property("label", get(&graph[graph_bundle].duration)); ->error write_graphviz_dp(ofs, graph, dp); } int main() { Graph graph; Graph::vertex_descriptor a = add_vertex(graph); Graph::vertex_descriptor b = add_vertex(graph); Graph::vertex_descriptor c = add_vertex(graph); Graph::vertex_descriptor d = add_vertex(graph); Graph::edge_descriptor e1 = (add_edge(a, b, graph)).first; Graph::edge_descriptor e2 = (add_edge(b, c, graph)).first; Graph::edge_descriptor e3 = (add_edge(c, d, graph)).first; Graph::edge_descriptor e4 = (add_edge(d, a, graph)).first; graph[e1].weight = 1.1; graph[e2].weight = 1.5; graph[e3].weight = 5.2; graph[e4].weight = 1.2; graph[graph_bundle].duration = 10; printGraph(graph, "graph.dot"); }

2 3

[Boost.Python] Exposing protected virtual member functions
by Vitaly Budovski 09 Nov '13

09 Nov '13

Hi, I'd like to expose some protected virtual member functions to Python, but I've been unable to find a practical way to do so. I've created a wrapper making the protected functions public, but have not been able to make further progress due to either compile errors or Python runtime exceptions. Is there something I've overlooked in the documentation? Please see code below. Thanks, Vitaly #include <boost/python.hpp> #include <string> using namespace boost::python; class base { //protected: public: virtual std::string f() { return "base::f"; } public: virtual std::string g() = 0; }; class derived: public base { public: std::string g() { return "derived::g"; } }; struct base_wrapper: base, wrapper<base> { std::string f() { if(override f = this->get_override("f")) { return f(); } return base::f(); } std::string default_f() { return base::f(); } std::string g() { return this->get_override("g")(); } }; BOOST_PYTHON_MODULE(protected) { class_<base_wrapper, boost::noncopyable>("base") .def("f", &base::f, &base_wrapper::default_f) .def("g", pure_virtual(&base_wrapper::g)) ; class_<derived, bases<base>, boost::noncopyable>("derived") .def("g", &derived::g) ; }

1 0

Converting Timestamp into hex of 8 bytes
by Rahul Mathur 09 Nov '13

09 Nov '13

Sorry for wrong post. I have a string Timestamp of 1383889129 which on ONLINE conversion from string to hex gives a 10 bytes as "31 33 38 33 38 38 39 31 32 39". But, the need is to have the TIMEStamp in 8 bytes only. OR any other way using either C++ or Boost where I can convert the string Timestamp of 1383889129 into 8 bytes hex value. Any clue ..!!

4 4

Handling JSON with Boost
by Jauhien Piatlicki 07 Nov '13

07 Nov '13

Hi all, I'm interested whether there exists a library for working with JSON in Boost that preserves data types? I know about property_tree, but if I read a valid JSON with it and then write it out it results in all fields being strings. This is because property_tree internally does not distinguish different value types. But I need something that can read e.g. {"number":42} get access to it and write it then in the same form. Thanks for answering, regards, Jauhien

4 4

Boost API System or Library Calls ..
by Rahul Mathur 06 Nov '13

06 Nov '13

Out of curiosity, wish to know if BOOST ASIO calls like - boost::asio::read() boost::asio::transfer_all() boost::asio::write() BOOST LEXICAL calls like - boost::lexical_cast<std::string>() BOOST POSIX TIME calls like - boost::posix_time::microsec_clock::universal_time() boost::posix_time::microsec_clock::ptime() are LIBRARY or SYSTEM calls. How to know which calls of BOOST are SYSTEM and LIBRARY calls? Thanks.

5 5

Safe to pass dereferenced anonymous shared_ptr?
by Chris Stankevitz 06 Nov '13

06 Nov '13

Q1: Is the below code dangerous? A1: Only if the anonymous shared_ptr is deleted before function f returns. Q2: Is the anonymous shared_ptr deleted before function f returns? A2: ? Thank you, Chris === #include <iostream> #include <boost/make_shared.hpp> void f(const int& i) { std::cout << "i = " << i << std::endl; } int main() { f(*boost::make_shared<int>(42)); return 0; }

4 7

[Containter] Compilation errors with 1.55 and VC2013
by Szymon Gatner 05 Nov '13

05 Nov '13

Hi, I am getting compilation errors with Container library after trying to build with Visual Studio 2013 and Boost 1.55 RC: Error 1 error C2228: left of '.select_on_container_copy_construction' must have class/struct/union c:\devel\boost_1_55_0\boost\intrusive\detail\has_member_function_callable_with.hpp 200 1 ConsoleApplication2 Minimalistic example to reproduce: #include <boost/signals2/connection.hpp> #include <boost/container/slist.hpp> #include <boost/container/vector.hpp> int main() { typedef boost::signals2::connection Connection; boost::container::vector<Connection> connections_; } Error happens wheter vector or slist is used (didn't try more containers) Regards, Szymon Gatner

3 3

[LOG]: set_filter error in tutorial_filtering.cpp and per-sink instance filtering
by Johan Lundberg 05 Nov '13

05 Nov '13

Dear all, (this is my first post to this list) I'm trying to make a boost log sink that would filter on a severity level which should be thread-safely modifiable while logging is ongoing. On my way there I get errors while trying to compile this tutorial example: http://www.boost.org/doc/libs/1_54_0/libs/log/example/doc/tutorial_filterin… The file as-is compiles and runs fine, but if I switch on the user-provided filtering function example by activating the block in #if 0 (and commenting out the other definition of init()) I get errors related to phoenix bind: The offending line is 155: sink->set_filter(phoenix::bind(&my_filter, severity, tag_attr)); I would be grateful for advice on how to get that working. More details on that at the end of the mail. Now, assuming I get past the errors, what is a good way to add a per-sink log level filter? I may have several instances of the sink. To clarify: I know this is not the right syntax, but If I could use a custom functor or a C++11 lambda for a filter I would do something like this: using ATseverity=std::atomic<severity_level> ; // (severity is an enum) shared_ptr<ATseverity> shared_log_level = std::make_shared<ATseverity>(warning); sink->set_filter( [=shared_log_level]->bool(severity_level const& level, std::string const& tag){ return level >= shared_log_level->get(); }); I could then store and modify the shared_log_level from elsewhere. Is there a right way to do this? cheers /Johan Details on the compilation error: I think this is the relevant part of the error message: /myboost154/include/boost/phoenix/core/preprocessed/expression_10.hpp:140:17: error: could not convert ‘boost::proto::make_expr<boost::phoenix::detail::tag::function_eval, boost::proto::domainns_::basic_default_domain, boost::phoenix::detail::function_ptr<2, bool, bool (*)(const boost::log::v2_mt_posix::value_ref<severity_level>&, const boost::log::v2_mt_posix::value_ref<std::basic_string<char> >&)>, boost::log::v2_mt_posix::expressions::attribute_keyword<tag::severity>, boost::log::v2_mt_posix::expressions::attribute_keyword<tag::tag_attr> >((* & a0), (* & a1), (* & a2))’ from ‘const type {aka const boost::proto::exprns_::basic_expr<boost::phoenix::detail::tag::function_eval, boost::proto::argsns_::list3<boost::proto::exprns_::basic_expr<boost::proto::tagns_::tag::terminal, boost::proto::argsns_::term<boost::phoenix::detail::function_ptr<2,bool, bool (*)(const boost::log::v2_mt_posix::value_ref<severity_level>&, const boost::log::v2_mt_posix::value_ref<std::basic_string<char>>&)> >, 0l>, boost::log::v2_mt_posix::expressions::attribute_keyword<tag::severity>, boost::log::v2_mt_posix::expressions::attribute_keyword<tag::tag_attr> >, 3l>}’ to ‘boost::proto::exprns_::basic_expr<boost::proto::tagns_::tag::terminal, boost::proto::argsns_::term<boost::phoenix::detail::function_ptr<2, bool, bool (*)(const boost::log::v2_mt_posix::value_ref<severity_level>&, const boost::log::v2_mt_posix::value_ref<std::basic_string<char> >&)> >, 0l>::proto_child0 { aka boost::phoenix::detail::function_ptr<2, bool, bool (*)(const boost::log::v2_mt_posix::value_ref<severity_level>&, const boost::log::v2_mt_posix::value_ref<std::basic_string<char> >&)>}’ }; I tried boost 1.54 and compiled with gcc 4.7.1 and 4.8.1 under Linux, with and without C++11 mode: g++ -std=gnu++11 -DBOOST_ALL_DYN_LINK boostlogexample.cpp -L $BOOST/lib/ -I $BOOST/include/ -pthread -lboost_system -lboost_log_setup -lboost_log \ -lboost_date_time -lboost_thread -lrt -lboost_filesystem

1 1

container with large number of elements and a small number of invalide elements
by MM 05 Nov '13

05 Nov '13

Hello, I have a need for a container where I can store 10^6 elements of type T (PODs) , a small number of which (say 100) can be invalid elements (ie special value of T) The 100 elements can be randomly distributed throughout the 10^6 elements. The container should be as close to a sequence container as possible. The required usages: 0. invalid elements can be set = 10^6 - 100 1. get number of valid elements in O(1) 2. get the max/min element of all valid elements 3. have a bidrectional iterator that iterators only over valid elements On a separate note, is there a version of multi_array where the number of dimensions is known at runtime? regards,

6 17

[1.55.0] Release candidate 2 available
by Marshall Clow 05 Nov '13

05 Nov '13

RC2 files for 1.55.0 are available at http://boost.cowic.de/rc/ As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: 93780777cfbf999a600f62883bd54b17 *boost_1_55_0_rc2.tar.gz d6eef4b4cacb2183f2bf265a5a03a354 *boost_1_55_0_rc2.tar.bz2 8aca361a4713a1f491b0a5e33fee0f1f *boost_1_55_0_rc2.zip 4e5bbc15fc8c80df8be428f8a5b5a823 *boost_1_55_0_rc2.7z Thanks! -- The release managers === Changes to release since rc1 === ---- Author: bemandawes Date: 2013-10-28 13:16:50 EDT (Mon, 28 Oct 2013) New Revision: 86503 URL: http://svn.boost.org/trac/boost/changeset/86503 Log: Merge 86426 from trunk. This should fix msvc test failures. ---- Author: bemandawes Date: 2013-10-29 09:15:03 EDT (Tue, 29 Oct 2013) New Revision: 86514 URL: http://svn.boost.org/trac/boost/changeset/86514 Log: Merge 84443 from trunk, to fix mpl assert failures for msvc in many libraries. ---- Author: eric_niebler Date: 2013-10-29 12:53:08 EDT (Tue, 29 Oct 2013) New Revision: 86516 URL: http://svn.boost.org/trac/boost/changeset/86516 Log: merge [86460] from trunk, fix bootstrap on mingw --- Author: jhunold Date: 2013-10-29 13:19:33 EDT (Tue, 29 Oct 2013) New Revision: 86519 URL: http://svn.boost.org/trac/boost/changeset/86519 Log: Merge 84402 from ^/trunk ------------------------------------------------------------------------ r84402 | jhunold | 2013-05-21 14:23:20 +0200 (Di, 21 Mai 2013) | 4 lines Fix: cache file loading and saving must use native paths Fixes caching at least on Windows platform ------------------------------------------------------------------------ --- Author: johnmaddock Date: 2013-11-01 05:32:55 EDT (Fri, 01 Nov 2013) New Revision: 86532 URL: http://svn.boost.org/trac/boost/changeset/86532 Log: Merge fix for dependency tracking. --- Author: danieljames Date: 2013-11-03 14:20:09 EST (Sun, 03 Nov 2013) New Revision: 86551 URL: http://svn.boost.org/trac/boost/changeset/86551 Log: Merge Visual C++ 12 fix for Unordered. --- Author: viboes Date: 2013-11-03 15:43:42 EST (Sun, 03 Nov 2013) New Revision: 86552 URL: http://svn.boost.org/trac/boost/changeset/86552 Log: Thread: merge [86540] to fix #9319.

8 12

boost::statechart "custom::reaction"
by rahul baruah 05 Nov '13

05 Nov '13

I am using boost::statechart. Instead of simple_state , using boost:: statechart::state. Now i am able to access the state_machine member variable. but when i am using custom::reaction or mpl::list it throught error. Code Snippet: struct NotReady : StateBase< NotReady, TopState > { //typedef NotReadyStateList reactions; NotReady (my_context ctx) : my_base( ctx ){ outermost_context().maxPairedDevices = 3; cout << outermost_context().maxPairedDevices << endl; } template< class MostDerived > sc::result react( const PairingBaseEvents< MostDerived > & ); public: typedef sc::custom_reaction< StartEvent > reactions; }; Error: ndefined reference to `boost::statechart::detail::safe_reaction_result NotReady::react<StartEvent>(PairingBaseEvents<StartEvent> const&)' regards Rahul Baruah

3 5

Concatenating int value to string ..??
by Rahul Mathur 05 Nov '13

05 Nov '13

All, I wish to have string (combination of string value and int value) which increments until 100. The code is as below - --- #include <iostream> #include <cstdlib> #include <string> #include <boost/lexical_cast.hpp> using namespace std; int main() { std::string TID = "Apple"; int TagID = 01, Value = 000 /* 100 IDs*/; std::string TagName, TagInitial, TagUId; TagName = TID + boost::lexical_cast<std::string>(TagID); // Should be as -> Apple01 TagInitial = TagName + boost::lexical_cast<std::string>(Value); // Should be as -> Apple01000 int i = boost::lexical_cast<int>(TagInitial) + 1; // Should be as -> Apple01001 TagUId = boost::lexical_cast<std::string>(i); std::cout << "Tag ID .." << TagUId << endl; } I get below error message - -- terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::bad_lexical_cast> >' what(): bad lexical cast: source type value could not be interpreted as target Aborted (core dumped) -- Any clue? ~ Thanks!!

7 11

XML Serialization
by Lasse Laursen 05 Nov '13

05 Nov '13

Hey People, I'm having some issues using BOOST's serialization capabilities in version 1.53. To be honest, the capability provided by boost seem to involve a lot of macro usage here and there, and despite the fact that I've read a number of tutorials, I still don't feel like I've got a solid grasp on how it all works. Probably because I keep running into compilation problems that I can't seem to figure out how to fix logically, I usually just end up trying a number of different macro's that other people recommend until it just works. Not ideal, I know, but... All my logical deduction usually leads me to solutions that don't work as I keep expecting. Case in point, I want to serialize some data as XML. Essentially it's just a struct with a few strings in it that is used as 'settings' data. I get the following error message during compilation in VS2012: Error 47 error C2664: 'boost::mpl::assertion_failed' : cannot convert parameter 1 from 'boost::mpl::failed ************boost::serialization::is_wrapper<T>::* ***********' to 'boost::mpl::assert<false>::type' d:\sdks\cinder\boost\boost\archive\basic_xml_iarchive.hpp 70 I found out, it had to do with any XML serilization requiring an additional 'name' to be added, pertaining to how XML works via the 'BOOST_SERIALIZATION_NVP' macro. I added that to all my strings - still nothing. Then I proceeded to comment out everything, so my struct is now just an empty shell: struct AEMSettings { // Superfluous since structs are fully public? friend class boost::serialization::access; AEMSettings() { } /*** * Serialize handles both input/output as the & operator doubles as both << and >> */ template<class Archive> void serialize( Archive &ar, const unsigned version ) { //ar & BOOST_SERIALIZATION_NVP( mSpeakersDeviceName ); //ar & BOOST_SERIALIZATION_NVP( mSpeakersDeviceDriverName ); //ar & BOOST_SERIALIZATION_NVP( mHeadphonesDeviceName ); //ar & BOOST_SERIALIZATION_NVP( mHeadphonesDeviceDriverName ); //ar & BOOST_SERIALIZATION_NVP( mMicrophoneDeviceName ); //ar & BOOST_SERIALIZATION_NVP( mMicrophoneDeviceDriverName ); } //std::string mSpeakersDeviceName; //std::string mSpeakersDeviceDriverName; //std::string mHeadphonesDeviceName; //std::string mHeadphonesDeviceDriverName; //std::string mMicrophoneDeviceName; //std::string mMicrophoneDeviceDriverName; }; Can anyone tell me what I am missing here? Regards, Gazoo

3 3

[Spirit] Parser omits certain characters
by beet 05 Nov '13

05 Nov '13

Dear all, For all of the following, please see the attached test-case. I would like to specify certain characterstics of variables in the following way: d(MY_DPAR_01,-10.3,12.8,100) . So in this case there is a type-identifyer (d stands for "double") and then, in parantheses, a variable name (or alternatively an index), a lower and upper boundary and some integer parameter. What is contained in the list depends on the parameter type. It should then be possible to specify any number of variables, including their properties, in a comma-separated list. So I need to dissect a string like the following: "d(MY_DPAR_01,-10.3,12.8,100), d(0,-10.3,12.8,100), i(SOME_IPAR_17, 0,5), b(SOME_BPAR)" The first step, as I see it, is to seperate the "outer" list and to store each variable description in a boost::tuple<char, std::string>, where the char holds the type identifier and the std::string the part inside the parantheses. Then, in the second step, I want to parse that string according to what is expected for this particular type. I have now run into the problem that the following rule in Boost.Spirit "swallows" all non-alphanumeric characters. qi::rule<std::string::const_iterator, std::string(), ascii::space_type> varSpec = +(alnum | '_' | ',' | '.' | '+' | '-'); varSpec stands for the text inside of the parantheses. I would have expected this to match any string of at least one of the listed character types (i.e. alpha-numeric, underscore, etc.). The rule qi::rule<std::string::const_iterator, boost::tuple<char, std::string>(), ascii::space_type> varString = char_("dfib") >> '(' >> varSpec >> ')'; is then meant to separate the type-descriptor from the description. Parsing with these yields the output d: MYDPAR01103128100 d: 0103128100 i: SOMEIPAR1705 b: SOMEBPAR on a Mac (Mavericks, Boost 1.54) i.e., only the alpha-numeric characters remain. So clearly my assumption about the "varSpec" rule is wrong. On a Ubuntu 13.10 System (Boost 1.54, g++ 4.8.1) I get the output d: d: i: b: which is equally wrong. Funnily, if I define varSpec simply as +char_, varSpec gives the correct result on the Mac (haven't tried Ubuntu), but varString does not match anything. So I am at a loss here and would appreciate some help. Best Regards and thanks for any help you can provide, Beet

3 4

[move] non-inline move constructor - won't compile
by Krzysztof Czainski 03 Nov '13

03 Nov '13

Hi, I've been trying to writhe a non-inline move constructor, to place its' implementation in a .cpp file. I'm having problems compiling my code, which I managed to simplify (after examining output of the preprocessor) down to this short code snippet: #include <boost/move/core.hpp> struct X { X( ::boost::rv< X >& x ); operator ::boost::rv<X>&() { return *static_cast< ::boost::rv<X>* >(this); } }; X::X( ::boost::rv< X >& x ) {} Compiling a .cpp file with the above contents gives me: a.cpp:41:1: error: prototype for 'X::X(boost::rv<X>&)' does not match any in class 'X' X::X( ::boost::rv< X >& x ) {} ^ a.cpp:35:8: error: candidates are: X::X(const X&) struct X ^ a.cpp:37:5: error: X::X(boost::rv<X>&) X( ::boost::rv< X >& x ); ^ I'm using MinGW gcc 4.8.1 in default -std mode (not-C++11 mode), boost-1.54. Am I doing something wrong here, or is it a bug in the compiler? Regards, Kris

3 4

Status of smart_assert
by Daan Baas 03 Nov '13

03 Nov '13

Howdy, I was searching for an intelligent assert facility with serveral debug levels, that asserts when it is run debug mode and throws in release mode. That is how i ran into Boost's SMART_ASSERT ( http://www.drdobbs.com/cpp/enhancing-assertions/184403745) However, I cannot find anything about it that was posted after 2003 nor can I find it on the Boost site. Does anyone know what happened to the project? Was is cancelled or did it make it into Boost under another name? Kind regards, Dano

2 2

[MSM] Parse error in Visual Studio 2012, not in Visual Studio 2013.
by Vincent N. Virgilio 02 Nov '13

02 Nov '13

Hello, In the code below, VS 2012 (target v110) underlines the macro name BOOST_MSM_EUML_EVENT_WITH_ATTRIBUTES in the last line with a wavy red line. Hovering a mouse over it shows a tooltip with error report: "Error: not a class or struct name" And if I recall correctly, the code does not compile. The error disappears in VS 2013 (v120). Though in VS 2013 (target v120), if I remove line "namespace msm = boost::msm;", the last line triggers an error again. This time with a different tooltip error report: "Error: name followed by '::' must be a class or namespace name." Am I using MSM incorrectly? Thanks, Vince // BEGIN euml.h #include <boost/msm/back/state_machine.hpp> #include <boost/msm/front/state_machine_def.hpp> #include <boost/msm/front/euml/euml.hpp> using namespace boost::msm::front::euml; namespace msm = boost::msm; BOOST_MSM_EUML_DECLARE_ATTRIBUTE(char*, charp_attr) BOOST_MSM_EUML_ATTRIBUTES((attributes_ << charp_attr), found_attrS) // next line gives error in VS 2012 (target v110), "Error: not a class or struct name". Not in VS 2013 (targ. v12). BOOST_MSM_EUML_EVENT_WITH_ATTRIBUTES(found, found_attrS) // END euml.h

3 3

[1.55.0] Release candidates available
by Marshall Clow 01 Nov '13

01 Nov '13

Release candidate files for 1.55.0 are available at http://boost.cowic.de/rc/ As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: ba1ae723b8b4d1edeca684bb14b3213b *boost_1_55_0_rc1.tar.gz 66e01be2f328c556835c42be7b6d17d5 *boost_1_55_0_rc1.tar.bz2 925bb6c96082033687e5742b4e77af7b *boost_1_55_0_rc1.zip 830d61075ecf7d5ebf7c0843033393bf *boost_1_55_0_rc1.7z Thanks! -- The release managers

9 16

Connecting signals where one is a member
by Chris Stankevitz 01 Nov '13

01 Nov '13

Hello, "Connect 4" fails to compile in the code below on g++ 4.6.3. Why does "Connect 4" not work while "Connect 2" does work? Connect 4 and 2 are the same thing as far as I can see. Thank you, Chris #include <boost/signals2/signal.hpp> typedef boost::signals2::signal<void( )> TDSignalA; typedef boost::signals2::signal<void(int)> TDSignalB; TDSignalA SignalA1; TDSignalB SignalB1; struct TSItem { TDSignalA SignalA2; TDSignalB SignalB2; }; int main() { TSItem Item; // Connect 1 SignalA1.connect(Item.SignalA2); // Connect 2 SignalA1.connect(TDSignalA::slot_type(&TSItem::SignalA2, &Item)); // Connect 3 SignalB1.connect(Item.SignalB2); // Connect 4 //SignalB1.connect(TDSignalB::slot_type(&TSItem::SignalB2, &Item, _1)); return 0; }

2 2

VS2005 compiler error in swap() definition after including flat_map.hpp
by Фёдор Трушкин 01 Nov '13

01 Nov '13

Dear All, I encountered a strange problem recently. It's perfectly reproducable under VS2005 even with the latest boost 1_55 rc. Here's the minimal reproducable example: // ----- //#include <boost/container/flat_map.hpp> #include <boost/container/detail/config_begin.hpp> #include <boost/intrusive/detail/memory_util.hpp> #include <boost/container/detail/memory_util.hpp> void swap(); // ------ In VS2005 this innocent piece of code yields: // ------ error C2365: 'swap' : redefinition; previous definition was 'formerly unknown identifier' // ------ VS2010 and GCC 4.5.2 have no complaints. Looks like VS2005 compiler is entangled somehow by BOOST_INTRUSIVE_HAS_MEMBER magic. Actually, it's quite a headache that with VS2005 one can't introduce swap() after, say, flat_map is included. Should this incompatibility be regarded as a bug? Best regards, Fedor Trushkin

2 1

[python] Crash right after initialization after ubgrading from 1.48 to 1.54
by Richi Lists 31 Oct '13

31 Oct '13

Now I * downloaded the python 2.75 source * compiled python 2.75 * compiled boost 1.54 against python 2.75 * built our software against these new libs -> The behavior is still the same. ob_type of the first python object I create is 0xffffffff, and trying to access it results in an immediate crash. This only applies to the debug version. Release is fine. Also, the behavior is the same, if I link the boost debug build against python release or debug. Rgds Richard

1 0

[Iostreams] Problem to chain gzip filter with home-made one
by Christophe hugon 31 Oct '13

31 Oct '13

Hello everybody, I meeting a execution issue while using the gzip filter of the Boost.Iostreams, combined with a buffered home-made one. I would like to know what is wrong in my way to do it. Firstly, my filter is a general DualFilter. It uses an buffer (of 16 bytes) before the transformation and the put/return. The categories of my filter is closable and dual_use. You can find code details in the end of this mail. My problem is ([a/b] means for the separates case with the letter a and b): If I take the 6 possibilities for write and read: 1) io::filtering_[i/o]stream stm; 2) stm.push(io::gzip_[de/]compressor()); 3) stm.push(my_[Unt/T]ransFilter); at the moment to use the stm.[get/put] to read/write 1)-2) works perfectly 1)-3) works perfectly 1)-3)-2) works perfectly 1)-2)-3) Stay stuck (obviously it is the one I really need) In fact in the last case, the gzip filter give this infinite loop to the put of the TransFilter : 31 -117 8 0 0 0 0 0 0 31 -117 8 0 0... A similar case occur with the bzip filter (even if it is a bit longer loop). What do I do wrong ? Somebody can help me ? Thanks in advance Christophe PS: here a simplified implementation of my filter template<typename Source> int get(Source& src) { //at the very beginning or if the buffer are fully returned if (fcurrent_pos == fblkLength) { fcurrent_pos =0; FillThem (src); //fill&transform the buffers with the right number of src.get() fc=CheckNSolveBufferIsEnd (src); //eventually prepare read the suffix of the file } //The effective return if (fcurrent_pos < fUntrans.size ()) return (int)fUntrans[fcurrent_pos++]; else return fc; } template<typename Sink> bool put(Sink& dest, int c) { if (c==EOF || c ==boost::iostreams::WOULD_BLOCK) return false; if (fUntrans.size () < fblkLength - 1) { fUntrans.AddData((char*)(&c),1); return true; } fUntrans.AddData((char*)(&c),1); fenc->TransformThis (fUntrans, fTrans, fKey); for (int i=0; i<fblkLength; i++) { if (!boost::iostreams::put(dest, (int)fTrans[i])) return false; } return true; } template<typename Sink> bool close (Sink &dest, BOOST_IOS::openmode m) { if (fFlushed) return true; PrepareSuffix (); for (int i=0; i< fblkLength*2; //suffix length i++) { if (!boost::iostreams::put(dest, (int)fTrans[i])) return false; } return true; }

1 0

BOOST ASIO: Having successful READ from the SERVER ..??
by Rahul Mathur 31 Oct '13

31 Oct '13

All, Initially after having done successful CONNECT with the server, sending successful LOGON message and finally KEEP ALIVE status, I tried to read from the server as - -- boost::asio::streambuf data; std::size_t length = boost::asio::read(socket, data, boost::asio::transfer_at_least(socket.available())); -- but was able to partially receive response from remote server which I don't think was completely successful, so to have COMPLETE RESPONSE from the server, tried with below modification - -- boost::system::error_code ec; boost::asio::streambuf data; std::size_t length = boost::asio::read(socket, data, boost::asio::transfer_all(), ec); --- but this modification doesn't work. Any clue what should be the proper way? Thanks!!

2 1

SIGSEGV when removing Ethernet/Serial Port cables
by Ulf Samuelsson 30 Oct '13

30 Oct '13

Have a boost enable application which communicated over USB serial ports and UDP If the cables are removed, the library fires off SIGSEGV signals. I am trying to catch exceptions but neither } catch (boost::system::system_error &e) { } catch (std::exception &e) { seems to work, and a SIGSEGV signal is sent when the port->get_io_service().reset(); statement is executed. How can I handle the problem, that the USB serial cable is removed? I know I can test for the existence of /dev/ttyUSBx in the file system or /sys/class/tty/ttyUSBx" I know I can write a SIGSEGV handler, but would like to know if there is any exception I can catch instead. Have the same problem when the Ethernet Cable is removed. How handle this? // Serial Port Read. // Reads a character or times out bool SerialPort::read_char(unsigned char& val) { val = c = '\0'; try { // After a timeout & cancel it seems we need // to do a reset for subsequent reads to work. port->get_io_service().reset(); // Asynchronously read 1 character. async_read(*port, buffer(&c, 1), boost::bind(&SerialPort::read_complete, this, placeholders::error, placeholders::bytes_transferred)); // Setup a deadline time to implement our timeout. timer->expires_from_now(boost::posix_time::milliseconds(1)); timer->async_wait(boost::bind(&SerialPort::time_out, this, placeholders::error)); // This will block until a character is read // or until the it is cancelled. port->get_io_service().run(); if (!read_error) { val = c; } else { val = 0; // Timeout } } catch (boost::system::system_error &e) { val = 2; DebugPort.WriteLine("Serial Port System Error", RED); return false; } catch (std::exception &e) { val = 1; DebugPort.WriteLine("Serial Port General Error", RED); return false; } return !read_error; } // Network port open bool UdpClient::Open(void) { udp::resolver resolver(io_service); udp::resolver::query query(udp::v4(), ipAddress, Port); try { Endpoint = *resolver.resolve(query); } catch (boost::system::system_error &e) { // We might not have a network connection return false; } if (!CreateSocket()) { return false; } if (!OpenSocket()) { return false; } init = true; return true; } Best Regards Ulf Samuelsson

2 1

Looking for official .m4 script
by Yaroslav Panych 30 Oct '13

30 Oct '13

HI Looking for official .m4 script to include it into distribution package in order to use its benefits while configuring/build software. Yes, I know about Tsuna's boost.m4 library but I never heard it was approved by Boost officials. Regards

2 1

[graph] ambiguity moving from 1.48 to 1.54
by Richi Lists 30 Oct '13

30 Oct '13

Hi, I'm in the process of upgrading from boost 1.48 to 1.54. We're using MSVC10. Most stuff that comes up is easy to fix, but this one gives me headaches: ...boost\boost_1_54_0\boost/pending/property.hpp(108): error C2752: 'boost::lookup_one_property_internal<PList,PropName>' : more than one partial specialization matches the template argument list with [ PList=boost::property<boost::edge_bundle_t,EdgeData>, PropName=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/pending/property.hpp(114): could be 'boost::lookup_one_property_internal<boost::property<Tag,T,Base>,Tag>' ...boost\boost_1_54_0\boost/pending/property.hpp(108): or 'boost::lookup_one_property_internal<boost::property<Tag,T,Base>,boost::edge_bundle_t>' ...boost\boost_1_54_0\boost/pending/property.hpp(108): or 'boost::lookup_one_property_internal<T,boost::edge_bundle_t>' ...boost\boost_1_54_0\boost/pending/property.hpp(158) : see reference to class template instantiation 'boost::lookup_one_property_internal<PList,PropName>' being compiled with [ PList=boost::property<boost::edge_index_t,int,boost::property<boost::edge_bundle_t,EdgeData>>, PropName=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/graph/detail/adjacency_list.hpp(1505) : see reference to class template instantiation 'boost::lookup_one_property<T,Tag>' being compiled with [ T=boost::property<boost::edge_index_t,int,boost::property<boost::edge_bundle_t,EdgeData>>, Tag=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/graph/detail/adjacency_list.hpp(2028) : see reference to class template instantiation 'boost::adj_list_helper<Config,Base>' being compiled with [ Config=boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config, Base=boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config> ] ...boost\boost_1_54_0\boost/graph/adjacency_list.hpp(267) : see reference to class template instantiation 'boost::vec_adj_list_impl<Graph,Config,Base>' being compiled with [ Graph=boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>, Config=boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config, Base=boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config> ] ..\graph_problem.cpp(52) : see reference to class template instantiation 'boost::adjacency_list<OutEdgeListS,VertexListS,DirectedS,VertexProperty,EdgeProperty>' being compiled with [ OutEdgeListS=boost::vecS, VertexListS=boost::vecS, DirectedS=boost::undirectedS, VertexProperty=VertexProps, EdgeProperty=EdgeProps ] ...boost\boost_1_54_0\boost/graph/adjacency_list.hpp(374): error C2182: '[]' : illegal use of type 'void' The code to reproduce is here: // boost #include <boost/graph/adjacency_list.hpp> #include <boost/graph/subgraph.hpp> #include <boost/graph/graph_traits.hpp> class EDGE {}; ////////////////////////////////////////////////////////////////////////// /// graph's vertex data struct VertexData { /// c-tor VertexData() { pos[0] = 0.0; pos[1] = 0.0; pos[2] = 0.0; } VertexData(const double* _pos) { pos[0] = _pos[0]; pos[1] = _pos[1]; pos[2] = _pos[2]; } /// data double pos[3]; }; ////////////////////////////////////////////////////////////////////////// /// graph's edge data struct EdgeData { /// c-tor EdgeData() : pEdge(NULL) {} EdgeData(EDGE* _pEdge) : pEdge(_pEdge) {} /// comparison for equality bool operator==(const EdgeData& dst) const { return pEdge == dst.pEdge; } /// data EDGE* pEdge; }; ////////////////////////////////////////////////////////////////////////// /// types typedef VertexData VertexProps; ///< custom vertex properties typedef boost::property<boost::edge_index_t, int, boost::property<boost::edge_bundle_t, EdgeData> > EdgeProps; ///< custom edge properties typedef boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS, VertexProps, EdgeProps> Type; ///< custom graph type ////////////////////////////////////////////////////////////////////////// /// unary predicate for testing if edge has duplicates in graph struct CheckDuplicateEdge { CheckDuplicateEdge(const Type& _g) : g(_g) {} // check if this edge has duplicates bool operator()(const Type::edge_descriptor& e) const { // do something return false; } private: const Type& g; }; ////////////////////////////////////////////////////////////////////////// void RemoveDuplicateEdges(Type& g) { remove_edge_if(CheckDuplicateEdge(g), g); } ////////////////////////////////////////////////////////////////////////// void main() { std::cout << "hello graph"; } ////////////////////////////////////////////////////////////////////////// I have tried to find a solution, and have searched the internet for hours. So, hopefully anybody here saw something similar, and solved it already. Rgds Richard

2 3

[python] Crash right after initialization after ubgrading from 1.48 to 1.54
by Richi Lists 30 Oct '13

30 Oct '13

I upgraded from boost 1.48 to .154 and I thought I was done after all the compiler errors and unit tests were ironed out. But now when I start our application, it terminates immediately. This only happens with the debug build. The release build is fine. I verified that it properly links to boost-python-vc100-mt-gyd-1_54.lib and python27_d.lib The python version (2.73) is still the same that we used with boost 1.48. I tracked it down to here: Py_Initialize(); m_mainModule = boost::python::import(__main__); <-- crash here With the debugger I found out that the string object is invalid. More specifically, ob_type is 0xffffffff. So I added this exception which always throws now: Py_Initialize(); const boost::pathon::str someRandomString("someRandomString"); if(reinterpret_cast<void*>(0xffffffff) == someRandomString.ptr()->ob_type) throw std::runtime_error("Python problem"); What could be going on here? Rgds Richard

1 0

[Geometry] Compilation errors after moving to 1.55b1
by Szymon Gatner 29 Oct '13

29 Oct '13

Moving to 1.55b1 from 1.54 gives: Error 1 error C2976: 'boost::geometry::strategy::transform::translate_transformer' : too few template arguments offending line is: typedef strategy::transform::translate_transformer<Point, Point> Translate; Regards, Szymon Gatner

3 3

[Asio] Compilation errors after switching to 1.55 b1
by Szymon Gatner 29 Oct '13

29 Oct '13

I am getting very weird error after moving from 1.54 to 1.55b1: 2>C:\devel\boost_1_55_0b1\boost/asio/detail/impl/socket_ops.ipp(2224): error C2653: 'Windows' : is not a class or namespace name (..\..\lib\hexine.cpp) and many more in that file, then: 2>C:\devel\boost_1_55_0b1\boost/asio/detail/buffer_sequence_adapter.hpp(36): error C2653: 'Windows' : is not a class or namespace name (..\..\lib\hexine.cpp) Than, many errors later: 2>C:\devel\boost_1_55_0b1\boost/asio/detail/winrt_async_manager.hpp(62): error C2039: 'IAsyncAction' : is not a member of 'Windows::Foundation' (..\..\lib\hexine.cpp) This looks like C++/CLI implementation. I assume this is related to "Implemented a limited port to Windows Runtime. This support requires that the language extensions be enabled." Does it mean that switch from 1.54 to 1.55 requires changes in project settings? Visual Studio 2012 Update 3, Windows 7 Regards, Szymon Gatner

2 1

[date_time] Not built with Visual Studio 2013
by Szymon Gatner 29 Oct '13

29 Oct '13

Hi, I just built boost 1.55 beta1 with recently released VC2013 on Win7 and for some reason date-time library was not built. I didn't sepcify it when building (just hit b2 to get everything). Regards, Szymon Gatner

2 3

[Boost-users] {exception][boost::throw_exception][BOOST_NO_EXCEPTIONS] implications
by Jeff Flinn 28 Oct '13

28 Oct '13

The project I've joined [ab]uses BOOST_NO_EXCEPTIONS in 1.55b as a means to intercept exceptions thrown via boost::throw_exception() and translate them to the project's single exception type derived from std::exception, leaving me a bit queasy. That queasiness is reinforced by #ifdef BOOST_NO_EXCEPTIONS #error This header requires exception handling to be enabled. #endif in ...\boost_1_55_0b1\boost\exception\detail\exception_ptr.hpp. This implies that BOOST_NO_EXCEPTIONS is used for more that just allowing client code to supply it's own throw_exception. Does boost and/or C++ provide a better customization point to safely translate exceptions thrown by boost? (note: we're using vs2013, generally the latest gcc, and clang) Thanks, Jeff

2 1

Boost Iterator Adaptor, with a mutable member variable is optimized out incorrectly
by Jonas Hörsch 28 Oct '13

28 Oct '13

hello everyone, i am not sure if this is actually a question about boost usage. but in the attached code i'm stuck with the unlikely situation, that gcc optimizes away my code and produces garbage for -On (n>=1), while it works like intended with -O0. attached is the minimal code, with which i could reproduce the problem: on linux i686 g++ version 4.6.3 and x86_64 g++ version 4.7.2, using packaged boost_1_54_0 in both instances, i see the following: when i compile the code with optimization (-O1) % ./iterator_adaptor_weirdness -1219536795 -1219536795 and without (-O0) % ./iterator_adaptor_weirdness 4 2 is this a bug in gcc? is it wrong to use mutable variables in conjunction with iterator_adaptor? any insight is appreciated, jonas P.S.: I know, that this particular instance of the problem, might be better solved using a transform_iterator, but unfortunately that approach doesn't scale.

2 2

[Asio] Example of the "passing the buck" pattern
by Sorin Fetche 28 Oct '13

28 Oct '13

Hi all, Although the pattern of "passing the buck" has been mentioned in several places in the context of Boost.Asio, I didn't find any self-contained example that demonstrates it in a reusable manner. So, after going through all the information I could find, I tried to create such an example that demonstrates it using C++03.I would like to ask for feedback related to the attached sample code (its "Test Infrastructure" and "Test Code" sections) and double check that it does indeed depict what "passing the buck" means. For future reference I'm listing here the all the sources of information I could find related to the pattern of "passing the buck". it also turned out that a helper utility is needed in order to wrap boost::bind created handlers and have the wrapper integrate with the Asio customization points. - N3388 paper: Using Asio with C++11 http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2012/n3388.pdf Chapter 7 describes this strategy to develop efficient and generic composed operations by letting their caller decide via the provided completion handler how the intermediary handlers should interact with the Asio customization points. - Thinking Asynchronously: Designing Applications with Boost.Asio https://github.com/boostcon/2011_presentations/raw/master/mon/thinking_asyn… The "Managing Complexity" chapter references the "pass the buck" strategy with slides 113 and 114 showing invocation and allocation hooks that call the hooks of a nested handler. - The implementation of higher level functions from Boost.Asio like async_read or async_write. The problem with them being an example of the pattern is that they are using custom classes for the intermediary operation handlers instead of relying on boost::bind to build them on the fly. - Asio Samples project http://sourceforge.net/projects/asio-samples For example its handler wrapper: context_alloc_handler and its usage in the echo server implementation. Thank you, Sorin Fetche

1 0

Preprocessor Recursion
by Mitch Besser 27 Oct '13

27 Oct '13

I'm trying to create a 'C' macro that will define and initialize static data. For example: STATIC_CONST_STRUCT ( A, a, MEMBER_DATA(CONST_STR, a, "Hello, a") MEMBER_DATA(CONST_STR, b, "Hello, b") MEMBER_STRUCT ( C, c, MEMBER_DATA(CONST_STR, d, "Hello, d") MEMBER_DATA(CONST_INT, e, 1) ) ); Would cause the 'C' preprocessor to create: static const struct A { CONST_STR a; CONST_STR b; struct C { CONST_STR d; CONST_INT e; } c; } = {"Hello, a", "Hello, b", {"Hello, d", 1}}; If I could use C++, this would be easy, but I'm using 'C' and it would be very helpful to have the initialization along with the declaration. I can't quite figure out how to make this work. My macros stop expanding. I suspect the recursive nature of the problem having an arbitrarily deep nesting is why. If it helps, I do have a maximum depth for the nesting. I've tried reading the docs and learn more about recursion, but I can't seem to find a solution. I suspect I'm close. Here's what I have so far (not claiming this is a good solution, so feel free to suggest anything that would make this better): #define MEMBER_DATA_TAG 0 #define MEMBER_STRUCT_TAG 1 #define MEMBER_TAG(MEMBER) BOOST_PP_SEQ_ELEM(0, MEMBER) #define MEMBER_DATA_TYPE(MEMBER_DATA) BOOST_PP_SEQ_ELEM(1, MEMBER_DATA) #define MEMBER_DATA_NAME(MEMBER_DATA) BOOST_PP_SEQ_ELEM(2, MEMBER_DATA) #define MEMBER_DATA_VALUE(MEMBER_DATA) BOOST_PP_SEQ_ELEM(3, MEMBER_DATA) #define MEMBER_STRUCT_TYPE(MEMBER_STRUCT) BOOST_PP_SEQ_ELEM(1, MEMBER_STRUCT) #define MEMBER_STRUCT_NAME(MEMBER_STRUCT) BOOST_PP_SEQ_ELEM(2, MEMBER_STRUCT) #define MEMBER_STRUCT_MEMBER_SEQ(MEMBER_STRUCT) BOOST_PP_SEQ_ELEM(3, MEMBER_STRUCT) #define MEMBER_DATA(TYPE, NAME, VALUE) ((MEMBER_DATA_TAG)(TYPE)(NAME)(VALUE)) #define MEMBER_STRUCT(TYPE, NAME, MEMBER_SEQ) ((MEMBER_STRUCT_TAG)(TYPE)(NAME)(MEMBER_SEQ)) #define IS_MEMBER_STRUCT(MEMBER_SEQ_ELEM) BOOST_PP_EQUAL(MEMBER_TAG(MEMBER_SEQ_ELEM), MEMBER_STRUCT_TAG) #define MEMBER_STRUCT_DECLARE(TYPE, NAME, MEMBER_SEQ) \ struct TYPE \ { \ BOOST_PP_SEQ_FOR_EACH(MEMBER_ELEM_DECLARE, BOOST_PP_EMPTY(), MEMBER_SEQ) \ } NAME #define MEMBER_ELEM_DECLARE(_r, _data, MEMBER_SEQ_ELEM) \ BOOST_PP_IIF \ ( \ IS_MEMBER_STRUCT(MEMBER_SEQ_ELEM), \ MEMBER_STRUCT_DECLARE \ ( \ MEMBER_STRUCT_TYPE(MEMBER_SEQ_ELEM), \ MEMBER_STRUCT_NAME(MEMBER_SEQ_ELEM), \ MEMBER_STRUCT_MEMBER_SEQ(MEMBER_SEQ_ELEM) \ ), \ MEMBER_DATA_DECLARE \ ( \ MEMBER_DATA_TYPE(MEMBER_SEQ_ELEM), \ MEMBER_DATA_NAME(MEMBER_SEQ_ELEM), \ MEMBER_DATA_VALUE(MEMBER_SEQ_ELEM) \ ) \ ); #define MEMBER_DATA_DECLARE(TYPE, NAME, VALUE) TYPE NAME #define MEMBER_VALUE_INIT(MEMBER_SEQ) \ BOOST_PP_SEQ_FOR_EACH_I(MEMBER_VALUE_INIT_DECLARE, BOOST_PP_EMPTY(), MEMBER_SEQ); #define MEMBER_VALUE_INIT_DECLARE(_r, _data, i, MEMBER_SEQ_ELEM) \ BOOST_PP_COMMA_IF(i) \ BOOST_PP_IIF \ ( \ IS_MEMBER_STRUCT(MEMBER_SEQ_ELEM), \ {MEMBER_VALUE_INIT(MEMBER_SEQ_ELEM)}, \ MEMBER_DATA_VALUE(MEMBER_SEQ_ELEM) \ ) #define STATIC_CONST_STRUCT(TYPE, NAME, MEMBER_SEQ) \ static const MEMBER_STRUCT_DECLARE(TYPE, NAME, MEMBER_SEQ) = \ { \ MEMBER_VALUE_INIT(MEMBER_SEQ) \ } Thanks. -- "лось и белка"

2 1

[geometry] Difference problem using boost geometry in 1.47
by Macumber, Daniel 26 Oct '13

26 Oct '13

I am using boost geometry in 1.47 and am pretty happy with it. However, I am having a problem performing an intersection that I can't figure out how to solve. When I difference two geometries, I am sometimes left with a result that uses reverse winding to subtract a piece of the result rather than providing a "clean" polygon with winding in one direction only. I've posted an image here that demonstrates this: http://oi42.tinypic.com/2elfp0o.jpg I get the feeling that if I were to use the right models this might somehow be fixed? If not is there an algorithm that can clean up the result? Here is the code I am using: typedef boost::geometry::model::d2::point_xy<double> BoostPoint; typedef boost::geometry::model::polygon<BoostPoint> BoostPolygon; typedef boost::geometry::model::ring<BoostPoint> BoostRing; BoostRing facePolygon; BoostRing otherFacePolygon; std::vector<BoostPolygon> faceDifferenceResult; .... boost::geometry::difference(facePolygon, otherFacePolygon, faceDifferenceResult); Any ideas are much appreciated, thanks a lot, Dan

2 1

MFC ASSERT in dllinit.cpp line 703 when using Boost.Log
by Simon Martin 25 Oct '13

25 Oct '13

Hi, I receive an assertion, when I try to use Boost.Log in one of my DLLs. I have a tools.dll, which contains my logging facility and I tried to implement to logging stuff with Boost.Log. This tools.dll is referenced by another helper.dll and both DLLs are used in the application. However, when I start the application, I get the famous ASSERT. This happens with Boost 1.54 and Boost.Log as well as with Boost 1.48 and Boost.Log 1.1. I've digged through lots of mailing lists and old postings today. What I understood so far, there's an issue with _pRawDllMain pointer in Boost.Thread, some suggested commenting it out in #include <boost/thread/detail/tss_hooks.hpp>, but this leads to memory leaks? This thread http://lists.boost.org/threads-devel/2009/06/0476.php said it would be fixed after 1.40, but I don't know, if this has been done. Dynamic linking to Boost as suggested here http://stackoverflow.com/questions/5355931/adding-include-boost-thread-mute… did not help. Any hints, what I should check? Thanks Simon

1 0

Reading & Parsing an .ini input file using BOOST... ..!!
by Rahul Mathur 25 Oct '13

25 Oct '13

All, Need small help. I have an "input.ini" input file which contains sample values like – --- App=| 5000 | 5001 | 0 | | | | 1BYTE | --- Starting with word (viz. App), followed by equal sign, than data within pipes (|) and blank space between two pipes ( | | ). Anything only between two pipes is to be read. I am looking to parse this input file. After parsing, assign each data to struct member type and than send this complete struct as socket. I am looking to use Boost property tree (or the needed one) and finally the ASIO to send the complete filled struct as stream. I tried below code to read complete data and it worked but - --- #include <iostream> #include <boost/property_tree/ptree.hpp> #include <boost/property_tree/ini_parser.hpp> #include <boost/filesystem/fstream.hpp> int main() { boost::property_tree::ptree pt; boost::property_tree::ini_parser::read_ini("input.ini", pt); std::cout << pt.get<std::string>("App") << std::endl; } --- But, how to read between pipes? Any clue?

4 5