June 2014 - Boost-users - lists.preview.boost.org

Re: [Boost-users] [Boost.Test] BOOST_CHECK_THROW behaviour
by Ghyslain Leclerc 23 Jun '14

23 Jun '14

On 23/06/2014 17:19:01 +1200, Gavin Lambert wrote >>You could try submitting that as a bug report in Trac (if it's not already there), or waiting for Gennadiy Rozental's response. Done. I hope I did it properly. Never reported a bug on Boost before. Quoted our discussion, so that some more info is available if necessary. >>In the meantime, you'd need to write a more intelligent test yourself. I will do it manually. Found someone on Stack Overflow who wanted the same behavior I expected and did it. In case it helps someone, here is the snippet: try { … code that should throw ... BOOST_ERROR( “<function_name> did not throw anything. Expected a <exception_name>” ); } catch( <exception_name> const& ) { //ok, this is the expected case } catch(…) { BOOST_ERROR( “<function_name> has throw an unknown exception, but a <exception_name> was expected.” ); } and the reference: http://stackoverflow.com/questions/21774387/boost-unit-testing-exceptions Thanks again for your help.

1 0

[Boost.Test] BOOST_CHECK_THROW behaviour
by Ghyslain Leclerc 23 Jun '14

23 Jun '14

Hello, Sorry for the long post! I have a simple class, which I am posting a fraction of here for discussion sake: ------------------------------------------------------ class AsciiTableRow { public: void setCol( std::size_t colIdx, std::string const& newContent ); void setCol( std::size_t colIdx, const char* newContent ); private: void checkIdxBound_( std::size_t colIdx, std::string const& msg ); std::vector< std::string > data_; bool maxColWidthIsValid_; } inline void AsciiTableRow::setCol( std::size_t colIdx, std::string const& newVal ) { checkIdxBound_( colIdx, "setCol" ); maxColWidthIsValid_ = false; data_[colIdx] = newVal; } inline void AsciiTableRow::setCol( std::size_t colIdx, const char* newContent ) { setCol( colIdx, std::string( newContent ) ); } inline void AsciiTableRow::checkIdxBound_( std::size_t colIdx, std::string const& msg ) { if( colIdx > data_.size() ) { throw std::out_of_range( "AsciiTableRow::" + msg + " : colIdx out of range" ); } } ------------------------------------------------------ The code above should be sufficient for the discussion and I don’t think any other code is relevant. Should you need to see my constructor and destructor, nothing is virtual so I just copy every value and set some defaults. Nothing fancy, but I will gladly post the code if need be. What I do then is try and devise a test to see if the exception is indeed thrown and if it is the correct one. Here is the code I use: AsciiTableRow testRow; // empty, so any index will throw BOOST_CHECK_THROW( testRow.setCol( 30, 12 ), std::out_of_range ); BOOST_CHECK_THROW( testRow.setCol( 30, 12l ), std::out_of_range ); This works fine. Then, I had the idea to check what the behaviour would be if the wrong exception is thrown. I simply changed the first call like this : BOOST_CHECK_THROW( testRow.setCol( 30, 12 ), std::runtime_error ); BOOST_CHECK_THROW( testRow.setCol( 30, 12l ), std::out_of_range ); I was under the impression (and I might be wrong here, sorry :-( ) that the Boost.Test library would (the numbers are not to mean that order, simply there for enumeration): 1- output a message saying the test has failed, that an exception was indeed thrown, but it was not a runtime_error 2- continue the rest of the test (single test currently) 3- terminate but I actually never get the message of step 1, the execution of the rest of the test seems never occur (the second BOOST_CHECK_THROW) and the program terminate because of my exception and not very gracefully. I guess my questions to the other users here are: - Was my understanding of what BOOST_CHECK_THROW does flawed ? - Is the behaviour normal ? - Is there an obvious problem I am missing with the design of my simple class ? In case it is useful, I am on OSX 10.9. I have compiled boost myself (sources of version 1.55 downloaded from the Boost site). I compile my work with C++11 enabled. I am using GCC 4.7 installed via Macports. Anyhow, any insight will be appreciated. As everyone is busy, thanks in advance to anybody for any time spent on this. Ghyslain

2 1

[asio] Closing async client socket after writing last message
by Nicholas Yue 23 Jun '14

23 Jun '14

Hi, I have adapted the chat tutorial example to send messages to a similar (chat tutorial based) server. It is currently working fine however due to the async_write(), I can't close the socket until after all the pending data has been processed on the server end. I have google and found discussion and mentioned about async_write() completion handler calling socket.close() however, I was not able to find specific detail as to how that is done. I want to avoid closing the connection prematurely as it is asynchronous writing. My current do_write() looks like this void DisplayChatClient::do_write(DisplayChatMessage msg) { DLOG(INFO) << "DisplayChatClient::do_write()"; bool write_in_progress = !write_msgs_.empty(); write_msgs_.push_back(msg); if (!write_in_progress) { boost::asio::async_write(socket_, boost::asio::buffer(write_msgs_.front().data(), write_msgs_.front().length()), boost::bind(&DisplayChatClient::handle_write, this, boost::asio::placeholders::error)); } } void DisplayChatClient::handle_write(const boost::system::error_code& error) { DLOG(INFO) << "DisplayChatClient::handle_write()"; if (!error) { write_msgs_.pop_front(); if (!write_msgs_.empty()) { boost::asio::async_write(socket_, boost::asio::buffer(write_msgs_.front().data(), write_msgs_.front().length()), boost::bind(&DisplayChatClient::handle_write, this, boost::asio::placeholders::error)); } } else { do_close(); } } Cheers -- Nicholas Yue Graphics - Arnold, Alembic, RenderMan, OpenGL, HDF5 Custom Dev - C++ porting, OSX, Linux, Windows http://au.linkedin.com/in/nicholasyue https://vimeo.com/channels/naiadtools

4 7

Boost Filesystem fails with FreeBSD 10, clang, and boost
by Ricky Huang 21 Jun '14

21 Jun '14

Hello all, I am currently migrating to a FreeBSD 10 system which comes default with clang language. I was able to compile Boost just fine. But when I include filesysstem.hpp to it, I get the following error message: > Linking CXX executable boost_clang_test > /usr/bin/ld: : invalid DSO for symbol `_ZN5boost6system15system_categoryEv' definition > /usr/local/lib/libboost_system.so.1.55.0: could not read symbols: Bad value > CC: error: linker command failed with exit code 1 (use -v to see invocation) > *** Error code 1 > > Stop. I am using cmake as my build system, could it be possible that I need to adjust my cmake variables? I am not sure if I've came to correct forum or should this be on FreeBSD, clang, CMake forum(s)? Please let me know if I am at the wrong place… The minimum amount of code to induce the problem are the following 2 files: CMakeLists.txt: cmake_minimum_required(VERSION 2.8) # Find and include Boost libraries find_package(Boost 1.55 COMPONENTS filesystem REQUIRED) include_directories(${Boost_INCLUDE_DIR}) set(LIBS ${LIBS} ${Boost_LIBRARIES}) set(BOOST_CLANG_TEST_SOURCES ${SOURCES} main.cpp) add_executable(boost_clang_test ${BOOST_CLANG_TEST_SOURCES}) target_link_libraries(boost_clang_test ${LIBS}) main.cpp: #include <stdio.h> #include <iostream> #include <sstream> #include <string> #include <boost/filesystem.hpp> using namespace std; using namespace boost::filesystem; const char *progname; int main(int argc, char **argv){ return 0; } Thanks in advance.

3 4

[graph] kamada_kawai_spring_layout warnings
by Michael Olea 21 Jun '14

21 Jun '14

Hi. I'm getting compiler warnings for subscripts out of range from kamada_kawai_spring_layout.hpp. I'm using boost_55_0, compiled with clang: Apple LLVM version 5.1 (clang-503.0.40) (based on LLVM 3.4svn) Target: x86_64-apple-darwin13.2.0 Thread model: posix ../../../../../include/boost/graph/kamada_kawai_spring_layout.hpp:110:49: warning: array index 2 is past the end of the array (which contains 2 elements) [-Warray-bounds] double denom = mat[0][0] * (mat[1][1] * mat[2][2] - mat[2][1] * mat[1][2]) The offending code is: template <> struct linear_solver<3> { template <typename Vec> static Vec solve(double mat[2][2], Vec rhs) { double denom = mat[0][0] * (mat[1][1] * mat[2][2] - mat[2][1] * mat[1][2]) - mat[1][0] * (mat[0][1] * mat[2][2] - mat[2][1] * mat[0][2]) + mat[2][0] * (mat[0][1] * mat[1][2] - mat[1][1] * mat[0][2]); double x_num = rhs[0] * (mat[1][1] * mat[2][2] - mat[2][1] * mat[1][2]) - rhs[1] * (mat[0][1] * mat[2][2] - mat[2][1] * mat[0][2]) + rhs[2] * (mat[0][1] * mat[1][2] - mat[1][1] * mat[0][2]); double y_num = mat[0][0] * (rhs[1] * mat[2][2] - rhs[2] * mat[1][2]) - mat[1][0] * (rhs[0] * mat[2][2] - rhs[2] * mat[0][2]) + mat[2][0] * (rhs[0] * mat[1][2] - rhs[1] * mat[0][2]); double z_num = mat[0][0] * (mat[1][1] * rhs[2] - mat[2][1] * rhs[1] ) - mat[1][0] * (mat[0][1] * rhs[2] - mat[2][1] * rhs[0] ) + mat[2][0] * (mat[0][1] * rhs[1] - mat[1][1] * rhs[0] ); Vec result; result[0] = x_num / denom; result[1] = y_num / denom; result[2] = z_num / denom; return result; } }; Looks like it should be: static Vec solve(double mat[3][3], Vec rhs) ... No?

1 0

stack unwinding with boost::fcontext
by David Frank 20 Jun '14

20 Jun '14

Context - 1.53.0 beta1 Context - 1.53.0 beta1 Context Struct fcontext_t and related functions Each instance of fcontext_t represents a context (CPU registers and stack space). View on www.boost.org Preview by Yahoo I'm wondering if there is any example showing how to do stack unwinding as described at the bottom of the above link as quoted below: " Stack unwinding Sometimes it is necessary to unwind the stack of an unfinished context to destroy local stack variables so they can release allocated resources (RAII pattern). The user is responsible for this task. " Thanks, Frank

2 1

[coroutine] await_emu example
by bbxiong 20 Jun '14

20 Jun '14

Hi, coroutine listers! The example is very awesome. but after reading though the source, i have a question: Is any limitation makes a push_coroutine must be caller and pull_coroutine must be the coro? This way, we have to yield out at the first line of coroutine function by calling caller(), just like the example showed: typedef coroutines::pull_coroutine<void> coro_pull; typedef coroutines::pull_coroutine<void> coro_push; auto asynchronous(F f) -> future<decltype(f())> { typedef promise<decltype(f())> CoroPromise; CoroPromise coro_promise; auto coro_future = coro_promise.get_future(); CurrentCoro current_coro = { make_shared<coro_pull>(std::bind([f](CoroPromise &coro_promise, coro_push &caller) { caller(); coro_stack.top().caller = &caller; coro_promise.set_value(f()); }, std::move(coro_promise), placeholders::_1); }; coro_stack.push(std::move(current_coro)); (*coro_stack.top().coro)(); coro_stack.pop(); return coro_future; } There is an extra call to caller() in coroutine function to jump out to prepare coro stack. Why not swap coro_pull and coro_push like this: auto asynchronous(F f) -> future<decltype(f())> { typedef promise<decltype(f())> CoroPromise; CoroPromise coro_promise; auto coro_future = coro_promise.get_future(); CurrentCoro current_coro = { make_shared<coro_push>(std::bind([f](CoroPromise &coro_promise, coro_pull &caller) { coro_stack.top().caller = &caller; coro_promise.set_value(f()); }, std::move(coro_promise), placeholders::_1); }; coro_stack.push(std::move(current_coro)); (*coro_stack.top().coro)(); coro_stack.pop(); return coro_future; } This still works on my machine, and has no extra "call to caller()", push_coroutine is not invoked until we explicitly run it (while pull_coroutine is invoked in constructor). So, my question is, s there any bad effects by using coroutines in this way? Thanks in advance!

2 1

How to use vertex mapping parameter in copy_graph
by Christoph Czurda 19 Jun '14

19 Jun '14

Hello! I want to create a straight line drawing of a planar graph that is not maximal planar. So I want to follow the instructions in http://www.boost.org/doc/libs/1_55_0/libs/graph/doc/straight_line_drawing.h… : "If you want to draw a graph g, you can create a copy g' of g, store a mapping m of vertices in g' to vertices in g, triangulate g', and then send g' in as the input to chrobak_payne_straight_line_drawing. The drawing returned can then be applied to g using m to translate vertices from one graph to another." But I'm stuck at the point where I need to make a copy of the graph and store the vertex-to-vertex mapping between the two graphs. I'm not good enough at C++ to understand from the documentation at http://www.boost.org/doc/libs/1_55_0/libs/graph/doc/copy_graph.html how to provide or access the last parameter orig_to_copy(Orig2CopyMap c) If someone could fill in the missing code in this function, I would be most thankful! void doSomething(MyGraph_t& g){ MyGraph_t g_copy; copy_graph(g, g_copy, [...???...]); // here I would like to access the Orig2CopyMap which stores the mapping }

1 0

Prominent Booster makes a big splash
by Robert Ramey 19 Jun '14

19 Jun '14

Notice that our own Matthias Troyer - important boost contributor made a big splash. http://www.theverge.com/2014/6/19/5824336/google-s-quantum-computer-just-fl… Robert Ramey -- View this message in context: http://boost.2283326.n4.nabble.com/Prominent-Booster-makes-a-big-splash-tp4… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

[archive] binary archive to memory stream
by Michael Powell 19 Jun '14

19 Jun '14

Hello, I am wrapping my head around archive, specifically binary archive. I'd like to serialize a vector of objects in such a way, I need to capture intermediate vector buffers (literally, to vector of uint8_t) so that I can determine their sizes. After which point the parent object(s) can serialize to file using more conventional, built-in file streams and such. It seems that binary_oarchive is helpful, but still requires there be an ostream or streambuf in the mix. Which, as an adapter layer, while helpful, it still leaves the question of extending streambuf to work with ostream to vector<uint8_t> open. I've read some docs on vectors and streambuf, but they all seem to be geared towards a known-size vector, istream and such. I am interested in a vector that can be streamed to, will grow in size, etc. Some clues seem to hint at underflow, overflow type issues? Any helpful pointers how to go about doing this? Thanks... Best regards, Michael Powell

4 5

[Python]compiling python library
by Miklós Tóth 19 Jun '14

19 Jun '14

Hi All, I want to embed python in my c++ code using boost.python library. As far as I understand I need python headers and libs. So my python27.lib is "incompatible". I want to compile it from code but I failed with mingw32. Could you give me some advice, or links to achieve my goal? I'm using gcc4.6.3 and qt 4.8.2 and boost 1.49 on windows 7. Thanks, Miklós

2 9

Possible problem with property tree and ini files
by Tim Burgess 19 Jun '14

19 Jun '14

Hi, I make the following call in my application, based on Boost 1.55.0: property_tree::ini_parser::read_ini(IniFile.GetFullPath().ToStdString(), IniTree); This works fine for .ini files up to 64kb, but throws an exception for larger files. Is this a known limitation and, if so, can anybody suggest an alternative that could cope with larger files, please? Best wishes. Tim Burgess

1 0

Splitting string at whitespace
by Florian Lindner 17 Jun '14

17 Jun '14

Hello, for some time now I've been trying to split a given string at whitespaces: "A B C" -> "A", "B", "C". I want to use boost/algorithm/string #include "boost/algorithm/string/split.hpp" #include "boost/algorithm/string/compare.hpp" std::vector<std::string> tokenize(const std::string& text ) { std::vector<std::string> tokens; boost::algorithm::split(tokens, text, boost::algorithm::is_equal(" ")); return tokens; } but compilation on gcc 4.9.0 / boost 1.55.0 just gives: test.cpp: In function 'std::vector<std::basic_string<char> > tokenize(const string&)': test.cpp:40:71: error: no matching function for call to 'boost::algorithm::is_equal::is_equal(const char [2])' boost::algorithm::split(tokens, text, boost::algorithm::is_equal(" ")); ^ test.cpp:40:71: note: candidates are: In file included from /usr/include/boost/algorithm/string/finder.hpp:24:0, from /usr/include/boost/algorithm/string/split.hpp:17, from test.cpp:7: /usr/include/boost/algorithm/string/compare.hpp:34:16: note: boost::algorithm::is_equal::is_equal() struct is_equal ^ /usr/include/boost/algorithm/string/compare.hpp:34:16: note: candidate expects 0 arguments, 1 provided /usr/include/boost/algorithm/string/compare.hpp:34:16: note: boost::algorithm::is_equal::is_equal(const boost::algorithm::is_equal&) /usr/include/boost/algorithm/string/compare.hpp:34:16: note: no known conversion for argument 1 from 'const char [2]' to 'const boost::algorithm::is_equal&' What is wrong there? Thanks, Florian

2 1

Boost Voronoi: Automate creation of graph for indoor navigation
by Stadin, Benjamin 16 Jun '14

16 Jun '14

Hi all, I’m provided with geometry data for buildings. The data contains info about the „feature“ type of a geometry (floor, room, stairway, elevator). Now I need to automate the creation of an indoor routing graph (room-to-room navigation) to minimize manual works later on. This is not for a robot scenario – the line parts will be shown on a map application. Therefore it should also look good (routing edges should be centered etc). This is new land for me and I’ll will be grateful for advise how to achieve this. My initial thought is that Voronoi may be helpful. But this alone will probably not be sufficient. Any pointers and examples welcome. Thanks Ben

2 1

[numeric conversation] behavior on std::numeric_limits<double>::max() + eps with numeric_cast
by Olaf Peter 16 Jun '14

16 Jun '14

Hello, at this time I play with numeric_cast and struggle with this: first: double z = std::numeric_limits<double>::max(); z += 2*std::numeric_limits<double>::epsilon(); // doesn't matter if 1*eps std::cout << z << std::endl; std::cout << boost::numeric_cast<double>(z) << '\n'; I would expect that an exception is thrown which isn't, the output is 1.79769e+308 1.79769e+308 and as second: If I write my onw UDT as shown in the tutorial like: namespace my { enum range_check_result { cNonFinite = 0, cInRange = 1, cNegOverflow = 2, cPosOverflow = 3 } ; //! Define a custom range checker template<typename Traits, typename OverFlowHandler> struct range_checker { typedef typename Traits::argument_type argument_type; typedef typename Traits::source_type S; typedef typename Traits::target_type T; //! Check range of integral types. static range_check_result out_of_range(argument_type value) { if(!std::isfinite(value.to_builtin())) // Neither infinity nor NaN. return cNonFinite; else if(value > boost::numeric::bounds<T>::highest()) return cPosOverflow; else if(value < boost::numeric::bounds<T>::lowest()) return cNegOverflow; else return cInRange; } static void validate_range(argument_type value) { BOOST_STATIC_ASSERT(std::numeric_limits<T>::is_bounded); OverFlowHandler()(out_of_range(value)); } }; .... struct bad_numeric_cast : virtual boost::exception, virtual std::bad_cast { ... }; struct rounding_error : virtual boost::exception, std::domain_error { rounding_error() : std::domain_error("NaN or Infinity") {} // XXX Better solution? virtual const char * what() const throw() { return "bad numeric conversion: can not be represented in the target integer type"; } }; struct negative_overflow : virtual bad_numeric_cast { .... }; struct positive_overflow : virtual bad_numeric_cast { .... }; struct overflow_handler { void operator()(range_check_result result) { if(result == cNonFinite) BOOST_THROW_EXCEPTION(rounding_error()); else if(result == cNegOverflow) BOOST_THROW_EXCEPTION(negative_overflow()); else if(result == cPosOverflow) BOOST_THROW_EXCEPTION(positive_overflow()); } }; it doesn't change the result - is this the right way? Thanks, Olaf

4 15

[asio]remote_endpoint throws exception
by higashiyama＠ncad.co.jp 16 Jun '14

16 Jun '14

Hi, I wrote TCP client program using boost::asio. OS: Debian GNU/Linux (x86_64) Boost: 1.49 I use asio::ip::tcp::socket::async_conect function and I called ip::tcp::socket::remote_endpoint at async_connect callback. Like this: // async_resolve handler void my_class::handle_resolve(const boost::system::error_code& error, boost::asio::ip::tcp::resolver::iterator endpoint_iterator) { if (!error) { socket_.async_connect(endpoint, strand_.wrap(...)); } } void my_class::handle_connect(const boost::system::error_code& error, boost::asio::ip::tcp::resolver::iterator endpoint_iterator) { if (!error) { // socket_ is my_class member. socket_.remote_endpoint(); // throw exception ! } ... } ip::tcp::socket::remote_endpoint sometime throws boost::system::system_error exception (error message is "remote_endpoint: Transport endpoint is not connected"). But async_connect is successfully completed. Why does remote_endpoint function throw exception? Thank you,

3 3

[boost::filesystem] Improvement/Suggestion for copy()
by Kyle Ketterer 15 Jun '14

15 Jun '14

Hello all, I was curious about the possible implementation of a progress call back for the copy() ( and related functions) in ::filesystem. With WinAPI you can use CopyFileEx with the possibility of a progress call back. It would be nice to have this ability "out of the box" in boost's namespace. This would be especially be useful for transferring larger files. Thoughts? Thanks

1 0

[geometry] how to using UDT
by Olaf Peter 15 Jun '14

15 Jun '14

Hello, I try to catch some failures using an own user defined type (UDT), but it doesn't compile e.g. it's hardcoded to use std::sqrt: boost_1_55_0/boost/geometry/strategies/cartesian/distance_pythagoras.hpp:165:13: error: no matching function for call to 'sqrt(core::my_type<double>)' further, is there a way to adapt templated points? BOOST_GEOMETRY_REGISTER_POINT_2D(point_type, point_type::coordiante_type, bg::cs::cartesian, x, y) got errrors. Thanks, Olaf

2 2

BGL performance: is pool_allocator possible and does it help iteration performance?
by Jeremy Murphy 15 Jun '14

15 Jun '14

Hi, I'm implementing a graph algorithm that requires a lot of repeated iteration over a graph as it builds it. So I wonder about cache misses. Is it possible to use a pooled allocator with the vector/list that gets used in the VertexList and EdgeList properties of adjacency_list? If there are other techniques for improving search/iteration performance (at the cost of insertion), I'd also be interested. Thanks! Jeremy

1 0

Boost Library Incubator recommended by steering committee
by Beman Dawes 14 Jun '14

14 Jun '14

The Boost Library Incubator (www.blincubator.com) is a web site conceived, designed, and implemented to alleviate the log-jam of libraries waiting to be reviewed by Boost. It helps new library submitters through the process, encourages development of an initial user base, and accumulates feedback in the form of comments, test results and Boost style reviews. The Boost Steering Committee has voted to recommend using the Boost Library Incubator as a prelude to getting a library reviewed by Boost. The Steering Committee recommends libraries currently in the Boost review queue be added to the Incubator by their developers. The relationship between Boost and the Incubator is purely informal, but there as several things Boosters can do to help make the Incubator a success. Authors can submit their libraries and respond to feedback. Users can experiment with submitted libraries and provide feedback on their experience. WordPress/php developers can contribute enhancements to the Boost Library Incubator implementation. The steering committee extends its thanks to Robert Ramey for launching the Boost Library Incubator! --The Boost Steering Committee

3 2

[utility] prior(it, n) for n being an unsigned type
by Adam Romanek 13 Jun '14

13 Jun '14

Hi! I've just finished investigating an issue in my code which led it to an assert/abort in certain cases. It turned out it was caused by the use of boost::prior(it, n) where n was unsigned int with some non-zero value. I found a question [1] on the mailing list back in 2004 which describes this issue. Additionally a solution to this problem was also posted in one of the replies [2]. My question is - why hasn't it made into the repository yet? The current behavior of boost::prior(it, n) when n is unsigned type is far from the one expected by its potential users so I think it should either be properly documented (warned) or fixed immediately. Am I not seeing something? WBR, Adam Romanek [1] http://lists.boost.org/Archives/boost/2004/02/61556.php [2] http://lists.boost.org/Archives/boost/2004/02/61578.php

1 0

std::limits compile errors with boost
by David Frank 13 Jun '14

13 Jun '14

Anyone has seen this?: In file included from ../include/boost/lockfree/detail/freelist.hpp:12, from ../include/boost/lockfree/queue.hpp:24, from test.cpp:2: /usr/local/powerpc-linux-gnu/include/c++/4.3.2/limits:287:22: error: macro "min" requires 2 arguments, but only 1 given Thanks

4 3

<boost::wave> Memory issue when using wave to run a file parser service
by Yong Huang 12 Jun '14

12 Jun '14

Hi all, I run into a memory issue when uses boost wave library. My application runs as a *service *that parses c files. In each individual run, it calls in the wave library through a sub-function call to process one file. I noticed that some memory is not released from wave after each run. The consequence is that the memory usage of the application keeps increasing till the system breaks down. I could reproduce the problem with a simple modification of the quick_start sample code shipped with the boost wave library. (attached below). My system is win64 and I see the problem with both 1.49 and 1.55. (built with b2 toolset=msvc-11.0 address-model=64 --build-type=complete) Something I found during the investigation: * The problem is gone if I build the application with link<static>. (Unfortunately, this is not an option for me) * From the memory usage pattern, it seems that there is a memory pool in place that should have been released after each turn. When repeatedly passing a 4MB file, the memory usage (taken from task maneger) is: 99,700K, 198,404K, 395,404K, 395408K, and 789,396K. When running with static link, the memory usage after each run is: 1800K, 2480K, 2480K, .... The maximum usage is 8300K. Is there any workaround with the problem? Thanks, Yong ------------------------------------- begin of the code --------------------------------------------------- #define BOOST_WAVE_SERIALIZATION 0 // enable serialization #define BOOST_WAVE_BINARY_SERIALIZATION 0 // use binary archives #define BOOST_WAVE_XML_SERIALIZATION 1 // use XML archives /////////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////////////// // Include Wave itself #include <boost/wave.hpp> /////////////////////////////////////////////////////////////////////////////// // Include the lexer stuff #include <boost/wave/cpplexer/cpp_lex_token.hpp> // token class #include <boost/wave/cpplexer/cpp_lex_iterator.hpp> // lexer class // sub-function that calls into wave int do_it(std::string filename) { boost::wave::util::file_position_type current_position; std::ifstream instream(filename); std::string instring; instream.unsetf(std::ios::skipws); instring = std::string(std::istreambuf_iterator<char>(instream.rdbuf()), std::istreambuf_iterator<char>()); typedef boost::wave::cpplexer::lex_token<> token_type; typedef boost::wave::cpplexer::lex_iterator<token_type> lex_iterator_type; typedef boost::wave::context<std::string::iterator, lex_iterator_type> context_type; typedef boost::wave::cpplexer::lex_token<> TokenType; typedef boost::wave::cpplexer::lex_iterator<TokenType> LexIterator; LexIterator aIter(instring.begin(), instring.end(), TokenType::position_type( filename.c_str()), boost::wave::language_support( boost::wave::support_cpp | boost::wave::support_option_long_long ) ); LexIterator endIter; for (; aIter != endIter; ) { (*aIter).get_value(); ++aIter; } return 0; } int main(int argc, char *argv[]) { int aNum = 1; for (;;) { do_it("c:\\memleak.txt"); std::cout << "Process the file again (0/1)? \n"; scanf("%d", &aNum); if (aNum == 0) { printf("\nYou entered zero so we stop!\n"); return 0; } } return 0; }

2 1

[python] TypeError when left side is noncopyable
by Igor R 12 Jun '14

12 Jun '14

I'm trying to expose an operator: struct my_class : boost::noncopyable {}; my_class &operator <<(my_class &e, int i); // in the module definition: class_<my_class, boost::noncopyable>("my_class") .def(self << int()) ; However, an attempt to invoke this operator in python causes the following error: TypeError: No to_python (by-value) converter found for C++ type: my_class I guess this is related to the fact my_class is noncopyable, isn't it? Is there a workaround for this issue? Thanks.

1 0

Meeting C++ Schedule released
by Jens Weller 12 Jun '14

12 Jun '14

Hello C++ & Boost Community, since last week the schedule for Meeting C++ is released: http://meetingcpp.com/index.php/schedule14.html Like last year Meeting C++ will feature 21 Talks in 3 Tracks, the conference will be opened by Scott Meyers, the closing Keynote will be given by Hartmut Kaiser. This years Theme Track is "Scientific Programming with C++". Also, if your are a student, you have the chance to participate in our student program which will give 50 Students access to the conference: http://meetingcpp.com/index.php/newsreader/items/meeting-cpp-2014-the-stude… kind regards, Jens Weller

1 0