December 2012 - Boost-users - lists.preview.boost.org

boost mpi message truncated error
by Kyle Joseph Berger 10 Dec '12

10 Dec '12

Hello, I am using boost mpi and I am having issues getting the message truncation error upon receiving a message. It is not always the same message, but it does always happen and I have been unable to figure out why. Here is the code of interest: //send particles for(int i=0;i<world.size();i++) { world.isend(i,i,particles_to_be_sent[i]); } //receive particles int receive_counter=0; boost::optional<boost::mpi::status> msginfo=world.iprobe(mpi::any_source, world.rank()); while (receive_counter<world.size()) { if (msginfo) { world.recv(msginfo->source(),world.rank(),particles_received); for (unsigned int i=0;i<particles_received.size();i++) { //add particle to its associated box in the particle grid particle_grid[particles_received[i].location_number()].insert(pair<int,Particle>(particles_received[i].global_part_num(),particles_received[i])); //check to see if the particle is contained within the box if (particles_received[i].position().y()>grid_locations[particles_received[i].location_number()].bottom() && particles_received[i].position().y()<grid_locations[particles_received[i].location_number()].top() && particles_received[i].position().x()>grid_locations[particles_received[i].location_number()].west() && particles_received[i].position().x()<grid_locations[particles_received[i].location_number()].east() && particles_received[i].position().z()>grid_locations[particles_received[i].location_number()].south() && particles_received[i].position().z()<grid_locations[particles_received[i].location_number()].north()) { //if particle is in box, then denote it as such in the particle properties particle_properties[particles_received[i].global_part_num()].particle_in_box()[particles_received[i].location_number()]=true; } } receive_counter++; } msginfo=world.iprobe(mpi::any_source, world.rank()); } I have left out variable definitions, but everything has the right type (including sending and receiving the same type-particles_to_be_sent[i] and particles_received are both vector<Particle>, where Particle is a type I've created). I thought that boost mpi is supposed to take care of making the receive buffer the right size on its own, but something isn't working here. Any help would be appreciated. Thanks, Kyle

2 1

[exception, system] using boost::system::error_code with boost::exception
by Marco Craveiro 10 Dec '12

10 Dec '12

Hi boost users, I'm a very happy user of both error_code and exception, but now I found a new use case which is to combine them. So I'm looking for an example of how to add an error_code into an exception and get it all nicely formatted with diagnostic_information. I can see how to use error_info [1] but not error_code. Any pointers would be appreciated. Cheers Marco -- So young, and already so unknown -- Pauli blog: http://mcraveiro.blogspot.com [1] http://www.boost.org/doc/libs/1_52_0/libs/exception/doc/tutorial_transporti…

2 1

Oven, regular and Boost
by Robert Jones 10 Dec '12

10 Dec '12

The 'Oven' port... http://dl.dropbox.com/u/1682460/git/OvenToBoost/libs/range/doc/html/range_e… makes much of its 'regular' operator, which is used to fix this problem using boost::lambda::_1; for_each_(r | filtered(_1 % 2 == 0), f); // Error! Can't default construct/copy assign for_each_(r | filtered(regular(_1 % 2 == 0)), f); // OK However, AFAICS this is not a problem in current-ishBoost (1.47). Has this ever been a problem? Can anyone comment on its general provenance? Thx, Rob.

3 5

[Range] Sliced... less than it could be.
by Robert Jones 10 Dec '12

10 Dec '12

I was disappointed to discover that I can't write range | filtered( some_predicate ) | sliced( n, m ) | ..... because sliced requires a random access range, and filtered produces a forward range! Piffle. It seems such a useful and obvious use-case. Is this restriction really necessary? Thx, Rob.

3 6

[overloaded_function] Why does it define f_ as const?
by Felipe Magno de Almeida 10 Dec '12

10 Dec '12

Hello, I'm trying to use overloaded_function. But unfortunately it defines f_ as const, which makes it not Assignable. Which seems to me, like a unnecessary limitation. I looked at the docs, but couldn't find any mention of assignment nor a rationale for this behavior. Is there a specific reason for overloaded_function to not be Assignable? Regards, -- Felipe Magno de Almeida

1 0

How to implement a custom hash function for __uint128_t?
by Scott Smedley 10 Dec '12

10 Dec '12

Hi all, How can I implement a hash function for a __uint128_t type that is compatible with Boost? I tried this: #include <boost/functional/hash.hpp> #include <unordered_set> #ifdef USE_64 typedef uint64_t MyId; // works fine #else typedef __uint128_t MyId; #endif inline std::size_t hash_value (const MyId &x) { // not a very good hash function, but I just want to get it working first! return static_cast<std::size_t>(x); } int main (int argc, char *argv[]) { std::unordered_set<MyId, boost::hash<MyId>> s; s.insert(42); } but I get 144 lines of errors (!) using g++-4.7.2 with Boost v1.51.0 on 64-bit Scientific Linux 6.2: error: no matching function for call to 'hash_value(const __int128 unsigned&)' Any pointers/help would be muchly appreciated. Scott. :)

3 2

Re: [Boost-users] text_file_backend with append and rotate (Victor Yankee)
by Victor Whiskey Yankee 09 Dec '12

09 Dec '12

Anyone, Would someone show me what I am doing wrong setting up this logger? I posted on the boost log site too but got no reply there either. Thanks, Vic > Hello everyone! > > How to create a text_file_backend that appends to existing log file and > also has rotation. The init function below always creates a new file every > time I run my program no matter the rotation settings. > > I am using svn version 774 (latest as of Dec 5, 2012. I tried many ways but > no joy. There is not an example like this, and there does not seem to be a > test for appending either, so maybe it is not supported, but that would be > bad. So I must be doing something wrong. > > Here is my function to set up the logger (using my custom severity levels): > > void init() > { > typedef boost::log::sinks::synchronous_sink<boost::log::sinks::text_file_backend > FileSink_t; > > boost::shared_ptr< FileSink_t> mySink(new FileSink_t( > boost::log::keywords::file_name = "%Y%m%d_%H%M%S_%5N.log" > , boost::log::keywords::rotation_size = 16 * 1024 * 1024 /3 > , boost::log::keywords::time_based_rotation =boost::log::sinks::file::rotation_at_time_point(0,0,0) > , boost::log::keywords::auto_flush = true > , boost::log::keywords::open_mode=(std::ios::out |std::ios::app) // seems to have no effect > )); > > // this method of setting append does not seem to work either, so Imust be doing something wrong > mySink->locked_backend()->set_open_mode(std::ios::out|std::ios::app); > > > mySink->locked_backend()->set_file_collector(boost::log::sinks::file::make_collector( > boost::log::keywords::target = "logs" > , boost::log::keywords::max_size = 16 * 1024 * 1024 > , boost::log::keywords::min_free_space = 2 * 16 * 1024 * 1024)); > > mySink->set_filter(boost::log::filters::attr< LOGGING::LOG_LEVEL("Severity", std::nothrow) >= LOGGING::warning); > mySink->locked_backend()->scan_for_files(); mySink->set_formatter( > boost::log::formatters::format("[%1% | %2%] %3%") > % boost::log::formatters::date_time< boost::posix_time::ptime("TimeStamp") > % boost::log::formatters::attr< LOGGING::LOG_LEVEL >("Severity",std::nothrow) > % boost::log::formatters::message()); > > boost::log::core::get()->add_sink(mySink); > boost::log::core::get()->add_global_attribute("TimeStamp",boost::log::attributes::local_clock()); > }

1 0

Boost Filesystem references in static/shared library
by Miguel Guedes 09 Dec '12

09 Dec '12

Hello List, I was using boost::filesystem previously in a project that built an executable but it has now grown and the requirements changed, and I now have to build a portion of the code as a library; preferably an so. Since I've started building code as a library I've been getting errors about invalid references to boost::filesystem. Here's how I'm building the library: $ g++ -DBOOST_ALL_NO_LIB -fPIC -lrt -lstdc++ -shared -pthread -lboost_system -lboost_thread -lboost_filesystem -L/usr/local/lib -ldl -lpthread -lstdc++ -lclang -o libomnis.so cout.o command_line_processor.o command_options_container.o commands_container.o command_base.o command_option_base.o command_option.o command_option_multi.o command_option_integer.o command_option_flag.o command_serve.o command_send.o command_help.o command_version.o register.o unsaved_file.o clang_parameters.o ustring_vector.o string_vector.o request_strings.o source_file.o reply.o omnisource.o omnis.o As you can see I've tried building it with (as well as without) the BOOST_ALL_NO_LIB macro defined. I then compile the main executable as given: $ g++ libomnis.so client.o omnis_command_line_processor.o request_handler.o request_handler_1.0.o request_parser.o server.o connection.o omnisource_loader.o omnis_app.o main.o -o omnis -pthread -lboost_system -lboost_thread -lboost_filesystem -L/usr/local/lib -ldl -lpthread -lstdc++ -lclang -L. -lomnis And this results in the following output: ./libomnis.so: undefined reference to `boost::filesystem2::detail::dir_itr_close(void*&, void*&)' ./libomnis.so: undefined reference to `boost::filesystem2::detail::dir_itr_first(void*&, void*&, std::string const&, std::string&, boost::filesystem2::file_status&, boost::filesystem2::file_status&)' ./libomnis.so: undefined reference to `boost::filesystem2::detail::dir_itr_increment(void*&, void*&, std::string&, boost::filesystem2::file_status&, boost::filesystem2::file_status&)' ./libomnis.so: undefined reference to `boost::filesystem2::detail::status_api(std::string const&, boost::system::error_code&)' ./libomnis.so: undefined reference to `boost::filesystem2::detail::not_found_error()' collect2: error: ld returned 1 exit status What am I missing here? PS: No errors occurred previously when I was just building the executable. -- Miguel

1 0

Why is cref required in this nested bind example?
by Chris Stankevitz 08 Dec '12

08 Dec '12

Hello, Can someone please explain why a cref is needed in the last line of the attached code snippet? Thank you, Chris === #include <boost/bind.hpp> #include <boost/function.hpp> using boost::bind; using boost::cref; using boost::function; void a() { } void b(function<void()>) { } void f() { //bind(b, bind(a)); bind(b, cref(bind(a))); }

4 7

boost shared_ptr with own dtor
by Philipp Kraus 08 Dec '12

08 Dec '12

Hello, I'm using boost::shared_ptr for using a C pointer. The ctor and all functions are called without any problem, but I have got a problem with the dtor. I need to call a C function on the dtor eg cptr_close( mypointer ) How can I this used with boost::shared_ptr? I need the cptr_close call on the pointer member of the dtor on the shared_ptr. Thanks a lot Phil

3 7