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

Inconsistent 1.55 downloads
by Rob Conde 11 Feb '14

11 Feb '14

The contents of boost_1_55_0.zip and boost_1_55.tar.bz2 are not consistent from: http://sourceforge.net/projects/boost/files/boost/1.55.0/ The tar.bz2 version compiles on linux, but the zip version does not. It complains about unknown modules when running b2. Rob Conde

2 2

How to compile boost with non-standard clang installations.
by Bill White 11 Feb '14

11 Feb '14

I am trying to build boost using clang. I know there are many sources of advice for doing this, but I can't seem to find a description of my exact system, and my attempts to adapt them have not been notably successful. Here are my configuration constraints: 1.) Clang is not the system compiler. It is compiled in installed in separately, with prefix $a/clang. So, $a/clang/bin/clang++ is the C++ compiler, and so forth. I am using the tip of the llvm/clang tree. 2.) I want to use libc++, which is installed in $a/clang/lib/libc++.so. I also want to use the libc++ abi, which is installed in $a/clang/lib/libcxxabi.so. 3.) The libc++ standard library include files are installed in $a/include/c++/v1. For most programs I compile with the options -nostdinc++ -isystem $a/clang/include/c++/v1 -std=c++11 and link with -stdlib=libc++ -L$a/clang/lib -lcxxabi This seems to work pretty well for most circumstances. However, I can't figure out how to translate this to boost 1.55.0. I tried: ./bootstrap.sh --with-toolset=clang --prefix=$a/clang ./b2 toolset=clang \ cxxflags="-std=c++11 -nostdinc++ -isystem $a/clang/include/c++/v1 -stdlib=libc++" \ linkflags="-stdlib=c++ -L$a/clang/lib" \ --prefix=$a/clang -j10 define=BOOST_SYSTEM_NO_DEPRECATED stage release However, the compilations failed. They all seemed to fail compiling type_traits. A typical failure is: clang-linux.compile.c++.without-pth bin.v2/libs/regex/build/clang-linux-3.0/release/threading-multi/icu.o In file included from libs/regex/build/../src/icu.cpp:20: In file included from ./boost/regex/config.hpp:43: In file included from ./boost/config.hpp:40: In file included from ./boost/config/select_stdlib_config.hpp:37: In file included from ./boost/config/no_tr1/utility.hpp:21: In file included from /local/billw/arch/glnxa64/clang/include/c++/v1/utility:157: In file included from /local/billw/arch/glnxa64/clang/include/c++/v1/__tuple:16: /local/billw/arch/glnxa64/clang/include/c++/v1/type_traits:3308:31: error: expected expression static const bool value = decltype(__test<_Tp>(nullptr))::value; ^ 1 error generated. "clang++" -c -x c++ -std=c++11 -nostdinc++ -isystem /local/billw/arch/glnxa64/clang/include/c++/v1 -O3 -finline-functions -Wno-inline -Wall -pedantic -pthread -fPIC -std=c++11 -nostdinc++ -isystem /local/billw/arch/glnxa64/clang/include/c++/v1 -DBOOST_ALL_NO_LIB=1 -DBOOST_REGEX_DYN_LINK=1 -DBOOST_SYSTEM_NO_DEPRECATED -DNDEBUG -I"." -o "bin.v2/libs/regex/build/clang-linux-3.0/release/threading-multi/icu.o" "libs/regex/build/../src/icu.cpp" ...failed clang-linux.compile.c++.without-pth bin.v2/libs/regex/build/clang-linux-3.0/release/threading-multi/icu.o... (Here $a is /local/billw/arch/glnxa64.) I apologize for the length of this, but I didn't know what to leave out. Thanks in advance.

1 1

[property_tree] hex integer translator
by Steve Hartmann 11 Feb '14

11 Feb '14

I am using property_tree to parse some xml files provided by outside sources. Within those xml files, sometimes integers are represented as decimal, others are represented as hex. Below is an example line where a property named Index uses a hex representation. <Index>#x7000</Index> So I tried to read this as an integer, with the following source line, but that throws an exception since it does not now how to translate the hex format from the xml file. deviceList.second.get<int>("Index") So my question is: is there a translator already included in the property_tree <that I just cannot find>, or do I need to write my own translator to deal with hex integers. Thanks, Steven

1 0

[container.string] why isn't memcpy used in the internals?
by Adam Romanek 11 Feb '14

11 Feb '14

Hi! Recently I've been trying to improve the performance of a code that makes heavy use of std::string (lots of constructions and destructions). My first thought was to use boost::container::string as it does not use reference-counting and uses small-string optimization, so it should be a perfect replacement for std::string when it comes to performance. Right... But the problem is that switching to boost::container::string only made things worse in the general case. I wrote "in the general case" because when strings to be held are small and SSO may be applied then boost::container::string outperforms std::string considerably. However, in the general case constructing boost::container::string from a C string is significantly slower than constructing std::string. My investigation showed that std::string uses memcpy to copy the C string into its buffer while boost::container::string does not. The memcpy comes from std::char_traits::copy so maybe boost::container::string could use it too? I guess the rest of operations from std::char_traits (find, move, assign etc.) could be utilized too as std::char_traits has been carefully crafted with performance in mind. Should I create a ticket for this issue? Best regards, Adam Romanek

2 4

How to use graph properties
by Paolo Bolzoni 11 Feb '14

11 Feb '14

Dear list, I need to keep track of two special vertices in my graph, So I thought to attach two vertex_descriptor as graph properties so have easy access any moment. Properties work fine for edges or vertices, but I cannot make them work for the graph. In the bottom is the C++ code. The graph has the two properties, now how I cat get and put them? This code does not work and the error messages are impossible to read for me. Otherwise that is the way to keep graph-wise properties? The code compiles removing the two puts, so I assume the error is there. Yours sincerely, Paolo #include <boost/graph/adjacency_list.hpp> namespace boost { enum graph_source_t { graph_source = 100 }; BOOST_INSTALL_PROPERTY (graph, source); enum graph_destination_t { graph_destination = 101 }; BOOST_INSTALL_PROPERTY (graph, destination); } typedef size_t graph_vertex; typedef boost::adjacency_list < boost::vecS, boost::vecS, boost::undirectedS, boost::no_property, boost::no_property, boost::property < boost::graph_source_t, graph_vertex, boost::property < boost::graph_destination_t, graph_vertex, boost::no_property>>> graph_t; static_assert(std::is_same<graph_vertex, boost::graph_traits<graph_t>::vertex_descriptor>::value, "Check out the vertex descriptor type!"); int main() { using namespace boost; graph_t g; add_edge(0, 1, g); put(graph_destination, g, 0); put(graph_source, g, 1); }

2 2

Re: [Boost-users] Boost-users Digest, Vol 3709, Issue 1
by Deepa S 11 Feb '14

11 Feb '14

Hi Gennadiy, Yes I would like to know how I need to write my own main in this case? Also I was wondering if there is any workaround that was found other than re-writing the main? Thanks, Deepa Here is what I am trying to do, may not be the best way, but trying to re-run the test suite ts1 - 3 consecutive times to test for repeatability: bool init_unit_test_suite(int argc, char **argv) { test_suite* ts1 = BOOST_TEST_SUITE( "SKUResourcesTestSuite" ); ts1->add( BOOST_TEST_CASE(&TestCase1) ); ts1->add( BOOST_TEST_CASE(&TestCase2) ); framework::master_test_suite().add( ts1 ); for(int i = 0;i < 3;i++) { framework::master_test_suite().add( ts1 ); } return true; } Deepa S <mslvada <at> yahoo.com> writes: > Wanted to check if there was any resolution to the error indicated > in this post: > http://lists.boost.org/boost-users/2010/06/59564.php > > I am running into a similar issue and would like to know how to resolve it. What exactly is the issue? Do you want to write your own main? Gennadiy On Monday, February 10, 2014 10:30 PM, "boost-users-request(a)lists.boost.org" <boost-users-request(a)lists.boost.org> wrote: Send Boost-users mailing list submissions to boost-users(a)lists.boost.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.boost.org/mailman/listinfo.cgi/boost-users or, via email, send a message with subject or body 'help' to boost-users-request(a)lists.boost.org You can reach the person managing the list at boost-users-owner(a)lists.boost.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Boost-users digest..." Today's Topics: 1. Re: Regarding post: Using Boost.Test in Iteration (Gennadiy Rozental) 2. [container.string] why isn't memcpy used in the internals? (Adam Romanek) 3. Re: [container.string] why isn't memcpy used in the internals? (Ion Gazta?aga) 4. [property_tree] Compilation Problem when #including <boost/property_tree/json_parser.hpp> (Adi Shavit) 5. Re: [property_tree] Compilation Problem when #including <boost/property_tree/json_parser.hpp> (Edward Diener) ---------------------------------------------------------------------- Message: 1 Date: Mon, 10 Feb 2014 07:16:06 +0000 (UTC) From: Gennadiy Rozental <rogeeff(a)gmail.com> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Regarding post: Using Boost.Test in Iteration Message-ID: <loom.20140210T081523-687(a)post.gmane.org> Content-Type: text/plain; charset=us-ascii Deepa S <mslvada <at> yahoo.com> writes: > Wanted to check if there was any resolution to the error indicated > in this post: > http://lists.boost.org/boost-users/2010/06/59564.php > > I am running into a similar issue and would like to know how to resolve it. What exactly is the issue? Do you want to write your own main? Gennadiy ------------------------------ Message: 2 Date: Mon, 10 Feb 2014 09:18:20 +0100 From: Adam Romanek <romanek.adam(a)gmail.com> To: "boost-users(a)lists.boost.org" <boost-users(a)lists.boost.org> Subject: [Boost-users] [container.string] why isn't memcpy used in the internals? Message-ID: <52F88B4C.40707(a)gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed Hi! Recently I've been trying to improve the performance of a code that makes heavy use of std::string (lots of constructions and destructions). My first thought was to use boost::container::string as it does not use reference-counting and uses small-string optimization, so it should be a perfect replacement for std::string when it comes to performance. Right... But the problem is that switching to boost::container::string only made things worse in the general case. I wrote "in the general case" because when strings to be held are small and SSO may be applied then boost::container::string outperforms std::string considerably. However, in the general case constructing boost::container::string from a C string is significantly slower than constructing std::string. My investigation showed that std::string uses memcpy to copy the C string into its buffer while boost::container::string does not. The memcpy comes from std::char_traits::copy so maybe boost::container::string could use it too? I guess the rest of operations from std::char_traits (find, move, assign etc.) could be utilized too as std::char_traits has been carefully crafted with performance in mind. Should I create a ticket for this issue? Best regards, Adam Romanek ------------------------------ Message: 3 Date: Mon, 10 Feb 2014 09:54:36 +0100 From: Ion Gazta?aga <igaztanaga(a)gmail.com> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] [container.string] why isn't memcpy used in the internals? Message-ID: <52F893CC.8020703(a)gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed El 10/02/2014 9:18, Adam Romanek escribi?: > Hi! > > Recently I've been trying to improve the performance of a code that > makes heavy use of std::string (lots of constructions and destructions). > My first thought was to use boost::container::string as it does not use > reference-counting and uses small-string optimization, so it should be a > perfect replacement for std::string when it comes to performance. > > Right... But the problem is that switching to boost::container::string > only made things worse in the general case. I wrote "in the general > case" because when strings to be held are small and SSO may be applied > then boost::container::string outperforms std::string considerably. > However, in the general case constructing boost::container::string from > a C string is significantly slower than constructing std::string. > > My investigation showed that std::string uses memcpy to copy the C > string into its buffer while boost::container::string does not. The > memcpy comes from std::char_traits::copy so maybe > boost::container::string could use it too? I guess the rest of > operations from std::char_traits (find, move, assign etc.) could be > utilized too as std::char_traits has been carefully crafted with > performance in mind. > > Should I create a ticket for this issue? Yes, please. I'm slowly trying to benchmark and improve Boost.Container performance, I haven't started optimizing basic_string so your ticket could be a good start point. In a quick inspection I see that the constructor is dispatched to basic_string::assign(It, It) and that assignment is not optimized at all. I should definitely write a assignment function that dispatches to std::char_traits::copy, as it could be used to optimize several member functions. Best, Ion ------------------------------ Message: 4 Date: Mon, 10 Feb 2014 12:29:39 +0200 From: Adi Shavit <adishavit(a)gmail.com> To: boost-users(a)lists.boost.org Subject: [Boost-users] [property_tree] Compilation Problem when #including <boost/property_tree/json_parser.hpp> Message-ID: <CAEWUs4h_ExtG5fG9L=x2p0hss_f=r_eVyrwoPaTN-RYyJPcH0g(a)mail.gmail.com> Content-Type: text/plain; charset="iso-8859-1" Hi, On VS2010, the following code does not compile if I restore the commented out #include: *// #include <boost/property_tree/json_parser.hpp>* struct X { enum Labels { ERROR, ONE, TWO, THREE }; // problem => ^^^^^ }; void f() { X::Labels r = X::ERROR; // problem => ^^^^^ } The VS2010 errors are: t.cpp(5): error C2059: syntax error : 'constant' t.cpp(5): error C2143: syntax error : missing ';' before '}' t.cpp(5): error C2238: unexpected token(s) preceding ';' t.cpp(7): error C2059: syntax error : '}' t.cpp(7): error C2143: syntax error : missing ';' before '}' t.cpp(7): error C2059: syntax error : '}' t.cpp(11): error C2589: 'constant' : illegal token on right side of '::' t.cpp(11): error C2059: syntax error : '::' It seems that somewhere inside the #include hierarchy the symbol ERROR is #defined to cause this code to fail. Renaming ERROR to something else avoids this problem. I think I tracked it into ' boost/spirit/home/classic/iterator/impl/file_iterator.ipp' but I'm not sure where exactly it is coming from. Adi

2 1

[MPL] implementing a "trait"?
by John M. Dlugosz 11 Feb '14

11 Feb '14

I want to mark certain types as being of a category of my invention (e.g. is_pretty), for purposes of using enable_if in the same manner is standard type traits such as is_arithmetic etc. The trait will be false by default, and I'd declare something to nominate types that should be seen to have that trait. I want it to follow inheritance; if B has been declared to be in my category, and C is derived from B, then C will also be in that category without needing to do anything more (though some way to turn it _off_ would be available). Can someone advise me on an easy way to structure this? Thanks, —John

6 10

retrieve data using associative property maps with boost::bimaps - failed
by Aniket Pugaonkar 10 Feb '14

10 Feb '14

Hi, I am facing compilation error in the following code. I do understand why the error is, but I could not find a good solution. Here is my code and question explained in detail *What I want to do:* I want to store a key- value pair where both key and its respective values is unique. But After I create a map, I want to use the VALUE and get the corresponding key. Essentially the best way is a bimap. So I do the following using associative property maps. *What I did:* 1. define a bimap (my_bimap ) 2. define associative property map for left view of bimap ( right_asso_bimap ) 3. define associative property map for right view of bimap ( left_asso_bimap ) 4. fill the bimap using *left* view associative property map. i.e. *use boost::put( left_asso_bimap, key, value )* 5. Now using *right* view associative property map, retrieve key for given value i.e. use *boost::get( right_asso_bimap, value )* My main motive is to use GET and PUT helper functions for insertion and retrieval from a bimap. Use PUT to insert in LEFT BIMAP. and Use GET to retrieve from RIGHT BIMAP. I know the error is because of non-const-ness of the iterators because of which I cannot de-reference using iterators. But what would be the alternative to this ? This link here explains how to define parameters for bimap in order for iterators to be mutable. http://www.boost.org/doc/libs/1_55_0/libs/bimap/doc/html/boost_bimap/the_tu… *But I get the error below: * In file included from main.cpp:2:0: /usr/local/include/boost/property_map/property_map.hpp: In instantiation of 'boost::associative_property_map<UniquePairAssociativeContainer>::value_type& boost::associative_property_map<UniquePairAssociativeContainer>::operator[](const key_type&) const [with UniquePairAssociativeContainer = boost::bimaps::views::list_map_view<boost::bimaps::relation::member_at::right, boost::bimaps::detail::bimap_core<int, boost::bimaps::list_of<long unsigned int>, mpl_::na, mpl_::na, mpl_::na> > boost::associative_property_map<UniquePairAssociativeContainer>::reference = const int&; boost::associative_property_map<UniquePairAssociativeContainer>::value_type = const int; boost::associative_property_map<UniquePairAssociativeContainer>::key_type = long unsigned int]': /usr/local/include/boost/property_map/property_map.hpp:357:54: required from 'Reference boost::get(const boost::put_get_helper<Reference, PropertyMap>&, const K&) [with PropertyMap = boost::associative_property_map<boost::bimaps::views::list_map_view<boost::bimaps::relation::member_at::right, boost::bimaps::detail::bimap_core<int, boost::bimaps::list_of<long unsigned int>, mpl_::na, mpl_::na, mpl_::na> > > Reference = const int&; K = int]' main.cpp:41:35: required from here /usr/local/include/boost/property_map/property_map.hpp:502:20: error: no match for 'operator[]' (operand types are 'boost::associative_property_map<boost::bimaps::views::list_map_view<boost::bimaps::relation::member_at::right, boost::bimaps::detail::bimap_core<int, boost::bimaps::list_of<long unsigned int>, mpl_::na, mpl_::na, mpl_::na> > >::C {aka boost::bimaps::views::list_map_view<boost::bimaps::relation::member_at::right, boost::bimaps::detail::bimap_core<int, boost::bimaps::list_of<long unsigned int>, *mpl_::na, mpl_::na, mpl_::na> >}' and 'const key_type {aka const long unsigned int}')* *return (*m_c)[k];* ^ Here my code. You can also see the code here on this website link:* http://coliru.stacked-crooked.com/a/59cbf2c564bf22b2 <http://coliru.stacked-crooked.com/a/59cbf2c564bf22b2>* #include <boost/bimap.hpp> #include <boost/property_map/property_map.hpp> #include <boost/bimap/property_map/set_support.hpp>#include <boost/bimap/list_of.hpp>#include <iostream> using namespace boost; int main() {typedef int vertex_descriptor_t;namespace bm = boost::bimaps;typedef bm::bimap< vertex_descriptor_t, bm::list_of<size_t> > vd_idx_bimap_t;typedef boost::associative_property_map<vd_idx_bimap_t::left_map> Lasso_vd_idx_bimap_t;typedef boost::associative_property_map<vd_idx_bimap_t::right_map> Rasso_vd_idx_bimap_t; // define bimap and assoc prop map for itvd_idx_bimap_t my_bimap;Lasso_vd_idx_bimap_t Lmy_asso_bimap(my_bimap.left );Rasso_vd_idx_bimap_t Rmy_asso_bimap(my_bimap.right); typedef typename vd_idx_bimap_t::value_type value_type; my_bimap.insert( value_type( 1, 100 ) ); my_bimap.insert( value_type( 2, 200 ) ); int z = 1; std::cout << "value = " << get ( my_bimap.left, z ) << std::endl; // prints correctly value = 100 // use left asso bimap for inserting to bimap boost::put( Lmy_asso_bimap, 3, 300 ); // correct // print bimap using *left view*for(auto t = my_bimap.left.begin(); t != my_bimap.left.end(); ++t) std::cout << t->first << " " << t->second << "\n"; int y = 2;*z **= boost::get( Lmy_asso_bimap, y ); // CORRECT OUTPUT for left view * std::cout << "value = " << z << std::endl; // Use right asso bimap to get key based on supplied value y = 100;*z = boost::get( Rmy_asso_bimap, y );* *// -->> ERROR HERE for right view* std::cout << "value = " << z << std::endl; // print map again. (*right view*)for(auto t = my_bimap.right.begin(); t != my_bimap.right.end(); ++t) std::cout << t->first << " " << t->second << "\n"; } Please let me know whether this is a potential bug, or the code semantics is incorrect. Or the current library cannot support this. ? Regards, Aniket Pugaonkar -- Thanks and regards, Aniket Pugaonkar

1 0

interprocess library deadlocked
by Hernán Leoni 10 Feb '14

10 Feb '14

Hi everyone, I'm using boost interprocess, and in some cases I get the interprocess library deadlocked, now I'm using internal timed locks, so, I get timeout exception Fortunately it is not that frequent, but it is an issue I need to resolve before going live, When this happens the shared memory cannot be used at all, so it needs to be removed and a new one must be created. I'm not sure yet why these internal mutexes are getting locked and never unlocked, it may be happening with some process being killed unexpectedly. I was wanting to see if anybody has already figured out which approach is better for this case. Thanks a lot, Hernan

1 0

Regarding post: Using Boost.Test in Iteration
by Deepa S 10 Feb '14

10 Feb '14

Hi Gennadiy/Gaurav, Wanted to check if there was any resolution to the error indicated in this post: http://lists.boost.org/boost-users/2010/06/59564.php I am running into a similar issue and would like to know how to resolve it. Thanks, Deepa

2 1

Attn: We need GSoC 2014 mentors for Boost!
by Niall Douglas 09 Feb '14

09 Feb '14

Dear Boost community, We are now just five days away from the closing date for Google Summer of Code 2014 registration. Last year's GSoC was a huge success, with Google funding work on Boost to the tune of $35,000 and paying over $8,000 in costs for three representatives of Boost to attend the GSoC mentor's summit at Google HQ. As part of filling in our application for 2014, we must supply to Google a list of potential GSoC mentors and potential GSoC projects for summer 2014. We do NOT want a repeat of 2012 when our application was rejected due to an insufficiently long list of prospective mentors and projects! At the time of writing, we have just *eight* prospective mentors, and we need more to be sure of being approved by Google! Therefore, if you think yourself able to mentor a student doing some work on Boost this summer, *please* consider adding a description of the proposed work item and your name to the list at https://svn.boost.org/trac/boost/wiki/SoC2014. If you want to know more about mentoring a Google Summer of Code funded student work before you nominate yourself, please feel free to ask on the main Boost developers mailing list boost(a)lists.boost.org. Thank you in advance for your time. Niall Douglas (Boost Google Summer of Code admin) Boris Schäling (Boost Google Summer of Code backup admin)

2 4

Boost.Asio ssl session reuse
by Krzysztof Trzepla 09 Feb '14

09 Feb '14

Hello, Is there a way to cache ssl session using Boost.Asio library, so that it can be later on reused with an abbreviated handshake? If so, I will appreciate any example. -- Regards, Krzysztof Trzepla

1 0

[Boost.Python] Wrapping a heterogeneous container
by Mario Lang 08 Feb '14

08 Feb '14

Hi. I have finally figured out how to (mostly) wrap a container of boost::variant<> in boost::python: http://blind.guru/boost_python-and-boost_variant.html The only problem left is the fact that I have to disable the use of reference semantics (proxies) in vector_indexing_suite. If I pass false for the NoProxy template parameter of vector_indexing_suite<>, I can no longer dereference container elements: >>> import bpv >>> bpv.make_vector()[0] Traceback (most recent call last): File "<stdin>", line 1, in <module> TypeError: No Python class registered for C++ class boost::variant<a, b, boost::detail::variant::void_, ...> However, I have a converter registered with to_python_converter<> that handles extracting the proper Python object from a variant. This works if NoProxy is true. However, if NoProxy is true, container elements do not have reference semantics, so in-place modifications will be lost. To me, it also looks like the TypeError raise above is sort of bogus, since I actually used to_python_converter to register conversion for this particular type. So I guess there are different types of registries involved, but the message does not indicate which registry is did refer to. Does anyone know if I can enable reference semantics for a heterogeneous container wrapped with indexing_suite? -- CYa, ⡍⠁⠗⠊⠕

1 0

Apology: RE: FileSystem: crash on OSX 10.8 with create_directory
by Tim Burgess 08 Feb '14

08 Feb '14

Hi, I must apologise for the stupidity of this question - if I'd have surrounded the call with a try/catch block I'd have figured it out sooner, so it's my own bad coding. Now all I need to figure out is how to programmatically create a directory with admin rights to get around the permission denied exception when trying to create files/directories within /library/application support/ on the Mac. Best wishes. Tim Burgess From: Boost-users [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Tim Burgess Sent: 07 February 2014 16:32 To: boost-users list Subject: [Boost-users] FileSystem: crash on OSX 10.8 with create_directory Hi, My app crashes out when I run the following: boost::filesystem::path myPath = wxStandardPaths::Get().GetConfigDir().ToStdWstring(); if (myPath.has_filename()) { myPath.remove_filename(); } // Add the company name and check for directory existence myPath /=wstrRBCompanyName; if (!boost::filesystem::exists(myPath)) { // The following test displays the path I'd expect to see (/library/Raised Bar in my case) wxMessageBox( myPath.native(), wstrErrorTitle, wxOK | wxICON_EXCLAMATION); // Create the required directory - displays a message box indicating deletion of a lock file here and the app exits boost::filesystem::create_directory(myPath); } This code appears to work fine on Windows, but fails on OSX 10.8. I should point out that I'm a blind programmer and I'm struggling to get Apple's VoiceOver utility to interact in a useful manner with Xcode 5's debugger and that's why I'm using a wxMessageBox to try and track down my failure. Best wishes. Tim

1 0

Generating PDF from QuickBook?
by Nat Goodspeed 08 Feb '14

08 Feb '14

I'm sorry, this must be a FAQ, but my Google search turned up mail from 2009 advising me to cd to libs/xxx/doc and run ../../../b2 pdf When I do that in a new modular Boost tree, I get the following errors: /Users/nat/boost-git/tools/build/src/tools/boostbook.jam:704: in construct from module object(boostbook-target-class)@1466 error: Cannot build documentation type 'pdf'. /Users/nat/boost-git/tools/build/src/build/targets.jam:1332: in class(a)basic-target.generate from module object(boostbook-target-class)@1466 /Users/nat/boost-git/tools/build/src/build/targets.jam:757: in generate-really from module object(main-target)@1471 /Users/nat/boost-git/tools/build/src/build/targets.jam:729: in class(a)main-target.generate from module object(main-target)@1471 /Users/nat/boost-git/tools/build/src/build/targets.jam:272: in class(a)project-target.generate from module object(project-target)@1460 /Users/nat/boost-git/tools/build/src/build-system.jam:707: in load from module build-system /Users/nat/boost-git/tools/build/src/kernel/modules.jam:289: in import from module modules /Users/nat/boost-git/tools/build/src/kernel/bootstrap.jam:139: in boost-build from module /Users/nat/boost-git/boost-build.jam:17: in module scope from module Is there some incantation I can apply to produce PDF output?

5 7

Confused with the example students_t_single_sample.cpp
by lizy10b 08 Feb '14

08 Feb '14

Hi there, I got a problem with the example "students_t_single_sample.cpp". The corresponding document is at: http://www.boost.org/doc/libs/1_55_0/libs/math/doc/html/math_toolkit/stat_t… Actally, I am confused with the mathematical logic rather than coding. The document presents a table that shows four hypothesis and tests, I am OK with the frist two, but for the third and the fourth ones: The original text: Hypothesis 3: the sample mean is less than the true mean Test 3: Reject if CDF of t > significance level, cdf(dist, t) > alpha I think p-value should be the integral from t to +Infinite: cdf(complement(dist, t)), So reject Hypothesis 3 when cdf(complement(dist, t)) < alpha, or say cdf(dist, t) > 1-alpha The original text: Hypothesis 4: the sample mean is greater than the true mean Reject if complement of CDF of t > significance level: cdf(complement(dist, t)) > alpha Similarly, the p-value should be the integral from -Infinite to t: cdf(dist, t), so reject Hypothesis 4 when cdf(dist, t) < alpha Is that right? Thanks. Zhiyu Li 2014-01-11 lizy10b

3 3

Newbie math question - roots of a distribution
by Benjamin Ward (ENV) 08 Feb '14

08 Feb '14

Hi Boost list, I've just installed and started to use boost for various problems and I've hit on a math problem I'd like to solve with it but I'm unsure on how to achieve it. I've gone through the documentation and have worked out how to easily compute the value from a cdf of a binomial distribution: #include <boost/math/distributions/binomial.hpp> #include <iostream> int main(){ int numbertrials = 4164; double successprob = 10e-6; int numbersuccesses = 75 boost::math::binomial mydistribution(numbertrials, successprob); double ans = cdf(mydistribution, numbersuccesses); std::cout << "The answer is " << ans << std::endl; return 0; } But now I'd like to find the root of the cdf, in other words, I want to find the value of successprob, from numbertrials and numbersuccesses and ans. Let's define ans as 0.95, by finding the root of cdf(mydistribution, numbersuccesses) - 1, using a root finding algorithm, I should be able to get the value of successprob. I've tried reading the roots documentation and looking at some examples but they use simple functions like x * sin(x) - there is only one variable and it is x, the binomial cdf has quite a few variables - number of trials, number of successes, probability of a single success. How can I define a function with boost::math::binomial that could be used with boosts root libraries to find the probability of success from the cdf, given I know ans, the number of successes, and the number of trials? Thanks, Ben W.

2 3

intrusive data structures with custom pointer class
by Matei David 07 Feb '14

07 Feb '14

Hi, I would like to use some boost::intrusive data structures (lists and trees) in a project I'm working on, but I ran into trouble trying to customize the existing classes to my needs. I decided to post my problem here, in case somebody knowledgeable of boost::intrusive *implementation details* can assist me. Thank you in advance. Background: In the project I'm working on, I need to manage a large number of objects in memory. The data is dynamic, so I need to be able to alloc&dealloc objects, in addition to keeping the objects in various lists&trees (for which I'm trying to use boost::intrusive). The RAM usage is large (think, >100GB), but the number of objects is smaller than 2^32. So one thing I'd like to do is to replace 8-byte raw pointers with 4-byte offsets into a deque. Doing this would solve the alloc/dealloc part. However, I now need to customize boost::intrusive containers to work with these 4-byte offsets instead of raw pointers. Initial solution that doesn't work: The most important limitation of these 4-byte indirect pointers (say to a type T) is that if you fully dereference one to T&, you can no longer reconstruct the indirect pointer (with operator&) in constant time. I.e. converting a real-pointer to indirect-pointer must be avoided. Thus, to make methods such as insert(T&) work, I had to roll out a reference-replacement class in addition to the pointer-replacement class. This is what I initially tried: - Given a type T, I rolled out pointer-replacement class Ptr<T> which holds a 4-byte int, and upon operator*, it produces a Ref<T> object. - A reference-replacement Ref<T> class similarly holds a 4-byte int, it implements operator& which produces the associated Ptr<T> object, and it also provides an operator T& () conversion which dereferences the object from the deque. - With these, I tried a custom traits struct looking like this: value_type = node := T node_ptr = pointer := Ptr<T> reference := Ref<T> Problems: Even though intrusive data structures are designed with the stated goal of leaving memory management to the user, intrusive lists&trees do allocate one extra node object that they use as "header". The code then takes the address (&) of this node and uses it in comparisons as a node_ptr. Because of this, my custom traits above could not compile. The only way I could find to make such containers work for my needs is basically a hack: define "node" to be a "pointer holder" object (neither member nor base of "value"), that allocates a "value" struct from the deque upon construction, that responds to operator& by returning the held pointer, and that otherwise acts as a value&. From what I could gather reading the docs, boost::intrusive containers allow some type of customization via the traits structs. However, I see some drawbacks: A. The traits classes seem to be there to support 2 existing customization types, namely the situation where "node" is a base class of "value" (implementing "base hooks") and the situation where "node" is a member of "value" (implementing "member hooks"). My situation represents a 3rd customization type. B. Even though the traits classes allow one to specify pointer types, it's possible (?) that under the hood the classes were designed to support only raw pointers and smart pointers, not the indirect offset-like pointers I need. Given A&B, it seems to me that in my customization I would need 2 different hacks. In the interest of code stability, I started wondering whether it is feasible to adapt existing boost::intrusive data to my needs (my first instinct) as opposed to simply rolling out my own rbtree. Questions: 0. Obviously: Have you seen a similar situation being dealt with before? Maybe there's an existing solution out there. My reluctance to rolling out my own intrusive rbtrees is due to the fact I'd have to test them, as opposed to using a known implementation. 1. Are boost::intrusive data structures supposed to support custom pointer classes (neither real pointers nor smart pointers)? For instance, in 1.46 I found this code in boost/intrusive/detail/tree_algorithms.hpp static node_ptr uncast(const_node_ptr ptr) { return node_ptr(const_cast<node*>( ::boost::intrusive::detail::boost_intrusive_get_pointer(ptr))); } So despite all the custom traits structs defining node_ptr and const_node_ptr, under the hood you have a line of code which assumes node_ptr can be constructed from node*. This is the expensive real-pointer to offset-pointer no-no I mentioned earlier. However, intrusive lists worked just fine in 1.46, which suggests this was simply a bug. 2. Can you point me to an example of defining and using custom pointer class with boost::intrusive? To enable compilation (of lists only) in 1.46, I had to customize std::iterator_traits (that part I understood), but then also boost::pointer_to_other and boost::intrusive::detail::smart_ptr_type. Even though the code compiled, I don't know what those are doing exactly, so I'm weary of them. 3. In 1.55, I have a few more pages of errors, I think mainly due to these pointer customization issues I don't fully understand. Something seems to have changed from 1.46 to 1.55. Is there a reference about what changed? Or is there a portable way of rolling out such a custom pointer class? Many thanks, M

2 1

BOOST Spirit Qi and lexer parsing of tokens
by Howard, Tony 07 Feb '14

07 Feb '14

Hi, I am starting a project that will use spriti Qi and a lexer. I am having difficulty getting something simple to parse correctly without having to add states and just want to make sure I am going about it the right way. I have a label that should parse the following example (where the label starts/ends with a single quote): 'this is my BOOLEAN 123label 12as' And then this label will be followed by zero or more blanks and then a number enclosed in parenthesis: ( 123 ) There are a number of arbitrary blanks in between the number and the enclosing parenthesis. I am ignoring blanks and whitespace in the default state of the lexer (assume blanks and white spaces are defined in the patterns referenced): this->self += lex::token_def<>("{BLANK}")[ lex::_pass = lex::pass_flags::pass_ignore ]; this->self += lex::token_def<>("{WS}")[ lex::_pass = lex::pass_flags::pass_ignore ]; The first problem I encountered was that if I ignored blanks in the initial state and tried to parse a label that contains blanks that I don't want to ignore, this was not possible. So I had to create a label state, to which the blank token is added, and am switching to this state initially using in_state[...] so that when we are done with the label, we return to the initial state, where blanks are once again ignored. This worked. The problem then came with the digits. Since the digits were already recognized in the label state, I could not also recognize them in the initial state. So I added a digit state, where the digit token is now defined and the label state has to transition to this state while parsing between letter, symbols and digits (and whatever else I decide is allowed to make up my label), and after I get out of the label state, my rule parses a parenthesis, goes into the digit state, and then parses the closing parenthesis. This works correctly for what I need now. The issue I am having is looking ahead. What if a rule needs lower case letters only? I will have to create a state for it, remove it from the label state, etc. I will have to do this for any core patterns, which to me seems like there should be an easier way to do this. Any suggestions, comments, concerns? I have code snippets if my approach is not clear. Thanks for your help. Tony.

1 0

FileSystem: crash on OSX 10.8 with create_directory
by Tim Burgess 07 Feb '14

07 Feb '14

Hi, My app crashes out when I run the following: boost::filesystem::path myPath = wxStandardPaths::Get().GetConfigDir().ToStdWstring(); if (myPath.has_filename()) { myPath.remove_filename(); } // Add the company name and check for directory existence myPath /=wstrRBCompanyName; if (!boost::filesystem::exists(myPath)) { // The following test displays the path I'd expect to see (/library/Raised Bar in my case) wxMessageBox( myPath.native(), wstrErrorTitle, wxOK | wxICON_EXCLAMATION); // Create the required directory - displays a message box indicating deletion of a lock file here and the app exits boost::filesystem::create_directory(myPath); } This code appears to work fine on Windows, but fails on OSX 10.8. I should point out that I'm a blind programmer and I'm struggling to get Apple's VoiceOver utility to interact in a useful manner with Xcode 5's debugger and that's why I'm using a wxMessageBox to try and track down my failure. Best wishes. Tim

1 0

Build: Clang, libstdc++ and Mac 10.8
by Tim Burgess 07 Feb '14

07 Feb '14

Hi, Due to the requirements of another library I have to use, I need to build Boost 1.55.0 on my Mac 10.8 system using the libstdc++ STL library, rather then libc++. I've now got myself royally confused with the build configuration process, as to where I specify toolsets and other options to achieve what I need. I'm failing with: Cd /users/tim/developer/libraries/boost_1_55_0 ./b2 clean sudo sh bootstrap.sh macosx-version=10.8 macosx-version-min=10.7 architecture=combined threading=multi link=static address-model=32_64 ./b2 toolset=clang -std=c++11 cxxflags="-stdlib=libstdc++" linkflags="-stdlib=libstdc++" I don't actually need the c++11 support. Can anybody advise, please? Best wishes. Tim Burgess

4 4

lexical_cast: fatal error
by Tim Burgess 07 Feb '14

07 Feb '14

Hi, Using 1.55.0 in my Xcode project with clang as the compiler I'm getting: "/boost/type_traits/is_fundamental.hpp:26:20: fatal error: recursive template instantiation exceeded maximum depth of 128"::boost::is_arithmetic<T>::value This error is triggered by me including lexical_cast.hpp. Has anybody else encountered this and, if so, found a solution, please? Best wishes. Tim Burgess

3 4

Boost Accumulators Tail
by Jon Lederman 06 Feb '14

06 Feb '14

Hi, I am a newbie to using boost accumulators. I have tried a simple program using examples from the boost website. I have something as simple as: #include <boost/accumulators/accumulators.hpp> #include <boost/accumulators/statistics/stats.hpp> #include <boost/accumulators/statistics/tail.hpp> include <boost/accumulators/statistics/tail_quantile.hpp> using namespace boost::accumulators; typedef accumulator_set<double, stats<tag::tail_quantile<right> > > accumulator_t_right; typedef accumulator_set<double, stats<tag::tail_quantile<left> > > accumulator_t_left; accumulator_t_right acc0( tag::tail<right>::cache_size = 1000 ); accumulator_t_right acc1( tag::tail<right>::cache_size = 1000 ); accumulator_t_left acc2( tag::tail<left>::cache_size = 1000 ); accumulator_t_left acc3( tag::tail<left>::cache_size = 1000 ); I am met with: /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:56:63: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:56:65: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:56:67: error: template argument 2 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:56:88: error: invalid type in declaration before ';' token /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:57:62: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:57:64: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:57:66: error: template argument 2 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:57:86: error: invalid type in declaration before ';' token /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:59:42: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:59:56: error: expected unqualified-id before '=' token /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:60:42: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:60:56: error: expected unqualified-id before '=' token /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:61:41: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:61:55: error: expected unqualified-id before '=' token /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:62:41: error: template argument 1 is invalid /usr/local/src/freeswitch/src/mod/applications/mod_soundtouch/mod_soundtouch.cpp:62:55: error: expected unqualified-id before '=' token Do I need to use a specific namespace for the right and left arguments? Where are these defined? Or is something else wrong. Any help would be greatly appreciated. Thank you in advance. Regards, Jon

1 0

Building Boost dynamically linkable gives linking error?
by Mehmet Emin Erük 06 Feb '14

06 Feb '14

I am trying to build boost and generate dll's but whatever I have done did not solve the problem. Here what I have got: Creating library bin.v2\libs\thread\build\msvc-11.0\dbg\adrs-mdl-32\async-excpt-on\thrd-mlt\boost_thread-vc110-mt-gd-1_53.lib and object bin.v2\libs\thread\build\msvc-11.0\dbg\adrs-mdl-32\async-excpt-on\thrd-mlt\boost_thread-vc110-mt-gd-1_53.exp thread.obj : error LNK2019: unresolved external symbol "__declspec(dllimport) public: __thiscall boost::gregorian::greg_month::greg_month(unsigned short)" (__imp_??0greg_month@gregorian@boost@@QAE@G@Z) referenced in function "public: __thiscall boost::gregorian::date::date(enum boost::date_time::special_values)" (??0date@gregorian@boost@@QAE@W4special_values@date_time@2@@Z) thread.obj : error LNK2019: unresolved external symbol "__declspec(dllimport) public: __thiscall boost::gregorian::greg_month::operator unsigned short(void)const " (__imp_??Bgreg_month@gregorian@boost@@QBEGXZ) referenced in function __catch$?make_external_thread_data@?A0x9b8c45f0@boost@@YAXXZ$0 bin.v2\libs\thread\build\msvc-11.0\dbg\adrs-mdl-32\async-excpt-on\thrd-mlt\boost_thread-vc110-mt-gd-1_53.dll : fatal error LNK1120: 2 unresolved externals Here some information about the setup: OS: W7 x64 Command to build: b2 -j%NUMBER_OF_PROCESSORS% --abbreviate-paths toolset=msvc-11.0 define=_SECURE_SCL=0 define=_CRT_SECURE_NO_DEPRECATE define=_CRT_SECURE_NO_WARNINGS --without-mpi address-model=32 variant=debug install link=shared runtime-link=shared threading=multi asynch-exceptions=on exception-handling=on extern-c-nothrow=off --prefix=..\boost cd .. I have CYGWin installed to the computer. *Here what I have tried till now;* *I have tried to build it in 32 bit command window (I am using jenkins to build boost) <tns:build tns:position='boost/boost_1_53_0' tns:command='C:\Windows\SysWOW64\cmd /C bootstrap.bat'/> <tns:build tns:position='boost/boost_1_53_0' tns:command='C:\Windows\SysWOW64\cmd /C build_boost-win32-debug-vs2012.cmd'/> *I wrote #define BOOST_ALL_DYN_LINK in the boost config/user.hpp *I wrote #define BOOST_ALL_NO_LIB in the boost config/user.hpp Is there anyone who experienced such a problem before? I really could not find anything in anywhere? -- bir gün herkes özgür olacak

1 1

[Move, Container] returning a move-only type
by John M. Dlugosz 05 Feb '14

05 Feb '14

The definition of boost::container::vector includes: BOOST_COPYABLE_AND_MOVABLE rather than BOOST_MOVABLE_BUT_NOT_COPYABLE if the element type cannot be copied. If I have: typedef boost::container::vector<MyType> VecType; VecType foo() { ⋮ return boost::move(some_other_vec); } VecType x= foo(); On a configuration with BOOST_NO_RVALUE_REFERENCES defined, this gives an error within VecType's copy constructor, since MyType cannot be copied. If I write VecType x= boost::move(foo()); then I get an error that no matching move is viable, as the move(T&) wants an lvalue as the argument. I can't write the return type of foo to be rv<VecType> because that can't be copied (fake move-out like auto_pointer) either; the members are declared private and never defined. I'm thinking that if VecType were BOOST_MOVABLE_BUT_NOT_COPYABLE then it would work as expected, since operator rv<VecType>& would be provided. How can I make this work without surgery on the Boost headers, and hopefully without something ugly like deriving from vector and having to re-define and forward all the various constructors and such? —John

2 1

[Range, Container] range of to-be-moved values ?
by John M. Dlugosz 05 Feb '14

05 Feb '14

Given a boost::container::vector<C> vec; I can write vec.push_back(move(value)) for a move-only (non-copyable) type C, instead of vec.push_back(value) as usual. But what would I write instead of vec.insert(pos,b,e) where (b,e] is a range of values? I'm particularly fond of Range concept, rather than loose iterators, so my function would be something that takes something that works as a Range; e.g. template <typename R> void prepend (const R& src_range) { auto b= boost::begin(src_range); auto e= boost::end(src_range); vec.insert(vec.begin(), b, e); } An rvalue reference to a range is not the same thing as a range of values that I may move from. I could have a differently named function and pass it a range with the implicit assumption that the function will modify the values, but is there a proper way to indicate that as part of the type? src_range can be, say, a pair of (non-const) iterator s as opposed to a pair of const_iterator s, or otherwise range_iterator<R>::type is dereferencable to a const value or not. But that itself is hard to see in general let alone overload against! If my function were called prepend_and_destroy_original(src) it might clue in the caller, but in general it is surprising that a parameter of (const R&) will modify the values. src itself doesn't change, and now refers to the same range of values which are now swapped out with defaults. With built-in pointers and even shared_ptr I can distinguish between an in-out parameter and a parameter that dereferences to a modifiable or const value. But what about with Ranges? Even with lower-case ranges provided by two iterators, How can you call something like boost::container::vector v2 (InIt_start, InIt_end); to convert a std::vector or some other container of move-only items? Only if the types match exactly can the vector(vector&&) constructor be used. —John

4 6

[iostreams] Why does boost::iostreams::copy close the Sink & Source?
by Sebastian Karlsson 05 Feb '14

05 Feb '14

What's the rationale behind this decision? I don't find it intuitive and reduces the utility of the function. In my case I read from a file into a stream which I want to later use, the signature of the device is important. But as soon as I've copied my data over to the device it's destroyed and along side the data goes down the drain. Kind regards, Sebastian Karlsson

1 0

Boost.MultiIndex. Using several keys together.
by syvyi 05 Feb '14

05 Feb '14

Hello, there's a structure struct person { std::string name; std::string surname; int birth_year; std::string craft; }; How to create such a multiindex container that dumping could performed using an index as name, surname and year triple alltogether? -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-MultiIndex-Using-several-keys-toge… Sent from the Boost - Users mailing list archive at Nabble.com.

3 4

boost::python wrapping a container of boost::variant?
by Mario Lang 05 Feb '14

05 Feb '14

Hi. Does anyone have ideas how to convince boost::python to wrap a container of boost::variant? I need this to generate useful python bindings for a project of mine which uses boost::variant internally. Since python supports heterogenous containers natively, it seems to me this should be doable. However, I have attempted to find a solution on my own on several occasions, and always failed. One obvious hint is the indexing_suite. However, now that I looked at it again, I think I know where the problem is: The get_item method of indexing_suite always returns a data_type object, which is value_type from the container in question. However, from what I can see, I'd need something that converts to boost::python::object, so that I can return whatever class is appropriate for the actual variant instance. In other words, get_item would need to return a boost::python::object instead of data_type. To illustrate in code what I want, here is a minimal example: #include <boost/python.hpp> #include <boost/variant.hpp> struct a { int x; }; struct b { double y; }; typedef boost::variant<a, b> variant; typedef std::vector<variant> vector; BOOST_PYTHON_MODULE(bpv) { using namespace boost::python; class_<a>("a", init<a>()) .def(init<>()) .def_readonly("x", &a::x); class_("b", init()) .def(init<>()) .def_readonly("y", &b::y); } The question is, how do I expose std::vector<variant> to Python? I guess I need to define a static_visitor that returns python objects, something like: struct to_python_visitor : boost::static_visitor<boost::python::object> { result_type operator()(a const &a_) const { // return a boost::python::object that refers to a_ } ... }; And hook that up to the container support somehow such that boost::apply_visitor(to_python_visitor(), variant_instance) is called whenever a value is retrieved from the container. Can this even be done with current Boost.Python? Any hints, maybe even concrete code examples, would be highly appreciated. -- CYa, ⡍⠁⠗⠊⠕

1 0

boost::none for boost::shared_ptr
by Chris Stankevitz 03 Feb '14

03 Feb '14

Hello, Does boost::shared_ptr offer a way to create an empty shared pointer without a lot of typing? Perhaps something analogous to boost::none. I am not using c++11. Thank you, Chris == using boost::none; using boost::optional; // declare f(optional<CSomeReallyLongClassName> Item); g(shared_ptr<CSomeReallyLongClassName> pItem); // invoke f with empty object f(none); // invoke g with empty object g(shared_ptr<CSomeReallyLongClassName>());

5 9

multi_array move semantics
by MM 03 Feb '14

03 Feb '14

Hello, 1. Browsing the multi_array doc, there doesn't seem to be a move ctor? 2. Has there been any attempt to implement one? 3. if not, and the authors are reading this, are there any hints to help with this? Regards MM

2 1

serialization: intrusive_ptr
by MM 03 Feb '14

03 Feb '14

Hello, I've found an implementation of a serializer for intrusive_ptr here: http://stellar.cct.lsu.edu/files/hpx_0.9.5/html/code/hpx/util/serialize_int… which i've injected into my boost lib. Hartmut Kaiser, I guess, would there be a proposal to add it serialization for future standard boost? Regards MM

1 0

inter thread communication with message queues
by Kenneth Adam Miller 31 Jan '14

31 Jan '14

I was reading about message queues with respect to inter thread communication, and I have a specific use case that I want to describe. It's rather complex, so I want to be careful that I design it so that there are no race conditions. Pool A is a buffer (of type 1) handle manager, that feeds buffer handles into thread set 1, and receives old handles from thread set 2. There is only one thread running pool A. Thread set 1 is a set of N threads that fill up buffers of type 1 with regular data. It pushes full buffers to thread set 2. Thread set 2 is a set of M threads that compress buffers of type 1 into data buffers of type 2. Once the buffer of type 1 has be used up, it is cleared and sent back to Pool A. It pulls fresh data buffers of type 2 from Pool B, and sends freshly filled compressed buffers to Pool C. Pool C does something with the buffer, clears it and hands it to Pool B. Can this be done safely using an implementation like what I see here: http://www.justsoftwaresolutions.co.uk/threading/implementing-a-thread-safe… ? Note 1: -= denotes a 1-to-N connect and =- a N-to-1 connect RR = Request reply relationship +--------------------------------------------------------------------+ | | v | Pool A RR Thread Set 1 Router X | PULL-PUSH --==> PULL-read-PUSH ==--> PULL-PUSH --==> PULL Thread Set 2 PUSH \ / +-compress-+ Pool B RR / \ PULL-PUSH --==> PULL PUSH ^ || | || +----------- PUSH-write-PULL <--==++ Pool C Questions: how do I have a separate thread poll on multiple shared message queues? Like, I have a thread pool that both receives requests for buffer handles and that receives old buffer handles to recycle.

3 3

asio/ip/udp.hpp causes MSVC to use std::bind
by Chris Stankevitz 30 Jan '14

30 Jan '14

boost 1.55 VS2012 Update 4 === Can you tell me why the following code fails to compile? It smells as if udp.hpp is causing MSVC to "using namespace std". PS: It doesn't happen if I remove the asio include. PPS: It doesn't happen if g() takes int vs std::string. PPPS: It doesn't happen on GCC Thank you, Chris === #include <boost/bind.hpp> #include <boost/asio/ip/udp.hpp> #include <string> using boost::bind; void g(const std::string&) { } int main() { bind(g, _1); return 0; } === 1>------ Build started: Project: OverlayServer, Configuration: Debug Win32 ------ 1> main.cpp 1>main.cpp(13): error C2668: 'std::bind' : ambiguous call to overloaded function 1> c:\program files (x86)\microsoft visual studio 11.0\vc\include\functional(1387): could be 'std::_Bind<_Forced,_Ret,_Fun,_V0_t,_V1_t,_V2_t,_V3_t,_V4_t,_V5_t,<unnamed-symbol>> std::bind<void,const std::string&,boost::arg&>(_Rx (__cdecl *)(const std::basic_string<_Elem,_Traits,_Alloc>),_Vx0_t)' [found using argument-dependent lookup] 1> with 1> [ 1> _Forced=true, 1> _Ret=void, 1> _Fun=void (__cdecl *const )(const std::string &), 1> _V0_t=boost::arg<1> &, 1> _V1_t=std::_Nil, 1> _V2_t=std::_Nil, 1> _V3_t=std::_Nil, 1> _V4_t=std::_Nil, 1> _V5_t=std::_Nil, 1> <unnamed-symbol>=std::_Nil, 1> I=1, 1> _Rx=void, 1> _Elem=char, 1> _Traits=std::char_traits<char>, 1> _Alloc=std::allocator<char>, 1> _Vx0_t=boost::arg<1> & 1> ] 1> c:\externalpackages\boost_1_55_0\boost\bind\bind_cc.hpp(26): or 'boost::_bi::bind_t<R,F,L> boost::bind<void,const std::string&,boost::arg>(R (__cdecl *)(B1),A1)' 1> with 1> [ 1> R=void, 1> F=void (__cdecl *)(const std::string &), 1> L=boost::_bi::list1<boost::arg<1>>, 1> I=1, 1> B1=const std::string &, 1> A1=boost::arg<1> 1> ] 1> while trying to match the argument list '(overloaded-function, boost::arg)' 1> with 1> [ 1> I=1 1> ] ========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

2 2

[accumulators] "count" NaN but otherwise ignore it?
by Rhys Ulerich 29 Jan '14

29 Jan '14

Is there a feature within accumulators to count NaNs on extraction but to otherwise ignore them? Maybe some sort of pass filter on incoming data? I've got a situation where I'd like to avoid NaNs clobbering otherwise useful information, but I still want to total count of data seen to include them. The context is a time integrator for a physics simulation. The integrator tracks the number of timesteps. When the physics kicks back a NaN (because something's amiss), I'd like to count the failed timestep but not have the NaN destroy the useful mean of timesteps taken thus far. Thanks, Rhys

2 1

Avoid transitive link dependency
by Dario Ramos 29 Jan '14

29 Jan '14

I wrote a serial port static library using boost::asio. I'm trying to use this library from another project I'm building as a dll, but this second project fails to link: it wants to link with Boost's sytem library. I used PIMPL to remove all boost #includes from my class header, and also tried setting Visual Studio's Link Library Dependencies option set to Yes in my static library. Neither approach worked. I know I can solve this by adding the boost binary directory to the library directories for my dll project, but I'd like to avoid that. The way I see it, only my static library should have a Boost link dependency. I'm using Visual Studio 2012, targetting to Windows XP (v110_xp toolset), and Boost 1.53.0. I asked about this at Stack Overflow if anyone's interested: http://stackoverflow.com/questions/21435666/avoid-link-dependency-with-boost -- *Darío Eduardo Ramos* Meditech S.A. www.meditech.com.ar (+54) 01147603300, Interno 31 Av. Julio A. Roca 3456 Florida Oeste, Bs.As. Argentina

1 1

[Container] any debug renderings for Visual Studio?
by John M. Dlugosz 29 Jan '14

29 Jan '14

When I try to inspect, say, a flat_vector (or anything built on top of it like a flat_map) I can't see the elements in the Watch display. Is there some way to teach VS about that? I seem to remember a long time ago having VS skip things like shared_ptr dereference when stepping through.

5 4

boost condition_variable internals
by Kenneth Adam Miller 29 Jan '14

29 Jan '14

I need to know that boost does not use any kind of locking, mutex or semaphore mechanism within the boost::condition_variable class. I don't know where it is in source, or I would check to make certain. Can someone provide me a yes no verification?

3 3

[offtopic] C++Now 2014 - hotel room available
by Hartmut Kaiser 29 Jan '14

29 Jan '14

I apologize for posting off-topic. Hey all, Unfortunately I will not be able to come to C++Now this year. However I already booked my hotel room at the Aspen Meadows (at the early bird rate!). If anybody is interested in using the hotel reservation, please get in contact with me offline. Thanks! Regards Hartmut --------------- http://boost-spirit.com http://stellar.cct.lsu.edu

1 0

C++Now 2014 Student Volunteer Program Accepting Applications - Please help spread the word
by Bryce Adelstein-Lelbach 28 Jan '14

28 Jan '14

C++Now, presented by Boost in-cooperation with the ACM, is an international gathering of C++ programmers. In 2013, the conference launched a Student/Volunteer Program to enable young coders to attend the conference. This year, we are again are inviting students with an interest in C++ programming to attend the May 12-17, 2014 conference in Aspen, CO as Student/Volunteers. The program is an excellent way for students with any interest in C++ to learn about C++ and make lasting connections with the C++ community. "Attending has truly been one of the most inspirational moments of my life. I had the opportunity to learn, socialize and have a great time in general. I hope to attend this conference yearly now and make a lasting impact in the computer science community." - Grant Mercer, 2013 C++Now Student/Volunteer. "All of the attendees had knowledge to share. Although the talks can be seen online, nothing can compare to being surrounded by over a hundred C++ enthusiasts." - Alex Duchene, 2013 C++Now Student/Volunteer Student/Volunteers will receive free registration. A limited number of applicants will receive stipends for travel expenses. Volunteers will handle various tasks during the conference, such as assisting presenters, setting up presentation rooms, running A/V equipment and helping with the conference picnic. Students from all fields of study are welcome to apply. However, this conference covers advanced C++ topics, and applicants should have ample experience with the C++ programming language and be familiar with general computer science topics. Applications will be accepted until February 15th, 2014. Application decisions will be sent out by March 18th, 2014. You can learn more about the program and some of the students who have participated at http://cppnow.org/volunteer. -- Bryce Adelstein-Lelbach, Student/Volunteer Chair

1 0

Forthcoming Boost.Fiber review
by Nat Goodspeed 28 Jan '14

28 Jan '14

Hi all, The review of Boost.Fiber by Oliver Kowalke is scheduled from Monday, January 6th to January 12th. ----------------------------------------------------- About the library: Boost.Fiber provides a framework for micro-/userland-threads (fibers) scheduled cooperatively. The API contains classes and functions to manage and synchronize fibers similar to Boost.Thread. Each fiber has its own stack. A fiber can save the current execution state, including all registers and CPU flags, the instruction pointer, and the stack pointer and later restore this state. The idea is to have multiple execution paths running on a single thread using a sort of cooperative scheduling (versus threads, which are preemptively scheduled). The running fiber decides explicitly when it should yield to allow another fiber to run (context switching). Boost.Fiberinternally uses coroutines from Boost.Coroutine <http://ok73.funpic.de/boost/libs/coroutine/index.html>; the classes in this library manage, schedule and, when needed, synchronize those coroutines. A context switch between threads usually costs thousands of CPU cycles on x86, compared to a fiber switch with a few hundred cycles. A fiber can only run on a single thread at any point in time. docs: http://ok73.funpic.de/boost/libs/fiber/doc/html/index.html<http://ok73.ok.funpic.de/boost/libs/coroutine/doc/html/> git: https://github.com/olk/boost-fiber src: http://ok73.ok.funpic.de/boost.fiber.zip<http://ok73.ok.funpic.de/boost.coroutine.zip> --------------------------------------------------- Naturally, you need not wait until the review formally opens to start reviewing the Fiber library. Please always state in your review whether you think the library should be accepted as a Boost library! Additionally please consider giving feedback on the following general topics: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? Nat Goodspeed Boost.Fiber Review Manager ------------------------------

7 25

making a geometry half closed
by Sean Farrow 27 Jan '14

27 Jan '14

Hi, >From what I understand any boost geometry is fully closed. Is there a way of making a geometry half closed, such that the bottom row is not included? I'm needing to do this with a Box2D$Values. Any help appreciated. Cheers Sean.

6 8

[Geometry] Issues when registering a windows point type
by Sean Farrow 26 Jan '14

26 Jan '14

Hi, I'm having an Issue when trying to register a windows POINT type. I'm registering the Point structure from windef.h as follows: BOOST_GEOMETRY_REGISTER_POINT_2D(POINT, LONG, boost::geometry::cs::cartesian, x, y); I'm getting the following in my build log: 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2870: 'boost' : a namespace definition must appear either at file scope or immediately within another namespace definition 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2913: explicit specialization; 'boost::geometry::traits::tag' is not a specialization of a class template 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2059: syntax error : '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '{' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2447: '{' : missing function header (old-style formal list?) 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2913: explicit specialization; 'boost::geometry::traits::dimension' is not a specialization of a class template 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2059: syntax error : '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C3083: 'mpl': the symbol to the left of a '::' must be a type 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2039: 'int_' : is not a member of 'boost' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2913: explicit specialization; 'boost::geometry::traits::coordinate_type' is not a specialization of a class template 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2059: syntax error : '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2913: explicit specialization; 'boost::geometry::coordinate_system' is not a specialization of a class template 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2059: syntax error : '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C3083: 'cs': the symbol to the left of a '::' must be a type 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2039: 'cartesian' : is not a member of 'boost::geometry' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2143: syntax error : missing ';' before '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2913: explicit specialization; 'boost::access' is not a specialization of a class template 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2059: syntax error : '<' 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): error C2267: 'set' : static functions with block scope are illegal 1>c:\work\pantheon studios\o2\disesupport\supportfunctions.cpp(37): fatal error C1903: unable to recover from previous error(s); stopping compilation I'd rather not have to define my own point structure if at all possible! Any help appreciated to solve this. Cheers Sean.

2 2

Help compiling/linking boost example
by zaphod742 . 25 Jan '14

25 Jan '14

I installed boost-all-dev using apt-get on Ubuntu. I am trying to compile this example program of sync_client.cpp http://www.boost.org/doc/libs/1_45_0/doc/html/boost_asio/example/http/clien… I get this error .. $ sudo g++ sync_client.cpp -I usr/local/include/boost -L /usr/local/lib /tmp/ccaVkCWp.o: In function `__static_initialization_and_destruction_0(int, int)': sync_client.cpp:(.text+0xd61): undefined reference to `boost::system::generic_category()' sync_client.cpp:(.text+0xd6d): undefined reference to `boost::system::generic_category()' sync_client.cpp:(.text+0xd79): undefined reference to `boost::system::system_category()' /tmp/ccaVkCWp.o: In function `boost::system::error_code::error_code()': sync_client.cpp:(.text._ZN5boost6system10error_codeC2Ev[_ZN5boost6system10error_codeC5Ev]+0x17): undefined reference to `boost::system::system_category()' /tmp/ccaVkCWp.o: In function `boost::asio::error::get_system_category()': sync_client.cpp:(.text._ZN5boost4asio5error19get_system_categoryEv[boost::asio::error::get_system_category()]+0x5): undefined reference to `boost::system::system_category()' /tmp/ccaVkCWp.o: In function `boost::asio::detail::posix_tss_ptr_create(unsigned int&)': sync_client.cpp:(.text._ZN5boost4asio6detail20posix_tss_ptr_createERj[boost::asio::detail::posix_tss_ptr_create(unsigned int&)]+0x19): undefined reference to `pthread_key_create' /tmp/ccaVkCWp.o: In function `boost::asio::detail::posix_thread::~posix_thread()': sync_client.cpp:(.text._ZN5boost4asio6detail12posix_threadD2Ev[_ZN5boost4asio6detail12posix_threadD5Ev]+0x26): undefined reference to `pthread_detach' /tmp/ccaVkCWp.o: In function `boost::asio::detail::posix_thread::join()': sync_client.cpp:(.text._ZN5boost4asio6detail12posix_thread4joinEv[boost::asio::detail::posix_thread::join()]+0x2b): undefined reference to `pthread_join' /tmp/ccaVkCWp.o: In function `boost::asio::detail::posix_tss_ptr<boost::asio::detail::call_stack<boost::asio::detail::task_io_service>::context>::~posix_tss_ptr()': sync_client.cpp:(.text._ZN5boost4asio6detail13posix_tss_ptrINS1_10call_stackINS1_15task_io_serviceEE7contextEED2Ev[_ZN5boost4asio6detail13posix_tss_ptrINS1_10call_stackINS1_15task_io_serviceEE7contextEED5Ev]+0x15): undefined reference to `pthread_key_delete' /tmp/ccaVkCWp.o: In function `boost::asio::detail::posix_tss_ptr<boost::asio::detail::call_stack<boost::asio::detail::strand_service::strand_impl>::context>::~posix_tss_ptr()': sync_client.cpp:(.text._ZN5boost4asio6detail13posix_tss_ptrINS1_10call_stackINS1_14strand_service11strand_implEE7contextEED2Ev[_ZN5boost4asio6detail13posix_tss_ptrINS1_10call_stackINS1_14strand_service11strand_implEE7contextEED5Ev]+0x15): undefined reference to `pthread_key_delete' collect2: ld returned 1 exit status How can I solve this error?

1 0

Fwd: [boost] Boost.Fiber review January 6-15
by Nat Goodspeed 23 Jan '14

23 Jan '14

SUMMARY The review of the proposed Boost.Fiber library ended on January 15, 2014. The verdict is: not in its present form. The lively discussions during the course of the review indicate considerable interest in this library, however, and every submitted review can be read as: perhaps, if certain changes are made. This produced a long list of suggestions, which constitutes the bulk of this report. On behalf of the Boost community, I would like to commend Oliver Kowalke for the work he has put into the Fiber library so far. I encourage him to continue to evolve this library and to bring it back for a mini-review. Rather than considering the subject library as a whole, a Boost mini-review zooms in on particular issues. That seems an appropriate tool to use for further evaluation of the Fiber library -- always subject, of course, to the approval of the Review Wizards. I received seven formal votes, abbreviated here: Niall Douglas: YES, IF Eugene Yakubovich: YES, AND Antony Polukhin: YES, IF Vicente J. Botet Escriba: NO, UNTIL Agustín K-ballo Bergé: NO, IN CURRENT STATE Hartmut Kaiser: NO, IN CURRENT FORM Thomas Heller: NO, IN CURRENT FORM I thank each of these reviewers, and indeed everyone who investigated the Fiber library and participated in the discussions. If you feel that I have misrepresented your position, or have omitted or garbled an important point, please respond: the archived mail thread should accurately reflect the will of the community, even if this message does not. >From the long list of suggestions, a few key themes emerged. PERFORMANCE Many respondents requested performance tests (specifics below). There were a number of suggestions about possible performance pitfalls in the current implementation, such as use of STL containers (with consequent heap allocations) and locks for thread safety. Several people suggested implementing performance tests *before* starting any such optimizations, which seems like sensible advice. Niall Douglas pointed out that picking some fixed number of fibers is less interesting than showing how resource consumption rises with the number of fibers: "I think it isn't unreasonable for a library to enter Boost if it has good performance *scaling* to load (e.g. O(log N)), even if performance in the absolute or comparative-to-near-alternatives sense is not great. "Absolute performance can always be incrementally improved later, whereas poor performance scaling to load usually means the design is wrong and you're going to need a whole new library with new API. "This is why I really wanted to see performance scaling graphs. If they show O(N log N) or worse, then the design is deeply flawed and the library must not enter Boost. Until we have such a graph, we can't know as there is no substitute for empirical testing." Given the truth of that last observation, I have recast certain suggested optimizations as requests for measured performance cost. Performance requests include: - empty function (create, schedule, execute, delete one fiber) - same with one yield - overhead of using a future object - tests as described in [1]; such tests should allow comparing with TBB, qthreads, openmp, HPX - construct and join a single fiber vs. construct and join a single thread (empty function) - construct and join several fibers vs. construct and join several threads (empty function) - construct and detach a single fiber vs. construct and detach a single thread (empty function) - cost of STL containers (std::vector, std::map, std::deque) - cost of fiber interruption support (same as next bullet?) - cost of thread safety: intra-thread fibers vs. inter-thread fibers - cost of round_robin_ws vs. round_robin scheduler - cost of building on Coroutine vs. building on Context Once Fiber performance tests are available, I trust the community will assist Oliver in running them on systems otherwise unavailable to him. There was some disagreement on whether it is essential for the Fiber library to attain a certain level of performance before it should be accepted. Respondents fall into two broad camps. Some see fibers as lightweight threads (without the overhead of kernel context switching). They assert that since the API is that of std::thread, the only reason to accept a separate library would be astounding performance. They note that on some hardware it is already plausible to run hundreds of thousands of concurrent std::threads. Others see fibers as a tool for organizing code based around asynchronous operations (rather than chains of callbacks). They assert that for such purposes, eliminating the kernel from context switching is sufficient performance guarantee. I myself maintain that cooperative context switching provides important functionality you cannot reasonably get from std::thread. The Fiber library attempts to address both use cases. (It was suggested that if its primary target were code organization, perhaps it should be renamed something other than "Fiber.") In any case, the community clearly wants to see Fiber performance data. Some requested CPU cycles to eliminate clock speed differences, also memory consumption. ARMv7 data was requested for "extra bonus points." DOCUMENTATION Many respondents requested additional documentation, or clarifications. Requests include: - Rationale page explaining what's there, what's not and why. Explain distinction between Coroutine library and Fiber library. If certain Fiber functionality is intended to support yet another library (rather than being complete in itself), call out what would need to be added. - A section on how to install and run the tests and examples. The need to embed in a Boost tree is implied but not stated. Mention the need to build the library and link with it. - Explain synchronization between fibers on different threads. Must the code take more care with this than with synchronizing fibers on the same thread? - Clarify that an exception raised by a fiber function calls std::terminate(), as with std::thread, rather than being consumed. - More clearly explain migrating a fiber from one scheduler to another. - Document async() in a way compatible with the C++ standard. - Clarify thread-local effect of set_scheduling_algorithm(). There was a request to put this function in a this_thread nested namespace to further clarify. - Move algorithm class documentation to "Extension" or "Customization" section. Clarify that it's not part of the baseline library functionality, but a customization point. - Document fiber::id. - Better document promise/future for void and R& (per C++ standard). - Document thread safety of each support class (or method, if it varies by method). - Document complexity guarantees per API. - Document exception safety per API. - Document supported architectures (perhaps link to Coroutine library's list); state minimum compiler versions. - Document the get/set overloads of thread_affinity() and priority() separately. Perhaps rename setters to set_thread_affinity() and set_priority(). - Explain how portable is fiber priority, if it's specific to a scheduler. What does priority mean when you migrate a fiber from one thread (one scheduler instance) to another? - Document the library's ASIO support. Link to Coroutine's ASIO yield functionality; ensure that ASIO yield is adequately explained. In particular, distinguish it from this_fiber::yield. - Better explain (and/or comment) publish-subscribe example, also other existing examples. In addition to the documentation requests above, there were requests for additional examples: - Simple example of ASIO callback implementation vs. the same logic using Fiber's ASIO support, a la [4]. - Example of a fiber pool. - Example of an arbitrary thread B filling a future on which a fiber in thread A is waiting. - Example of an arbitrary thread B posting to an asio::io_service running fibers in thread A. - Either defend fibers::condition_variable from spurious wakeups in existing examples, OR document the stronger condition_variable guarantee. - Example of M:N threading with ASIO. That might involve either one io_service per CPU, with fiber migration; or a single io_service with run() calls from each CPU, grouping fibers for each CPU into strands. - Example of one thread with many fibers making service requests on a pool of worker threads performing blocking calls. - Example of using thread_specific_ptr to manage lifespan of user-specified scheduler. - Example of the owner of a fiber changing the fiber's thread affinity vs. the fiber itself. When would you use each tactic? - Load example programs into an Examples appendix so that Google searches can turn up library documentation. LIBRARY API - Four people overtly approved the close parallel with std::thread and its support classes. - Allocating a default scheduler object, rather than specifying a default template param, was praised. - Three people called out the set difference between Boost.Thread features and Fiber (e.g. future::get_exception_ptr()). One wants these implemented immediately; another says they can be added later; the third simply requests that they be documented, with rationale. - Two people frowned on introducing operator bool methods not found in std::thread or Boost.Thread. - C++11 support was mentioned, notably Boost macros such as BOOST_RV_REF and BOOST_EXPLICIT_OPERATOR_BOOL. Also mentioned were: C++11 idioms; C++11 std::thread patterns; move construction; initializer lists; rvalue this overloads; deleting operators. - The fiber constructor and async() should accept a move-only callable. - At least for a C++11 compiler, fiber constructor and async() should accept variadic parameters. These should support move-only types, like Boost.Thread. C++03 support for variadic parameters would be nice, but is less important. - Every API involving time point or duration should accept arbitrary clock types, immediately converting to a canonical duration type for internal use. - Queues should support value_pop() returning item by value. This supports an item type without a default constructor. - Nested scoped_lock typedef has been deprecated in thread library. Remove in Fiber library. - Align the return type of shared_future::get() with the standard. In general, ensure that parameter types and return types are aligned with the standard. - A couple of people were bothered by the use of types in the detail namespace as parameters or return values in the algorithm API. (I note, however, that extending e.g. Boost.Range can involve touching its detail namespace. A customization point for a library may be a bit of a gray area.) - There was a suggestion to rename algorithm to scheduler. In that case, presumably set_scheduling_algorithm() could be renamed set_scheduler(). - There was a request to rename round_robin_ws to round_robin_work_stealing. - A couple of people consider the algorithm API too monolithic, pointing to redundancies in the round_robin, round_robin_ws and asio round_robin implementations. They suggested teasing out distinct classes, so that (for instance) a user-coded scheduler might be able to override a single method to respect fiber priority. In fact Eugene Yakubovich offered to experiment with refactoring the algorithm class this way. - There was a request for set_scheduling_algorithm() to return the previous pointer. (It might be useful for the requester to explain the anticipated use case. An earlier iteration of set_scheduling_algorithm() did return the previous pointer; Oliver intentionally changed that behavior.) - fiber_group got one thumbs-up and two thumbs-down. Options: retain; improve to use move support rather than fiber*; discard. There is an opportunity to improve on thread_group; naturally there is risk in diverging from thread_group. - Request deferred futures for lazy evaluation. - There was a suggestion to introduce a global object to coordinate thread-specific fiber schedulers, in the hope that the global object could perform all relevant locking and the thread-specific fiber schedulers could themselves be thread-unsafe. - There was a request to unify steal_from() and migrate_to() into a single method. I infer that this is predicated on the previous suggestion. - Request future::then() et al, per [5]. (Someone please clarify the present status of N3784?) - Request enriched barrier support per [6] and [7]. (Someone please clarify the present status of N3817?) - There are two fiber properties specific to particular schedulers: thread_affinity (used only by round_robin_ws) and priority (as yet unused by any scheduler). What if a user-coded scheduler requires a fiber property that does not yet exist? Is there a general approach that could subsume the present support for thread_affinity and priority, in fiber and this_fiber? Could the initial values for such properties be passed as part of the fiber constructor's attributes parameter? - One use case was surfaced that may engage the previous bullet: the desire to associate a given fiber with any of a group of threads, such as the set of threads local to a NUMA domain or physical CPU. IMPLEMENTATION - Replace std::auto_ptr with boost::scoped_ptr. The former produces deprecation warnings on GCC. - Reduce redundancy between try_lock() and lock(). - boost::fibers::asio::detail::yield_handler::operator()() calls algorithm::spawn() before algorithm::run(). Does this allow the scheduler to choose the next fiber to run, e.g. a user-coded scheduler that respects fiber priority? - Add memory transaction support to spinlock a la [8]. - Intel TSX lock avoidance would be nice. LINKS [1] https://github.com/STEllAR-GROUP/hpx/tree/master/tests/performance [2] http://stellar.cct.lsu.edu/pubs/isc2012.pdf [3] http://stellar.cct.lsu.edu/pubs/scala13.pdf [4] https://ci.nedprod.com/job/Boost.AFIO%20Build%20Documentation/Boost.AFIO_Do… [5] http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2013/n3784.pdf [6] http://www.boost.org/doc/libs/1_55_0/doc/html/thread/synchronization.html#t… [7] http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2013/n3817.html#barrier_… [8] https://github.com/BoostGSoC/boost.afio/blob/master/boost/afio/detail/ Nat Goodspeed Boost.Fiber Review Manager ________________________________

1 0

[Graph] initialize_vertex change vertex color (visitor)
by The Maschine 23 Jan '14

23 Jan '14

Hi all, I would like to build a visitor (for dikstra) with the initialise_vertex acting as 'colour map' modifier. I want to exclude some vertices from the search based on a condition. So I want to set some vertices 'black' in the init part of the algorithm. class dijkstra_2step : public boost::default_dijkstra_visitor { public: dijkstra_2step(std::vector<Weight> dists, double threshold): distances(dists), threshold(threshold) {} // THIS PART IS NOT CORRECT!!!! // void initialize_vertex(boost::graph_traits <unGraph>::vertex_descriptor u, const unGraph& g){ if( distances[u] > threshold ) color[u] = black; // ?????? } ////////// std::vector<Weight> distances; double threshold; }; Any help for the above visitor? How to I access the colour? I couldn't find something online. Thanks

3 12

[qi::phrase_parse] how to retain minus as string
by Uthpal Urubail 23 Jan '14

23 Jan '14

I am trying to store minus as the string in the key. I think its getting conflict with to(a-z) symbol key = +qi::char_(".a-zA-Z0-9") | qi::lexeme[('[' >> +qi::char_("@ + - * . ( , ) $ a-zA-Z0-9 ") >> ']')]; value = +qi::char_("a-zA-Z0-9") ; I have the string like below: std::string input = “[(@vala - @valb) * @valc]:real"; Can someone help me to resolve this? Regards, UJ

2 1

Build boost using CMake
by Chameera Wijebandara 23 Jan '14

23 Jan '14

Hi, I am new to the Boost. And I have to build Boost libraries with CMake. Any Suggestions? -- Thanks, Chameera

2 1

Server failing when BOOST_ASIO_ENABLE_HANDLER_TRACKING is active
by Gonzalo Garramuno 22 Jan '14

22 Jan '14

Trying to track down when the socket closes on my program, I turned on asio debugging. Server fails when BOOST_ASIO_ENABLE_HANDLER_TRACKING is active. Its stack trace is: g Scanning dependencies of target mrViewer [ 9%] Building CXX object mrViewer/src/CMakeFiles/mrViewer.dir/core/mrvServer.o Linking CXX executable ../../../bin/mrViewer [100%] Built target mrViewer # gga@gga-laptop1 /media/Linux/code/applications/mrViewer # mar ene 01 10:04 [8] $ mrViewer /media/Linux/movies/Tangled.mov [opengl] Using hardware shader profile: NVidia NV30 [opengl] mrViewer supports YUV images through shaders. [psql] Connected to database 'assets2d_development'. The code: boost::asio::io_service io_service; tcp::endpoint listen_endpoint(tcp::v4(), s->port); server rp(io_service, listen_endpoint, s->ui); io_service.run(); fails with: @asio|1357088683.200673|0|socket@0x7f303d8d2d00.close Got signal 11, faulty address is 0x1, from 0x4d222a [bt] Execution path: [bt] mrViewer(_ZN5boost4asio6detail15task_io_service16shutdown_serviceEv+0x3a) [0x4d222a] [bt] mrViewer(_ZN5boost4asio6detail15task_io_service16shutdown_serviceEv+0x3a) [0x4d222a] [bt] mrViewer() [0x4dd2c9] [bt] mrViewer(_ZN3mrv13server_threadEPKNS_10ServerDataE+0x408) [0x4e0e58] [bt] /usr/lib/libboost_thread.so.1.48.0(+0xdda9) [0x7f304e191da9] [bt] /lib/x86_64-linux-gnu/libpthread.so.0(+0x7e9a) [0x7f304f7f0e9a] [bt] /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f304da1ccbd] without the #define the code works reliably except sometimes the socket closes unexpectedly.

4 5

Re: [Boost-users] Interprocess: rbtree_best_fit ABI change from 1.53 to 1.55 ?
by Arunski, Kevin 22 Jan '14

22 Jan '14

> El 22/01/2014 16:14, Arunski, Kevin escribió: > > Hello, I have been using a boost::interprocess::map stored in a > > managed mapped file segment, using the rb_best_fit allocation > > algorithm. > > > > I have observed that starting in boost 1.55, I cannot interoperate > > with instances of this allocator that were created by programs > > compiled by boost 1.53. > > > > It appears that the size of a boost::intrusive multiset object has > > changed, and this changes the representation of the rb_best_fit > > algorithm within the shared memory region. > > Sorry about that. That must be a bug. Can you tell me please the sizes > you get in both versions? Sure. In code compiled with boost 1.53, this type: boost::interprocess::rbtree_best_fit<boost::interprocess::null_mutex_family,boost::interprocess::offset_ptr<void,__int64,unsigned __int64,0>,0> has size = 56 member m_header.m_imultiset has size 32 In code compiled with boost 1.55: the same type has size = 72 member m_header.m_imultiset has size 48

2 1