November 2008 - Boost-users

[Math] Accuracy test failures on QNX 6.4.0
by Niklas Angare 27 Nov '08

27 Nov '08

Some of the accuracy tests fail on QNX 6.4.0 x86 with GCC 4.2.4 and Dinkumware 5. Some tests such as test_carlson have a peak error 10%-40% higher than expected while for example test_ellint_1 has huge errors. I'm not sure how to proceed. I'd appreciate of someone could take a look at the trunk regression test results (runner NA-QNX640-qcc). -- Regards Niklas Angare

2 17

Trouble with asio using lib pion and boost.function
by Stephan Menzel 27 Nov '08

27 Nov '08

Hi there, I'm not actually sure this is a boost problem but I'll give it a try and ask for your opinion. Boost version is 1.36 I have a daemon here realising a web interface using pion and asio. Pion is also using asio. On Linux everything runs fine and I have my web interface, but using the same code on windows / vs 2005 I get the weirdest errors such as " Unhandled exception at 0x1004e55d in myapp.exe: 0xC0000005: Access violation reading location 0xcdcdcdd1." The debugger tells me this happens in function_template.hpp: #if !BOOST_WORKAROUND(BOOST_MSVC, < 1300) template<typename R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_PARMS> typename BOOST_FUNCTION_FUNCTION< R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_ARGS>::result_type BOOST_FUNCTION_FUNCTION<R BOOST_FUNCTION_COMMA BOOST_FUNCTION_TEMPLATE_ARGS> ::operator()(BOOST_FUNCTION_PARMS) const { if (this->empty()) boost::throw_exception(bad_function_call()); return static_cast<vtable_type*>(vtable)->invoker (this->functor BOOST_FUNCTION_COMMA BOOST_FUNCTION_ARGS); } #endif It says the static_cast rises the exception (?!) Damn MS. Anyway, I debugged into it and the last thing the debugger stepped into was in asio: size_t run(boost::system::error_code& ec) { if (::InterlockedExchangeAdd(&outstanding_work_, 0) == 0) { ec = boost::system::error_code(); return 0; } call_stack<win_iocp_io_service>::context ctx(this); size_t n = 0; while (do_one(true, ec)) if (n != (std::numeric_limits<size_t>::max)()) ++n; return n; } I don't know which line though. Can anyone help shed some light on this? Thanks and greetings... Stephan

2 4

[dataflow]Component Library
by Duane McKinney 27 Nov '08

27 Nov '08

Is there already or are there plans for components to be shared. I know this is a little outside the scope, but it does seem to be closely related. For example if I need an RGB to YUV component, it is likely that it already exists, and if not, it is likely that someone will need it after me.

2 4

How to make custom property map compatible with bgl_named_params?
by Tim Keitt 26 Nov '08

26 Nov '08

I have a custom (internal) property map defined as: struct edge_weight_extractor_map { typedef float8 value_type; typedef float8 reference; typedef WeightedEdgeTableSQLGraph::edge_descriptor key_type; typedef boost::readable_property_map_tag category; }; inline edge_weight_extractor_map::value_type get(const edge_weight_extractor_map& map, const edge_weight_extractor_map::key_type& key) { return key.edge_weight; } The map works fine as an internal property map with eg dijkstra_shortest_paths. However, I cannot get it to work when it is used in a bgl_named_params expression like: boost::brandes_betweenness_centrality(g, centrality, boost::weight_map(get(boost::edge_weight, g))); The error looks like: /usr/include/boost/property_map.hpp:25: error: no type named 'key_type' in 'struct boost::bgl_named_params<edge_weight_extractor_map, boost::edge_weight_t, boost::no_property>' /usr/include/boost/property_map.hpp:27: error: no type named 'reference' in 'struct boost::bgl_named_params<edge_weight_extractor_map, boost::edge_weight_t, boost::no_property>' /usr/include/boost/property_map.hpp:28: error: no type named 'category' in 'struct boost::bgl_named_params<edge_weight_extractor_map, boost::edge_weight_t, boost::no_property>' Do I need to specialize a traits template? Can someone post an example of how this works? (I've tried to specialize bgl_named_params itself, but unsuccessfully.) Thanks. THK -- Timothy H. Keitt http://www.keittlab.org/

1 1

program_options: Allowing unknown options in config files
by Kevin Channon 26 Nov '08

26 Nov '08

Hi, I'm trying to use the Program Options library to parse command line arguments and config files. I can do this with moderate success. I am currently stuck on a solution to the problem of unrecognized options that are present in a config file. It seems that parse_config_file() does not have an obvious way to not abort reading a config file once it finds something that isn't declared in the options description. Any help on how to get around this (or just a kind of "hey you missed this bit of the manual" kind of pointer) would be very welcome. Thanks, Kevin

1 0

Maintenance Guidelines wiki page (Revison 8)
by vicente.botet 26 Nov '08

26 Nov '08

Hi, I have updated the Maintenance Guidelines wiki page. I have take in account the minor proof-reading modifications from Steve Watanave which I have overridden by error (thanks Steve!),and added a lot of thinks, maybe too much :) Please read this mail completely before jumping to the page. This page need to be completed and reworked. Please be free to add new sections to the page or improve the current ones directly or post your comments or suggestions to this list. Here it is the table of contents. [C] 1. Motivation [C] 2. Introduction [C] 1. User code breaking cases [I] 2. Versioning individual Boost libraries [O] 3. Deprecating features [I] 4. Cross version testing [C] 3. Documentation [O] 1. Tag Boost library with specific library version [developer] [O] 2. Make feature request for each feature [developer, user] [I] 3. Include the tracked tickets on the Release notes [developer] [I] 4. List the test cases associated to the trac system tickets [developer] [I] 5. List the dependency upon other Boost library [developer] [O] 6. Document behavior differences between release and debug variants [O] 7. Document behavior differences between toolsets [developer] [C] 4. Coding [O] 1. Don't use using directives [user] [C] 2. Avoid the use of include all features files at the /boost level [user] [C] 3. Be careful with the use of using namespace header files [developer] [C] 4. Don't overload functions that are used by the TR1 (using using)[developer] [O] 5. Avoid include-all-features files at the /boost level [C] 6. Don't refine functions overloading without ensuring the same behavior [C] 7. Avoid the inclusion of symbols at the boost or boost::detail namespace [C] 8. Avoid different external behavior depending on the variant release or debug [O] 9. Avoid to change interfaces [developer] [O] 1. Don't delete files prematurely [developer] [O] 2. Don't delete namespaces prematurely [developer] [O] 3. Don't delete classes/functions/variables prematurely [developer] [O] 4. Don't modify functions prototypes prematurely [developer] [O] 5. Remove the deprecated features on a given release [developer] [C] 5. Test [C] 1. Test headers [developer] [C] 1. Include each header files twice [developer] [I] 2. include each couple of header files in both orders [C] 3. Include all header files [C] 4. link all the header files twice [C] 2. Don't forget to test [C] 1. The implicitly generated member functions [C] 2. The removed default member functions when you declare a constructor [C] 3. The deleted (private) default member functions [C] 4. The explicit constructors [C] 5. The implicit constructors or conversions [C] 6. The const-ness of variables, function parameters and function return types [I] 3. Separate the functional test from the unit test (implementation test) [I] 4. Track regression test on the trac system tickets [I] 5. Preserve the functional test from the preceding versions [developer] [C] 6. Inspections [O] 1. Announce new library features [developer] [C] 2. Inspect the code [developer, user, RM] [C] 3. Check that every modification has been documented [developer, user, RM] [C] 4. Check that every modification has been tested [developer, user, RM] [U] 7. Developer guidelines [U] 8. User guidelines [U] 9. Release manager guidelines Tasks I plan to do: 1* Add the motivation 2* Add other examples of user code break in 2.1 3* Update the Developer, User and Release manager guidelines cross reference (7,8,9) 4* Separate on 4 pages. Main, Documentation, Coding, Test and Inspections 5* Complete incomplete guidelines [C] and improve some of them [I] Note that some guidelines applied to the developer library could be done by other people that would need SVN writtig permission. In particular the Functional Test and some documentation reports. (See below points C and E) Other points I see: A* See if the Jamfile to test the headers from Steve can be adapted to tests (Steve?) 5.1.2 Include each couple of header files in both orders B* See if the difference between source and binary compatibility is desirable (ALL) C* See if the functional test can be written by a team of interested users to don't overload the developers. These tests could in addition include multi-library tests and be stored in a directory independent from the developers one (Interested users?) D* See which configurations (deprecated or not) could be tested by the current test team. (Release Manager Team?) E* See how the following points can be automated and where this information can be included in the documentation. A separated page make possible that this task could be done by other people than the developer. (Some one knowing the Track system?) 3.3. Include the tracked tickets on the Release notes 3.4. List the test cases associated to the trac system tickets F* See how the dependencies between libraries can be automated (Release Manager Team?, CMAKE project?) 3.5. List the dependency upon other Boost library In order to know if I'm going on the right direction, I would like the interested people reply to this post stating for each section, if it is OK[O], must be completed[C]/improved [I], must be updated [U], must be removed [R] and of course any idea is welcome. You can access it directly at "https://svn.boost.org/trac/boost/wiki/Guidelines/MaintenanceGuidelines" or at "https://svn.boost.org/trac/boost/wiki" >Guidelines >Maintenance Guidelines Best regards, Vicente

3 2

How to handle an adjacency matrix on large size
by Navin Shaw 26 Nov '08

26 Nov '08

Dear Friends, I am a fresh programmer and I have assigned a work where I need to handle a graph of very large size(around lakhs by lakhs matrix) with a huge amount of data. I heard that using boost library this can be done very easily, If anybody have any idea kindly guide me. Thanks in advance................ -- Thanks & Regards--- Navin Shaw

1 0

Re: [Boost-users] class API return a const&, optional, shared_ptr
by Emre Turkay 26 Nov '08

26 Nov '08

I see you have three nice options here, 1. Return a shared_ptr. 2. Return a reference and throw exception for invalid indexes. 3. Return an optional. I think all three of them are nice and safe. emre > I have a class A that holds a container of objects of type B indexed by int > for e.g. > > Typically, I have a function of A > > class A { > public: > const B& getB(int <index> ) const; > private: > std::map< int, shared_ptr<B> > listOfBs_; > }; > > The implementation of getB could be > > const B& A::getB(int <index> ) const > { > //what to do if index not found > } > > A number of solutions I think of: > > 1. > boost::optional<const B&> getB(int <index> ) const > { > } > > > 2. > change signature to > const B* getB(int <index> ) const > > > 3. > Define a > static const B notFoundB; > const B& A::getB(int <index> ) const > returns notFoundB when not found and have API users compare return value > with notFoundB > > > I suppose the cheapest in overhead/performance is 2... optional must have > some time overhead? > > Are there advised solutions? Favourites in terms of style? > > Regards,

1 0

hash_map use with serialization in Visual Studio 2005
by Mauricio Gomes 25 Nov '08

25 Nov '08

Dear all, I am learning to use the serialization library and I am having some difficulties in using hash_map. In the code below if I change to std::map it works fine. #include <cstddef> // NULL #include <iomanip> #include <iostream> #include <fstream> #include <string> #include <hash_map> #include <boost/archive/tmpdir.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/utility.hpp> #include <boost/serialization/list.hpp> #include <boost/serialization/hash_map.hpp> #include <boost/serialization/map.hpp> #include <boost/serialization/assume_abstract.hpp> template <typename StringType, typename SettingKeyType, typenameSettingValueType> void save_setting (StringType application_name, SettingKeyType key, SettingValueType value) { typedef stdext::hash_map<SettingKeyType, SettingValueType> Config; static stdext::hash_map<StringType, Config> app_configs; // key = application_name, value = settings map. Config cfg = app_configs[application_name]; cfg[key] = value; StringType file_name = application_name + ".cfg"; // Compose configuration file name. std::ofstream ofs (file_name.c_str ()); boost::archive::text_oarchive oa (ofs); oa << cfg; } int main(int argc, char *argv[]) { save_setting<std::string, int, std::string> ("test_app", 1, "value1"); } However if I try to use stdext::hash_map then it fails to compile with the following error: 1>c:\boost_1_37_0\boost\serialization\access.hpp(109) : error C2039: 'serialize' : is not a member of 'stdext::hash_map<_Kty,_Ty>' 1> with 1> [ 1> _Kty=int, 1> _Ty=std::string 1> ] It is most likely related to the fact that hash_map is not in the std namespace. I have searched the boost-users list and I have found this: http://lists.boost.org/boost-users/2005/08/13183.php but it refers to a GNU/Linux environment. Has this been addressed in 1.37 (that's the version I am using) ? Any suggestions ? Thank you in advance for your help. -- Mauricio Gomes

1 0

[dataflow][iostreams]Can someone Compare/Contrast
by Duane McKinney 25 Nov '08

25 Nov '08

If it has already been done I appologize. They seem very similar to me. Is it just a matter of style? It seems like dataflow lends itself better to dealing with blocks/frames of data than the iostreams library. I would appreciate any clarification though.

2 2