November 2009 - Boost-users

Determine Parent Class from boost::any
by Yan Cheng Cheok 23 Nov '09

23 Nov '09

May I know is there any way to determine parent class from boost::any? #include <iostream> #include <boost/any.hpp> class a { public: virtual ~a() {} }; class b : public a { }; bool is_class_a(const boost::any& any) { return boost::any_cast<a>(&any) != 0; } bool is_class_a_v2(const boost::any& any) { try { boost::any_cast<a>(any); return true; } catch(const boost::bad_any_cast &) { return false; } } int main() { std::cout<< is_class_a(b())<< std::endl; // return 0. but i wish to return 1. std::cout<< is_class_a(a())<< std::endl; // return 1. std::cout<< is_class_a_v2(b())<< std::endl; // return 0. but i wish to return 1. std::cout<< is_class_a_v2(a())<< std::endl; // return 1. }

2 1

[ASIO] Timer question
by OvermindDL1 23 Nov '09

23 Nov '09

Greetings. I programming library that I use (WT) had a bug in their asynchronous timer handling, and although they fixed it, they are thinking of moving over to Boost::ASIO timers since they already use Boost::ASIO for the networking stack. However, an interesting question was posed about how Boost::ASIO timers handle the time changing on the host computer (which is a possible scenario for this library due to its up-time requirements). Below is the basic part of the thread in question. Does anyone know the answer? I attempted a quick look through the timer source, however I cannot really seem to tell since I know not the internals of ASIO nor the posix datetime library in Boost... On Fri, Nov 20, 2009 at 12:55 AM, Pau Garcia i Quiles <pgquiles(a)elpauer.org> wrote: > On Fri, Nov 20, 2009 at 8:05 AM, Koen Deforche <koen(a)emweb.be> wrote: >> Hey all, >> >> 2009/11/19 OvermindDL1 <overminddl1(a)gmail.com>: >>>>> Speaking about WTimer, I think its implementation is wrong on Unix and >>>>> could be more precise on Windows (although I'm not sure more precision >>>>> is required; I've not checked if Wt::Time is used server-side). >>>> >>>> I haven't looked, but is there any reason that boost::asio::deadline_timer isn't being used to back WTimer? I'm pretty sure this addresses all of the concerns listed below. -sc >>>> >>>> >>>> http://www.boost.org/doc/libs/1_41_0/doc/html/boost_asio/reference/deadline… >>> >>> Much of Boost already handles a lot of things better, since boost is >>> linked in anyway then we should use them, they have already been well >>> debugged and optimized. >> >> That is very true. I guess a motivation at the time was to avoid the >> dependency on boost date_time for such a seemingly easy thing. But in >> the meantime, we are relying on boost date_time anyway, and it is >> clear that it is not that seemingly easy ? > > Actually, it is pretty easy. You've got 90% of the code there. Take a > look at the OSG file I said. Fixing this on Unix is straightforward, > fixing on Windows is easy too. > > Further, I think using boost date_time is not the right solution, as > it does not provide timers - i. e. you'll be basing the timer on the > clock time (hh:mm:ss) instead of processor ticks or some other entity > independent of clock time. If timers are based on hh:mm:ss - > hh':mm':ss' differences, they will fail when you change the computer > clock time. > > I have only looked slightly into the implementation of Asio timers and > they do not use gettimeofday but they don't use clock_gettime either, > therefore I wouldn't use them without carefully looking into the > implementation.

3 2

[BOOST-users] UBLAS X-<X> operationin ublas?
by Arman Khalatyan 23 Nov '09

23 Nov '09

Hello, how can I rewrite following expression to UBLAS? template <class T> boost::numeric::ublas::matrix<T> SubstractMeans( boost::numeric::ublas::matrix<T> &m, boost::numeric::ublas::vector<T> &vMeans) { ublas::matrix<T> mout(m.size1(), m.size2()); for(size_t i=0;i<m.size1();i++) for(size_t j=0;j<m.size2();j++) { mout(i,j)=m(i, j)-vMeans(i); } //cout<<mout<<endl; return mout; } thanks, Arman.

1 0

[Spirit.Qi] I want to modify _val, how?
by HT4N - 23 Nov '09

23 Nov '09

So, I was able to modify _1 (type of struct ptnode) by using a functor() (e.g. [_val = func(x, _1)] where the function will take a reference of _1 and do some modification to it and return (typeof(_1)&). Can we also modify _val placeholder on the fly through semantic action? i.e. once _val = _1 executed, then modify the _val object _________________________________________________________________ Hotmail: Trusted email with Microsoft's powerful SPAM protection. http://clk.atdmt.com/GBL/go/177141664/direct/01/ http://clk.atdmt.com/GBL/go/177141664/direct/01/

3 2

Problem with exception and property tree in 1.41.0.
by Jakub Herkel 23 Nov '09

23 Nov '09

Hello, I have tried 1.41.0, especially PropertyTree. I used cross compiler arm-linux-g++ version 3.2.2 and example from property tree (debug_settings.cpp). My compilation has failed with these output : In file included from /usr/local/arm-linux/arm-linux/include/boost/thread/future.hpp:12, from /usr/local/arm-linux/arm-linux/include/boost/thread.hpp:24, from /home/jakub/projects/tigris/agent/src/main/cpp/agent/Agent.cpp:10: /usr/local/arm-linux/arm-linux/include/boost/exception_ptr.hpp:43: error: looser throw specifier for `virtual boost::exception_ptr::~exception_ptr()' /usr/local/arm-linux/arm-linux/include/boost/exception/detail/exception_ptr_base.hpp:27: error: overriding `virtual boost::exception_detail::exception_ptr_base::~exception_ptr_base() throw ()' /usr/local/arm-linux/arm-linux/include/boost/exception_ptr.hpp:43: error: looser throw specifier for `virtual boost::exception_ptr::~exception_ptr()' /usr/local/arm-linux/arm-linux/include/boost/exception/detail/exception_ptr_base.hpp:27: error: overriding `virtual boost::exception_detail::exception_ptr_base::~exception_ptr_base() throw ()' /usr/local/arm-linux/arm-linux/include/boost/exception_ptr.hpp:43: error: looser throw specifier for `virtual boost::exception_ptr::~exception_ptr()' /usr/local/arm-linux/arm-linux/include/boost/exception/detail/exception_ptr_base.hpp:27: error: overriding `virtual boost::exception_detail::exception_ptr_base::~exception_ptr_base() throw ()' In file included from /usr/local/arm-linux/arm-linux/include/boost/property_tree/xml_parser.hpp:19, from /home/jakub/projects/tigris/agent/src/main/cpp/agent/Agent.cpp:17: /usr/local/arm-linux/arm-linux/include/boost/property_tree/detail/xml_parser_read_rapidxml.hpp: In function `void boost::property_tree::xml_parser::read_xml_internal(std::basic_istream<typename Ptree::key_type::value_type, std::char_traits<typename Ptree::key_type::value_type> >&, Ptree&, int, const std::string&)': /usr/local/arm-linux/arm-linux/include/boost/property_tree/detail/xml_parser_read_rapidxml.hpp:122: error: parse error before `>' token There are two problems: 1) One in file boost/exception/detail/exception_ptr_base.hpp 2) Second in file boost/property_tree/detail/xml_parser_read_rapidxml.hpp I have compared old and new version of file exception_ptr_base.hpp and there is one difference. The new version added a virtual destructor. When I removed this line, the first problem was solved. My question is (because I'm not an expert in Boost) : Can I remove this virtual destructor without any consequence? For the second problem I think that STL library which is used in arm-linux toolchain is a little bit old. But maybe I'm wrong. As a temporary fix I also removed this line. But I hope that there is a better solution how to fix it. Could someone show me the way how I can solve it? Thank you in advance best regards jakub

3 6

unit test framework XML output - CDATA sections?
by Maróy Ákos 22 Nov '09

22 Nov '09

Hi, I have a suggestion for the XML output generated by the boost unit test framework: currently, if tests fail, the message detailing the failure of the test is put in verbatim as a test element into the XML output. This can be problematic, if the output contains XML-like structures, like text that looks like an XML element, or invalid characters. thus, I'd suggest enclosing such output in a CDATA section, that would ensure that the generated XML file is indeed always well formed XML. see more here: http://www.w3.org/TR/REC-xml/#sec-cdata-sect - but basically this would mean printing the failures like this: <![CDATA[ reason of the failure... ]]> Akos

2 2

[test] utf now supports log/report sinks
by Gennadiy Rozental 22 Nov '09

22 Nov '09

Hi, Based on users' requests I've added 2 new runtime parameters: '--log_sink=' or '-k ' or BOOST_TEST_LOG_SINK: Allows to set the log sink. Valid values are 'stdout' (default), 'stderr', and anything else which is treated as file name. '--report_sink=' or '-e ' or BOOST_TEST_REPORT_SINK: Allows to set the report sink. Valid values are 'stderr' (default), 'stdout', and anything else which is treated as file name. Gennadiy

1 0

[Spirit.Qi] How do I get Rule Name/Id through Semantic Action / Phoenix variable
by HT4N - 22 Nov '09

22 Nov '09

I want to create a generic Parse Tree from a simple expression grammar (see below code snippet). Using the Mini_XML as the reference, I was able to create a Parse Tree close to what I want. If I run my simple program and parse an expressions "[foo=1 bar=2]", for example, it will printout the following Parse Tree: [foo=1 bar=2] ------------------------- Parsing succeeded <nonterminal> { <nonterminal> { <value_str>foo</value_str> <value_int>1</value_int> } <nonterminal> { <value_str>bar</value_str> <value_int>2</value_int> } } The next thing I want to achieve is to have the Parse Tree looks very similar to the one produced in Spirit.Classic. In Spirit.Classic for each node it has RuleId for which we associate it with a RuleName so that when we are iterating the node of the Parse Tree we can printout the RuleName. In other words, instead I want the above Parse Tree to look like the following : <Option> <Expression> <ParameterName>foo</ParameterName> <ParameterValue>1</ParameterValue> </Expression> <Expression> <ParameterName>bar</ParameterName> <ParameterValue>2</ParameterValue> </Expression> </Option> In other words, what I need to know is how I can get the Rule Name/Id through semantic actions/phoenix variable. Any pointer is appreciated. Thanks again. HT Code Snippet: namespace client { struct ptnode; typedef boost::variant< nil, unsigned int, string, boost::recursive_wrapper<ptnode> > ptnode_var; /////////////////////////////////////////////////////////////////////////// // Our AST /////////////////////////////////////////////////////////////////////////// struct ptnode { ptnode() : data(nil()) { } template<typename T> ptnode(const T& data) : data(data) { } ptnode_var data; vector<ptnode_var> children; }; } // We need to tell fusion about our mini_xml struct // to make it a first-class fusion citizen //[tutorial_xml1_adapt_structures BOOST_FUSION_ADAPT_STRUCT( client::ptnode, (client::ptnode_var, data) (std::vector<client::ptnode_var>, children) ) //] namespace client { template <typename Iterator> struct ExpressionGrammar : qi::grammar<Iterator, ptnode(), ascii::space_type> { ExpressionGrammar() : ExpressionGrammar::base_type(Option) { using qi::_val; using qi::_1; using qi::char_; using qi::alnum; using qi::alpha; using qi::lexeme; using qi::raw; using qi::uint_; using phoenix::at_c; using phoenix::push_back; Option = char_('[') >> *Expression [push_back(at_c<1>(_val), _1)] >> char_(']'); Expression = ParameterName [push_back(at_c<1>(_val), _1)] >> (char_('=')>> ParameterValue [push_back(at_c<1>(_val), _1)]); ParameterName = lexeme[+(alnum) [_val += _1]]; ParameterValue = uint_ [_val = _1]; Option.name("Option"); Expression.name("Expression"); ParameterName.name("ParameterName"); ParameterValue.name("ParameterValue"); } qi::rule<Iterator, ptnode(), ascii::space_type> Option; qi::rule<Iterator, ptnode(), ascii::space_type> Expression; qi::rule<Iterator, string(), ascii::space_type> ParameterName; qi::rule<Iterator, int(), ascii::space_type> ParameterValue; }; /////////////////////////////////////////////////////////////////////////// // Print out the mini xml tree /////////////////////////////////////////////////////////////////////////// int const tabsize = 4; void tab(int indent) { for (int i = 0; i < indent; ++i) { printf_s(" "); } } struct ptnode_printer { ptnode_printer(int indent = 0) : indent(indent) { } void operator()(ptnode const& node) const; int indent; }; struct ptnode_var_printer : boost::static_visitor<> { ptnode_var_printer(int indent = 0) : indent(indent) { } void operator()(const ptnode& var) const { ptnode_printer(indent+tabsize)(var); } void operator()(const nil&) const { // Print node type ... tab(indent); printf_s("<nonterminal>\n"); } void operator()(const string& val) const { tab(indent+tabsize); printf_s("<value_str>%s</value_str>\n", val.c_str()); } void operator()(const int& val) const { tab(indent+tabsize); printf_s("<value_int>%d</value_int>\n", val); } int indent; }; void ptnode_printer::operator()(const ptnode& node) const { boost::apply_visitor(ptnode_var_printer(indent), node.data); tab(indent); printf_s("{\n"); BOOST_FOREACH(const ptnode_var& child, node.children) { boost::apply_visitor(ptnode_var_printer(indent), child); } tab(indent); printf_s("}\n"); } } _________________________________________________________________ Windows 7: I wanted simpler, now it's simpler. I'm a rock star. http://www.microsoft.com/Windows/windows-7/default.aspx?h=myidea?ocid=PID24…

2 3

[test] Malloc exceptions when using test framework on 64bit OS X 10.6.1
by Vrai Stacey 22 Nov '09

22 Nov '09

Hi, I've been making my first forays in to using Boost on the new OS X release and have encountered some strange behaviour on 64bit machines. Testing with Boost 1.37.0 and 1.40.0 I've found that everything compiles as expected, but attempting to run any unit tests results in an error of the form: *** No errors detected test_exceptions(92517) malloc: *** error for object 0x3000100200460: pointer being freed was not allocated *** set a breakpoint in malloc_error_break to debug Running the test through gdb gives the following stack trace: Program received signal SIGABRT, Aborted. 0x00007fff83f98ff6 in __kill () (gdb) bt #0 0x00007fff83f98ff6 in __kill () #1 0x00007fff8403a072 in abort () #2 0x00007fff83f51095 in free () #3 0x000000010003668a in boost::unit_test::framework_impl::clear (this=0x1000698a0) at framework.ipp:133 #4 0x0000000100036788 in boost::unit_test::framework_impl::~framework_impl (this=0x1000698a0) at framework.ipp:122 #5 0x00000001000386f2 in __tcf_1 () at framework.ipp:222 #6 0x00007fff83f5d274 in __cxa_finalize () #7 0x00007fff83f5d18c in exit () #8 0x0000000100006e93 in start () at unit_test_parameters.ipp:76 This is using the vanilla GCC build that comes with the operating system (see below), default compile flags and strictly single threaded test applications. Using built-in specs. Target: i686-apple-darwin10 Configured with: /var/tmp/gcc/gcc-5646~6/src/configure --disable-checking --enable-werror --prefix=/usr --mandir=/share/man --enable-languages=c,objc,c++,obj-c++ --program-transform-name=/^[cg][^.-]*$/s/$/-4.2/ --with-slibdir=/usr/lib --build=i686-apple-darwin10 --with-gxx-include-dir=/include/c++/4.2.1 --program-prefix=i686-apple-darwin10- --host=x86_64-apple-darwin10 --target=i686-apple-darwin10 Thread model: posix gcc version 4.2.1 (Apple Inc. build 5646) Some Googling suggests it might be problem with a missing virtual destructor but irritatingly, none of the threads I've found reach any kind of conclusion. I was hoping the collective genius of the list would be able to inform me whether: 1. This is a known bug; I'm happy to put my debugging hat on and dive in, but not if it's already being investigated by people more qualified to do so. 2. There are any work arounds for the problem. 3. There's no problem at all; I'm being a colossal idiot and have overlooked something obvious. Thanks, vrai.

9 28

Re: [Boost-users] obtaining the day of the week as a string
by Igor R 22 Nov '09

22 Nov '09

> Is there a way of obtaining the day ofthe week as a string in the local > language using boost.datetime? > > If yes what is the best way of doing this? > Look at this example: http://www.boost.org/doc/libs/1_41_0/doc/html/date_time/examples.html#date_…

1 0