- Boost-users - lists.boost.org

Re: [Boost-users] [Serialization] is_serializable trait?
by Roman Perepelitsa 23 Jul '07

23 Jul '07

> Chris Fairles" <chris.fairles(a)gmail.com> on Fri, 20 Jul 2007 18:03:18 -0400 > How can you determine (compile-time) that a class has a serialize function? > > struct foo { > template <typename Archive> > void serialize (Archive &, const unsigned int); > }; > > "&foo::serialize" isn't a type and I can't figure out a way to make it > a type (besides a typedef in the class) and also a constant > expression. > > Tried to play around with this as well but theres no substitution > failure here and static functions can't be constant expressions > (commonnnn c++0x!!) > > struct serialize { > template <typename T> > static void value(T const&) { void(T::*f)(archive::xml_oarchive &, > const unsigned int) = &T::template serialize<archive::xml_oarchive>; > (void)f; } > }; > > Basically, in the code below I want to see "GoodGood" printed out and > the only thing I can change is Q. > > struct A{ > void f(int,int){} > }; > struct B{}; > > template <class T,class F=void> > struct Q{enum{value=false};}; > > template<class T> struct Q<T,void(T::serialize)(int,int)>{enum{value=true};}; > > int main(){ > if(Q<A>::value) { cout << "Good";} > if(!Q<B>::value){ cout << "Good";} > } > > Help/hints/remotely possible? You can try something like this. #include <iostream> #include <iomanip> typedef char yes_type; struct no_type { char padding[8]; }; template <class T> struct HasSerializeMember { typedef void (T::*SerializationFun)(int &, unsigned); template <SerializationFun> struct A {}; template <class Q> static yes_type Test(A<&Q::serialize>*); template <class Q> static no_type Test(...); static const bool value = sizeof(Test<T>(0)) == sizeof(yes_type); }; struct A { template <class Stream> void serialize(Stream & strm, unsigned version) {} }; struct B { }; struct C { void serialize() {} }; int main() { using namespace std; cout << boolalpha << HasSerializeMember<A>::value << '\n' // true << HasSerializeMember<B>::value << '\n' // false << HasSerializeMember<C>::value << endl; // false } Regards, Roman Perepelitsa Deutsche Bank Moscow +7 (495) 660-74-08 --- This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. Please refer to http://www.db.com/en/content/eu_disclosures.htm for additional EU corporate and regulatory disclosures.

3 3

[Boost.Test] warning C4265 when including <boost/test/unit_test.hpp> and friends
by Pierre-Jules Tremblay 23 Jul '07

23 Jul '07

Just trying out Boost 1.34.1RC3, I found this in <boost/test/detail/config.hpp>, lines 65-70: #ifdef __GNUC__ #define BOOST_TEST_PROTECTED_VIRTUAL virtual #else #define BOOST_TEST_PROTECTED_VIRTUAL #endif perhaps the condition should be changed to include MSVC8(SP1), to eliminate multiple copies of this annoying warning: xxxxx.hpp(54) : warning C4265: 'boost::unit_test::test_observer' : class has virtual functions, but destructor is not virtual Cheers, pj

2 2

Re: [Boost-users] Problem with unit test BOOST_AUTO_TEST_MAIN
by Sakharuk, Vlad (GMI Development) 23 Jul '07

23 Jul '07

I have found the problem (and the answer). Someone recompiled TinyXml lib from executable with main function included. And that function was called instead of test one. Gennadiy, Thanks for help. Regards Vlad. -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Gennadiy Rozental Sent: Thursday, July 19, 2007 1:47 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Problem with unit test BOOST_AUTO_TEST_MAIN "Sakharuk, Vlad (GMI Development)" <Vlad_Sakharuk(a)ml.com> wrote in message news:AAA38056FF35DE4CB4A65657784B08FAA9E0A6@MLNYC730MB.amrs.win.ml.com.. . > Yes, if I remove all lib it does run successfully. > I identified that tests run from TinyXML lib (opensource). > > I have to link with other libraries because the code I have to test > depends on them. I am just creating tast case to narrow down the > problem. > > I will look in 1.34.o but probably I could not. > > I will apriceate any solution how to disable particular testing suit > (from TinyXML). I could not physically remove those test from the lib. > > Thanks Vlad This is impossible naturally in current version of Boost.Test. I will look into allowing you to do this in 1.35.0. The only thing you can try to do (in 1.34) is massage master test suite somehow in test module init function. Gennadiy _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users -------------------------------------------------------- This message w/attachments (message) may be privileged, confidential or proprietary, and if you are not an intended recipient, please notify the sender, do not use or share it and delete it. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Merrill Lynch. Subject to applicable law, Merrill Lynch may monitor, review and retain e-communications (EC) traveling through its networks/systems. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or error-free. This message is subject to terms available at the following link: http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you consent to the foregoing. -------------------------------------------------------- "Please consider the environment before printing this email."

1 0

iterator_facade and custom reference types
by Daniel Oberhoff 23 Jul '07

23 Jul '07

hi I am struggling with iterator_face. I have a container of bits with a smart reference type. This is because the exposed type is bool, but there is never a full boolean variable around for an index, just a single bit. So whenever a reference to a bool is requested I want to pass my smart reference that knows how to get and set this bit. But iterator_facade is too smart for this: it wants to make a true reference to a supplied Value type, which results in references to temporaries. Is there any way to tell iterator_facade to use my type as a ref type? Daniel Oberhoff

2 1

shared_ptr + vector + std::sort
by bringiton bringiton 23 Jul '07

23 Jul '07

hi guys, just wondering if you have any shared_ptr tips for sorting vectors [code] std::vector< boost::shared_ptr<int> > v; //.. insert items into vector std::sort(v.begin(), v.end()); [/code] OK this doesn't sort by integer. I was hoping shared_ptrs < operator actually used the < operator of it's template type. however, i guess it's sorting by pointer value... the following works OK: [code] template <class T> struct X { bool operator()(const boost::shared_ptr<T>& a, const boost::shared_ptr<T>& b) { return *a < *b; } }; int main() { std::vector< boost::shared_ptr<int> > v; //.. insert items into vector std::sort(v.begin(), v.end(), X<int>()); return 0; } [/code] Does anyone have a better way of doing this? Maybe boost already defines a function similar to my X function. Maybe there is a way without using a predicate...

2 1

Non-throwing version of filesystem::create_directories?
by Olaf van der Spek 23 Jul '07

23 Jul '07

Hi, Is there a non-throwing version of filesystem::create_directories? Before I used normal C/POSIX APIs that returned an error value and I don't want to wrap each call in a try/catch statement. Olaf

1 0

[fusion/tr1] Include path problem?
by Kimon Hoffmann 23 Jul '07

23 Jul '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, while setting up the include paths as specified in the TR1 documentation I stumbled across a Boost.Fusion related compile error (included indirectly from <xpressive/proto/proto.hpp> in my code). After stepping through the include hierarchy, I was able to extract this minimal example, that demonstrates the problem: - -------- START: Test.cpp -------- // Originally included from <boost/xpressive/proto/proto_fwd.hpp> #include <boost/spirit/fusion/sequence/tuple_forward.hpp> // Originally included from #include <boost/xpressive/proto/op_base.hpp> #include <boost/spirit/fusion/sequence/tuple.hpp> - -------- END: Test.cpp -------- With MSVC++ 8.0 SP1 This file fails to compile with certain include path orderings while it compiles fine with others. The error messages are: ...\boost\spirit\fusion\sequence\make_tuple.hpp(23) : error C2027: use of undefined type 'boost::fusion::tuple<>' ...\boost\spirit\fusion\sequence\make_tuple.hpp(23) : error C2079: 'boost::fusion::make_tuple' uses undefined struct 'boost::fusion::tuple<>' ...\include\boost\spirit\fusion\sequence\make_tuple.hpp(24) : error C2514: 'boost::fusion::tuple<>' : class has no constructors Working include path orderings: 1. Any include path ordering that doesn't contain Boost.TR1 2. VC++ Standard Library, regular Boost, Boost.TR1 3. VC++ Standard Library, Boost.TR1, regular Boost Non-working include path orderings: 1. Boost.TR1, regular Boost, VC++ Standard Library [*] 2. Regular Boost, Boost.TR1, VC++ Standard Library [*] Preferred Boost.TR1 setting Generally speaking I would not consider this a problem, as I can simply order the include directories accordingly, but, as far as I understand it, Boost.TR1 can not be expected to work correctly with any of these settings. Regards, Kimon -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.4 (MingW32) iD8DBQFGn3ceIISObEqaPr4RAiszAJ9ipiY+qK+lLg8uDPnHEZejyA5ShwCfQ0fC 1vOMnumqPnguAgtv7co2N8c= =MIsY -----END PGP SIGNATURE-----

3 12

[IBD] Weekly news: 2007-07-11 to 2007-07-17
by Matias Capeletto 22 Jul '07

22 Jul '07

================================================== Hi Boost, Once a week, we will send you a brief of our work for the Improving Boost Docs project. We need more hands, join us! We are waiting you on the boost-docs list. Project Wiki: http://svn.boost.org/trac/boost/wiki/ImprovingBoostDocs RSS news feed: http://beta.boost.org/development/ibd_news.rss Community Blog: http://improving-boost-docs.blogspot.com ================================================== A bussy week ------------------------------------------------------------------------------------- Sorry for the delay, It was a busy week. I have started to promote IBD in Argentina... we may get some friendly hands from my country in the future. Kudos to all IBDers... the boost-docs list activity is truly impressive. 2007-07-17 ------------------------------------------------------------------------------------- * Proofreading of the IBD Wiki is almost complete Kudos to Jake and Paul! more... http://tinyurl.com/3234qk * Kai joins IBD He will be working at the "Trac Syntax Coloring Project". Thanks you so much for jumping in! more... http://tinyurl.com/3d6mwe 2007-07-16 ------------------------------------------------------------------------------------- * IBD welcomes Kai Brüning It will be great to count with his help. Lets work to make him feel comfortable in the Boost community. more... http://tinyurl.com/2to8bg 2007-07-15 ------------------------------------------------------------------------------------- * Translation project moved to Helping Boost Authors The translation project is now part of the new section "Helping Boost Authors". We hope this section will become the main communication channel between Boost Authors and IBD developers. more... http://svn.boost.org/trac/boost/wiki/HelpingBoostAuthors 2007-07-14 ------------------------------------------------------------------------------------- * Helping Boost Authors We have a new resource at IBD: a section in the project Wiki called "Helping Boost Authors". This section is a bazaar of friendly hands. We are teachers, experienced users, native English speakers and artists. We want to offer our help to Boost authors to improve their documentation. more... http://svn.boost.org/trac/boost/wiki/HelpingBoostAuthors 2007-07-13 ------------------------------------------------------------------------------------- * Documentation Best Practices The first guideline was added to the section "Documentation Best Practices". There are currently more documents to add here, we will try to fill this section with useful information for Boost Authors. more... http://svn.boost.org/trac/boost/wiki/DocumentationBestPractices 2007-07-12 ------------------------------------------------------------------------------------- * Proofreading of IBD Wiki starts Jake and Paul first project will be proofread the IBD Wiki pages. more... http://tinyurl.com/3786jf 2007-07-11 ------------------------------------------------------------------------------------- * Glyn joins the community Blog We are eagerly waiting to read you. more... http://improving-boost-docs.blogspot.com

1 0

[program_options] incorrect documentation
by Bryan Green 22 Jul '07

22 Jul '07

In the documentation for program_options, under the "Library Overview"/"Syntactic Information", the documentation appears to be incorrect. For the following option description: ("verbose", value<string>()->zero_tokens(), "verbosity level") It states that: "the user may either provide a single token for the value, or no token at all." I wrote a test program based on the options given in this section of the manual, and I get different behavior for the 'verbose' option. In fact, the actual behavior is almost intuitive: the option is unusable. No matter what you do, its an error. Seems natural since you're saying the option has a value associated with it, yet no tokens may be given to specify the value. It also turns out that 'multitoken()' does not allow you to pass multiple tokens, but it does raise an exception if you specify anything after the first value, i.e.: ./a.out --email blah@blah me@somewhere exception: in option 'email': multiple values not allowed ./a.out --email blah@blah --help exception: in option 'email': multiple values not allowed Here's my test program: <code> #include <boost/program_options.hpp> #include <string> #include <iostream> using namespace boost::program_options; using namespace std; int main(int argc,char *argv[]) { options_description desc; desc.add_options() ("help", "produce help message") ("compression", value<string>(), "compression level") ("verbose", value<string>()->zero_tokens(), "verbosity level") ("email", value<string>()->multitoken(), "email to send to") ; variables_map vm; try { store(parse_command_line(argc, argv, desc), vm); notify(vm); } catch (exception &e) { cout << "exception: " << e.what() << endl; return -1; } if (vm.count("help")) cout << desc << endl; if (vm.count("compression")) cout << "compression " << vm["compression"].as<string>() << endl; if (vm.count("verbose")) cout << "verbose " << vm["verbose"].as<string>() << endl; if (vm.count("email")) cout << "email " << vm["email"].as<string>() << endl; return 0; } </code> -Bryan

1 0

[boost.python] "Thin wrappers" for member functions
by Marcin Kalicinski 21 Jul '07

21 Jul '07

It's possible to expose free functions through "thin wrappers": void f() { //... } void f_wrapper() { // ... some extra functionality to be done when called from Python only f(); } def("f", f_wrapper); But is it possible to write "thin wrappers" for member functions (without modifying the wrapped class)? struct foo { void f(); }; // f_wrapper defined somehow somewhere class_<foo>("foo") .def("f", f_wrapper???) Thanks, Marcin

2 1