March 2006 - Boost-users - lists.preview.boost.org

Re: [Boost-users] [tuples] make_tuple_over a type-sequence
by EMalenfant＠interstarinc.com 29 Mar '06

29 Mar '06

From: Yuval Ronen [mailto:ronen_yuval@yahoo.com] > > Is there a meta-function that creates a tuple type which > contains the types contained in a type-sequence? (I'm talking > about a similar thing to variant's make_variant_over). > We use a generator like this: template<typename Typelist> struct TupleGen : boost::mpl::reverse_fold< Typelist, boost::tuples::null_type, boost::tuples::cons<boost::mpl::_2, boost::mpl::_1> > { };

3 2

[tuples] make_tuple_over a type-sequence
by Yuval Ronen 29 Mar '06

29 Mar '06

Is there a meta-function that creates a tuple type which contains the types contained in a type-sequence? (I'm talking about a similar thing to variant's make_variant_over).

2 3

Re: [Boost-users] Dynamically creating classed by name?
by David Klein 29 Mar '06

29 Mar '06

boost-users-bounces(a)lists.boost.org wrote: > boost-users-bounces(a)lists.boost.org wrote: >> Is there any way that, using Boost, you can dynamically create >> classes by name, similar to the MFC RUNTIME_CLASS macros? I was >> hoping that the boost::serialize libs might be able to help with >> this. >> >> For example, I want to be able to do something like: >> >> CMyClass* pClass = CREATE_CLASS_BY_NAME("CMyClass"); >> >> Can Boost help me to achieve this? >> >> >> >> _______________________________________________ >> Boost-users mailing list >> Boost-users(a)lists.boost.org >> http://lists.boost.org/mailman/listinfo.cgi/boost-users > > hi rob, > > i guess you're looking for some kind of reflection library. > there is none so far in boost, but there were some proposals in the > past. > > see: http://lists.boost.org/Archives/boost/2006/01/99686.php > > HTH dave > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users oh well, i just read about MFC's RUNTIME_CLASS and it seems more like a factory than a whole reflection library. there used to be an implementation of a generic factory class in the old boost vault on sourceforge. but i couldn't find it in the new one. maybe someone can upload it again? sorry for the noise before.

1 0

Re: [Boost-users] Dynamically creating classed by name?
by David Klein 29 Mar '06

29 Mar '06

boost-users-bounces(a)lists.boost.org wrote: > Is there any way that, using Boost, you can dynamically create > classes by name, similar to the MFC RUNTIME_CLASS macros? I was > hoping that the boost::serialize libs might be able to help with this. > > For example, I want to be able to do something like: > > CMyClass* pClass = CREATE_CLASS_BY_NAME("CMyClass"); > > Can Boost help me to achieve this? > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users hi rob, i guess you're looking for some kind of reflection library. there is none so far in boost, but there were some proposals in the past. see: http://lists.boost.org/Archives/boost/2006/01/99686.php HTH dave

1 0

[Test] Incorrect assertion count/Warning when tests do not call one of the Test Tools?
by Kevin Wheatley 29 Mar '06

29 Mar '06

I was just wondering if there is an option to generate an error/warning if a test function called by the test library does no 'asserts' (or any of the BOOST_* test tools). I know there is the option to output the number of assertions and the number of tests, but as we don't always have the one assert per test style of testing here, we can't use that directly to catch tests that do no testing, the detailed ouput level in this case does something like this: Running 5 test cases... unknown location(0): fatal error in "read_one_line_file": std::runtime_error: not implemented Test suite "Master Test Suite" failed with: 8 assertions out of 8 passed 4 test cases out of 5 passed 1 test case out of 5 failed Test case "construct_default_object" passed with: 1 assertion out of 1 passed Test case "no_such_file" passed with: 2 assertions out of 2 passed Test case "blank_filename" passed with: 2 assertions out of 2 passed Test case "read_empty_file" passed with: 2 assertions out of 2 passed Test case "read_one_line_file" aborted with: 1 assertion out of 1 passed The problem comes from (sample): BOOST_AUTO_TEST_CASE( construct_default_object ) { CSVFileReader defaultObject; boost::ignore_unused_variable_warning(defaultObject); } I'd like to say I don't see any asserts here, and would like a warning, or at least to note that I did not call an assert. In fact it looks like BOOST_AUTO_TEST_CASE() is adding an additional test to all of my test cases. Is this by design? Thanks Kevin -- | Kevin Wheatley, Cinesite (Europe) Ltd | Nobody thinks this | | Senior Technology | My employer for certain | | And Network Systems Architect | Not even myself |

2 4

bjam and gprof
by Dean Bushey 29 Mar '06

29 Mar '06

Hi, I'm trying to generate profile information on the test cases. I see that I can use the -sBUILD=profile option to bjam. This places the resulting gmon.out in the current directory, and lots of other information in the <boost>/bin/..... directory. To run gprof I need the executible file, along with the gmon.out file. Is this kept somewhere? Is there a way to tell bjam to save the executible file somewhere for use with gprof? Thanks Dean Dean Bushey --------------------------------- Blab-away for as little as 1¢/min. Make PC-to-Phone Calls using Yahoo! Messenger with Voice.

1 0

Re: [Boost-users] Known issues loading lots of objects withxml_iarchive
by Christopher Gillett 29 Mar '06

29 Mar '06

See I foolishly assumed that all your code was bug free :-) lol...sorry for the nebulous report - here is something a bit more concrete. The scenario as stated is loading a large number of objects that have been previously serialized into a number of files. After around 150 files have been processed I hit an archive exception and stream error. This occurs only when using the XML archive...if I use the binary archive I can process as many files as I'd like (and in fact I've tested many more than I actually need). The approach is "textbook" in that I've seen this same pattern many times in sample code, tutorials, etc. The only difference in the code between binary and xml serialization is my load function: XML: #include <boost/archive/xml_oarchive.hpp> #include <boost/archive/xml_iarchive.hpp> void load( MyObjType &obj, const char *fileName ) { std::ifstream ifs( fileName ); assert( ifs.good() ); boost::archive::xml_iarchive ia( ifs ); ia >> BOOST_SERIALIZATION_NVP( obj ); } Binary: #include <boost/archive/binary_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> void load( MyObjType &obj, const char *fileName ) { std::ifstream ifs( fileName ); assert( ifs.good() ); boost::archive::binary_iarchive ia( ifs ); ia >> obj; } Each class I'm serializing has the usual "boilerplate" in the class definition, for example: #include <boost/serialization/nvp.hpp> #include <boost/serialization/utility.hpp> #include <boost/serialization/list.hpp> #include <boost/serialization/version.hpp> class MyClass : public MyClassBaseObject { friend class boost::serialization::access; friend std::ostream & operator<<(std::ostream &os, const MyClass &MyClassObj); template<class Archive> void serialize(Archive &ar, const unsigned int version) { ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP( MyClassBaseObj ); ar & BOOST_SERIALIZATION_NVP( thing1 ); ar & BOOST_SERIALIZATION_NVP( thing2 ); ar & BOOST_SERIALIZATION_NVP( etc ); } ... }; I initially thought this was a file management problem, but since I can load as many files as I'd like with the binary archive I suspect a bug in the XML archive code. While I can live with binary serialization if necessary, my goal is XML serialization. Any thoughts on this? I can put together a totally contrived example program that will demonstrate the failure if necessary. I'm hoping to hear that it's something stupid on my part. Thanks, Chris Gillett > Date: Mon, 13 Mar 2006 18:40:32 -0800 > From: "Robert Ramey" <ramey(a)rrsd.com> > Subject: Re: [Boost-users] Known issues loading lots of objects > withxml_iarchive? > To: boost-users(a)lists.boost.org > Message-ID: <dv5aeu$l6e$1(a)sea.gmane.org> > > There are no "known" issues. In fact the system has no known bugs > either. LOL - of course the function of this list is to address the > problem. I haven't heard of anyone having this problem. You might > investigate a little by seeing if occurs just in xml or in other > archives > as well. > > Robert Ramey > > Christopher Gillett wrote: >> I am serializing a LOT of objects, one per file. I can create >> XML for >> many thousands of objects with no issue. However, when trying to >> load a few hundred objects (using a separate program, btw) I am >> hitting a boost::archive::archive_exception with a stream error while >> loading around the 161st file, which I believe based on stack traces >> is coming from within >> boost::archive::basic_xml_grammar<char>::parse_end_tag (). >> >> The approach I'm taking is fairly textbook, so I'm curious if there >> are known issues with loading a lot of files, etc. >> >> Any advice appreciate. >> >> Thanks, >> Chris Gillett

2 1

How to handle slightly different command line interface when using boost::program_options
by Peng Yu 29 Mar '06

29 Mar '06

The program_options package ease the way to implement the program cmd interface. But what I found is that as the interface getting bigger the main() program gets longer and longer and is hard to read, because a lot of conditional branch statements have to be used. In general, the declaration of the options are at the beginning of the program and the use of the options are at somewhere afterwards. This cause the program hard to maintain. I'm wondering if there is any OO way to handle this problem. Then the main program could be easily maintained and understood.

1 0

Using multi_array for OpenGL textures
by Matthias Biedermann 28 Mar '06

28 Mar '06

Hi there, I'd like to use 3D volume date stored as multi_array with OpenGL as texture. What is the best way to do this without copying, as the default ordering seems to be incompatible with OpenGL's column-major layout? I've read something about custom storage orderings etc. but my first tries did not work right away... Or is this related to the quite recent bug regarding resizing multi_arrays with non-default ordering (from Mar 6)? Thanks in advance for your help, Matthias

2 3

[mpl] integer for_each
by Yuval Ronen 28 Mar '06

28 Mar '06

Hi. I have a function which is templated with an integer template <int N> void f() { ... } Within this function, I wish to use mpl::for_each to call another integer-templated (say, g()) function, with all numbers from 0 to N-1. How can I achieve this? I thought of making a mpl::vector<int_<0>, int_<1>, ..., int_<N-1> >, and then use mpl::for_each to iterate it, but I don't know how to generate this mpl::vector. How can I do that? Or maybe there's a better way? BTW: I first thought of using mpl::vector_c, but it seems that mpl:for_each doesn't support mpl::vector_c, only mpl::vector. Is this true? Thanks, Yuval

2 2