- Boost-users - lists.boost.org

errors serializing pointers on Fedora Core 5
by Dan Bikel 25 Aug '06

25 Aug '06

The following tiny program generates two compiler errors on FC 5 (which comes with g++ 4.1.1 and boost 1.33.1): #include <iostream> #include <fstream> #include <string> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/string.hpp> using namespace std; using namespace boost; int main(int argc, char **argv) { string * pFoo = new string("foo"); ofstream ofs("strfile"); { archive::text_oarchive oa(ofs); oa << pFoo; } } Error No. 1: /usr/include/boost/archive/detail/oserializer.hpp:567: error: invalid application of -1òøsizeofòù to incomplete type òøboost::STATIC_ASSERTION_FAILURE<false> Error No. 2: /usr/include/boost/serialization/access.hpp:109: error: -1òøstruct std::basic_string<char, std::char_traits<char>, std::allocator<char> >òù has no member named òøserializeòù Error No. 1 can be fixed by making pFoo a const pointer type, i.e., by changing the declaration/assignment statement to be string *const pFoo = new string("foo"); I cannot figure out why such a change is necessary, as the serialization library documentation does not indicate that it is necessary to make pointers const in order to serialize them (as is borne out by the demo.cpp code). Also, I cannot figure out how to fix Error No. 2. Any help would be much appreciated! Thanks, Dan.

2 4

[MPL/Fusion?]Call a 0-args template function on members of a sequence
by Edd Dawson 24 Aug '06

24 Aug '06

Hello all, I have the following: template<typename T> void f() { /* ... */ } typedef boost::mpl::vector<int, long, double> my_vec; I'm looking for a way to call f<T>() for all types T in my_vec. I initially hoped that either an mpl algorithm or fusion::for_each might be able to accomplish this for me, but after reading through their docs this doesn't appear to be the case. Am I correct in this conclusion? If not, I'd really appreciate someone giving me a quick pointer as to what I need to do. Currently I'm using the specialization/static-recursion trick. Something like: template<typename types, int idx, int size> struct helper { static void call_f() { using namespace boost::mpl; f< at<types, int_<idx>::type >::type >(); helper<types, idx + 1, size>::call_f(); // recurse } }; template<typename types, int size> struct helper<types, size, size> { static void call_f() { } // end recursion }; void call_f_with_each_type() { enum { num_types = size<my_vec>::type::value }; helper<my_vec, 0, num_types>::call_f(); } It works fine, but I'm wondering if there's a more succinct and readable way. Kind regards, Edd

2 5

No thread safe libraries built for Darwin when using release 1.33.1?
by Rush Manbert 24 Aug '06

24 Aug '06

Hi All, This has been discussed before, but I can't find an answer, so here is another bjam build question. When I build the Boost libraries using the VC8 toolset on the PC, I create thread safe (-mt) versions of the libraries. When I build the Boost libraries on Darwin, there are never any thread safe library versions built, even if I specify "-sBUILD=<threading>multi" on the bjam command line. I have found threads that discuss use of the thread safe version of the regex lib, so I know that it exists and gets built on some platform(s). Why are there no thread safe libraries built for the Darwin toolset? On the surface, this seems like a bug, but I figure there must be some reasonable explanation. (Oh, please, let there be a reasonable explanation! :-) Regards, Rush

3 4

boost.test and Cruise Control
by Matt Schuckmann 24 Aug '06

24 Aug '06

Has anybody out there integrated tests built on boost.test with Cruise Control or Cruise Control .NET. We've started creating Unit Tests and we are just starting to use Cruise Control for continuous build integration and I'd like to have Cruise Control run and report on the unit tests. A quick look around Cruise Control only shows integration with NUnit. I don't know how tough or easy it is to make something workable with Boost.Test Sorry if this has already been addressed I did try to search for it but came with nothing in my new reader and the web interface is down. Thanks Matt S.

1 0

shared_ptr implementation question
by Ovanes Markarian 24 Aug '06

24 Aug '06

I have a question regarding shared_ptr implementation: currently shared_ptr is implementated so, that it contains: T * px; // contained pointer detail::shared_count pn; // reference counter I am interested to know philosophical thoughts why this kind of decision was made to implement the shared_ptr class this way? I assumed to save space which is consumed by such a pointer one could move T pointer (T*) to shared_count class. All shared_ptr classes would contain only one instance or (may be better pointer to counter): detail::shared_count* pn; // reference counter and the approach to retrieve pointer would be: pn->px //since px would be moved to pn This would reduce the size of shared_ptr<T> to sizeof(detail::shared_count*) which is at least by sizeof(detail::shared_count) smaller of may be more if alignment has to used. Are there any arguments which prevent shared_ptr from such implementation scenario? With Kind Regards, Ovanes Markarian

3 2

[Boost-Threads] Assertion during a second call to thread::join()
by Aubrey, Jason 24 Aug '06

24 Aug '06

Hi, Does anyone know the reason for the behavior in the following example program? This example attempts to simulate the scenario where two threads wait on a third thread. However, I realize that the following is actually one thread waiting on a second thread twice. Regards, Jason My Environment: MSVC 8.0, boost v1.33.1 #include <boost/thread/thread.hpp> #include <boost/thread/xtime.hpp> void f() { } void main() { boost::thread t(f); t.join(); t.join(); // Causes an assertion }

3 4

Debian-Boost linking errors in Cleversafe project
by Matt England 24 Aug '06

24 Aug '06

(This msg is getting cross-posted to the Boost and Debian users' lists. I hope that's ok.) Summary: Our builds are experiencing link-time problems with Boost libraries only on Debian systems. Details below. Does anyone have any suggestions on the root causes and fixes for this? If not, possibly other things we can do to better determine what's causing these symptoms? Details: We experience link-time errors on some Debian systems when linking against boost libraries that are built from the same automated source procedure (as per <http://svn.cleversafe.org/dscore/trunk/>) as the binary that links against them. Since our software and the Boost software/libs are all "built from source" using the same compiler/tools, we hope that we are removing any compatibility problems like C++ ABI incompatibilities between g++ revs (<http://autopackage.org/apbuild-apgcc.php#cxx>), etc. Details here: http://www.cleversafe.org/bugs/show_bug.cgi?id=239 Would anyone have any suggestions/pointers for help? For details on how to build our stuff from source, see: http://www.cleversafe.org/wiki/Project_Build_System#Building_from_source Note that we don't experience these errors when building on other systems, including the CentOS system detailed here: http://www.cleversafe.org/bugs/show_bug.cgi?id=239#c1 -Matt Dispersed Storage: http://www.cleversafe.org

2 1

Preprocessor library: inrement a macro
by Arno Schödl 24 Aug '06

24 Aug '06

Hello, I simply want to increment a macro, like that: #define VARIABLE 7 FunctionA( VARIABLE ); // VARIABLE expands to 7 #define VARIABLE BOOST_PP_INC(VARIABLE) FunctionB( VARIABLE ); // VARIABLE expands to 8 #define VARIABLE BOOST_PP_INC(VARIABLE) FunctionC( VARIABLE ); // VARIABLE expands to 9 #define VARIABLE BOOST_PP_INC(VARIABLE) ... Of course, as I have written it, it does not work. What's the right way? Thanks a lot! Arno P.S. I don't want to iterate. FunctionA/B/C are all different constructs. I simply want to avoid forgetting to adjust all numbers when making a change to the chain of calls.

2 1

[program_options] MS command line option
by Simon Reye 23 Aug '06

23 Aug '06

Is there any way to support the Microsoft 'standard' for command line options i.e '/' instead of '-'. We've got legacy code that requires command line options to be in this format while at the same time attempting to implement boost in a lot of our new code. Simon

2 3

calling map::insert through bind.
by Minkoo Seo 23 Aug '06

23 Aug '06

Hi list. I've got a question on calling map::insert using bind. What I'd like to do is to copy all members in m into m2: #include <iostream> #include <algorithm> #include <map> #include <utility> #include <boost/lambda/algorithm.hpp> #include <boost/lambda/lambda.hpp> #include <boost/lambda/bind.hpp> using namespace std; using namespace boost::lambda; int main() { map<string, int> m; map<string, int> m2; m["a"] = 1; m["b"] = 2; m2["c"] = 3; for_each(m.begin(), m.end(), bind(&map<string, int>::insert, &m2, bind(&make_pair, bind(&map<string, int>::value_type::first, _1), bind(&map<string, int>::value_type::second, _1)))); return EXIT_SUCCESS; } Unfortunately, the above code fails to compile and I have no idea why it fails. I'm using g++ and the errors are as follows: g++ map_test.cpp map_test.cpp: In function 'int main()': map_test.cpp:26: error: no matching function for call to 'bind(<unknown type>, const boost::lambda::lambda_functor<boost::lambda::lambda_functor_base<boost::lambda::action<2, boost::lambda::function_action<2, boost::lambda::detail::unspecified> >, boost::tuples::tuple<const std::basic_string<char, std::char_traits<char>, std::allocator<char> > std::pair<const std::basic_string<char, std::char_traits<char>, std::allocator<char> >, int>::* const, const boost::lambda::lambda_functor<boost::lambda::placeholder<1> >, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type> > >, const boost::lambda::lambda_functor<boost::lambda::lambda_functor_base<boost::lambda::action<2, boost::lambda::function_action<2, boost::lambda::detail::unspecified> >, boost::tuples::tuple<int std::pair<const std::basic_string<char, std::char_traits<char>, std::allocator<char> >, int>::* const, const boost::lambda::lambda_functor<boost::lambda::placeholder<1> >, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type> > >)' Any idea on this? Sincerely, Minkoo Seo -- View this message in context: http://www.nabble.com/calling-map%3A%3Ainsert-through-bind.-tf2150829.html#… Sent from the Boost - Users forum at Nabble.com.

3 2