March 2008 - Boost-users - lists.boost.org

[1.35.0] Release Candidate 2 available
by Beman Dawes 21 Mar '08

21 Mar '08

1.35.0 Release Candidate 2 (RC2) is available at http://boost.cowic.de/rc/ RC2 fixes all of the serious issues reported for RC1. Note that links to the web site from the RC appear to be broken because the restructured web site will not go live until the actual release ships. In the meantime, the new website content is available at beta.boost.org. Please take the time to download RC2, and try to install it on your system. Kick the tires, and report any problems or issues you discover. Report by posting to either boost(a)lists.boost.org or boost-users(a)lists.boost.org. For a list of known issues, see http://svn.boost.org/trac/boost/wiki/MergeOneDotThreeFiveDotZero. Thanks, --Beman

1 0

boost::asio Visual Studio 8 - fatal error C1189: #error : WinSock.h has already been included
by Julian Cain 21 Mar '08

21 Mar '08

I have upgraded to VS 8.0 Standard and when compiling ASIO errors on the following: 1>Unknown compiler version - please run the configure tests and report the results 1>C:\tor\trunk\lib\boost\boost/asio/detail/socket_types.hpp(27) : fatal error C1189: #error : WinSock.h has already been included Is this a known problem? thanks jc

4 4

[boost-users][test] Problems calling ::system() (... child has exited ...)
by Darren Garvey 21 Mar '08

21 Mar '08

Hello all, I'm trying to use system calls in a test that uses Boost.Test. I've been trying to use Julio Merino Vidal's Boost.Process sandbox code (which I have working most of the time), but I've also tried the ::system() call too. The following super-simple test case causes the prepended error... The fact that I can't get this working makes me wonder if I'm doing something forbidden? Platform: ----------------- Using built-in specs. Target: i486-linux-gnu Configured with: ../src/configure -v --enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.1.3 --program-suffix=-4.1--enable-__cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable-mpfr --enable-checking=release i486-linux-gnu Thread model: posix gcc version 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2) ----------------- Error: ----------------- Running 1 test case... unknown location(0): fatal error in "wget_test_1": child has exited; pid: 1000; uid: 11541; exit value: 0 *** 1 failure detected in test suite "wget_tests" EXIT STATUS: 201 ----------------- Program: ----------------- #define BOOST_TEST_MODULE broken #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_CASE( broken_test ) { ::system("ls"); } ----------------- Regards, Darren

1 0

[InterProcess] Compatibility with Boost multi_array
by sankar lingam 20 Mar '08

20 Mar '08

Can I create Boost multi_array in shared memory using Boost.InterProcess..? Can some one explain as to how this can be done.? I tried providing the Boost InterProcess Allocator to the Boost multi_array but it gives the expected compiler error of casting an offset<ptr> to T* , when the multi_array tries to convert return type of the allocator allocate function(which is an offset<ptr>) to T*. Can this problem be overcome by writing a custom stl compatible allocator wrapper over the interprocess allocator? Regards, Sankar

2 6

[addressof] Can't obtain the address of an instance of an anonymous struct
by Jean-Francois Bastien 20 Mar '08

20 Mar '08

We're running into an issue with addressof as described at: https://savannah.cern.ch/bugs/?33071 Basically this fails under GCC 3.4.4 as well as GCC 4.3: struct // NoName { } noName; void * fail() { return addressof(noName); } The error message is "no matching function for call to `addressof(<anonymous struct>&)'". MSVC 2005 SP1 compiles and runs this with no problem. The problem seems to be that GCC can't deduce the templated type because the struct is anonymous. I'm not sure whether this is because it's non-standard behavior that MSVC accepts and GCC doesn't or because GCC has a bug. Either way it would be nice if addressof had a workaround for the "broken" GCC versions, but I'm not sure what the best workaround would be. Axel suggested in the above link to C-style cast to char &, but this causes a problem in the (unlikely) event where operator char &() is declared in the anonymous struct. We could use template magic to see if operator char &() is declared, but I'm not very comfortable with such a huge workaround. If the discussion points towards a bug in GCC I'll file a bug report with them, but either way addressof isn't working with GCC 3.4.4 and 4.3 (at least), and it would be nice of Boost to offer a workaround.

1 0

Re: [Boost-users] Boost-users Digest, Vol 1577, Issue 2
by Abir Basak 20 Mar '08

20 Mar '08

> > I don't have access to that compiler. I tried running similar code in > Visual C++ 6.5 and Visual C++ express 8, and they both worked. I can > try adding something to the unit tests but that will take me a while > to work through. At the end of this email is the code I tested with, > can you try it on your compiler? If it works, I'm doing something > different to you, do you have any idea what? > > thanks, > > Daniel > > #include <boost/unordered_map.hpp> > #include <boost/tuple/tuple.hpp> > #include <iostream> > > int main() { > typedef boost::unordered_multimap<char, int> Mymap; > Mymap c1; > > c1.insert(Mymap::value_type('a', 1)); > c1.insert(Mymap::value_type('a', 2)); > c1.insert(Mymap::value_type('b', 3)); > c1.insert(Mymap::value_type('c', 4)); > > Mymap::iterator it,end; > boost::tie(it,end) = c1.equal_range('a'); > for( ; it!=end; ++it) { > std::cout<<(*it).second<<" "; > } > } Yes, MSVC 9 (i.e 2008) runs it properly, but MSVC 7.1 (2003 . NET) fails to compile it. It looks like, for some reason node_ = data::next_group(node_); from increment_group() tries to construct data, which is a typedef to BOOST_UNORDERED_TABLE_DATA even while using static function for that. just changing the statement to node_ = next_group(node_); (hash_table_impl.hpp cvs line no 301) works. so it seems, msvc 7. either doesn't support calling static function by class name typedef, or the data typedef is confused with some other (like BOOST_UNORDERED_TABLE) note node_ = BOOST_UNORDERED_TABLE_DATA::next_group(node_); also works fine. i tried to produce some simple test case for this, like class outer{ public: typedef outer outer_t; inline static void func(){} class inner{ void func1(){ outer_t::func();// func() this works;or outer::func() works } }; }; this doesn't work in msvc 7.1 but works in 9.0 . i am not sure if it is valid in standard though. The error is, E:\lib\boost\trunk\boost\unordered\detail\hash_table_impl.hpp(301): error C2039: 'hash_table_data_equivalent_keys<std::allocator<std::pair<char const ,int> > >' : is not a member of 'boost::unordered_detail::hash_table_data_equivalent_keys<Alloc>::iterator_base' with [ Alloc=boost::unordered_detail::hash_types_equivalent_keys<std::pair<const char,int>,char,boost::hash<char>,std::equal_to<char>,std::allocator<std::pair<const char,int>>>::value_allocator ] thanks abir

2 1

Re: [Boost-users] unordered_set insert returns const_iterator ?
by Abir Basak 20 Mar '08

20 Mar '08

> > I don't have access to that compiler. I tried running similar code in > Visual C++ 6.5 and Visual C++ express 8, and they both worked. I can > try adding something to the unit tests but that will take me a while > to work through. At the end of this email is the code I tested with, > can you try it on your compiler? If it works, I'm doing something > different to you, do you have any idea what? > > thanks, > > Daniel > > #include <boost/unordered_map.hpp> > #include <boost/tuple/tuple.hpp> > #include <iostream> > > int main() { > typedef boost::unordered_multimap<char, int> Mymap; > Mymap c1; > > c1.insert(Mymap::value_type('a', 1)); > c1.insert(Mymap::value_type('a', 2)); > c1.insert(Mymap::value_type('b', 3)); > c1.insert(Mymap::value_type('c', 4)); > > Mymap::iterator it,end; > boost::tie(it,end) = c1.equal_range('a'); > for( ; it!=end; ++it) { > std::cout<<(*it).second<<" "; > } > } Yes, MSVC 9 (i.e 2008) runs it properly, but MSVC 7.1 (2003 . NET) fails to compile it. It looks like, for some reason node_ = data::next_group(node_); from increment_group() tries to construct data, which is a typedef to BOOST_UNORDERED_TABLE_DATA even while using static function for that. just changing the statement to node_ = next_group(node_); (hash_table_impl.hpp cvs line no 301) works. so it seems, msvc 7. either doesn't support calling static function by class name typedef, or the data typedef is confused with some other (like BOOST_UNORDERED_TABLE) note node_ = BOOST_UNORDERED_TABLE_DATA::next_group(node_); also works fine. i tried to produce some simple test case for this, like class outer{ public: typedef outer outer_t; inline static void func(){} class inner{ void func1(){ outer_t::func();// func() this works;or outer::func() works } }; }; this doesn't work in msvc 7.1 but works in 9.0 . i am not sure if it is valid in standard though. The error is, E:\lib\boost\trunk\boost\unordered\detail\hash_table_impl.hpp(301): error C2039: 'hash_table_data_equivalent_keys<std::allocator<std::pair<char const ,int> > >' : is not a member of 'boost::unordered_detail::hash_table_data_equivalent_keys<Alloc>::iterator_base' with [ Alloc=boost::unordered_detail::hash_types_equivalent_keys<std::pair<const char,int>,char,boost::hash<char>,std::equal_to<char>,std::allocator<std::pair<const char,int>>>::value_allocator ] thanks (sorry for double posting ... last message subject line was wrong) abir

1 0

unordered_set insert returns const_iterator ?
by Abir Basak 20 Mar '08

20 Mar '08

Hi while i am using boost unordered_set , and insert a new key value pair, it returns the iterator as a const one. In fact when i looked at the code, the iterator is a typedef to implementation::const_iterator ... so there is no way to gen an mutable iterator from the set, including begin() and end() is it a standard behavior or i am missing something ? thanks abir

3 3

ignoring function arguments with Boost.Bind?
by Christoph Duelli 20 Mar '08

20 Mar '08

Please consider the following scenario: class Caller { public: Caller(...); typedef boost::function<void (int)> MyCallback; MyCallback callback; }; class Callee { public: Callee(); void do_it() {...}; }; Caller caller; Callee callee; // does not work as do_it does not match MyCallback caller.callback = boost::bind(&Callee::do_it, &callee, ...); The function do_it() is not interested in the arguments passed to the callback. I obviously could make the function take these arguments and just ignore them... but I was wondering: is it possible to leave do_it() as it is and work some magic with Boost.Bind that 'drops' the unwanted argument? Basically create an anonymous function object that calls callee without the argument it gets passed itself. (Similarly to Qt signals/slots: the slot may have fewer arguments than the signal.) Thank you and best regards Christoph

2 1

Re: [Boost-users] boost serialization & module boundaries
by Robert Ramey 20 Mar '08

20 Mar '08

I'm not sure from the title of the post what the question is. Never-the-less, I'll provide an answer. Up until now, there have problems when serialization code was spread accross and/or repeated in distinct runtime modules. Also, there was a problem when serialization was invoked simultaneously from different threads. Both these problems are related to the fact that the library has to keep a table of which types have the serialization code loaded. This has now been addressed in the recently checked in version in the trunk. Serialization code can be in dlls. Also code can be repeated accross DLLs. Personally, I would prefer to organize my code so that this doesn't happen but apparently some people found this inconvenient. I have tested this by hand with my VC 7.1 system. The tests consist of making separate DLLS and running the executable all from the same directory. I have been unable to figure out how to do such a thing with bjam so as far as linux is concerned its untested. Robert Ramey

1 0