September 2011 - Boost - lists.preview.boost.org

[config] explicit call of operator templates / msvc bug workaround
by Stefan Strasser 10 Oct '12

10 Oct '12

Hi, is there already a macro in Boost.Config that allows you to insert the "template" keyword in the following code for those compilers that require it? struct functor{ template<class T> void operator()(); }; template<class F> void g(F f){ f.template operator()<int>(); //(1) works with GCC and comeau f.operator()<int>(); //(2) works with MSVC } this apparently only applies to operator calls. MSVC accepts both (1) and (2) for regular functions. (instantiating the operator template with an object of type T as argument, like mpl::for_each does, is not an option. I can't construct T.) thanks, Stefan

6 10

status of sp_counted_base_solaris.hpp
by Graham Bennett 07 Jun '12

07 Jun '12

Hi all, I'm interested in getting boost::shared_ptr to use atomic ops on Solaris. I notice there is an implementation using the OS-provided atomics in boost/detail/sp_counted_base_solaris.hpp, but it doesn't seem to be included. According to this thread: http://lists.boost.org/Archives/boost/2006/07/107144.php, it seems the reason might be that it wasn't possible to come up with a proper preprocessor condition to include it on the right versions of Solaris. I'm considering modifying my boost installation to include this header on Solaris - does anyone know of any issues with the code, or can they confirm that it has been well tested? Are there any other reasons for its omission? Thanks, Graham -- Graham Bennett

4 7

[contract] syntax redesign
by Lorenzo Caminiti 04 May '12

04 May '12

Hello all, After a long time of inactivity for Boost.Contract, I am considering how to re-design its macro syntax to make it simpler. I can eliminate a large number of extra parenthesis based on a few pp tricks I've learned in processing Boost.Local's `const bind(type)& var`. Plus, I can provide (optional) support for variadics macros so to use comma-separated pp tuplets instead of parenthesized pp sequences on preprocessors with variadics. For example, I _think_ (but I've not implemented this yet) that I can simplify the syntax like this: #include <contract.hpp> #include <vector> #include "myvector/pushable.hpp" // Wrapper class that adds contracts to std::vector. CONTRACT_CLASS( template( typename T ) class (myvector) extends( public pushable<T> ) // Subcontracting. ) { CONTRACT_CLASS_INVARIANT( empty() == (size() == 0) // More invariants here (comma separated)... ) CONTRACT_FUNCTION( public void (push_back)( (const T&) element ) precondition( size() < max_size() // More preconditions here (comma separated)... ) postcondition( auto old_size = CONTRACT_OLDOF(size()), // Old value. size() == old_size + 1 // More postconditions here (comma separated)... ) ) { vector_.push_back(element); // Implementation. } // Rest of the class here (with more contracts if needed)... public: typedef typename std::vector<T>::size_type size_type; size_type size(void) const { return vector_.size(); } size_type max_size(void) const { return vector_.max_size(); } bool empty(void) const { return vector_.empty(); } const T& back(void) const { return vector_.back(); } private: std::vector<T> vector_; }; For a side-by-side comparison with N1962 syntax: http://svn.boost.org/svn/boost/sandbox/contract/libs/contract/doc/html/inde… For a comparison of the entire std::vector (plus Boost.ConceptCheck concepts): http://svn.boost.org/svn/boost/sandbox/contract/libs/contract/doc/html/cont… For a comparison for all Boost.Contract features (plus Boost.ConceptCheck concepts and Boost.Parameter named parameters): http://svn.boost.org/svn/boost/sandbox/contract/libs/contract/doc/html/cont… (Don't pay attention to the rest of the docs that are way out of date.) A few notes: * Extra parenthesis are still needed around class and function names. * Extra parenthesis are still needed around types (like for the parameter `(T) element`) but only if the type is not a C++ keyword (no parenthesis needed aournd int, double, void, etc) or it contains a symbol (parenthesis needed if & or * are used `(int&) x`, etc). * In postconditions, the `auto old_size = CONTRACT_OLDOF(size())` only copies the vector size (and not the entire vector :)) ). Similarly, `auto result = return` can be used to access the return value for non-void functions. (The `auto ...` is parsed by the pp and it's implemented using Boost.Typeof so no C++0x is required.) * Template declaration (but not instantiation) angular parenthesis <> are replaced by round parenthesis (). All of this _should_ be implementable in pure ISO C++. What do you think? Based on the discontent that the previous "parenthesized" syntax generated among Boosters, the key question is: Would you consider this new syntax usable?? Thanks, --Lorenzo

5 45

vsix - template_profiler extension
by Brandon Kohn 27 Feb '12

27 Feb '12

Hello, I've put together a Visual Studio 2010 extension package which wraps the functionality of Steven Watanabe and Domagoj Saric's template_profiler to create a template profile tool. It also adds a command to profile the build time of a project or a single C++ file. Both functions are accessed from the context menu in the solution explorer tree-view. It's still fairly young and could use some kicking about. Are there any VC10 users here that would be interested in giving it a try? Cheers, Brandon

5 16

[move] new rvalue reference emulation
by Jeffrey Lee Hellrung, Jr. 19 Feb '12

19 Feb '12

I've recently discovered (well, probably rediscovered) an alternate rvalue reference emulation, one which allows rvalues ***of any arbitrary (move-emulation-enabled) type*** to be captured in C++03 and identified as rvalues, rather than references-to-const (there is a drawback, though, which I'll get to below; as usual, there's no "free lunch"). This is in contrast to the current move emulation framework Ion has created in Boost.Move which, at best, allows you to capture rvalues of specific types. That is, if you can enumerate the types for which you want to enable rvalue capture, then you're good. For example, if you want to create a generic unary function some_fn, and you know you want to at least capture rvalues of type X, you could do: template< class T > void some_fn(T&); // captures (lvalue) references-to-non-const, explicitly created emulated rvalue references, *and* (lvalue) references-to-(X const) template< class T > typename boost::disable_if< boost::is_same<T,X> >::type void some_fn(T const &); // captures (lvalue) references-to-const *except* (lvalue) references-to-(X const) void some_fn(rv<X>&); // captures rvalues of X as emulated rvalue references However, if Y is any type other than X, the above overloads will capture an rvalue of Y as an (lvalue) reference-to-(Y const). This is one of the shortcomings of using the emulated rvalue reference rv<T>& as a function argument: it can't bind to rvalues in a context where the template parameter T is deduced. Now I don't know if the following proposition is a good idea or not, but I think it is worth considering. Again, we would like the emulated rvalue reference rv<T>& to bind to rvalues, but the problem is the deduction of the T template parameter. Instead of using rv<T>& to capture arbitrary rvalues, I'll use a kind of type-erased emulated rvalue reference, generic_rv<...> (for lack of a better name at the moment), which will just store a void pointer to the original, pre-type-erased object. Of course, we need some way to recover the object back from the void pointer, and that's what the "..." template parameters are for. This is where the (unfortunate) runtime overhead comes in, as some dispatching mechanism (e.g., function pointers) needs to be used to effect the recovery of the original, pre-type-erased object. The important difference between rv<T> and generic_rv<...> is that there's no T in the template parameter list "..." of generic_rv; more to the point, its template parameters are not deduced within the context of binding function arguments of some_fn, hence the compiler will be allowed to look for conversions from arbitrary rvalues to generic_rv<...>. Hopefully that gives enough of an introduction to the idea. Further details are probably best left to a code snippet. Comments? Specifically, should this be considered as an addition to Boost.Move's current emulation framework? Has anyone seen this technique before? ---------------- #include <iostream> #include <boost/type_traits/is_convertible.hpp> #include <boost/utility/enable_if.hpp> // rv<T>& emulates T&& template< class T > class rv : public T { rv(); rv(rv const &); rv& operator=(rv const &); }; template< class T > struct has_move_emulation : boost::is_convertible< T, rv<T>& > { }; // *** NEW EMULATED RVALUE REFERENCE *** // generic_rv< F, Result > also emulates an rvalue reference, but in a kind of // type-erased way. The actual object referred to is referenced by a void // pointer. To be able to cast the void pointer back to a pointer-to-object, we // need to package the void pointer together with some kind of dispatching // mechanism. We use a function pointer to an instantiation of cast_forward<T>. template< class F, class Result = void > struct generic_rv { typedef Result (*p_cast_forward_type)(F, void*); generic_rv(p_cast_forward_type p_cast_forward, void* p) : mp_cast_forward(p_cast_forward), m_p(p) { } Result cast_forward(F f) const { return (*mp_cast_forward)(f, m_p); } template< class T > static Result cast_forward(F f, void* const p) { return f(static_cast< rv<T>& >(*static_cast< T* >(p))); } private: p_cast_forward_type mp_cast_forward; void* m_p; }; // X is just a typical move-emulation-enabled class. struct X { X() { std::cout << "X::X()" << std::endl; } X(X const &) { std::cout << "X::X(X const &)" << std::endl; } X(rv<X>&) { std::cout << "X::X(rv<X>&)" << std::endl; } X& operator=(X) { std::cout << "X::operator=(X)" << std::endl; return *this; } X& operator=(rv<X>&) { std::cout << "X::operator=(rv<X>&)" << std::endl; return *this; } operator rv<X>&() { std::cout << "X::operator rv<X>&()" << std::endl; return *static_cast< rv<X>* >(this); } operator rv<X> const &() const { std::cout << "X::operator rv<X> const &() const" << std::endl; return *static_cast< rv<X> const * >(this); } // *** NEW CONVERSION OPERATOR *** // X provides an implicit conversion to generic_rv. template< class F, class Result > operator generic_rv< F, Result >() { std::cout << "X::operator generic_rv< F, Result >()" << std::endl; return generic_rv< F, Result >( &generic_rv< F, Result >::template cast_forward<X>, static_cast< void* >(this) ); } }; struct some_fn { template< class T > void operator()(T&) const { std::cout << "some_fn::operator()(T&) const" << std::endl; } template< class T > typename boost::disable_if< has_move_emulation<T> >::type operator()(T const &) const { std::cout << "some_fn::operator()(T const &) const" << std::endl; } // Notice that this overload requires template parameter deduction, hence // the compiler cannot apply an implicit conversion to rv<T>& from rvalues // of move-emulation-enabled types. In other words, this overload can only // bind to *explicitly* created emulated rvalue references, not to "real" // rvalues :( template< class T > void operator()(rv<T>&) const { std::cout << "some_fn::operator()(rv<T>&) const" << std::endl; } // *** NEW FUNCTION OVERLOAD TO CAPTURE RVALUES *** // Since this overload requires no template parameter deduction, the // compiler *can* apply an implicit conversion to generic_rv<...> from // rvalues of move-emulation-enabled types. Yes, there will be some runtime // overhead from the function pointer dispatching, but that will often be // preferable to copying x! void operator()(generic_rv< some_fn > const x) const { std::cout << "some_fn::operator()(generic_rv< some_fn >) const" << std::endl; return x.cast_forward(*this); } }; template< class T > T make() { return T(); } int main(int argc, char* argv[]) { int a = 0; int const b = 0; some_fn()(a); // some_fn::operator()(T&) const some_fn()(b); // some_fn::operator()(T const &) const some_fn()(make< int >()); // some_fn::operator()(T const &) const X x; // X::X() X const y; // X::X() some_fn()(x); // some_fn::operator()(T&) const some_fn()(y); // some_fn::operator()(T&) const some_fn()(make<X>()); // X::X() // X::operator generic_rv< F, Result >() // some_fn::operator()(generic_rv< some_fn >) const // some_fn::operator()(rv<T>&) const return 0; } ---------------- - Jeff

10 33

[preprocessor] Warning: Incoming
by Paul Mensonides 18 Jan '12

18 Jan '12

I'm about to commit a relatively significant changeset to the pp-lib. This is the result of work by Edward Diener and I to add limited variadic support to the pp-lib. Edward can comment in more detail regarding the extent of those changes, but they do not add ground-up support. However, there are significant changes to heavily used low-level facilities such as BOOST_PP_TUPLE_ELEM (sometimes even complete implementation replacements). We have put in a great deal of effort to avoid introducing new errors or warnings in either the building of the Boost binaries or in the regression tests. However, we may have missed some issues. Regards, Paul Mensonides

11 42

Is there any interest in command line interpreter class templates?
by alexey.tsoy 29 Dec '11

29 Dec '11

Hello everybody, The file command_line.zip in the containers directory in Boost’s vault contain command line interpreter library. There you can find simple tutorial with examples and draft documentation. The library supports variable amount of option parameters as well as automatic type casting. Note that their types may be diferent. The library doesn't require special linking because it contains headers only. You only need to type in a few lines of code to start using it. Regards Alexey -- View this message in context: http://boost.2283326.n4.nabble.com/Is-there-any-interest-in-command-line-in… Sent from the Boost - Dev mailing list archive at Nabble.com.

4 6

[lockfree] review
by Gordon Woodhull 05 Nov '11

05 Nov '11

Hi Tim, Hopefully I'll have at least a partial review. I only know enough about the subject to be a little dangerous. (One course in concurrent data structures, one successfully implemented lockfree hash table - except for the memory reclamation. Unspecified barriers and ABA problems. :p ) Considering the difficulty of the subject, you seem to have carefully chosen a small set of algorithms where at least there is consensus that the sources you are using have solved the problem correctly. I apologize if this has already been covered, but I don't see anything in the documentation about which platforms and compilers this has been tested on. These algorithms are extremely susceptible to compiler and processor optimizations. Anyway, I thought I would try to run the tests, and maybe try changing the testing methodology a little, but I got a few compiler errors, below. Defining uint fixed most of these, but bench_1.cpp also appears to be out of date? Of course I also have my lack of understanding of boost build library linking to contend with. Hope to try again later. Cheers, Gordon ~/lib/boost_lockfree/libs/lockfree/test ~/src/boost-trunk grodo-mcwoohu:test gordon$ bjam ...found 141 targets... ...updating 21 targets... darwin.compile.c++ ../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi/bench_1.o g++: unrecognized option '-no-cpp-precomp' bench_1.cpp: In function 'void test_fifo_pop()': bench_1.cpp:67:23: error: no matching function for call to 'boost::lockfree::fifo<long int>::dequeue(long int*)' ../../../boost/lockfree/fifo.hpp:214:10: note: candidate is: bool boost::lockfree::detail::fifo<T, freelist_t, Alloc>::dequeue(T&) [with T = long int, freelist_t = boost::lockfree::caching_freelist_t, Alloc = std::allocator<long int>] bench_1.cpp: In function 'void test_fifo_pop_unsafe()': bench_1.cpp:77:30: error: no matching function for call to 'boost::lockfree::fifo<long int>::dequeue_unsafe(long int*)' ../../../boost/lockfree/fifo.hpp:255:10: note: candidate is: bool boost::lockfree::detail::fifo<T, freelist_t, Alloc>::dequeue_unsafe(T&) [with T = long int, freelist_t = boost::lockfree::caching_freelist_t, Alloc = std::allocator<long int>] bench_1.cpp: In function 'void test_stack_pop()': bench_1.cpp:87:19: error: no matching function for call to 'boost::lockfree::stack<long int>::pop(long int*)' ../../../boost/lockfree/stack.hpp:179:10: note: candidate is: bool boost::lockfree::stack<T, freelist_t, Alloc>::pop(T&) [with T = long int, freelist_t = boost::lockfree::caching_freelist_t, Alloc = std::allocator<long int>] bench_1.cpp: In function 'void test_stack_pop_unsafe()': bench_1.cpp:97:26: error: no matching function for call to 'boost::lockfree::stack<long int>::pop_unsafe(long int*)' ../../../boost/lockfree/stack.hpp:207:10: note: candidate is: bool boost::lockfree::stack<T, freelist_t, Alloc>::pop_unsafe(T&) [with T = long int, freelist_t = boost::lockfree::caching_freelist_t, Alloc = std::allocator<long int>] "g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -dynamic -no-cpp-precomp -gdwarf-2 -fexceptions -fPIC -I"../../.." -c -o "../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi/bench_1.o" "bench_1.cpp" ...failed darwin.compile.c++ ../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi/bench_1.o... ...skipped <p../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi>bench_1 for lack of <p../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi>bench_1.o... ...skipped <p../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi>bench_1.run for lack of <p../../../bin.v2/libs/lockfree/test/bench_1.test/darwin-4.5.3/debug/threading-multi>bench_1... darwin.compile.c++ ../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi/fifo_test.o g++: unrecognized option '-no-cpp-precomp' fifo_test.cpp:103:18: error: 'uint' does not name a type fifo_test.cpp: In member function 'void fifo_tester<freelist_t>::add()': fifo_test.cpp:116:14: error: 'uint' was not declared in this scope fifo_test.cpp:116:19: error: expected ';' before 'i' fifo_test.cpp:116:26: error: 'i' was not declared in this scope fifo_test.cpp:116:31: error: 'nodes_per_thread' was not declared in this scope fifo_test.cpp: In member function 'void fifo_tester<freelist_t>::run()': fifo_test.cpp:187:1: error: 'nodes_per_thread' was not declared in this scope "g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -dynamic -no-cpp-precomp -gdwarf-2 -fexceptions -fPIC -I"../../.." -c -o "../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi/fifo_test.o" "fifo_test.cpp" ...failed darwin.compile.c++ ../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi/fifo_test.o... ...skipped <p../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi>fifo_test for lack of <p../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi>fifo_test.o... ...skipped <p../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi>fifo_test.run for lack of <p../../../bin.v2/libs/lockfree/test/fifo_test.test/darwin-4.5.3/debug/threading-multi>fifo_test... darwin.link ../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi/freelist_test ld: library not found for -lboost_thread collect2: ld returned 1 exit status "g++" -o "../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi/freelist_test" "../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi/freelist_test.o" -lboost_thread -lboost_unit_test_framework -g ...failed darwin.link ../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi/freelist_test... ...skipped <p../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi>freelist_test.run for lack of <p../../../bin.v2/libs/lockfree/test/freelist_test.test/darwin-4.5.3/debug/threading-multi>freelist_test... darwin.compile.c++ ../../../bin.v2/libs/lockfree/test/ringbuffer_test.test/darwin-4.5.3/debug/threading-multi/ringbuffer_test.o g++: unrecognized option '-no-cpp-precomp' ringbuffer_test.cpp:209:14: error: 'uint' does not name a type ringbuffer_test.cpp: In member function 'void ringbuffer_tester::add()': ringbuffer_test.cpp:225:14: error: 'uint' was not declared in this scope ringbuffer_test.cpp:225:19: error: expected ';' before 'i' ringbuffer_test.cpp:225:26: error: 'i' was not declared in this scope ringbuffer_test.cpp:225:31: error: 'nodes_per_thread' was not declared in this scope ringbuffer_test.cpp: In member function 'void ringbuffer_tester::run()': ringbuffer_test.cpp:284:1: error: 'nodes_per_thread' was not declared in this scope ringbuffer_test.cpp: In member function 'void ringbuffer_tester_buffering::add()': ringbuffer_test.cpp:315:14: error: 'uint' was not declared in this scope ringbuffer_test.cpp:315:19: error: expected ';' before 'i' ringbuffer_test.cpp:315:26: error: 'i' was not declared in this scope ringbuffer_test.cpp:315:31: error: 'nodes_per_thread' was not declared in this scope ringbuffer_test.cpp: In member function 'void ringbuffer_tester_buffering::run()': ringbuffer_test.cpp:386:1: error: 'nodes_per_thread' was not declared in this scope

8 63

Re: [boost] [lockfree] review
by Boehm, Hans 05 Nov '11

05 Nov '11

[Dave Abrahams pointed me at this thread.] Alexander Terekhov <terekhov <at> web.de> writes: > > Dave Abrahams wrote: > > > > on Thu Aug 25 2011, Alexander Terekhov <terekhov-AT-web.de> wrote: > > > > > Ilya Sokolov wrote: > > >> > > >> > > >> AFAIK, in the example above another thread can't observe reordering > > >> without introducing UB. > > > > > > I meant "observed" as in "check the sequence of writes in the generated > > > code". > > > > Do we care about what the generated code looks like? IIUC what counts > > is the sequence of memory states that another another thread can see. > > > > > But int can be simply changed to relaxed atomic<int> so that another > > > thread can observe reordering without UB. > > > > Sure. > > I meant that for the example upthread, with an int instead of relaxed > atomic<int>, another thread's code triggering UB does not prevent us > from observing reordered writes in the writing thread's code (in > generated code), by looking at it. > > Changing int to relaxed atomic<int> simply removes UB but does not > change the relevant code in the writing thread. > > regards, > alexander. > I agree with Alexander and most of the recent postings. If I write mx.lock(); x = 1; mx.unlock(); my.lock(); y = 1; my.unlock(); There is nothing to prevent the stores to x and y from becoming visible out of order at the hardware level. Enforcing ordering here would add needless and substantial cost. But a data-race-free program cannot observe such reordering. (One reference for that is Sarita Adve's and my PLDI 08 paper. Unofficial TR version at http://www.hpl.hp.com/techreports/2008/HPL-2008- 56.html. I believe Mark Batty et al have a much more complete and formal proof.) If we change x and y to be atomic variables, and the stores to be memory_order_relaxed, then an observer can tell the difference. I.e. if x and y are initially zero, and another thread does r1 = y; r2 = x; (or the acquire versions), it can see r1 = 1 and r2 = 0, since there are no synchronizes-with relationships between threads, and thus nothing to enforce ordering. If you use memory_order_relaxed, or even the acquire/release forms, things do get MUCH more complicated. The equivalent rules for other languages are an interesting story. I believe Posix implentors assume the above rules. I have a PPoPP 2007 paper that explains why that's not exactly what Posix actually states. OpenMP 3.0 has stricter rules which are also not consistently obeyed, and are somewhat fixed in OpenMP 3.1. (See my MSPC 11 paper for some details.) Java has arguably the same rules as C++ in this respect. In my mind, the main way in which sequential consistency differs from the cheaper acquire/release semantics that Alexander wants is that a) Dekker's algorithm (which by itself isn't that practically interesting), plus a number of real lock-free algorithms (which do matter) don't work. b) In my mind, acquire/release is a far more difficult model to explain to non- experts than the model for C++ with only sequentially consistent atomics. You can no longer reason about interleaving thread actions, either for purposes of determining the existence of a data-race, nor when defining what data-race- free programs mean. This was extensively discussed in C++ meetings starting in about 2005. The current solution with default-sequentially-consistent atomics, but explicit support for weaker ordering, was the resulting committee compromise. Hans

6 19

Determining interest in Delegates submission
by Jared McKee 05 Nov '11

05 Nov '11

I would like to determine interest in a possible library submission. I have written some code which converts a member function and object pointer combination to a free function for use in C style callbacks. It has some limitations which I will describe shortly. The basic idea is to be able to convert: R (B::*)(T0,T1,…) and B* to R (*)(T0,T1,…) Here's an example usage: class foo { int x; int test(int v) {return x + v;} }; foo* a = new foo; a->x = 3; int (*f)(int) = delegate(a, &foo::inc); cout << f(4) << endl; // outputs "7" When I first attempted this I used a technique where I allocated some executable memory in the heap and dynamically created a free function to call the member function for the instance. It was fun attempting that but, as you can imagine, as soon as I used the optimizer flags on my compiler I would get segment faults :) My second approach was to use templates and the Boost Preprocessor library to create appropriate static arrays of pointers to free functions matching the required signatures and then hand those pointers out each time the delegate() function is called. The two downsides to this are: you can only call delegate() for a certain function signature a limited number of times, and it adds a lot of potentially unused code to your executable. Here's the code (I've stripped out error checking for the sake of illustration): #include <boost/preprocessor.hpp> #ifndef _delegate_h_ #define _delegate_h_ #define MAX_DELEGATES 256 #define DELEGATE_PARAMS 0 #include "delegate_template.hpp" #undef DELEGATE_PARAMS #define DELEGATE_PARAMS 1 #include "delegate_template.hpp" #undef DELEGATE_PARAMS // … #endif // delegate_template.hpp #define DELEGATE_STORE BOOST_PP_CAT(DELEGATE_STORE,**DELEGATE_PARAMS) template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > struct DELEGATE_STORE { typedef B* objectPointer; typedef R (B::*memberFunction)( BOOST_PP_ENUM_PARAMS(DELEGATE_**PARAMS, T) ); typedef R (*freeFunction)( BOOST_PP_ENUM_PARAMS(DELEGATE_**PARAMS, T) ); static int delegateCount; static int delegateMax; static objectPointer delegateObjectPointer[MAX_**DELEGATES]; static memberFunction delegateMemberFunction[MAX_**DELEGATES]; template <int i> static R func( BOOST_PP_ENUM_BINARY_PARAMS(**DELEGATE_PARAMS, T, v) ) { return (delegateObjectPointer[i]->***delegateMemberFunction[i])( BOOST_PP_ENUM_PARAMS(DELEGATE_**PARAMS, v) ); } static const freeFunction delegateFreeFunction[MAX_**DELEGATES]; static freeFunction getDelegate(objectPointer o, memberFunction m) { int delegateIndex = delegateCount++; delegateObjectPointer[**delegateIndex] = o; delegateMemberFunction[**delegateIndex] = m; return delegateFreeFunction[**delegateIndex]; } }; template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > int DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::delegateCount = 0; template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > int DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::delegateMax = MAX_DELEGATES; template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > typename DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::memberFunction DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) > ::delegateMemberFunction[MAX_**DELEGATES] = {0}; > template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > typename DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::objectPointer DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) > ::delegateObjectPointer[MAX_**DELEGATES] = {0}; > #define FUNC_POINTER(z, n, params) &DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::template func< n >, template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > const typename DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::freeFunction DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) > ::delegateFreeFunction[MAX_**DELEGATES] = { > BOOST_PP_REPEAT(MAX_DELEGATES, FUNC_POINTER,) }; template <class B, class R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, class T) > R (*delegate( B* b, R (B::*f)( BOOST_PP_ENUM_PARAMS(DELEGATE_**PARAMS, T) ) ))( BOOST_PP_ENUM_PARAMS(DELEGATE_**PARAMS, T) ) { return DELEGATE_STORE<B,R BOOST_PP_ENUM_TRAILING_PARAMS(**DELEGATE_PARAMS, T) >::getDelegate(b, f); } // end Thankyou very much if you managed to read all the way to here! By the way, this is my first active participation in any open source project :) Jared

3 5