December 2006 - Boost-users

Boost Lamdba: Binding to overloaded operator and templated functions???
by Haroon Khan 02 Dec '06

02 Dec '06

Is it possible to bind to the following functions using boost.lamdba and if so how? class some_class { public: some_class operator % (int x) { return this } template<typename T> void foo(T x){ } }; I use the examples given in online-documentation but it gives me a lot of errors. I'm using VC8 on windows XP. Thanks, Haroon

2 4

newbie: shared_ptr member function customer deleter???
by Travis Parks 02 Dec '06

02 Dec '06

Hello: I just started using some boost today. I am trying to use a non-static member function as the deleter in a shared_ptr. I was looking at the source code and I don't think it is supported. This is what I had: shared_ptr<Environment> environment(Environment::createEnvironment(), Environment::terminateEnvironment); string userName = "monkey_bottoms"; string password = "blahblahblah" shared_ptr<Connection> connection( environment->createConnection( userName, password), &Environment::terminateConnection); This didn't work at all! When I looked at the source I saw why. Note that Environment::terminateConnection is a static method. However, this source works fine for the static Environment::createEnvironment method. I am working in Visual Studio 8 with the OCCI. For the time being, I have made a temporary solution. I am hoping someone can help me find a less customized solution. Plus, I am thinking having a pointer to an object that I am trying to manage kind of defeats the point. Here you go: #include "occi.h" #include <boost/shared_ptr.hpp> #include <string> #include <functional> // retarded temporary solution template <class TCLASS, class TRETURN, class TARG> class call_redirect_t : public std::unary_function<TARG, TRETURN> { TCLASS * instance; // barf I have a pointer to a managed resource! TRETURN (TCLASS::*func) (TARG); public: call_redirect_t(TCLASS * instance, TRETURN (TCLASS::*func) (TARG)) : instance(instance), func(func) { } TRETURN operator()(TARG arg) { return (instance->*func)(arg); } }; // convenience function template <class TCLASS, class TRETURN, class TARG> call_redirect_t<TCLASS, TRETURN, TARG> call_redirect(TCLASS * instance, TRETURN (TCLASS::*func)(TARG)) { return call_redirect_t<TCLASS, TRETURN, TARG>(instance, func); } int main(int argc, char * argv[]) { // lazy using namespace std; using namespace oracle::occi; using namespace boost; shared_ptr<Environment> environment( Environment::createEnvironment(), // gets Environment * Environment::terminateEnvironment); // static method string userName = "flipflop"; string password = "memoryrizor"; shared_ptr<Connection> connection( environment->createConnection(userName, password), // gets Connection * call_redirect( environment.get(), // pull the pointer out; instant poo &Environment::terminateConnection) // non-static method ); } Well, now that I think about it; since the temporary call_redirect will only exist in the scope of the shared_ptr (who handles my resource) it doesn't matter that I let it go, and parameter evaluation order is not an issue. Take a look at the operator() in call_redirect_t. I had to put parenthesis around: (instance->*func) I am think I have forgotten the syntax. Hmph. Weep, I am not perfect. Thanks for any input, folks. Travis Parks

2 2

Boost Graph - Sub Graph Bundled Properties Error
by John Paulus 02 Dec '06

02 Dec '06

Could someone explain with the code below generates an error? The code worked fine with a standard boost graph (ie not a subgraph<>)...... I can create the sub graph, but cannot access the bundled properties data: typedef adjacency_list<listS, listS, undirectedS, VertexData, property<edge_index_t, int, EdgeData> > GraphType; typedef subgraph<GraphType> SubGraphType; SubGraphType SubGraph; SubGraphType::vertex_descriptor V = add_vertex(SubGraph); SubGraph[V].ID=1; ////ERROR!!!!!! Sincerely, John

3 3

Building boost 1.33.1 64-bit using VC-8
by Jason Nye 02 Dec '06

02 Dec '06

Hello all, First of all, I want to say boost has been totally awesome for the programming team in which I work. Kudos to everyone contributing! For skimmers, #3 is the one that has a question related to it. I'm trying to build boost 1.33.1 64-bit using VC-8 (team suite) and I've run into some (minor) issues. They are: 1. The vc-8_0-x86_amd64-tools.jam does not refer to the correct vcvars batch file. It specifies $(VC_ROOT)\bin\x86_amd64\VCVARSamd64.bat but there isn't any such file. The proper file is $(VC_ROOT)\bin\x86_amd64\vcvarsx86_amd64.bat 2. The thread library doesn't link because of the ice_wrapper used in once.cpp (in compare_exchange line 119). Replacing the compare_exchange with an actual call to InterlockedCompareExchange did the trick. I believe this is because it is an intrinsic function. 3. This one is the most confusing to me: the abi prefix headers perform #pragma pack(push, 8). No warnings pop up in 32-bit mode because 8 bytes is the default alignment. However, in 64-bit mode, the default alignment is 16 bytes. Why would boost ever want to enforce a specific packing? Is it safe for me to simply comment these out? I *really* don't like the idea of enforcing an 8 byte alignment when the default is 16 bytes. Any comments here? Thanks! Jason.

3 3

Re: [Boost-users] boost::weak_ptr and boost::intrusive_ptr
by Nat Goodspeed 02 Dec '06

02 Dec '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sebastian Redl > Sent: Friday, December 01, 2006 1:23 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] boost::weak_ptr and boost::intrusive_ptr > > The use case here is this: > > NodePtr base = make_tree(); > NodePtr child = base.get("path/to/child"); > base.reset(); > Now, unless the parent knows that it owns the child and notifies it, the > parent pointer of child is dangling. A weak pointer can detect this > situation. [Nat] Okay. > The reason I see for there being no weak_intrusive_ptr is that it is > quite simply impossible to implement while staying intrusive. Let's > remember what a weak pointer does: after the death of the object it > points to, it knows that it is now invalid. There being no notification > of pointers when other pointers get destructed, there's only one way to > do this: to keep the refcount for the object alive, so that the weak > pointer can check it. > But if the refcount is part of the object, as is the case for > intrusive_ptr, it's impossible to keep the refcount alive while > destroying the object. > > In other words, the appeal of intrusive_ptr is that it doesn't require > any resources outside the object it points to. However, this restriction > makes it impossible to implement an efficient weak pointer. [Nat] Not to be argumentative, but... It seems possible to implement weak_ptr notification by building a list of weak_ptr instances referencing a given object. If we don't want that list to consume additional heap memory, the list could itself be intrusive in the weak_ptr objects. If we want it to be efficient, we build a doubly-linked list. (This may call for a policy-based implementation so the consumer can decide which overhead is least noxious.)

6 5

Boost::wave, sample driver and how to use it
by mosfet 02 Dec '06

02 Dec '06

Hi, I don't undertand how wave works. I am trying to use it on a simple C header file. I have included System and user include paths and some defines but when I start wave sample driver wave.exe, nothing happens. What I want is to convert all the defines into native types, for instance. By the way the config file doesnt' seem to like windows path while it works on the command line. foo.h //-----// VOID foo(BYTE bTest, CHAR szTest); ULONG foo(LPBYTE bTest, CHAR szTest); I would like to get void foo(unsigned char bTest, char szTest); unsigned long foo(BYTE * bTest, char szTest); but maybe I haven't understood what a preprocessor was. If someone could help, it's really urgent.

3 4

interprocess::list and destructors of stored objects
by alexmark 02 Dec '06

02 Dec '06

Hi to all! For the container type boost::interprocess::list<T, A> I need the destructor of stored object T be called when erase() is called for a corresponding list item. By default the T destructor is not called, memory only is deallocated. What is the correct way to force boost::interprocess::list to invoke destructors for contained objects while list item or the container itself is deleted? TYA.

3 2

Re: [Boost-users] boost::weak_ptr and boost::intrusive_ptr
by Nat Goodspeed 01 Dec '06

01 Dec '06

________________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Gottlob Frege Sent: Friday, December 01, 2006 3:25 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] boost::weak_ptr and boost::intrusive_ptr > On 12/1/06, Nat Goodspeed <ngoodspeed at solidworks dot com> wrote: > > It seems possible to implement weak_ptr notification by building a list > > of weak_ptr instances referencing a given object. If we don't want that > > list to consume additional heap memory, the list could itself be > > intrusive in the weak_ptr objects. If we want it to be efficient, we > > build a doubly-linked list. (This may call for a policy-based > > implementation so the consumer can decide which overhead is least > > noxious.) > > Interestingly, since you define the intrusive_ptr_add_ref and > release, you can probably implement the list management without > changing intrusive_ptr. [Nat] That's true, you could. But that would impose the requirement on everyone who wanted to use intrusive_ptr with this new variety of weak_ptr. Perhaps intrusive_ptr could do something implicit before calling intrusive_ptr_release. Maybe that could be implemented as an intrusive_ptr policy. > Would you prefer to have the list of weak_ptrs kept on the > intrusive_ptr (on the actual 'intruded' object, actually), or > have a global map of intrusive_ptr object to weak_ptr list? [Nat] I was thinking of storing the head pointer in the object itself. But it might be more consistent with the present intrusive_ptr implementation to call an intrusive_ptr_get_first_weak_ptr function, requiring the consumer to decide where to put it.

1 0

Re: [Boost-users] Boost::wave, sample driver and how to use it
by Nat Goodspeed 01 Dec '06

01 Dec '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Sebastian Redl > Sent: Friday, December 01, 2006 3:37 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Boost::wave, sample driver and how to use it > > mosfet wrote: > > VOID foo(BYTE bTest, CHAR szTest); > > ULONG foo(LPBYTE bTest, CHAR szTest); > > > If this is Windows, those are typedefs, not macros. The preprocessor > does not affect them. [Nat] <off-topic> Might gccxml be a better answer for the OP? Caveat: AFAIK, gccxml doesn't deal well with MFC headers and doesn't yet have its own equivalents. </off-topic>

1 0

Re: [Boost-users] Boost::wave, sample driver and how to use it
by Nat Goodspeed 01 Dec '06

01 Dec '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of mosfet > Sent: Thursday, November 30, 2006 3:07 PM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] Boost::wave, sample driver and how to use it > > foo.h > //-----// > > VOID foo(BYTE bTest, CHAR szTest); > ULONG foo(LPBYTE bTest, CHAR szTest); > > I would like to get > > void foo(unsigned char bTest, char szTest); > unsigned long foo(BYTE * bTest, char szTest); > > but maybe I haven't understood what a preprocessor was. > If someone could help, it's really urgent. [Nat] You mentioned Windows. Does that mean you're using Microsoft C++? Wouldn't it work to use something like cl /E ... or cl /P ... ?

1 0