January 2005 - Boost-users - lists.preview.boost.org

RE: [Boost-users] Re: a question on boost::is_function
by Foster, Gareth 06 Jan '05

06 Jan '05

Interesting stuff, thanks all that replied. I think was just trying to related the function type to the syntax seen in other types, the fact is that the function type is a bit special I suppose, so I should really just take it by itself. I think I've seen all the possible uses for it thanks to the replies I've had here anyway. It's always interesting to see syntax from the murkier corners of the C++ repertoire :) Thanks again, Gaz -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of John Maddock Sent: 06 January 2005 10:45 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Re: a question on boost::is_function > Could I do ... > > int (int) * p_my_function_pointer; > > Or some other such madness? I'm reading MSDN about the indirection > operator > at the moment, maybe that will help to clear this up in my head. Can you > recommend a page on "function types". You can certainly do it via a typedef: typedef int mysig(int); mysig* p_my_function; // declares function pointer mysig myproc; // declares a function called myproc with signature mysig. John. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

boost::intrusive_ptr and std::list::iterator
by Ernst Murnleitner 06 Jan '05

06 Jan '05

Hello Readers, I have a question regarding boost::intrusive_ptr and std::list::iterator. I have a std::list<boost::intrusive_ptr<CDataset> > which holds my datasets. My question: Is the increment counter of the intrusive_ptr incremented only by doing the following (assigning a iterator to a dataset)? (I could try myself, but I do not know what happens with other compilers/libraries) typedef List std::list<boost::intrusive_ptr<CDataset> >; List list; ... // List holds datasets ... // Get an iterator: List::iterator iter = list.begin(); // Is the Increment counter of the dataset incremented by this action? // Or does the iterator not necessarily keep a pointer to the dataset directly? It would be nice, if somebody could answer this question. In the meantime I try out, what my implementation makes (gcc, linux). Greetings Ernst

2 2

RE: [Boost-users] Re: a question on boost::is_function
by Foster, Gareth 06 Jan '05

06 Jan '05

> boost::is_function<int(int)>::value > I know it's confusing. Most C++ programmers never see a function type > in the wild. You can use remove_pointer to turn a pointer-to-function > into a function. Could I do ... int (int) * p_my_function_pointer; Or some other such madness? I'm reading MSDN about the indirection operator at the moment, maybe that will help to clear this up in my head. Can you recommend a page on "function types". Gaz -- Dave Abrahams Boost Consulting http://www.boost-consulting.com _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

4 3

Boost Logo Contest
by Joel 06 Jan '05

06 Jan '05

Hello, Boost (www.boost.org) is an open, collaborative effort that includes thousands of programmers from the C++ community at large. Boost is hoping to retire the existing logo (http://www.boost.org/boost.png) and adopt a new one. We are now launching a competition for the creation of a new logo. The competition is open to all designers, both professional and non- professional. The final design must be submitted as a scalable vector format, preferably SVG or EPS. A free, open source, scalable vector graphics editor can be found here: http://inkscape.org/. Alternative bitmap versions (small, medium and large icons, web graphics, etc.) of the logo may also be submitted to augment the main vector format. A JPEG or PNG version, to be placed in a web page, must also be submitted for judging. The Boost community will vote on submitted entries. Voting shall be based on Instant Runoff Voting (http://www.fairvote.org/irv/whatis2.htm), meaning that each individual voter shall choose 5 favorites, in order of preference. In addition to proper acknowledgement, the winning designer will receive: + the Boost Logo Designer T-shirt (printed with the winning logo design), plus, + a signed copy of the MPL book by David Abrahams and Aleksey Gurtovoy, C++ Template Metaprogramming: Concepts, Tools, and Techniques from Boost and Beyond (http://www.awprofessional.com/titles/0321227255) Entries will be accepted on or before February 15, 2005. I will act as the logo contest manager. Please email all entries to me (djowel at gmail dot com). I shall post all submitted entries to Boost's web site. The competition will close on March 1, 2005 and the winning entry will be announced soon thereafter. + The logo must be immediately identifiable with the spirit of Boost. + The logo's concept must be simple and clear and should have relevance to the organization. + The logo should not be too detailed and should work in low resolution. + Any logo submission that is blatantly racist, sexist, obscene, or otherwise discriminatory will be rejected. + The logo must be the original work of the submitter. + The logo must be free of any copyright or other intellectual property claims. + The winning logo shall be distributed under the Boost Software License, Version 1.0. (www.boost.org/LICENSE_1_0.txt). Regards, Joel de Guzman http://www.boost-consulting.com http://spirit.sf.net _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

1 0

RE: [Boost-users] status of boost::signal multi-threading support
by Frank Maddin 05 Jan '05

05 Jan '05

> If you're using a signal from two threads concurrently, you will most > likely see random crashes and other bad behavior. Signals keep a lot of > state, even when they are being invoked, so just about anything can > trip it up. > > > I am curious because, I am wondering in my ignorance, what are the > > components of the boost::signal library and how they can be > > multi-thread > > capable. This ofcourse is answered by reviewing the code. > > Ideally, a thread-safe Signals would allow you to invoke in several > threads at one, disconnect signals in several threads concurrently, > etc. > > Doug Just to make sure I understand, this is only a problem if you are using the *same* signal from two threads concurrently. This is what I've thought, but the phrase *If you're using a signal* makes me want to check. Not trying to nit-pick, just trying to make sure there isn't some situation I don't understand. Thanks, Frank

2 1

bind (excessive?) argument copying
by Trent Hill 05 Jan '05

05 Jan '05

I was recently surprised to find that boost::bind seems to copy arguments a number of times when called. For example, this program: class C { public: C() {} C(const C &c) : data(c.data) { std::cout << "C(const C &c)\n"; } boost::array<char, 4000> data; }; void foo(const C &) {} int main(int a_argc, char **a_argv) { C c; boost::bind(&foo, c); return EXIT_SUCCESS; } results in this output (Visual C++ .NET with /O2): C(const C &c) C(const C &c) C(const C &c) C(const C &c) Four copies - I had hoped to only see a single copy. Logically, there doesn't seem to be a need for more than one copy. Is there an implementation issue that requires this many copies? I was hoping to use bind with some moderately weighted objects, however the number of copies that seem to occur may be prohibitive for me. (I realise that dynamically allocating objects and passing around pointers is an alternative but I'd rather avoid that if possible.) Any thoughts? Thanks, Trent.

4 3

handling exceptions in Boost threads
by Steve McKay 05 Jan '05

05 Jan '05

Is there a generally accepted way to handle exceptions in Boost threads? I have a main thread which spawns 4 children and waits for them all to finish with boost::thread_group::join_all. I want to know if any of them threw an exception (which I would catch within the child thread before returning from that thread). Steve F. McKay Social Security Administration Room 760 Altmeyer 6401 Security Boulevard Baltimore, Maryland 21235 (410)965-3014

1 0

a question on boost::is_function
by Kevin Lee 05 Jan '05

05 Jan '05

should the following code return: true or false? boost::is_function<int (*)(int)>::value if the result is false, why? I don't understand it. if the result is true, the code of boost_1_32_0/libs/type_traits/test/is_function_test.cpp line 42 expect it to be false. could someone help me with this? ----------------------------------------- This message and its attachments may contain privileged and confidential information. If you are not the intended recipient(s), you are prohibited from printing, forwarding, saving or copying this email. If you have received this e-mail in error, please immediately notify the sender and delete this e-mail and its attachments from your computer.

4 3

question about breadth_first_search
by Matthias Linkenheil 05 Jan '05

05 Jan '05

Dear all, i want to use the breadth_first_search algorithm to store the visited nodes e.g. in a list for access in the main function. Has anyone an idea to handle this? In my code i print out the visited nodes. The examples in the Boost User Guide about the visitor concept are sometimes not so easy to understand. ... using namespace boost; template <typename VertexNameMap> class bfs_name_printer:public default_bfs_visitor { public: bfs_name_printer(VertexNameMap n_map) : m_name_map(n_map){} template <typename Vertex, typename Graph> void discover_vertex(Vertex u, const Graph& ) const { std::cout<<get(m_name_map, u) <<" "; } private: VertexNameMap m_name_map; }; ... main(){ ... typedef property_map<Graph, vertex_index_t>::type VertexNameMap; VertexNameMap id = get(vertex_index, g); bfs_name_printer<VertexNameMap> vis(id); breadth_first_search(fg, s, visitor(vis)); ... Regards, Matthias

3 2

read_write_mutex assertion on multiprocessor Wintel
by jkharris7＠comcast.net 05 Jan '05

05 Jan '05

As I reported earlier, read_write_mutex asserts in debug mode (VC71), apparently seeing false wake-ups on a condition variable. I was able to do this with just a simple group of threads locking a RWM in tight loops, with one writer and a few readers. The new information is that I am runing this on a dual-Xeon with hyper-threading enabled (so four apparent CPU's), and I'm thinking that maybe boost libs don't get exercised on this kind of hardware very much. Is is possible that a condition::notify_all() happens on each processor? Has anyone else seen this? For now, I just removed the asserts and rebuilt. Synchronization seems to be unaffected in my real-world stress test of RWM. jh

2 1