August 2008 - Boost - lists.preview.boost.org

Removing class type from member function signature. [probably: function_types] [repost]
by Fabio Fracassi 29 Aug '08

29 Aug '08

Hi, I'm sorry, this is a repost of a message I send to boost user, with a somewhat more restrictive subject. I didn't get any input there, and I hope that someone can help me here. I'm currently trying to wrap (or forward) a call to boost::signal<A_func_Signature>::connect(), to - amongst other things - insulate users from using boost::bind. So what I have is something like this: template< typename Signature, class Reciever > void connect( const signal_id& id, const Signature& sig, const Reciever* rcv ) { typedef ???? AdaptedSignature; get_signal< AdaptedSignature > ( id ) -> connect(boost::bind(sig, rcv, _1)); } now this code is supposed to be called like this: connect( some_signal_id, &SomeClass::some_member, someClass_instance ); where SomeClass::some_member has the same function signature as the signal which is associated with some_signal_id. e.g. boost::signal<void (bool)> signal; struct SomeClass { void some_member(bool b); }; So my question is how can I construct the "AdaptedSignature" type from the template deduced "Signature"? From what I gather from the documentation, boost function_types might hold the answer to my question, but I cannot figure out how. So is what I am trying to do possible, and is function_types the right tool? TIA Fabio

1 1

strange memcpy example
by Jens Guether 29 Aug '08

29 Aug '08

Hi all together, I found a very strange example of the memcpy() function on the web page: http://www.boost.org/community/generic_programming.html ---------------------------------------------------------- void* memcpy(void* region1, const void* region2, size_t n) { const char* first = (const char*)region2; const char* last = ((const char*)region2) + n; char* result = (char*)region1; while (first != last) *result++ = *first++; return result; } ---------------------------------------------------------- Perhaps I missed something essential, but usually the memcpy() function returns a pointer to the destination (region1 in this case). The above version of memcpy() returns region1+n. regards Jens ---- Visit us at IFA 2008 / Berlin (Germany), 29 August - 3 September 2008 / Hall 5.3 Booth 17 http://www.hhi.fraunhofer.de/de/veranstaltungen/aktuelle-veranstaltungen/if… IBC 2008 / Amsterdam (NL), 12 - 16 September 2008 / Hall 8 Booth C81 http://www.hhi.fraunhofer.de/index.php?id=2399&L=1 insite 2008 / Johannesburg (South Africa), Sandton Convention Centre / 15 - 17 September 2008 http://www.insitex.co.za

1 0

Memoryleaks in boost::pool_allocator
by Paul Bormans 29 Aug '08

29 Aug '08

Some time ago i had noticed some strange behavior in the boost pool_allocator regarding its memory usage. Now that i've found these great mailinglists ;) i've reproduced the error and with some additional analysis i can post the thing here. My environment: - VS2005 - Windows XP - Boost 1_34_1 with BOOST_ALL_DYN_LINK set (if it makes a difference) - Running release build The test code: { std::pair<int, int> pairStart(m_spLeakDetection->ReportLeaks(_T("boost test 1a"))); { std::vector<int> v; std::pair<int, int> pairInterim2(m_spLeakDetection->ReportLeaks(_T("boost test 1b"))); //0 bytes (only stack memory used) for (int i = 0; i < 1000; ++i) { v.push_back(5); } std::pair<int, int> pairInterim3(m_spLeakDetection->ReportLeaks(_T("boost test 1c"))); //4264 bytes in 1 block v.clear(); } std::pair<int, int> pairInterim4(m_spLeakDetection->ReportLeaks(_T("boost test 1d"))); //0 bytes } { std::pair<int, int> pairStart(m_spLeakDetection->ReportLeaks(_T("boost test 2a"))); { std::vector<int, boost::pool_allocator<int> > v; std::pair<int, int> pairInterim2(m_spLeakDetection->ReportLeaks(_T("boost test 2b"))); //0 bytes (only stack memory used) for (int i = 0; i < 1000; ++i) { v.push_back(5); } std::pair<int, int> pairInterim3(m_spLeakDetection->ReportLeaks(_T("boost test 2c"))); //16312 bytes over 7 blocks; 136, 264, 520, 1032, 2056, 4104, 8200 v.clear(); } std::pair<int, int> pairInterim4(m_spLeakDetection->ReportLeaks(_T("boost test 2d"))); //still 16312 bytes over 7 blocks boost::singleton_pool<boost::pool_allocator_tag, sizeof(int)>::release_memory(); std::pair<int, int> pairEnd(m_spLeakDetection->ReportLeaks(_T("boost test 2e"))); //8112 over 6 blocks; 136, 264, 520, 1032, 2056, 4104 { std::vector<int, boost::pool_allocator<int> > v; for (int i = 0; i < 10; ++i) { v.push_back(5); } std::pair<int, int> pairInterim3(m_spLeakDetection->ReportLeaks(_T("boost test 2f"))); //8112 over 6 blocks; 136, 264, 520, 1032, 2056, 4104 v.clear(); } { std::vector<int, boost::pool_allocator<int> > v; for (int i = 0; i < 1000; ++i) { v.push_back(5); } std::pair<int, int> pairInterim3(m_spLeakDetection->ReportLeaks(_T("boost test 2g"))); //24504 bytes over 7 blocks; 136, 264, 520, 1032, 2056, 4104, 16392 v.clear(); } } >From these tests; - why does the boost::pool_allocator reserve 16312 bytes for 1000 int's while the default stl vector consumes (only) 4264 bytes? - why does a call to release_memory( ) not remove all the memory allocations done by the pool_allocator? But only the last block? - why does the size of each new block gets duplicated in size (even after release_memory)? Does this make sense to any of you? Best regards, Paul _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

2 3

[config] boost::ulong_long_type issue for Borland 5.6.4?
by Gennaro Prota 29 Aug '08

29 Aug '08

John, is BOOST_HAS_LONG_LONG correctly defined for Borland 5.6.4? Many test failures seem related to this (I'm referring to trunk). Thanks, -- Genny

2 1

Re: [boost] String Algorithms Library: Case insensitive compareUTF-8
by Graham 28 Aug '08

28 Aug '08

>> Martin Lutken wrote: >>> Anyone who knows how this could be made possible? >>> I suppose I need a locale facet like the std::ctype, but which works for >>> UTF-8, and not just for ASCII a-z,A-Z. I guess the information in a >table >>> like this (http://www.unicode.org/Public/UNIDATA/CaseFolding.txt) >>> could be used. >>> >> >> This might not work out-of-the-box. StringAlgo lib is designed around the >sequences >> od characters. Since UTF-8 have variable character with encoding, >algotrithms >> in the library would not work as expected. >> >> To make it working, you will need a container with iterators, that will >> iterate over meta-characters, not bytes. >> >>> If it's better/easier just to convert the string to UTF-32 before doing >case >>> insensitive compares, replaces I could live with that. >> >> If you meant UTS-32 and you have a corresponding locale implementation, >than >> this approach is a viable solution. >> >> Sorry, what is UTS-32 ? I tried to Google it: 351 results, with none of >them >> looking like char encoding related. >> >> I found this article on Wikipedia on UTF-32/UCS-4: >> http://en.wikipedia.org/wiki/UTF-32 >> >> Is it not what I need ? >> I suspect that many people must have ran into similar problems. Perhaps >we should >> add a 32 bit string class to Boost. And until I get a better >understanding, I will >> keep calling it UTF-32 :-) >> > > >Sorry, I mixed up it a little. I meant UCS-4 a.k.a fixed-width encoding. I >was not >aware that UTF-32 id de-facto the same. > >Anyway, the statement about usability with StringAlgo still holds. It can >work with >any fixed-size encoding, as long as you have the corresponding locales. > >It could theoretically work also with variable-with characters, provided >you >have a container/localte framework, that allows to operate on >metacharacters. >I'm not sure how efficient it will be, though. > >Best regards, >Pavol. MArtin, The Unicode library I posted in the vault will do what you want for arbitrary characters. It allows you to take two UTF-32 unicode strings and compare them at different comparison levels [e.g. exact, case insensitive etc], and includes the calls necessary for iterating characters [which may be more than a single 4 byte character [e.g. surrogates can be 3 x 'UTF-32' numbers] and then of course the ability to iterate graphemes. It will allow you to do a case insensitive comparison using the full Unicode character library specification. The only thing we did not have time to do was do the string wrapper class. Feel free to work on that ! Thanks. Yours, Graham

1 0

FSM Review
by Phil Endecott 28 Aug '08

28 Aug '08

Rather than repeating myself too much here I refer the review manager to my two previous posts where I implemented a couple of simple problems using this library. My initial expectation was that I would find the library useful for these problems, but it became clear that the resulting code was no better than an ad-hoc implementation without the use of the library. I was hoping that in the time since then the author or some other reviewer would produce an example that would convince me that there are classes of problems where this library is useful, but no such example has appeared. > What is your evaluation of the design? The main mode of use of this library doesn't seem to offer enough beyond what's possible with a more ad-hoc implementation style to be worthwhile. I note that it also has a more declarative mode (with a state transition table) which I did not evaluate properly. > What is your evaluation of the implementation? I have not evaluated the implementation. > What is your evaluation of the documentation? It is sufficient. > What is your evaluation of the potential usefulness of the library? I have not been convinced that this library is useful. > Did you try to use the library? With what compiler? Did you have any > problems? I did not try to compile any of my code. > How much effort did you put into your evaluation? A glance? A quick reading? > In-depth study? A few hours. > Are you knowledgeable about the problem domain? Not especially. Actually in a previous life I knew a lot about implementing and optimising FSMs in hardware, but that doesn't really have a lot to do with most software implementations. > And, finally, every review should answer this question: Do you think the library should be accepted as a Boost library? No, not unless some convincing examples of applications where it is useful can be produced. Regards, Phil.

1 0

Re: [boost] UTF-8 conversion etc. (Sebastian Redl) (Sebastian Redl)
by Graham 28 Aug '08

28 Aug '08

>> As I said to Phil, Rogier and I completed a Unicode character library >> for >> >> Release under boost, but never submitted it to Boost as we had intended >> to >> >> release it with a string library built on it, and never had time to do >> the >> >> second part of the work. >> >Post it, and we'll do the second part. It's open-source. > >Sebastian Sebasian, As requested, I have posted a Unicode character support library in the boost vault as Unicode_lib.zip. Please feel free to use this under the boost license. It would be great if somebody had the time to develop the existing character support to ad the string wrappers that we had intended to - but ran out of time doing. It should be well documented, but if you want any questions answered, please feel free to ask and I'll try and help! Thanks. Yours, Graham

6 7

Another website docpage nit
by Jonathan Biggar 28 Aug '08

28 Aug '08

In the page: http://www.boost.org/doc/tools.html the description of Wave says: A Standards conformant C/C++ preprocessor usable on top of any other compiler. Usable for instance for the debugging of the expansion of macros in your code or as a replacement for your build in preprocessor. ^^^^^^ built- -- Jon Biggar Floorboard Software jon(a)floorboard.com jon(a)biggar.org

2 1

[smart_ptr] release()
by Phil Bouchard 28 Aug '08

28 Aug '08

As it turns out I think a release() member function should be added to the smart pointer. Its purpose is to give away ownership of the object. See the following pseudo-code: template <typename T> class smart_ptr { ... smart_ptr(element_type *); value_type * get(); element_type * release() { element_type * __p = pointer_ref(); pointer_ref() = 0; if (-- counter_ref()) return 0; else return __p; } }; It follows the same idea of its predecessor auto_ptr<> but will return a null pointer if the object is still shared thus not ready to be deleted. This way the allocator will be much easier to support either by calling this function before deallocate(), inside it or by using a specialized wrapper class to handle it: template <typename T> class allocator { ... typedef smart_ptr<T> pointer; void deallocate(pointer & __p) { delete p.release(); } }; -Phil

1 0

Website documentation page nit
by Jonathan Biggar 28 Aug '08

28 Aug '08

The docuemntation page on the website: http://www.boost.org/doc/ only has two subbullets (Boost Jam and Boost Build) under the "Tools" bullet. But if you click on the Tools bullet, you get a longer list: Boost.Build Boost.Jam Regression Inspect BoostBook bcp QuickBook Wave Shouldn't those all appear on the doc page. It took me a while to find the BoostBook documentation because of this. -- Jon Biggar Floorboard Software jon(a)floorboard.com jon(a)biggar.org

1 0