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

[signals] 2 (and a half?) bugs: slot leaks, access violation
by Ames Andreas 04 Aug '05

04 Aug '05

Hi, I'm a newbie so please be forbearing when I have made wrong assumptions. I have attached two simple test programs simple.cc and simple1.cc to illustrate the (presumable) bugs. I currently use VC 7.1 and Boost 1.32.0 on WinXP. The access violation in simple.cc (when built with NDEBUG) occurs when I try to emit a signal without slots. Documentation seems to state that in this case the return value of the emission is undefined, which seems to imply that the actual call should work (which it should anyway, IMHO). When I compile with _DEBUG defined I get the following assertion thrown at me: <assertion> Assertion failed: first != last, file boost/boost/last_value.hpp, line 23 </assertion> simple1.cc shows a very basic slot leakage, you will see 9 constructor calls but only 8 destructor calls for the handler. This could be caused by having pos.impl_->group->second.erase(pos.impl_->slot_); commented out in named_slot_map.cpp:230. The half bug I mentioned in the subject is the one which I am least sure about. simple1.cc shows that my callback functor is copied eight times during connect (that obviously implies seven destructor calls). Isn't that a performance issue? TIA, aa -- Andreas Ames | Programmer | Comergo GmbH | Voice: +49 69 7505 3213 | andreas . ames AT comergo . com

2 1

[OT] Information regarding template (instantiation) and compiler/linker performance
by Ames Andreas 04 Aug '05

04 Aug '05

Hi, although not directly boost related, I think the needed expertise is certainly here and the answer to my question might be of common interest. Can someone point me to information resources about differences in handling (removal of duplicated class instantiations, name resolution rules for identifiers in template definitions etc.) of and compiler/linker performance regarding templates? I'm mainly interested in gcc and vc7; I've already read the discussion about frepos in the gcc manual but found nothing about vc7. For both of them but also for other tool chains I would be happy about further and more detailed resources. TIA, aa -- Andreas Ames | Programmer | Comergo GmbH | Voice: +49 69 7505 3213 | andreas . ames AT comergo . com

1 0

Notes on installing Boost with Mac OS X
by Graeme Mathieson 04 Aug '05

04 Aug '05

Hi, I'm pretty new to using Boost. Actually, I haven't really used it at all yet. :-) But I've managed to get it installed on Mac OS X 10.4 and usable with Boost.Build v2 and I thought I'd share my experiences in case it's of any use to others. You can find them here: http://woss.name/2005/08/03/boosting-mac-os-x/ In particular, I'm curious to know if the boost.jam file I created is even vaguely the right way to go about getting Boost to work with Boost.Build or if I've just missed something obvious. Cheers, Graeme. -- Mail: mathie(a)woss.name | Web: http://woss.name/ AIM: Math1e | PGP: 1024D/D72F2737

1 0

optional + serialize
by Jasper, Alexander 03 Aug '05

03 Aug '05

Hi Jeff, All! thanks a lot for the reply. I assume the bug fix is in 1.33? I'm still having issues with boost.optinal and boost.serialize working together (in 1.32). Is for deserialization of boost::optional<T> T required to have default ctor?. This prevents boost::optional<boost::posix_time::time_period> from being serialized. The extraction code fails to compile. I would an ugly way around this (please see my last email). Does anybody have ideas for better solutions? I've attached a modified version of #include <boost/serialization/optional.hpp> to this email, that contains code changes to make to work with types w/o T having a default ctor. But I'm really unhappy with that solution and I don't know what I broke with this change... thanks in advance Alex > -----Ursprüngliche Nachricht----- > Von: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org]Im Auftrag von > Jeff Garland > Gesendet: Freitag, 29. Juli 2005 15:40 > An: boost-users(a)lists.boost.org > Betreff: Re: [Boost-users] questions concerning optional + date_time > +serialize > > > On Fri, 29 Jul 2005 15:21:38 +0200, Jasper, Alexander wrote > >... > > Another thing is that I'm unable to serialize not_a_date_time > > ptimes. The extractions fails here. I'm using the binary > archive. Is > > that supported? The documentation of date time only states > > compatibility with text and XML archive... > > There was a recent (jun 21) bug-fix in date-time that was > preventing correct > serialization of special values -- even with text archives -- > so it's likely > that's what you are seeing. I haven't tried the binary > archive, but there's > no reason it shouldn't work. I believe you should be able to > just pickup the > latest version of gregorian/greg_serialize.hpp and > posix_time/time_serialize.hpp to try it out. > > Jeff > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

2 1

date_time | VC71 | dynamic link
by John Smith 03 Aug '05

03 Aug '05

I have been trying to use Boost Date_Time library with my code. I used boostjam to compile the source; the command used was: bjam -s"TOOLS=vc-7_1" --with-date_time install. The following libraries were built (among others): boost_date_time-vc71-mt-1_32.dll 45,056bytes boost_date_time-vc71-mt-1_32.lib 26,164bytes libboost_date_time-vc71-mt-1_32.lib 521,764bytes boost_date_time-vc71-mt-gd-1_32.dll 77,824bytes boost_date_time-vc71-mt-gd-1_32.lib 26,312bytes libboost_date_time-vc71-mt-gd-1_32.lib 1,380,694bytes My project is setup as follows: Runtime Library: Multi-threaded Debug DLL (/MDd) Additional Include Directories: $(BOOSTHOME)\include\boost-1_32 Additional Library Directories: $(BOOSTHOME)\lib Additional Dependencies: boost_date_time-vc71-mt-gd-1_32.lib Using the above setup I could only link if the static library (libboost_date_time-vc71-mt-gd-1_32.lib) was also available. If not, I kept getting LINK : fatal error LNK1104: cannot open file 'libboost_date_time-vc71-mt-gd-1_32.lib' However, when this file is available, I am able to link and use the DLL (checked DLL usage using depends.exe). Does this make sense? I was under the impression that I should only need the import library. Also, how does it know the name if the library? Thanks.

3 4

problem with float or double serialization with binary archive VC 7.1 boost1.32 and 1.33
by Stephane Martineau 03 Aug '05

03 Aug '05

Hye, this little program fail in serialization of a float with binary archive but success with xml or text archive ?? as i am not an expert, i have now idea why ??? i have the save problem with double and like with float, only depending of the value .... it fails under windows XP, VC++7.1 but success under Linux + gcc thanks for helping , Stéphane. // problemeserialisation.cpp // #include <iostream> #include <fstream> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/binary_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/archive/xml_iarchive.hpp> void test_binary (float & f) { using namespace std ; std::ofstream ofs("val.bin"); boost::archive::binary_oarchive oa(ofs,std::ios::binary); oa<< f ; ofs.close(); std::cout << "Valeur serialisee binary " << f << std::endl ; std::ifstream ifs("val.bin"); boost::archive::binary_iarchive ia(ifs,std::ios::binary); float f1; ia >> f1; if (ifs.fail()) { cout << "badbit " << ( ifs.rdstate( ) & ios::badbit ) << endl; cout << "failbit " << ( ifs.rdstate( ) & ios::failbit ) << endl; cout << "eofbit " << ( ifs.rdstate( ) & ios::eofbit ) << endl; } std::cout << "Valeur deserialisee binary "<< f1 << std::endl ; ifs.close(); return ; } void test_text (float & f) { std::ofstream ofs("val.txt"); boost::archive::text_oarchive oa(ofs); oa << f ; ofs.close(); std::cout << "Valeur serialisee Texte " << f << std::endl ; std::ifstream ifs("val.txt"); boost::archive::text_iarchive ia(ifs); float f2 ; ia >> f2; std::cout << "Valeur deserialisee Texte "<< f2 << std::endl ; ifs.close(); return ; } void test_xml (float & f) { std::ofstream ofs("val.xml"); boost::archive::xml_oarchive oa(ofs); oa << BOOST_SERIALIZATION_NVP(f) ; ofs.close(); std::cout << "Valeur serialisee Xml " << f << std::endl ; std::ifstream ifs("val.xml"); boost::archive::xml_iarchive ia(ifs); float f2 ; ia >> BOOST_SERIALIZATION_NVP(f2); std::cout << "Valeur deserialisee Xml "<< f2 << std::endl ; ifs.close(); return ; } int main (int argc, const char * argv[]) { float f1 = (float)2.411667 ; test_text(f1); test_xml(f1); test_binary(f1); return 0; } output : Valeur serialisee Texte 2.41167 Valeur deserialisee Texte 2.41167 Valeur serialisee Xml 2.41167 Valeur deserialisee Xml 2.41167 Valeur serialisee binary 2.41167 badbit 0 failbit 2 eofbit 1 Valeur deserialisee binary -1.07137e+008

2 2

Re: [Boost-users] [1.33.0] Release candidate #1 is available
by Richard Jennings 03 Aug '05

03 Aug '05

The date_time library fails to compile with Borland C++ 5.6.4 for want of the recursive functions such as localtime_r etc. Borland needs adding to the list of compilers not supporting these functions in date_time/compiler_config.hpp. I built RC1 by replacing Spirit 1.8.x supplied with RC1 with Spirit 1.6 and then did: > bjam "-sTOOLS=borland" --without-wave stage The wave library requries Spirit 1.8.x so I chose to exclude it from the build. Richard -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Doug Gregor Sent: 01 August 2005 18:04 To: boost-users(a)lists.boost.org Subject: [Boost-users] [1.33.0] Release candidate #1 is available Boost 1.33.0 release candidate #1 is now available here: http://www.osl.iu.edu/~dgregor/boost-1.33.0/ There are two sets of files important for this release: - boost_1_33_0_rc1.*: The actual release candidate - boost_1_33_0_(docbook|fo|man).*: The documentation in other formats Please download and unpack the archives, browse the documentation, try the release with your favorite compiler, etc. Thanks! Doug Gregor 1.33.0 Release Manager _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

4 5

Fw: shared_array EXC_BAD_ACCESS
by Christopher Hunt 03 Aug '05

03 Aug '05

Hi Rene/Peter, I'd be happy to test this out in my environment. Question is: where can I get 1.33 from? Is there a single-drop of it available right now or should I CVS it? It might be useful to provide release candidates from the general download area on sourceforge. I understand that we're at release candidate #1 right now. I can't see where this is available (I've searched the archives too). Update: I just did a CVS export with an rtag for 1.33 and I've been getting a "cvs export: [00:44:01] waiting for david_abrahams's lock in /cvsroot/boost/boost/tools/build/test/startup/boost-root" for the last hour. It isn't easy to get hold of 1.33 is it? Cheers, -C On 03/08/2005, at 5:18 AM, boost-users-request(a)lists.boost.org wrote: > Rene Rivera wrote: >> Peter Dimov wrote: >>> I'm forwarding the following post from boost-users >>> >>> http://lists.boost.org/boost-users/2005/08/12938.php >>> >>> since it describes a pretty severe configuration problem on Mac OS X >>> that apparently leads to inconsistent definitions of >>> BOOST_HAS_THREADS and all kinds of interesting ODR violations. >> >> I think that problem is fixed. The original post doesn't mention the >> Boost version so it's hard to be sure. The checks for threading and >> the type of target getting built is considerably more complicated now >> than just the TARGET_CARBON check :-) >> >> It would be awesome if the original poster could test with 1.33 to >> make sure it's fixed. * * * Showpiece your Palm device and instantly access world times using our beautiful 3 dimensional globe http://www.classactionpl.com/

3 2

random number library | variate_generator | dynamic
by John Smith 02 Aug '05

02 Aug '05

Hello: I am trying to use boost::variate_generator so that the generator and distribution can be set dynamically at run time. Currently, I am doing this using the following from http://www.boost.org/libs/random/random_speed.cpp Classes: RandomGenBase, DynamicRandomGenerator, GenericRandomGenerator Function: make_dynamic(URNG & rng, const Dist& d) Then I am able to do: boost::mt19937 rng; iosim::GenericRandomGenerator<int> g_int; g_int.set(iosim::make_dynamic(rng, boost::uniform_int<>(-2,4))); (use g_int()...) g_int.set(iosim::make_dynamic(rng, boost::geometric_distribution<>(0.5))); (use g_int()...) Is there a better way of doing this? Thanks.

1 0

Re: [Boost-users] multi_array_ref timing issue
by Sohail Somani 02 Aug '05

02 Aug '05

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Grant Goodyear > Sent: Tuesday, August 02, 2005 1:12 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] multi_array_ref timing issue > > > Sohail Somani wrote: [Tue Aug 02 2005, 02:28:09PM CDT] > > I think the indexing method you use is the best for your > application. > > > > I hope someone can correct me. > > Thank you very much. I _definitely_ appreciate it! No prob bob. Just helping out a fellow multi_array sucker^H^H^H^H^H^Huser

1 0