- Boost-users - lists.boost.org

Subject: Re: [Containers] Marshalling deques
by Brian Wood 26 Aug '11

26 Aug '11

Michael Goldshteyn: > I think that having a ?configurable? deque class template > is one of the best things that could be offered by a boost > Containers library. I have always wondered why noone > wrote a version of the deque class template that allowed > the configuration of such things as the page / block size, > rather than have it be determined by a compile time > formula based on the size of the data type being stored. > It would seem to me that the page / block size is one of > the more critical pieces affecting the size / speed > tradeoffs in a deque?s performance when there are many > insertions taking place. The thread I mentioned has a link to a segmented_array container which may be of interest. I agree that what you suggest would be an improvement. Shalom, Brian Wood Ebenezer Enterprises http://webEbenezer.net <http://webebenezer.net/>

1 0

[filesystem] posix path on cygwin disappeared - how to make it work?
by Frédéric Bron 25 Aug '11

25 Aug '11

How can I use the posix path on cygwin? I am trying to upgrade from 1.43 but macros BOOST_POSIX_API and BOOST_POSIX_PATH are now deprecated. Why forcing WINDOWS_API while it worked well before? Regards, Frédéric

1 1

boost::mutex::scoped_lock fails with res = 22 (EINVAL)
by Andrej van der Zee 25 Aug '11

25 Aug '11

Hi, I am trying to understand why the acquiring of a lock fails on error code 22 (EINVAL). After Googling I find one of two causes: 1) The mutex was created with the protocol attribute having the value PTHREAD_PRIO_PROTECT and the calling thread's priority is higher than the mutex's current priority ceiling. The mutex is created by the main-thread and the lock is acquired by a thread from a boost::thread_group. But in many cases the lock is acquired without error, so I guess this is not the cause. 2) The value specified by mutex does not refer to an initialized mutex object. The mutex is default constructed by boost::mutex's constructor so I expect it to be initialized properly. Both causes *seem* not to apply to my case. What am I missing here? I am using boost 1.46.1 on Ubuntu amd64. Cheers, Andrej

1 0

Re: [Boost-users] [type traits] Extension to operator operatordetection -> choose your preferred naming
by Alex Perry 25 Aug '11

25 Aug '11

on Tue Aug 23 2011, Fr?d?ric Bron <frederic.bron-AT-m4x.org> wrote: > We have come to three proposed lists of names (A, B and C): > https://svn.boost.org/trac/boost/wiki/GuideLines/Naming/OperatorTraitN > ames I have included the corresponding names of standard keywords, > standard function objects from <functional>, boost Proto and boost > Operators. Another vote for C (though having done nothing on this not sure my vote really counts for anything) - only issue I have with C is :- has_negate surely has_unary_minus needs to be there as an alias for this at the very least. Alex

1 0

[Boost.Test]feature request to list testcase tree
by Tan, Tom (Shanghai) 25 Aug '11

25 Aug '11

I am a fan of Boost.Test. As a habbit, I organize my code this way: // file: Foo.h Class foo { // implementations }; #ifdef ENABLE_TEST namespace test_cases { BOOST_AUTO_TEST_CASE(foo_tests) { // test predicates of class foo } } //namespace test_cases #endif //end of file Foo.h With this structure, I can always carry test cases with any class I write, I just need to copy the file around. When I need to test it there is a universal test main.cpp //File test_main.cpp #define BOOST_TEST_MAIN #include <boost/test/unit_test.hpp> #define ENABLE_TEST #include "foo.h" //#include other includes //End of File test_main This pattern has helped me to use TDD easily. The only issue is that test cases are scattered among each header files. Usually I do not know exactly how many test cases/suites are included. If there is a {--list|-l } option to print a tree of registered test suites and test cases, so that one can quickly look them up here and filter the test cases to be run, it would be a great feature.

1 0

Boost.Serialization: Error on Calling Class Method during serialization process
by Hazrat Pradipta Ranjali 25 Aug '11

25 Aug '11

I have the following problem when I try to call method in "TestSerialize" class during serialization process. Here is my code: class TestSerialize { public: std::string GetVal() { return Val + "abc"; } void SetVal(std::string tVal) { Val = tVal.substr(0, 2); } protected: std::string Val; friend class boost::serialization::access; template<class Archive> void save(Archive & ar, const unsigned int version) const { using boost::serialization::make_nvp; std::string tVal = GetVal(); // Error here ar & make_nvp("SC", tVal); } template<class Archive> void load(Archive & ar, const unsigned int version) { using boost::serialization::make_nvp; std::string tVal; ar & make_nvp("SC", tVal); SetVal(tVal); } BOOST_SERIALIZATION_SPLIT_MEMBER(); }; int main() { TestSerialize tS; std::ofstream ofs("test.xml"); boost::archive::xml_oarchive oa(ofs, boost::archive::no_header); oa << BOOST_SERIALIZATION_NVP(tS); ofs.close(); return 0; } The error that I encountered is: 'TestSerialize::GetVal' : cannot convert 'this' pointer from 'const TestSerialize' to 'TestSerialize &' This error only happens on "save" but not "load" I wonder why I get this error. I would like to know what Boost.Serialization do such that we have these two different behaviors. I use Boost Library 1.47.0 Regards,

2 1

Re: [Boost-users] [Containers] Marshalling deques
by Brian Wood 25 Aug '11

25 Aug '11

Ion Gazta?aga: > >El 22/08/2011 19:44, Brian Wood escribi?: >> Would you be willing to add a function that gives that? Tia. > > Yes, I think we need to think about a useful interface for that, maybe > something like local_iterator in unordered containers or something like > segmented iterator described by Matt Austern in his "Segmented Iterators > and Hierarchical Algorithms" paper If I could get the number of elements in the first chunk and the size of the chunks, that would be sufficient. (I believe the size of the chunks is constant within a deque instance.) I find it disappointing that the latest version of the standard doesn't have anything related to this and would rather just get something working than figuring out what would be a flexible, general interface. Shalom, Brian Wood Ebenezer Enterprises http://webEbenezer.net

3 2

[build] cannot specify different version of gcc
by Brodie Thiesfield 25 Aug '11

25 Aug '11

Hi, I'm trying to build boost on CentOS 5.5 using the /usr/bin/gcc44 compiler and I'm running into difficulties with documentation not fitting reality. How do I make this work? What I've tried... Note: I've removed /usr/bin/gcc and /usr/bin/g++ to force errors when gcc44 and g++44 are not used. Using ./bootstrap.sh Building Boost.Jam with toolset ... tools/jam/src/### ### No toolset specified. Please use --toolset option. Setting CC=/usr/bin/gcc44 and CXX=/usr/bin/g++44 (note that this won't work if gcc and g++ exist because then it will default to the gcc toolkit and not the gcc-4.4 toolkit that I want). This builds a number of things (looking at the bootstrap.log) but still gives the following error. Ignoring the error and calling ./bjam then fails. Building Boost.Jam with toolset ... tools/jam/src/### ### No toolset specified. Please use --toolset option. Using ./bootstrap.sh --toolset=gcc gives an error: unrecognized option: --toolset=gcc Using ./bootstrap.sh --with-toolset=gcc gives an error: Building Boost.Jam with toolset gcc... Failed to build Boost.Jam -> bootstrap.log shows that it is trying to build with gcc, which doesn't exist, therefore error. Using ./bootstrap.sh --with-toolset=gcc-4.4 gives an error: Building Boost.Jam with toolset gcc... Failed to build Boost.Jam -> bootstrap.log shows ### Unknown toolset: gcc-4.4 Creating ~/user-config.jam as: import toolset.using ; using gcc : 4.4 : g++44 ; export BOOST_USER_CONFIG=~/user-config.jam Then calling the same ./bootstrap.sh command lines above show no difference. How in the world do I get this thing to use /usr/bin/gcc44 and /usr/bin/g++44 to compile? I think that I'm going to have to create local links ./gcc -> /usr/bin/gcc44 and add the current directory to the path just to get this to build the way that I want it to. This seems crazy. Regards, Brodie

3 3

Re: [Boost-users] [Containers] Marshalling deques
by Brian Wood 24 Aug '11

24 Aug '11

Ion Gazta?aga: El 24/08/2011 20:06, Brian Wood escribi?: >> >> If I could get the number of elements in the first chunk and the >> size of the chunks, that would be sufficient. (I believe the size >> of the chunks is constant within a deque instance.) > > Not necessarily, but I think all implementations do that. The question > is whether we should have a standard-friendly and implementation > agnostic interface or an interface just for boost::deque implementation. How about both? First a boost::deque implementation and then if someone is inclined to delve into this further a more general implementation that might replace the earlier implementation. Since this wasn't addressed by C++11, I'm not sure how long it will take to get attention. If the interface we come up with has weaknesses, someone else will figure out how to improve upon it. Regards, Brian Wood Ebenezer Enterprises http://webEbenezer.net

1 0

Boost::Test: Registering Positive and Negative Tests
by Jeffrey Walton 24 Aug '11

24 Aug '11

Hi All, Our project currently uses Boost ::Test and auto registers the tests with BOOST_AUTO_TEST_CASE. We have both positive an negative tests, and would like to register accordingly so we can selectively turn off negative tests (for example, turn of negative tests when running Valgrind due to Signal/Noise ratio). Is this supported? Jeff

2 1