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

[boost.typeof] Using boost::bind with BOOST_AUTO
by Roberto Gim�nez 06 Jun '09

06 Jun '09

I have the following code: #include <boost/typeof/typeof.hpp> #include <boost/bind.hpp> void test() { } int main(int argc, char** argv) { BOOST_AUTO(a, boost::bind(&test)); } But when I try to compile, I get errors: CC -m64 -erroff=wvarhidemem,hidevf,hidevfinvb -errtags=yes -library=stlport4 -mt -c -g -I/home/gimenero/include -o build/Debug/Sun12-Solaris-Sparc/main.o main.cc CC -m64 -erroff=wvarhidemem,hidevf,hidevfinvb -errtags=yes -library=stlport4 -mt -c -g -I/home/gimenero/include -o build/Debug/Sun12-Solaris-Sparc/main.o main.cc "main.cc", line 11: Error, nomatchoverin: Could not find a match for boost::type_of::encode<boost::type_of::V, boost::type_of::T> (boost::_bi::bind_t<void, void(*)(), boost::_bi::list0>) needed in main(int, char**). "main.cc", line 11: Error, nomatchoverin: Could not find a match for boost::type_of::encode<boost::type_of::V, boost::type_of::T> (boost::_bi::bind_t<void, void(*)(), boost::_bi::list0>) needed in main(int, char**). "main.cc", line 11: Error, nomatchoverin: Could not find a match for boost::type_of::encode<boost::type_of::V, boost::type_of::T> (boost::_bi::bind_t<void, void(*)(), boost::_bi::list0>) needed in main(int, char**). ... many times. Does BOOST_AUTO supports boost::bind?

3 7

Does Boost.Test support std::wstring log messages?
by Tan, Tom (Shanghai) 06 Jun '09

06 Jun '09

Is it possible to use Boost.Test macros with std::wstring directly like this: BOOST_TEST_MESSAGE(L"test log"); ? I tried but it does not work. I find that \test\utils\basic_cstring\basic_cstring.hpp has a templated string implementation. It seems to imply that wide char are supported by Boost.Test. Can anyone please let me know how? Thanks. tom

2 1

[Boost] [Test] [1.38] "can't use testing tools before framework is initialized"
by Greg Christopher 06 Jun '09

06 Jun '09

Hi, I am trying to integrate boost.test into an existing test program with its own framework. Therefore I'd like to use that program's main. I am currently running into this runtime error when calling BOOST_WARN_MESSAGE(1, "string"). I have tried three different things to avoid this condition: 1) I declared a TEST_SUITE using the BOOST_AUTO_TEST_SUITE() and END() macros. Probably not what I want because I want my own main to keep control, but figured it would generate a main that would allow the initialization to happen. 2) I called init_unit_test_suite( argc, argv ). Even though it was called, I am still getting this. 3) I defined a cpp_main(). This never got called. I was hoping that the existence of this symbol would cause initialization to happen elsewhere. My desired control flow is to have this test program in control and to use the BOOST logging tools. I am trying to slip in boost.test with a minimum of changes to this program. I have hooked its logging function to call boost logging facilities instead; just can't get past this problem. Thanks for any help! Greg

3 9

[mpi] could all_reduce handle MPI_IN_PLACE ?
by Alain Miniussi 05 Jun '09

05 Jun '09

Hi, There doesn't seems to be an elgant way to do an in place all_reduce on array right now (Although I only stumbled on the issue after I moved to openmpi 1.3.2). That is, if all_reduce_impl is called with (in_values == out_values) MPI is allowed to compain with a MPI_ERR_BUF. Of course, one can call all_reduce with static_cast<T*>(MPI_IN_PLACE) but I do not find that solution very elegant. Would it be a problem to modify all_reduce_impl into: template<typename T, typename Op> void all_reduce_impl(const communicator& comm, const T* in_values, int n, T* out_values, Op /*op*/, mpl::true_ /*is_mpi_op*/, mpl::true_ /*is_mpi_datatype*/) { if (in_values == out_values) { in_values = static_cast<T*>(MPI_IN_PLACE); } BOOST_MPI_CHECK_RESULT(MPI_Allreduce, (const_cast<T*>(in_values), out_values, n, boost::mpi::get_mpi_datatype<T>(*in_values), (is_mpi_op<Op, T>::op()), comm)); } It worked for me, and if there is an agreement I can propose a patch. Regards, -- Alain Miniussi http://www.oca.eu/alainm Equipe CRIMSON https://crimson.oca.eu +33 492 00 30 09

1 0

[proto] declaring result type for callable contexts
by Maurizio Vitale 05 Jun '09

05 Jun '09

Eric, It seems like result types must be defined inside a struct eval, but examples in the documentation show the form commented out below. Is this a recent change? Also, in absence of a result_type declaration inside struct eval {...} the type defaults to int. Would it be possible to make it void or some_ugly_looking_type_that_tell_me_i_is_all_wrong? int might happen to be what the user wants, until he decide to change it and hides the problem. template<typename Expr> struct my_context : proto::callable_context<const my_context<Expr> > { struct eval {typedef double result_type;}; // doesn't work (anymore?) //typedef double result_type; };

3 5

Bug Sprint Report for Thursday, June 4th
by Marshall Clow 05 Jun '09

05 Jun '09

The boost bug sprint started Friday, May 29th, and runs through Sunday June 7th. Thursday was a very good day - let's keep it up! The bug sprint will be over Sunday night. Status <https://svn.boost.org/trac/boost/report/1> 7:00 AM PDT 5/29: 797 open tickets. 7:45 AM PDT 5/30: 772 open tickets. 6:00 AM PDT 5/31: 761 open tickets. 6:00 AM PDT 6/2: 747 open tickets. 6:00 AM PDT 6/3: 748 open tickets. 6:00 AM PDT 6/4: 739 open tickets. 8:00 AM PDT 6/5: 730 open tickets. Activity: 5/29: 6 new tickets, 77 modifications, 25 tickets closed 5/30: 2 new tickets, 42 modifications, 17 tickets closed 5/31: 2 new tickets, 27 modifications, 8 tickets closed 6/1: 2 new tickets, 42 modifications, 17 tickets closed 6/2: 5 new tickets, 18 modifications, 5 tickets closed. 6/3: 4 new tickets, 19 modifications, 2 tickets closed. 6/4: 3 new tickets, 42 modifications, 24 tickets closed. [ Numbers may not add up because I am counting via two different methods, using data that is time-stamped in two different time zones. Also, tickets can be closed and reopened. ] What can you do? See <https://svn.boost.org/trac/boost/wiki/BugSprintJune2009> for details. I've added four new reports to the trac: Open ticket count by owner: <https://svn.boost.org/trac/boost/report/18> Open ticket count by component: <https://svn.boost.org/trac/boost/report/19> Open ticket count by ticket type: <https://svn.boost.org/trac/boost/report/20> Open ticket count by milestone: <https://svn.boost.org/trac/boost/report/21> There's been a lot of discussion on the list about tickets; and there are about a dozen patches that have been tested and are ready to be applied. If you're looking to help, you can look at some of the older bugs, say, the ones for Boost 1.36, and determine if they still apply. If they do, note that in the ticket. There are 35 tickets that are still open that are marked as "to be fixed in Boost 1.36" See <https://svn.boost.org/trac/boost/report/3> for a list of open tickets by milestone. --------- Sign-up sheet -------- The following people have committed to participate in the bug sprint. Please add you name here, along with what you can do. 1) Marshall Clow (coordination, Mac OS X characterization) 2) Beman Dawes (filesystem, system, timer) 3) Doug Gregor (function, graph, logic, mpi, numeric, property_map, signals) 4) "joel falcou" <joel.falcou(a)u-psud.fr> ( linux gcc testing/VC 2008 testing, patching) 5) Michael Caisse <boost(a)objectmodelingdesigns.com> (bug fixing) 6) Jeremiah Willcock <jewillco(a)osl.iu.edu> (will test and commit graph patches) 7) David Abrahams (iterator, python) 8) Sohail Somani (patches for g++ 4.2.4) 9) Steven Watanabe <watanabesj(a)gmail.com> (bug fixes, como) 10) vicente.botet" <vicente.botet(a)wanadoo.fr> 11) Stjepan Rajko <stjepan.rajko(a)gmail.com> (fusion) 12) Daniel Wallin (parameter, ...) -- -- Marshall Marshall Clow Idio Software <mailto:marshall@idio.com> It is by caffeine alone I set my mind in motion. It is by the beans of Java that thoughts acquire speed, the hands acquire shaking, the shaking becomes a warning. It is by caffeine alone I set my mind in motion.

1 0

BGL: how to count and get the connected components in the spanning_tree
by arm2arm 05 Jun '09

05 Jun '09

Hello, I am trying to get connected components in the MST. I have some undirected graph: ----------- BEGIN SNIP ------------------------ typedef adjacency_list < vecS, vecS, undirectedS, no_property, property < edge_weight_t, MyFloat > > Graph; typedef property_map<Graph, edge_weight_t>::type EdgeWeightMap; Graph graphFOF; ... Fill the graph.. //Get the MST property_map < Graph, edge_weight_t >::type weight = get(edge_weight, graphFOF); std::vector < Edge > spanning_tree; std::cout << "Building MST:." ; kruskal_minimum_spanning_tree(graphFOF, std::back_inserter(spanning_tree)); std::cout << " .. done " << std::endl; // // Here is problem: How to use MST edges for filtering the graphFOF?? positive_edge_weight<EdgeWeightMap> filter(meanW, get(edge_weight, graphFOF)); filtered_graph<Graph, positive_edge_weight<EdgeWeightMap> > fg(graphFOF, filter); std::vector<int> component(num_vertices(fg)); int num = connected_components(fg, &component[0]); ------------END SNIP--------------------------- How to count and get the connected components in the spanning_tree? Thank you beforehand. Arman. -- View this message in context: http://www.nabble.com/BGL%3A-how-to-count-and-get-the-connected-components-… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

Re: [Boost-users] About "transitive_closure" method of Boost graph.
by Amanullah Yasin 05 Jun '09

05 Jun '09

sorry for disturbing, i think you mean test example given on http://www.boost.org/doc/libs/1_39_0/libs/graph/example/transitive_closure.… Yes i tried this example but it is not using this method. void transitive_closure(const Graph& g, GraphTC& tc, G_to_TC_VertexMap g_to_tc_map, VertexIndexMap index_map) Thanks again Amanullah YASIN Master2 (ECD) Ecole Polytechnique Université de Nantes, France. Mob# 06 47 62 89 96 Fix # 02 72 00 26 45 amanyasin(a)gmail.com On Fri, Jun 5, 2009 at 10:57 AM, Vladimir Prus <vladimir(a)codesourcery.com>wrote: > On Friday 05 June 2009 12:52:37 you wrote: > > Thanks Vladimir Prus, for your reply. Please guide me how i can use this > > method > > Please reply to the list, not personally. > > > > > > > void transitive_closure(const Graph& g, GraphTC& tc, > > G_to_TC_VertexMap g_to_tc_map, VertexIndexMap > index_map) > > > > > > *specially how i can declare and use * > > > > * "G_to_TC_VertexMap" & "* > > > > *VertexIndexMap"* > > I have suggested you look at the testcase -- did you do that? > > - Volodya >

1 0

[multi-index] Container not safe against reentrant clear() / erase() ?
by Sebastian Theophil 05 Jun '09

05 Jun '09

Hi, I have encounted a scenario that triggers an access violation when removing reference counted objects from a container. Assume I have a (for simplicity's sake, global) multi_index_container g_multiindex containing reference counted objects of type CObject: boost::multi_index_container< ordered_unique< shared_ptr<CObject> > > g_multiindex; In general, an object of type CObject may be destroyed at any time and the destructor makes sure to remove any reference held in g_multiindex: class CObject { ~CObject() { // if object is explicitly destroyed, remove from container g_multiindex.erase( this ); } }; Since ownership of CObject is shared, g_multiindex may hold the last reference to a CObject instance. g_multiindex.clear then calls - delete_all_nodes() - delete_all_nodes(root()) - this->final_delete_node_(static_cast<final_node_type*>(x)); - which first calls the destructor on x (i.e. it calls ~CObject() on the root() node) At this point, the multi_index_container is not in a valid state. The root() node is being deleted but the container still contains the root() node and points to it, empty() would certainly still return false. Calling g_multiindex.erase with a compatible key crashes in my case. In general, it seems to me that it would be safer to implement clear() by doing something like: void clear() { multi_index_container cpy; swap(cpy, *this); cpy.clear(); } This would remove all elements from the container before actually deleting them. I'm unsure though how the same behavior could be achieved when erasing a single element. The C++ standard does not seem to define how clear() and erase() should behave in this regard, but I cannot image that my application scenario is so unusual that this problem has not been solved a million times. So I'm curious if the behavior of multi_index_containers is by design and maybe someone could be so kind to enlighten me why this is so. Is there any better way to achieve what I'm trying to do? Thank you very much, Sebastian -- Sebastian Theophil . stheophil(a)think-cell.com Software Engineer think-cell Software GmbH . Invalidenstr. 34 . 10115 Berlin, Germany http://www.think-cell.com . phone +49-30-666473-10 . toll-free (US) +1-800-891-8091 Directors: Dr. Markus Hannebauer, Dr. Arno Schoedl . Amtsgericht Berlin-Charlottenburg, HRB 85229

4 8

About "transitive_closure" method of Boost graph.
by Amanullah Yasin 05 Jun '09

05 Jun '09

Bonjour Dear I want to check whether there is a directed path between two given nodes (A and C) using Boost graph. I also want to access intermediate edges to examine PDAG properties. I test void transitive_closure(const Graph& g, GraphTC& tc) method. But i want to check between two given nodes. Now i want to ask you - void transitive_closure(const Graph& g, GraphTC& tc, G_to_TC_VertexMap g_to_tc_map, VertexIndexMap index_map) In the above method, how i can declare G_to_TC_VertexMap g_to_tc_map, VertexIndexMap index_map? http://www.boost.org/doc/libs/1_39_0/libs/graph/doc/transitive_closure.html - How i can examine the intermediate edges? or is there any other easy way? I will be thankful to you Best Regards Amanullah YASIN Master2 (ECD) Ecole Polytechnique Université de Nantes, France. Mob# 06 47 62 89 96 Fix # 02 72 00 26 45 amanyasin(a)gmail.com

2 2

Refactoring template functions. Needing help with compile time if statements.
by Ryan McConnehey 05 Jun '09

05 Jun '09

I'm trying to refactor my code to remove some code duplication. Currently I have a template that takes a pointer to store information, a map with the information and a text string. I've specialized the function to account for some fundamental types. As these cases have increased I notice the only difference between the function is how I'm accessing the iterator for the map. template<> void Customize::addVariableType(TiXmlElement *const xml, std::map<string, bool> const& map, string const& mapType) const { std::map<string, bool>::const_iterator itor; for (itor = map.begin(); itor != map.end(); ++itor) { itor->second; } } template<typename T> void Customize::addVariableType(TiXmlElement *const xml, std::map<string, T> const& map, string const& mapType) const { std::map<string, T>::const_iterator itor; for (itor = map.begin(); itor != map.end(); ++itor) { itor->second.base_type_value(); } } I'm using stlSoft's true_typedef for some custom types. STLSOFT_GEN_OPAQUE(fundamental_type); typedef stlsoft::true_typedef<unsigned short, fundamental_type> uint16_type; Most of the maps are comprised of these custom types. That's why the generic template accesses the underlying value of the iterator's second value. Also this accounts for the specialization for fundamental types. Since both the template and the specialization template have the same functionality but access the second map value differently I'm trying to refactor to have another function retrieve iterator value. Below is what I've tried so far. template<> bool accessValue(bool const& value) const { return value; } //V is the underlining type of W. If W was uint16_type then V should be unsigned short. template<typename V, typename W> typename V accessValue(typename W const& value) const { return value.base_type_value(); } I understand that I've moved the difficulty out of addVariableType into accessValue. Unfortunately, accessValue gives an error that typename V cannot be deduced. Also for each fundamental type I have to add an overload to accessValue. This turns into the same problem I'm facing with addVariableType but with less code to copy. Is there a way to have a template function that has an if statement to choice how to access the value? I've tried to use BOOST_TYPEOF_TPL and boost::is_class but was unsuccessful in making it work. Any help would be really helpful. Ryan

1 0

Box-Muller transform Gaussian Random Number Generator
by Enrique Fernández 05 Jun '09

05 Jun '09

Hello, Boost random library implements GRNG (Gaussian Random Number Generator) with Box-Muller transform (code here<http://www.boost.org/doc/libs/1_39_0/boost/random/normal_distribution.hpp>). I want to discuss about this line of code: _cached_rho = sqrt(-result_type(2) * log(result_type(1)-_r2)); I don't know why log is applied to 1-r2, instead of r2 directly, yielding _cached_rho = sqrt(-result_type(2) * log(_r2)); According with [1<http://www.cse.cuhk.edu.hk/%7Ephwl/mt/public/archives/papers/grng_acmcs07.p…>, 2 <http://mathworld.wolfram.com/Box-MullerTransformation.html>, 3<http://en.wikipedia.org/wiki/Box-Muller_transformation>] the uniform random variable r2 is always used directly (never 1-r2). I'd like to know why 1-r2 is used. I've tried to compare with other scientific libraries like GSL<http://www.gnu.org/software/gsl/manual/html_node/The-Gaussian-Distribution.…>, Scythestat <http://scythe.wustl.edu/api/rng_8h-source.html> and LTILib<http://ltilib.sourceforge.net/doc/html/index.shtml>, but they use the Box-Muller polar form (or Marsaglia polar method) or even Ziggurat algorithm, which are supposed to be more efficient. Why isn't polar form implemented? I'd like to know why 1-r2 is used anyway. Regards, Enrique

2 1

What will be in v1.40?
by Robert Dailey 05 Jun '09

05 Jun '09

Is there a document online that covers new features that will be introduced in Boost v1.40? If not, can someone provide a brief overview? Thanks.

2 1

Re: [Boost-users] Boost-users Digest, Vol 2016, Issue 2
by Tron Thomas 04 Jun '09

04 Jun '09

> Date: Thu, 4 Jun 2009 09:44:34 +0100 > From: "John Maddock" <john(a)johnmaddock.co.uk> > Subject: Re: [Boost-users] Compile error using boost's tr1::bind > To: <boost-users(a)lists.boost.org> > Message-ID: <059F08EC9A4049B194EBF6D5CA604C4C@acerlaptop> > Content-Type: text/plain; format=flowed; charset="Windows-1252"; > reply-type=original > > >> What is needed for the code to compile and run properly? >> > > Which Boost version and is Boost installed in /usr/include or somewhere > similar? > > Basically this is a known problem for some Boost versions when Boost is > installed in your *system search path* - the issue is that the Boost headers > are in a location that is searched *after* the libstdc++ headers and that > breaks the header-forwarding mechanism used by Boost.TR1. > > The solution is probably to upgrade to the latest Boost release, or if all > else fails, the SVN Trunk. > > HTH, John. > I'm not sure which version of boost this is. It would be whatever the most current version of boost is for Fedora 10, as I update my system on a regular basis. There is a /usr/include/boost directory on my system. I would assume that is the version of boost being used by the compiler.

2 1

unit tests: auto tests, missing main, redef'ed init_unit_test_suite
by Hicham Mouline 04 Jun '09

04 Jun '09

g++3.4 - boost1.37 Hello, I have n translation units .cpp and m auto test suites spread over the n files. For e.g. file ni.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite1) BOOST_AUTO_TEST_CASE(case1) BOOST_AUTO_TEST_SUITE_END() file nj.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite1) BOOST_AUTO_TEST_CASE(case2) BOOST_AUTO_TEST_CASE(case3) BOOST_AUTO_TEST_SUITE_END() file nk.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite2) BOOST_AUTO_TEST_CASE(case1) BOOST_AUTO_TEST_CASE(case2) BOOST_AUTO_TEST_SUITE_END() The .cpp files are compiled with -DBOOST_TEST_MAIN= AFAIU, I am then not required to provide the "init_unit_test" function nor the "main" function. I then try to link an executable: g++ -m64 -Wall -Wextra -Wsign-compare -Werror -Wno-unused -std=c++98 -pedantic -Wno-long-long -pipe -g -lboost_unit_test_framework-gcc34-mt-d-1_37 -o executable *.o but i get this: nk.o: In function `init_unit_test_suite(int, char**)': ...boost/test/unit_test_suite.hpp:188: multiple definition of `init_unit_test_suite(int, char**)' ni.o: ...boost/test/unit_test_suite.hpp:188: first defined here for each object file and /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/crt1.o(.text+0x21): In function `_start': : undefined reference to `main' Regards,

3 4

Re: [Boost-users] [string_algo] replace string with tokens
by Michael Schuster 04 Jun '09

04 Jun '09

>> I'm not sure what you mean with 'replace command'. >> boost::algorithm::replace_all() simply replaces a substring with another.If >> you mean boost::algorithm::split() you can pass a predicate as the third >> parameter (boost::algorithm::is_punct() might help). Well I didnot find replace_all in usage of the bib. Now that you tell me, i can find it. But here I see no way to tell replace_all that my strings are not seperated by spaces but +*() etc. a+3*sin(b) So what I had expected is like this: string str="a+3*sin(b)"; replace_all(str,"b", "Beta", "+*()") // now str="a+3*sin(Beta)"; Perhaps a task for tokenizer(?) >For what you are doing, perhaps regex_replace in Boost.Xpressive might >be more useful? I'll have a look Thanks Michael Schuster

1 0

boost::thread - Why does this code not work?
by Henrik Berg 04 Jun '09

04 Jun '09

Hi all, I have some problems with the boost::thread library, I've been staring at this code for a couple of days now but still have no idea what's wrong with it. The code starts a thread, the thread waits at a barrier, and command is transfered between the main thread and the new thread a couple of times (using two successive calls to barrier->wait() for each transition). This works fine most of the times, but sometimes the thread for some reason does not start. And when the main thread then waits at the barrier, the program obiously locks up. Here's the code: #include <iostream> #include <boost/thread.hpp> #include <boost/thread/barrier.hpp> class ThreadObject { public: boost::barrier *barrier; boost::thread *tacticsThread; ThreadObject() { barrier = new boost::barrier(2); std::cerr << "Main thread: Initiating ThreadObject\n"; tacticsThread = new boost::thread(*this); std::cerr << "Main thread: ThreadObject initiated\n"; } void operator()() { std::cerr << "Thread 2: Started\n"; // Waiting for the main thread to release this thread: barrier->wait(); std::cerr << "Thread 2: Processing.\n"; std::cerr << "Thread 2: Waiting for the main thread...\n"; // Release the main thread (which is waiting at the barrier): barrier->wait(); // Wait at the barrier until the main thread releases this thread again: barrier->wait(); std::cerr << "Thread 2: Continues.\n"; std::cerr << "Thread 2: Completed, now releasing the main thread.\n"; // Release the main thread (which is waiting at the barrier): barrier->wait(); } }; int main() { for (int i = 0; i < 100; i++) { std::cerr << std::endl << "Starting new test, i = " << i << std::endl; ThreadObject obj; std::cerr << "Main thread: Transferring command to Thread 2\n"; obj.barrier->wait(); obj.barrier->wait(); std::cerr << "Main thread: Command returned. Transfer command back to Thread 2 once more\n"; obj.barrier->wait(); obj.barrier->wait(); std::cerr << "Main thread: completed, everything ok\n"; } } In main(), a new ThreadObject is created, this will start a new thread. Then command is transfered back and forth a few times (with pairs of barrier->wait())). Since the code usually works fine, I've wrapped it all in a for loop, in order to repeat it multiple times. After a few iterations, the program locks up. The number of successful iterations before this happens varies, here is an example of output from a run which locked after two iterations: Starting new test, i = 0 Main thread: Initiating ThreadObject Main thread: ThreadObject initiated Main thread: Transferring command to Thread 2 Thread 2: Started Thread 2: Processing. Thread 2: Waiting for the main thread... Main thread: Command returned. Transfer command back to Thread 2 once more Thread 2: Continues. Thread 2: Completed, now releasing the main thread. Main thread: completed, everything ok Starting new test, i = 1 Main thread: Initiating ThreadObject Main thread: ThreadObject initiated Thread 2: Started Main thread: Transferring command to Thread 2 Thread 2: Processing. Thread 2: Waiting for the main thread... Main thread: Command returned. Transfer command back to Thread 2 once more Thread 2: Continues. Thread 2: Completed, now releasing the main thread. Main thread: completed, everything ok Starting new test, i = 2 Main thread: Initiating ThreadObject Main thread: ThreadObject initiated Main thread: Transferring command to Thread 2 So, why does not the new thread start up in the last test? Is there some problem with my code, or is there some bug in boost::thread? I'm compiling this using gcc-3.4.4 and boost 1.39.0 under cygwin. Thanks for any help, Henrik

2 1

[Unknown] Pedicate wrappers around std algorithms
by Robert Jones 04 Jun '09

04 Jun '09

Hi All, Does this kind of wrapper exist in Boost anywhere? template < class Iterator, class T > bool exists( Iterator begin, Iterator end, const T & value ) { return find( begin, end, value ) != end; } Thanks, - Rob.

2 1

Why does bind have overhead?
by Joseph Garvin 04 Jun '09

04 Jun '09

I've read that bind can have significant overhead unless your parameters are small (because then it can use the small buffer optimization). But why does it have anymore overhead than copying the parameters and assigning a function pointer? I suspect support for something (maybe member function pointers + multiple inheritance?) is responsible but I'm not sure what. I tried looking through bind.hpp but trying to figure out what it's actually doing amongst all the compiler workarounds and different parameter versions it's difficult to tell what's going on.

2 1

[Function] Potential bug with ref-wrapped function objects?
by Dave Bonora 04 Jun '09

04 Jun '09

Hello group, Boost.Function is exhibiting differing behavior in different boost releases: //-------------------------- struct F { void operator()() { } }; F obj; // assume obj remains in scope for duration of sample boost::function<void()> f1 = boost::ref( obj ); boost::function<void()> f2 = f1; assert( f1 == boost::ref( obj ) ); assert( f2 == boost::ref( obj ) ); // Passes in 1.35.0, Fails in 1.39.0 boost::function<void()> f3 = boost::ref( obj ); boost::function<void()> f4; f3.swap( f4 ); assert( f3 != boost::ref( obj ) ); assert( f4 == boost::ref( obj ) ); // Passes in 1.35.0, Fails in 1.39.0 //--------------------------- I'm assuming 1.35.0 is demonstrating the correct behavior whereas 1.39.0 is exhibiting a bug, but I thought I'd defer to the group before I reported it. Am I correct?

2 1

Compile error using boost's tr1::bind
by Tron Thomas 04 Jun '09

04 Jun '09

This code: #include <iostream> #include <algorithm> #include <boost/tr1/functional.hpp> using std::ostream; class Value { public: Value() : m_amount(0) {} void IncreaseBy(int amount) { m_amount += amount; } void WriteTo(ostream& stream) const { stream << m_amount; } private: int m_amount; }; ostream& operator <<(ostream& stream, const Value& value) { value.WriteTo(stream); return stream; } int main() { Value sum; int values[] = { 1, 2, 3, 4, 5 }; std::for_each(values, values + 5, std::tr1::bind(&Value::IncreaseBy, sum)); std::cout << "The value is " << sum << std::endl; return 0; } Produces compile error output like the following when compiled on Fedora 10 (Intel): In file included from /usr/include/boost/tr1/functional.hpp:8, from ValueSum.cpp:3: /usr/include/boost/tr1/detail/config.hpp:60:26: error: no include path in which to search for utility /usr/include/boost/mem_fn.hpp: In member function ‘R& boost::_mfi::dm<R, T>::operator()(T&) const [with R = void ()(int), T = Value]’: /usr/include/boost/bind.hpp:221: instantiated from ‘R boost::_bi::list1<A1>::operator()(boost::_bi::type<R>, F&, A&, long int) [with R = void (&)(int), F = boost::_mfi::dm<void ()(int), Value>, A = boost::_bi::list1<int&>, A1 = boost::_bi::value<Value>]’ /usr/include/boost/bind/bind_template.hpp:32: instantiated from ‘typename boost::_bi::result_traits<R, F>::type boost::_bi::bind_t<R, F, L>::operator()(A1&) [with A1 = int, R = void (&)(int), F = boost::_mfi::dm<void ()(int), Value>, L = boost::_bi::list1<boost::_bi::value<Value> >]’ /usr/lib/gcc/i386-redhat-linux/4.3.2/../../../../include/c++/4.3.2/bits/stl_algo.h:3791: instantiated from ‘_Funct std::for_each(_IIter, _IIter, _Funct) [with _IIter = int*, _Funct = boost::_bi::bind_t<void (&)(int), boost::_mfi::dm<void ()(int), Value>, boost::_bi::list1<boost::_bi::value<Value> > >]’ ValueSum.cpp:31: instantiated from here /usr/include/boost/mem_fn.hpp:359: error: invalid use of non-static member function What is needed for the code to compile and run properly?

3 2

Re: [Boost-users] unit tests: auto tests, missing main, redef'ed init_unit_test_suite
by Hicham Mouline 04 Jun '09

04 Jun '09

From: Hicham Mouline [mailto:hicham@mouline.org] Sent: 04 June 2009 14:03 To: 'boost-users(a)lists.boost.org' Subject: unit tests: auto tests, missing main, redef'ed init_unit_test_suite g++3.4 - boost1.37 Hello, I have n translation units .cpp and m auto test suites spread over the n files. For e.g. file ni.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite1) BOOST_AUTO_TEST_CASE(case1) BOOST_AUTO_TEST_SUITE_END() file nj.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite1) BOOST_AUTO_TEST_CASE(case2) BOOST_AUTO_TEST_CASE(case3) BOOST_AUTO_TEST_SUITE_END() file nk.cpp-------------------- #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(suite2) BOOST_AUTO_TEST_CASE(case1) BOOST_AUTO_TEST_CASE(case2) BOOST_AUTO_TEST_SUITE_END() The .cpp files are compiled with -DBOOST_TEST_MAIN= AFAIU, I am then not required to provide the "init_unit_test" function nor the "main" function. I then try to link an executable: g++ -m64 -Wall -Wextra -Wsign-compare -Werror -Wno-unused -std=c++98 -pedantic -Wno-long-long -pipe -g -lboost_unit_test_framework-gcc34-mt-d-1_37 -o executable *.o but i get this: nk.o: In function `init_unit_test_suite(int, char**)': ...boost/test/unit_test_suite.hpp:188: multiple definition of `init_unit_test_suite(int, char**)' ni.o: ...boost/test/unit_test_suite.hpp:188: first defined here for each object file and /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/crt1.o(.text+0x21): In function `_start': : undefined reference to `main' Regards, ------------------------------------------------------------------------ After rereading, I defined -DBOOST_TEST_MAIN= for only 1 of the source files. The objects compile ok. But linking fails with the same error. Regards,

1 0

[GUID] newbie's question
by bidi_bulle＠yahoo.com 04 Jun '09

04 Jun '09

Hello, I am actively searching for GUID librairy in the version 7. Perhaps I miss completely my search but I can't find it anymore. Can someone help me and provide me a link to this librairy ? Thanks Bob

2 1

Bug Sprint Report for Wednesday, June rd
by Marshall Clow 04 Jun '09

04 Jun '09

The boost bug sprint started Friday, May 29th, and runs through Sunday June 7th. Wednesday was not a good day in terms of bug closing; the good news is that we've already (as of 6 AM) closed more bugs on Thursday than all of Wednesday. Status <https://svn.boost.org/trac/boost/report/1> 7:00 AM PDT 5/29: 797 open tickets. 7:45 AM PDT 5/30: 772 open tickets. 6:00 AM PDT 5/31: 761 open tickets. 6:00 AM PDT 6/2: 747 open tickets. 6:00 AM PDT 6/3: 748 open tickets. 6:00 AM PDT 6/4: 739 open tickets. Activity: 5/29: 6 new tickets, 77 modifications, 25 tickets closed 5/30: 2 new tickets, 42 modifications, 17 tickets closed 5/31: 2 new tickets, 27 modifications, 8 tickets closed 6/1: 2 new tickets, 42 modifications, 17 tickets closed 6/2: 5 new tickets, 18 modifications, 5 tickets closed. 6/3 4 new tickets, 19 modifications, 2 tickets closed. [ Numbers may not add up because I am counting via two different methods, using data that is time-stamped in two different time zones. Also, tickets can be closed and reopened. ] What can you do? See <https://svn.boost.org/trac/boost/wiki/BugSprintJune2009> for details. I've added four new reports to the trac: Open ticket count by owner: <https://svn.boost.org/trac/boost/report/18> Open ticket count by component: <https://svn.boost.org/trac/boost/report/19> Open ticket count by ticket type: <https://svn.boost.org/trac/boost/report/20> Open ticket count by milestone: <https://svn.boost.org/trac/boost/report/21> There's been a lot of discussion on the list about tickets; and there are about a dozen patches that have been tested and are ready to be applied. If you're looking to help, you can look at some of the older bugs, say, the ones for Boost 1.36, and determine if they still apply. If they do, note that in the ticket. There are 35 tickets that are still open that are marked as "to be fixed in Boost 1.36" See <https://svn.boost.org/trac/boost/report/3> for a list of open tickets by milestone. --------- Sign-up sheet -------- The following people have committed to participate in the bug sprint. Please add you name here, along with what you can do. 1) Marshall Clow (coordination, Mac OS X characterization) 2) Beman Dawes (filesystem, system, timer) 3) Doug Gregor (function, graph, logic, mpi, numeric, property_map, signals) 4) "joel falcou" <joel.falcou(a)u-psud.fr> ( linux gcc testing/VC 2008 testing, patching) 5) Michael Caisse <boost(a)objectmodelingdesigns.com> (bug fixing) 6) Jeremiah Willcock <jewillco(a)osl.iu.edu> (will test and commit graph patches) 7) David Abrahams (iterator, python) 8) Sohail Somani (patches for g++ 4.2.4) 9) Steven Watanabe <watanabesj(a)gmail.com> (bug fixes, como) 10) vicente.botet" <vicente.botet(a)wanadoo.fr> 11) Stjepan Rajko <stjepan.rajko(a)gmail.com> (fusion) 12) Daniel Wallin (parameter, ...) -- -- Marshall Marshall Clow Idio Software <mailto:marshall@idio.com> It is by caffeine alone I set my mind in motion. It is by the beans of Java that thoughts acquire speed, the hands acquire shaking, the shaking becomes a warning. It is by caffeine alone I set my mind in motion.

1 0

[bind maybe?] find_if of map mapped_type.
by Robert Jones 04 Jun '09

04 Jun '09

If I'm iterating over a map with, say find_if, and my predicate function accepts the mapped_type of the map, is there a mechanism to pick out the pair.second value inline? class C; map< int, C > m; bool pred( const C & ); // correct version of this intent. find_if( m.begin( ), m.end( ), pred( second ) ); Thanks, - Rob.

2 2