December 2011 - Boost-users

ASIO + SSL + no shared cipher
by Luke Walker 20 Dec '11

20 Dec '11

Hi All,I've taken the SSL server example "as is" but when I attempt to send a request from (Safari, Firefox, Chrome or IE) the handshake fails with "no shared cipher". I've researched the issue for several days and made a few changes. I use OpenSSL_add_all_algorithms and SSL_CTX_set_cipher_list to attempt to satisfy the cipher's but I get the same error. I also used OpenSSL > s_client to try and negotiate a cipher but it fails as well. One assumption I've assembled is that Chrome, Safari and NSURLRequest (Cocoa API) all use the same cipher list. Again I am using the ssl server example and have changed the context SSL Version but still no success. Any help would be greatly appreciated. Luke

1 0

Selecting a random neighbor
by George Bezerra 20 Dec '11

20 Dec '11

Hi, I'm totally new to BGL. I would like to know what is the best way to select a random neighbor of a vertex in an undirected graph without having to iterate through all neighbors. Can anyone help me with that? Thanks! George. -- George Bezerra Doctoral Student Department of Computer Science University of New Mexico, USA

2 1

[Regex] anyone seeing performance dip w/regex matching moving from 1.47->1.48 w/gcc?
by Kopet, Gary 20 Dec '11

20 Dec '11

Hi, I just pulled down boost regex 1.48 and built it with gcc 4.3 on CentOS 5.5. My internal regex performance tests are showing > 30% performance degradation between versions 1.47 and 1.48. I wanted to ask if anyone else has seen anything like this? We've been using the same makefile to built boost regex for a long time, and we haven't converted to bjam. So I am wondering if this could be due to our particular build environment? TIA, Gary Protected by Websense Hosted Email Security -- www.websense.com

2 2

boost test framework logging xml to file
by Jarosław Bober 20 Dec '11

20 Dec '11

Hello, is there a way to get test case name from global fixture? What I would like to have is separate xml file report for every test case. I'd like to do it this way: #include <string> #include <fstream> struct LogToFile { LogToFile() { std::string logFileName(boost::unit_test:: framework::master_test_suite().p_name); logFileName = current_test_case_name ; logFile.open(logFileName.c_str(), std::ios_base::app); boost::unit_test::unit_test_log.set_stream(logFile); } ~LogToFile() { logFile << "</TestLog>" << std::flush; logFile.close(); boost::unit_test::unit_test_log.set_stream(std::cout); } std::ofstream logFile; }; BOOST_GLOBAL_FIXTURE(LogToFile); Can I do that? Regards

2 1

[Interprocess] managed_shared_memory problem in 64-bit windows (v. 1.48.0)
by Michel Lestrade 20 Dec '11

20 Dec '11

Hi, I am trying to make a shared memory block between 2 Intel Fortran programs and thought to use a small C/C++ lib file to accomplish this: Fortran can use the ISO_C_BINDING module to make its pointers point to C objects but not vice versa and there is a vast Fortran code base which we have no intention of porting. There is a native pipe mechanism but given the amount of data I will be exchanging, disk access is something I want to avoid. The application is FEM modeling and the sparse matrix can be several hundred MBs; as the program evolves and the matrix changes, I expect hundreds of GBs of data in total I/O. Seems like it would be better to have a shared memory region and let the OS manage things as much as possible. Starting small to test the idea, I am running into trouble with something which seems easy enough to do in the examples but which only seems to work in 32-bit. Here is an incomplete fragment (got stuck early): #include <string> #include <boost/interprocess/managed_shared_memory.hpp> using namespace boost::interprocess; struct shm_remove { shm_remove(const char *input) { name = std::string(input); shared_memory_object::remove(name.c_str()); } ~shm_remove() {shared_memory_object::remove(name.c_str());} std::string name; } *remover; extern "C" void alloc_mumps(const int n, const int nnz, const char *name) { int size_shm = sizeof(int)*(2+2*nnz) + sizeof(double)*(nnz+2*n); remover = new shm_remove(name); managed_shared_memory managed_shm(create_only,name,size_shm); // Trying to do the equivalent of *i = new int int *i = managed_shm.construct<int>(anonymous_instance)[1](n); } Calling the function with n=3 and nnz=6 (a very small value compared to my final application) works OK when the application is compiled under 32-bit. Compiling under 64-bit produces an access violation error in win32_api.hpp: inline long interlocked_compare_exchange(long volatile *addr, long val1, long val2) { return BOOST_INTERLOCKED_COMPARE_EXCHANGE(addr, val1, val2); } values in debugger: addr=0, val1=1 val2=0. I do not know enough about Boost to know whether this is a design limitation, a bug or just an issue with my initial build. Just for reference, this is what I used: bjam --toolset=msvc-9.0 address-model=64 --build-type=complete stage Any suggestions ? Michel Lestrade Crosslight Software

2 2

[Boost] [Program Options]: Special character # in configuration file
by Bruno Matos 20 Dec '11

20 Dec '11

Hi, There is any way to use # in a configuration parameter value without being interpreted as comment initiator? Example: url=http://www.example.com/#sometopic Tia. Regards. -- Bruno Matos

1 0

help figuring out msg
by Jacques C. Richard 20 Dec '11

20 Dec '11

Can anyone help decipher this message? Is it really trying to allocate that size? Is gcc really using malloc under all these boost code? :) malloc: *** mmap(size=380554168512811008) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug terminate called after throwing an instance of 'std::bad_alloc' what(): std::bad_alloc [jacques-richards-mac-mini:00700] *** Process received signal *** [jacques-richards-mac-mini:00700] Signal: Abort trap (6) [jacques-richards-mac-mini:00700] Signal code: (0) [jacques-richards-mac-mini:00700] [ 0] 2 libSystem.B.dylib 0x00007fff8449b1ba _sigtramp + 26 [jacques-richards-mac-mini:00700] [ 1] 3 ??? 0x0030303700837a50 0x0 + 13563811672062544 [jacques-richards-mac-mini:00700] [ 2] 4 libstdc++.6.dylib 0x00007fff809c65d2 __tcf_0 + 0 [jacques-richards-mac-mini:00700] [ 3] 5 libobjc.A.dylib 0x00007fff84609b4d _objc_terminate + 120 [jacques-richards-mac-mini:00700] [ 4] 6 libstdc++.6.dylib 0x00007fff809c4ae1 _ZN10__cxxabiv111__terminateEPFvvE + 11 [jacques-richards-mac-mini:00700] [ 5] 7 libstdc++.6.dylib 0x00007fff809c4b16 _ZN10__cxxabiv112__unexpectedEPFvvE + 0 [jacques-richards-mac-mini:00700] [ 6] 8 libstdc++.6.dylib 0x00007fff809c4bfc _ZL23__gxx_exception_cleanup19_Unwind_Reason_CodeP17_Unwind_Exception + 0 [jacques-richards-mac-mini:00700] [ 7] 9 libstdc++.6.dylib 0x00007fff809c4efb _Znwm + 87 [jacques-richards-mac-mini:00700] [ 8] 10 Mg 0x0000000100113608 _ZN9__gnu_cxx13new_allocatorI12lattice_typeE8allocateEmPKv + 70 [jacques-richards-mac-mini:00700] [ 9] 11 Mg 0x000000010011362f _ZNSt12_Vector_baseI12lattice_typeSaIS0_EE11_M_allocateEm + 37 [jacques-richards-mac-mini:00700] [10] 12 Mg 0x0000000100113658 _ZNSt6vectorI12lattice_typeSaIS0_EE20_M_allocate_and_copyIPS0_EES4_mT_S5_ + 38 [jacques-richards-mac-mini:00700] [11] 13 Mg 0x000000010011376a _ZNSt6vectorI12lattice_typeSaIS0_EE7reserveEm + 114 [jacques-richards-mac-mini:00700] [12] 14 Mg 0x0000000100113881 _ZNK5boost13serialization3stl11reserve_impISt6vectorI12lattice_typeSaIS4_EEEclERS6_m + 33 [jacques-richards-mac-mini:00700] [13] 15 Mg 0x000000010011738b _ZN5boost13serialization3stl15load_collectionINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS6_EENS1_17archive_input_seqIS4_S8_EENS1_11reserve_impIS8_EEEEvRT_RT0_ + 189 [jacques-richards-mac-mini:00700] [14] 16 Mg 0x000000010011740c _ZN5boost13serialization4loadINS_3mpi15packed_iarchiveE12lattice_typeSaIS4_EEEvRT_RSt6vectorIT0_T1_EjN4mpl_5bool_ILb0EEE + 32 [jacques-richards-mac-mini:00700] [15] 17 Mg 0x0000000100117438 _ZN5boost13serialization4loadINS_3mpi15packed_iarchiveE12lattice_typeSaIS4_EEEvRT_RSt6vectorIT0_T1_Ej + 42 [jacques-richards-mac-mini:00700] [16] 18 Mg 0x0000000100117471 _ZN5boost13serialization11free_loaderINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS5_EEE6invokeERS3_RS7_j + 55 [jacques-richards-mac-mini:00700] [17] 19 Mg 0x0000000100117496 _ZN5boost13serialization10split_freeINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS5_EEEEvRT_RT0_j + 35 [jacques-richards-mac-mini:00700] [18] 20 Mg 0x00000001001174bb _ZN5boost13serialization9serializeINS_3mpi15packed_iarchiveE12lattice_typeSaIS4_EEEvRT_RSt6vectorIT0_T1_Ej + 35 [jacques-richards-mac-mini:00700] [19] 21 Mg 0x00000001001174f4 _ZN5boost13serialization13serialize_adlINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS5_EEEEvRT_RT0_j + 55 [jacques-richards-mac-mini:00700] [20] 22 Mg 0x0000000100117529 _ZNK5boost7archive6detail11iserializerINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS6_EEE16load_object_dataERNS1_14basic_iarchiveEPvj + 51 [jacques-richards-mac-mini:00700] [21] 23 libboost_serialization.dylib 0x0000000100533eb9 _ZN5boost7archive6detail14basic_iarchive11load_objectEPvRKNS1_17basic_iserializerE + 473 [jacques-richards-mac-mini:00700] [22] 24 Mg 0x00000001001178e3 _ZN5boost7archive6detail21load_non_pointer_typeINS_3mpi15packed_iarchiveEE13load_standard6invokeISt6vectorI12lattice_typeSaIS9_EEEEvRS4_RKT_ + 45 [jacques-richards-mac-mini:00700] [23] 25 Mg 0x000000010011790c _ZN5boost7archive6detail21load_non_pointer_typeINS_3mpi15packed_iarchiveEE6invokeISt6vectorI12lattice_typeSaIS8_EEEEvRS4_RT_ + 39 [jacques-richards-mac-mini:00700] [24] 26 Mg 0x0000000100117930 _ZN5boost7archive4loadINS_3mpi15packed_iarchiveESt6vectorI12lattice_typeSaIS5_EEEEvRT_RT0_ + 34 [jacques-richards-mac-mini:00700] [25] 27 Mg 0x000000010011795a _ZN5boost7archive6detail15common_iarchiveINS_3mpi15packed_iarchiveEE13load_overrideISt6vectorI12lattice_typeSaIS8_EEEEvRT_i + 40 [jacques-richards-mac-mini:00700] [26] 28 Mg 0x000000010011797f _ZN5boost3mpi15packed_iarchive13load_overrideISt6vectorI12lattice_typeSaIS4_EEEEvRT_iN4mpl_5bool_ILb0EEE + 35 [jacques-richards-mac-mini:00700] [27] 29 Mg 0x00000001001179ac _ZN5boost3mpi15packed_iarchive13load_overrideISt6vectorI12lattice_typeSaIS4_EEEEvRT_i + 42 [jacques-richards-mac-mini:00700] [28] 30 Mg 0x00000001001179d8 _ZN5boost7archive6detail18interface_iarchiveINS_3mpi15packed_iarchiveEErsISt6vectorI12lattice_typeSaIS8_EEEERS4_RT_ + 42 [jacques-richards-mac-mini:00700] [29] 31 Mg 0x0000000100117a0c _ZN5boost3mpi6detail21serialized_irecv_dataISt6vectorI12lattice_typeSaIS4_EEE11deserializeERNS0_6statusE + 40 [jacques-richards-mac-mini:00700] *** End of error message *** mpirun noticed that job rank 0 with PID 699 on node jacques-richards-mac-mini.local exited on signal 15 (Terminated). 1 additional process aborted (not shown) jacques-richards-mac-mini:SBF_T jacques$ Suspected cause: reqs.push_back( world.isend( Sender , SendTag , &Neighbor2Proc ) ) ; where std::vector< T > Neighbor2Proc ; Thanks!

3 4

Handler requirement for Asynchronous Operations in boost::asio
by Tu Hoang Do 19 Dec '11

19 Dec '11

It is specified in http://www.boost.org/doc/libs/1_47_0/doc/html/boost_asio/reference/AcceptHa… that a handler for `async_accept()` must satisfy the following function signature: void accept_handler( const boost::system::error_code& ec) { ... } However, in the `Daytime.3` example, using boost::bind, the handler can be specified as much parameter as desired, as long as it does not exceed the limit of `boost::bind` (which is 9 arguments at maximum): class tcp_server { public: tcp_server(boost::asio::io_service& io_service) : acceptor_(io_service, tcp::endpoint(tcp::v4(), 13)) { start_accept(); } private: void start_accept() { tcp_connection::pointer new_connection = tcp_connection::create(acceptor_.get_io_service()); acceptor_.async_accept(new_connection->socket(), boost::bind(&tcp_server::handle_accept, this, new_connection, boost::asio::placeholders::error)); } void handle_accept(tcp_connection::pointer new_connection, const boost::system::error_code& error) { if (!error) { new_connection->start(); } start_accept(); } tcp::acceptor acceptor_; }; Why is that possible? I thought even with `boost::bind`, the exact function signature still has to be matched. Notice the `handle_accept()` funciton and how it is used in `async_accept()`. The full code listing is here: http://www.boost.org/doc/libs/1_48_0/doc/html/boost_asio/tutorial/tutdaytim…

2 2

LINK : fatal error LNK1561: entry point must be defined
by ankit master 19 Dec '11

19 Dec '11

Hello, I am trying to run this simple example and I run into issues, could somebody please help me understand what am I doing incorrect here. #include <boost/test/unit_test.hpp> using namespace boost::unit_test; //____________________________________________________________________________// // you could easily implement test cases as a free functions // this test case will need to be explecetely registered in test tree void free_test_function() { // reports 'error in "free_test_function": test 2 == 1 failed' BOOST_CHECK(2 == 1); // non-critical test => continue after failure // reports 'unknown location(0): fatal error in "free_test_function": memory access violation // d:\source code\boost\libs\test\example\unit_test_example_02.cpp(25): last checkpoint' int* p = (int*)0x01; BOOST_CHECK( *p == 0 ); } //____________________________________________________________________________// test_suite* init_unit_test_suite( int, char* [] ) { framework::master_test_suite().p_name.value = "Unit test example 02"; // register the test case in test tree and specify number of expected failures so // this example will pass at runtime. We expect 2 errors: one from failed check and // one from memory acces violation framework::master_test_suite().add( BOOST_TEST_CASE( &free_test_function ), 2 ); return 0; } 1>------ Build started: Project: unitTestPrac2, Configuration: Debug Win32 ------ 1>Linking... 1>LINK : fatal error LNK1561: entry point must be defined 1>Build log was saved at "file://c:\Users\amaster\Desktop\AVM_Documents\CppPractise\unitTestPrac2\unitTestPrac2\Debug\BuildLog.htm" 1>unitTestPrac2 - 1 error(s), 0 warning(s) ========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ========== Thank you Ankit -- >From the withered tree, a flower blooms. - Zen -- >From the withered tree, a flower blooms. - Zen

1 0

Re: [Boost-users] Did he misunderstand you?
by John Wiegley 19 Dec '11

19 Dec '11

>>>>> Wayne Johnson <wjohnsonga(a)gmail.com> writes: > I'm new to the Boost libraries and becoming reacquainted with C++ after many > years of using other languages. As recommended in section 5.1 of the > "Getting Started on Windows", I visited the BoostPro website to find a > Windows 7 installer for v1.48.0 to use with VS C++ 2010. In response to a > followup message to BoostPro sales about the lack of an installer for the > latest version, I was told: >> "We don't have an installer at this time, and we are shortly about to >> remove that section from our website. The Boost build instructions can be >> found here ..." > Will there be no more Windows binary installers for the libraries or did I > misunderstand? Hi Wayne, What BoostPro does not have for 1.48 -- and has never made free to the public for any version -- are 64-bit installers for Windows. That was a paid service which we are now discontinuing. We may choose to make 64-bit installers freely available, time permitting, but we haven't finalized that decision yet. It's a bit of work to provide that service, and our plates have been overfull with paying work recently. -- John Wiegley BoostPro Computing http://www.boostpro.com

1 0