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

Boost.Interprocess accessing an unordered_map in shared memory.
by lazylabs＠gmail.com 12 Mar '13

12 Mar '13

Hi, I am trying to create a unordered_map in shared memory using Boost Interprocess library. Here, is the code, which I'm trying to use (taking examples from Boost Interprocess documentation): #include <boost/interprocess/managed_shared_memory.hpp> #include <boost/interprocess/allocators/allocator.hpp> #include <functional> #include <boost/functional/hash.hpp> #include <boost/unordered_map.hpp> #include <iostream> #include <string> #include <boost/interprocess/containers/string.hpp> namespace bipc = boost::interprocess; typedef bipc::allocator<char, bipc::managed_shared_memory::segment_manager> CharAllocator; typedef bipc::basic_string<char, std::char_traits<char>, CharAllocator> ShmemString; struct Person { int age; ShmemString name; double salary; Person(int i, double sal, const char* s, const char_allocator& a) : age(i), name(s, a), salary(sal) { } void print() {} } typedef ShmemString KeyType; typedef Person MappedType; typedef std::pair< KeyType, MappedType > MapPersonType; typedef bipc::allocator< MapPersonType, bipc::managed_shared_memory::segment_manager > ShMemAllocator; typedef boost::unordered_map< KeyType, MappedType, boost::hash<KeyType>, std::equal_to<KeyType>, ShMemAllocator > PersonMap; In the main reader program, I create a map in the shared memory and I insert an object in this map: int main() { bipc::managed_shared_memory segment(bipc::create_only, "MySharedMemory", 65536); PersonMap *persons = segment.construct<PersonMap>("MyHashMap") ( 3, boost::hash<ShmemString>(), std::equal_to<ShmemString>() , segment.get_allocator<MapPersonType>()); char_allocator alloc(segment.get_allocator<char>()); Person p1(20, 10000, "ABC", alloc); persons->insert(MapPersonType(ShmemString("Person1", alloc), p1)); } However, when I am trying to access the map in the reader program, I need to use a syntax like persons->at(ShmemString("H", segment.get_allocator<char>())).print(); However, I would prefer to do this with a std::string, which results in compilation errors: persons->at(std::string("H")).print(); Is it possible to write the above statement, i.e. accessing the map allocated in shared memory with std::string? Thanks.

3 4

Using SSL
by ustulation 11 Mar '13

11 Mar '13

Hello. this will be the first time i'll be writing an SSL and am using boost::asio::ssl. I've gone through the articles over the net and somewhat understand how ssl works. I've created 3 files at server end as of now: privateKey.pem, unsignedCertificate.csr and signedCertificate.pem. I used the server's privateKey.pem to sign the certificate (hence the generated 3rd file). Hence Server itself is a Certification Authority in this case (CA). Now what i understand from theory is: <1> During handshake client will request server's certificate. <2> on reception the client must verify if the ceritficate is signed by the mentioned trusted Certification Authority (CA) in the certificate. Since CA will have used a private key to sign the certificate client will need CA's public key stored somewhere on its machine. Use that public key file to verify that mentioned CA did indeed sign it. <3> if it is, then verify the certificate as a whole. In my case, the server had used it's own private key to self-sign the certificate. questions: <1> if CA was one of the trusted third parties the O.S would have stored it's public key in default locations. In that case i guess all i needed was ssl::context::set_default_verify_paths() and boost would have used all files there to match against the CA and use its public key. In this case however i need Server's public key file. Am i right? How do i get that (server's public key)? Do i use one of the 3 files above at the client end too? Is it load_verify_file("..") that i need to call pointing to server's public key? <2> context::set_verify_callback : does this get called during handshake? The handler must return bool. Suppose i return false, who will check it and what will it do? What i am guessing is: handshake -> blocks and runs a pre-verification test -> calls the supplied handler -> if handler returns true, handshake is successful otherwise either the handshake throws or fills the error code with non-zero value depending on what flavour of handshake(...) was called. Is this correct? <3> if i use ssl::rfc2818_verification("hostname") as verify_callback handler, what do i pass in place of hostname if there is no hostname (ie., i only know the server ip address). <4> do i have to use chain of files in this case? Because the server would just send one certificate file - the one that is signed right? -- View this message in context: http://boost.2283326.n4.nabble.com/Using-SSL-tp4644012.html Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

[graph][parallel] Problem distributed property map synchronization
by bminano 11 Mar '13

11 Mar '13

Hello, I need help. I have a distributed graph with a distributed property map for an edge value. The problem is that the synchronize procedure does not synchronize my ghost edges values. I have a loop that reads from the property, synchronizes, modifies its value and synchronizes it. What I see is that the value is not being synchronized for the ghost edges. for (int t = 0; t < 2; t++) { myfile<<rank<<":Time "<<t<<endl; BGL_FORALL_VERTICES(v, g, Graph) { for (boost::tie(ei_i,edge_end_i) = in_edges(v, g); ei_i != edge_end_i; ++ei_i) { remote_key_e k(rank, *ei_i); myfile<<"value "<<d_index[source(*ei_i, g)]<<" -> "<<d_index[target(*ei_i, g)]<<" valor "<<get(d_ecash, k)<<endl; } } synchronize(d_ecash); BGL_FORALL_VERTICES(v, g, Graph) { for (boost::tie(ei,edge_end) = out_edges(v, g); ei != edge_end; ++ei) { remote_key_e k(rank, *ei); put(d_ecash, k, t + 1); } } synchronize(d_ecash); } Could anyone tell me what I am doing wrong? I attach the full code if someone want to look at them. <nabble_a href="sync.cpp">sync.cpp -- View this message in context: http://boost.2283326.n4.nabble.com/graph-parallel-Problem-distributed-prope… Sent from the Boost - Users mailing list archive at Nabble.com.

3 3

[Graph][parallel] Scale-free graph generator without self-loops
by bminano 11 Mar '13

11 Mar '13

Hello, Does anyone knows if there is possible to generate a scale-free graph that does not generate self-loops (edges with the same vertex for source and target)? I am using the unique_rmat_iterator, but it generates some self-loops. I could remove them, but for the graph remains inconsistent because of a bug. I will apreciate your help. -- View this message in context: http://boost.2283326.n4.nabble.com/Graph-parallel-Scale-free-graph-generato… Sent from the Boost - Users mailing list archive at Nabble.com.

3 5

can't get first example in boost tutorial to work
by 7stud 11 Mar '13

11 Mar '13

I'm trying to install boost on mac osx 10.4.7. I followed the tutorial "Getting Started on Unix Variants" found here: http://www.boost.org/doc/libs/1_35_0/ more/getting_started/unix-variants.html Following the instructions in the tutorial, I installed boost in /usr/local as instructed, so my directory structure is: /usr/local/boost_1_34_1/boost But when I run the first example in the tutorial: #include <boost/lambda/lambda.hpp> #include <iostream> #include <iterator> #include <algorithm> int main() { using namespace boost::lambda; typedef std::istream_iterator<int> in; std::for_each(in(std::cin), in(), std::cout << (_1 * 3) << " " ); } I get the error(among others): error: boost/lambda/lambda.hpp: No such file or directory If I change the include path to: #include </user/local/boost_1_34_1/boost/lambda/lambda.hpp> I get the error(among others): error: boost/lambda/core.hpp: No such file or directory But core.hpp is in the lambda directory: /usr/local/boost_1_34_1/boost/lambda$ ls algorithm.hpp control_structures.hpp lambda.hpp bind.hpp core.hpp loops.hpp casts.hpp detail numeric.hpp closures.hpp exceptions.hpp switch.hpp construct.hpp if.hpp I'm using the Xcode 2.4 IDE to run my C++ programs. I also tried to execute the program from the command line with the original include path: #include <boost/lambda/lambda.hpp> but I get the same error: ~/2testing/dir1$ g++ prog1.cpp prog1.cpp:1:35: error: boost/lambda/lambda.hpp: No such file or directory prog1.cpp: In function 'int main()': prog1.cpp:8: error: 'boost' has not been declared prog1.cpp:8: error: expected namespace-name before ';' token prog1.cpp:8: error: '<type error>' is not a namespace prog1.cpp:11: error: '_1' was not declared in this scope I get the same errors again with the full include path: #include </user/local/boost_1_34_1/boost/lambda/lambda.hpp> $ g++ prog1.cpp In file included from prog1.cpp:1: /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:14:33: error: boost/lambda/core.hpp: No such file or directory /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:21:52: error: boost/lambda/detail/operator_actions.hpp: No such file or directory /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:22:61: error: boost/lambda/detail/operator_lambda_func_base.hpp: No such file or directory /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:23:63: error: boost/lambda/detail/operator_return_type_traits.hpp: No such file or directory /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:26:45: error: boost/lambda/detail/operators.hpp: No such file or directory /usr/local/boost_1_34_1/boost/lambda/lambda.hpp:30:46: error: boost/lambda/detail/member_ptr.hpp: No such file or directory prog1.cpp: In function 'int main()': prog1.cpp:8: error: 'boost' has not been declared prog1.cpp:8: error: expected namespace-name before ';' token prog1.cpp:8: error: '<type error>' is not a namespace prog1.cpp:11: error: '_1' was not declared in this scope What path should I be using? Or, should I set an environment variable somewhere?

6 16

adjacency matrix subgraph enumeration
by headupinclouds 09 Mar '13

09 Mar '13

What is the best way in boost to enumerate all connected subgraphs of an adjacency matrix? I know in advance that all nodes of the graph are connected and that it is cyclic. For the following graph: (0 <-> 1) (1 <-> 2) (2 <-> 3) (3 <-> 0) I want to enumerate the node list of all unique connected subgraphs: (0, 1) (0, 1, 2) (0, 1, 2, 3) (1, 2) (1, 2, 3) (2, 3) (0, 3) (0, 3, 2) -- View this message in context: http://boost.2283326.n4.nabble.com/adjacency-matrix-subgraph-enumeration-tp… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

is_foo_type naming rule
by Niitsuma Hirotaka 09 Mar '13

09 Mar '13

There are two type is_foo_type functor For example is_double can be --------- template<typename T> struct is_double_template : boost::is_same<T,double> {}; --------- or --------- struct is_double_runtime { template<typename T> bool operator()(T x){return false;} bool operator()(double x){return true;} }; --------- Are there any standard naming rule distinguish these functor? is_foo_runtime is_foo_compiletime ? The compile-time functor can convert to runtime factor using : --------- template<typename IsTemplateMplLambda> struct is_templated_type_to_runtime { typedef struct is_templated_type_runtime { template<typename T> bool operator()(T) { return boost::mpl::apply<IsTemplateMplLambda,T>::type::value; } } type; }; --------- usage --------- typedef boost::mpl::lambda<is_double_template<boost::mpl::_1 > >::type is_double_mpl_lambda; double x=1.1; int i=2; is_templated_type_to_runtime<is_double_mpl_lambda>::type is_double_runtime_object; std::cout << is_double_runtime_object(x) << std::endl; std::cout << is_double_runtime_object(i) << std::endl; ---------

3 3

newbie needs help with a linker problem
by Phil Huffman 09 Mar '13

09 Mar '13

Hello, I just installed Boost and GMP on my Mac, using all defaults. I want to use GMP via Boost, but there is a problem with the linker. I'm sure that clang is getting a bit lost looking for GMP's object modules. Any help will be gratefully accepted. Thanks. When I compile the below program: #include <boost/multiprecision/gmp.hpp> using namespace boost::multiprecision; int main() { mpz_int v = 1; for (unsigned i = 1; i <= 1000; ++i) { v *= i; } std::cout << "1000! = " << v << std::endl; // prints 1000! return 0; } using this command: c++ -I /usr/local/boost_1_53_0 mpz1.cpp -o mpz1 -v I get this: Apple clang version 4.1 (tags/Apple/clang-421.11.66) (based on LLVM 3.1svn) Target: x86_64-apple-darwin12.2.0 Thread model: posix "/usr/bin/clang" -cc1 -triple x86_64-apple-macosx10.8.0 -emit-obj -mrelax-all -disable-free -disable-llvm-verifier -main-file-name mpz1.cpp -pic-level 1 -mdisable-fp-elim -relaxed-aliasing -masm-verbose -munwind-tables -target-cpu core2 -target-linker-version 134.9 -v -resource-dir /usr/bin/../lib/clang/4.1 -I /usr/local/boost_1_53_0 -fmodule-cache-path /var/folders/hj/zc5pjqjd6gxgq2sw77xghkyh0000gn/T/clang-module-cache -fdeprecated-macro -fdebug-compilation-dir /Users/56phil/Dev/boostTest -ferror-limit 19 -fmessage-length 272 -stack-protector 1 -mstackrealign -fblocks -fobjc-runtime-has-arc -fobjc-runtime-has-weak -fobjc-dispatch-method=mixed -fobjc-default-synthesize-properties -fcxx-exceptions -fexceptions -fdiagnostics-show-option -fcolor-diagnostics -o /var/folders/hj/zc5pjqjd6gxgq2sw77xghkyh0000gn/T/mpz1-KE6NKq.o -x c++ mpz1.cpp clang -cc1 version 4.1 based upon LLVM 3.1svn default target x86_64-apple-darwin12.2.0 ignoring nonexistent directory "/usr/include/c++/4.2.1/i686-apple-darwin10/x86_64" ignoring nonexistent directory "/usr/include/c++/4.0.0" ignoring nonexistent directory "/usr/include/c++/4.0.0/i686-apple-darwin8/" ignoring nonexistent directory "/usr/include/c++/4.0.0/backward" #include "..." search starts here: #include <...> search starts here: /usr/local/boost_1_53_0 /usr/include/c++/4.2.1 /usr/include/c++/4.2.1/backward /usr/local/include /usr/bin/../lib/clang/4.1/include /usr/include /System/Library/Frameworks (framework directory) /Library/Frameworks (framework directory) End of search list. "/usr/bin/ld" -demangle -dynamic -arch x86_64 -macosx_version_min 10.8.0 -o mpz1 /var/folders/hj/zc5pjqjd6gxgq2sw77xghkyh0000gn/T/mpz1-KE6NKq.o -lstdc++ -lSystem /usr/bin/../lib/clang/4.1/lib/darwin/libclang_rt.osx.a Undefined symbols for architecture x86_64: "___gmp_get_memory_functions", referenced from: boost::multiprecision::backends::gmp_int::str(long, std::_Ios_Fmtflags) const in mpz1-KE6NKq.o "___gmpz_clear", referenced from: boost::multiprecision::backends::gmp_int::~gmp_int() in mpz1-KE6NKq.o "___gmpz_get_str", referenced from: boost::multiprecision::backends::gmp_int::str(long, std::_Ios_Fmtflags) const in mpz1-KE6NKq.o "___gmpz_init", referenced from: boost::multiprecision::backends::gmp_int::operator=(long) in mpz1-KE6NKq.o boost::multiprecision::backends::gmp_int::gmp_int() in mpz1-KE6NKq.o "___gmpz_mul_ui", referenced from: boost::multiprecision::backends::eval_multiply(boost::multiprecision::backends::gmp_int&, unsigned long) in mpz1-KE6NKq.o "___gmpz_set_si", referenced from: boost::multiprecision::backends::gmp_int::operator=(long) in mpz1-KE6NKq.o ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation)

2 1

building with iostreams and a non-standard zlib location
by James Amundson 09 Mar '13

09 Mar '13

Hi, I'm trying to do a bjam build of boost packages including iostreams. I have zlib installed under /unusual/location ./bjam include=/unusual/location/include install fails because it cannot find -lz. Obviously, I need to specify that libz is in /unusual/location/lib, but I don't see how to do that. I realize this seems obvious. Which part of the documentation am I missing? Thanks, Jim Amundson

4 3

VS2012 C4503 warning (decorated name length exceeded, name was truncated) with boost 1.52
by legalize+jeeves＠mail.xmission.com 08 Mar '13

08 Mar '13

[Please do not mail me a copy of your followup] I'm using boost 1.52. If you see this thread on stack overflow: <http://stackoverflow.com/questions/9669071/warning-c4503-when-using-pragma-…> You'll see a comment down at the bottom where someone has increased the warning level for their project and they are getting tons of these C4503 warnings when using boost includes. I have verified on my project that if I don't change the warning level and don't specifically disable this warning project-wide, then including using boost in my project does not generate the warning. However, if I increase the warning level beyond the default, to 3 in my case whereas the poster on stack overflow had used 4, then I get the warning. I, like the commenter on stack overflow, would like to know if there's a boost-friendly way to tell boost to disable the warning for boost headers where the boost authors have decided it's innocuous, while still retaining the high warning setting for my own code. Any suggestions? Thanks in advance. -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

2 2