- Boost-users - lists.boost.org

[Boost-Interest][General] Will Boost C++ work for my device control software
by Boost lzw 20 Oct '11

20 Oct '11

Hi everyone, I am soon to start a project and am wondering if I can use Boost C++ for it. Basically, I am going to develop software to control a variety of devices. The "Big Picture" of it is this: the software will "talk" to each of these devices via their drivers from different "interfaces" - USB port, serial port, etc., and collecting measurement data from these devices and controlling them. My "Big Picture" questions are: is it possible for me to use Boost C++ for these tasks. If so, how? Any links to where I should look at in the Boost documentation would be welcomed. Are there any specific Boost library that I should study before starting the project. Thanks in advance. Best Robert

2 1

serialization of STL containers with non-default allocator
by Jonathan Bishop 20 Oct '11

20 Oct '11

In my code I have used a non-default allocator MyAlloc for the STL classes. For instance, typedef std::basic_string< char, std::char_traits<char>, MyAlloc<char> > string ; Archive ar; ar & s; // compile error Is there a way to make this work. I know there is a builtin serialization for std::string which I would like to use. Thanks

1 0

[Boost.Thread] thread started after constructor
by gast128 20 Oct '11

20 Oct '11

Hello all, I have a basic question about threads: can one assume that the thread is started after the constructor call (or when using create_thread in thread groups) or does one has to use an extra synchronization primitive (e.g. a barrier)? Example (real world is more complex): void g(int* x) { int x1 = *x; } void f() { int x = 0; boost::thread thd(&g, &x); //<-- thread is started or not? }

1 1

[thread]N of args for thread::thread with args
by Marcello Puligheddu 20 Oct '11

20 Oct '11

In http://www.boost.org/doc/libs/1_47_0/doc/html/thread/thread_management.html Thread Constructor with arguments<http://www.boost.org/doc/libs/1_47_0/doc/html/thread/thread_management.html…> [...] Note: Currently up to nine additional arguments a1 to a9 can be specified in addition to the function f. But in Launching threads<http://www.boost.org/doc/libs/1_47_0/doc/html/thread/thread_management.html…>is clearly stated "There is an unspecified limit on the number of additional arguments that can be passed." -- marcello puligheddu Cagliari ,Sardegna, Italia +393203090251

3 4

boost-python stringifying
by immanuel litzroth 20 Oct '11

20 Oct '11

Hello, I've got a object and I need to stringify it in boost-pyton bridge.... actually just the equivalent of calling str(obj) in python. I have a hard time finding out how to do this from the documentation. Thanks in advance for any pointers, Immanuel

1 0

Is there Text Diff in Boost Unit Testing?
by Alexander Feldman 19 Oct '11

19 Oct '11

Hi Guys, I am now using Boost Unit Testing as well and I am quite happy. I am writing, among other things, a compiler and I would benefit if in BOOST_CHECK_EQUAL there is some kind of a text diff... Is there such a facility? Thanks, -- Alex

2 1

Proposal mpl::switch_case
by Hite, Christopher 19 Oct '11

19 Oct '11

My use case is a fast protocol decoder. I've described the protocol in mpl and generate the instructions I need to decode/encode/store packets. During decoding I have to be able to take a run time id and map it to the correct compile time generated code. Traditionally such code was written as a switch/case. The obvious solution is to use metaprogramatic recursion, for example mpl::for_each to check the id against each value. You would hope the compiler would inline all the recursion, detect the ifs on an id, and do general switch/case optimization. Unfortunately the code generated by gcc and it's not as good as I hoped. They actually generate a function for every ~6 levels of recursion and just do a bunch of if/equal/else. boost::variant has this same problem. They use BOOST_PP_REPEAT to generate an actual switch/case statement(visitation_impl.hpp). The compiler does a good job with the swich/case. It removes redundant cases and then optimizes using array lookups and binary searches depending on size. Good job! 1) Could we factor this logic out into a utility like mpl::switch_case? 2) What would the best interface look like? 3) What other bits of code could benefit from it? phoenix::switch_? Meta State Machine? Chris

2 1

[serialization] unregistered_class exception that wasn't occurring before
by Igor R 19 Oct '11

19 Oct '11

Hello, I'm serializing the following type: (pseudo-definition) multi_index_container< variant< shared_ptr<template1<arg1> >, shared_ptr<template2<arg2> >, etc... > > It works well when I make and load archives with the current version of my app, but when I try to open an archive made by the pervious version it throws unregistered_class for one of my classes. The both versions use boost 1.47, and the differences between them seem to me absolutely unrelated to serialization (as far as I can see). Unfortunately I can't provide here code snippets, so my question is theoretical: what could break the backward compatibility in such a way? What should I look into? Thanks.

2 4

[shared_ptr] implicit conversion to bool problem
by Steckmann, Sven 19 Oct '11

19 Oct '11

Dear list, I have a problem with the implicit conversion to bool in the shared_ptr class. I want to use operator overloading for an easy to use interface. I have 2 Instances of a class with a common Interface and I want to define an operator-class (OpAnd of type K) which can handle the operator. By compiling this little test below, I got: g++ -I/home/steckmann/proj/libs/boost -O0 -g3 -Wall -c -fmessage-length=0 -MMD -MP -MF"test.d" -MT"test.d" -o "test.o" "../test.cpp" ../test.cpp: In function 'int main()': ../test.cpp:49: error: ambiguous overload for 'operator&&' in 'a && b' ../test.cpp:49: note: candidates are: operator&&(bool, bool) <built-in> ../test.cpp:40: note: boost::shared_ptr<K> operator&&(boost::shared_ptr<K>, boost::shared_ptr<K>) This is quite clear to me, but how can I avoid this without touching the shared_ptr? Thanks for any comments, Sven #include <boost/shared_ptr.hpp> using namespace boost; class K { public: virtual bool value() = 0; }; class Value :public K { public: Value( bool i ) : var( i ) {} bool value() { return var; } bool var; }; class OpAnd : public K { public: OpAnd( boost::shared_ptr<K> l, boost::shared_ptr<K> r ) :l(l), r(r) {} bool value() { return l->value() && r->value(); } boost::shared_ptr<K> l; boost::shared_ptr<K> r; }; boost::shared_ptr<K> operator&&( boost::shared_ptr<K> l, boost::shared_ptr<K> r) { return shared_ptr<K>( new OpAnd(l, r ) ); } int main() { boost::shared_ptr<Value> a( new Value(true) ); boost::shared_ptr<Value> b( new Value(false) ); boost::shared_ptr<K> c = a && b; // Now use c here for some stuff c->value(); // returns false b->var = true; c->value(); // returns true return 0; }

2 1

Why does the creating of distributed adjacency list cost so much time?
by Yan 19 Oct '11

19 Oct '11

Hi all, I'm learning to use the distributed adjacency list and distributed dijkstra SP-algorithm. I use some test data contains about 3,800,000 edges and 2,800,000 vertices. I use a pair of iterators to pass the data to the distributed adjacency list, like that: Graph g(reader.begin(), reader.end(), reader.weight_begin(), reader.num_vertices(), pg, dist); "reader" is an user-defined class which supplys required data (edges, vertices, weights). "pg" is defined like that "mpi_process_group pg;" "dist" is supplied by the reader like that "graph::metis_distribution dist = reader.distribution();". It is based on a metis partition file coordinating the test data. I run 2 processes on my notebook computer(Intel core2 2.0G, Ram 2G, WinXPsp3). I also test the same data using a sequential bgl dijkstra SP-algorithm on this computer. The result is that: Parallel bgl: Creating of Graph g costs *10080 seconds*; dijkstra computing costs 28.281 seconds. Sequential bgl: Creating of Graph g costs *4.563 seconds*; dijkstra computing costs 1.703 seconds. As you see, when using distributed adjacency list, it costs so much time to create the graph. Is it nomal? I have done some more comparison between 2-processes parallel dijkstra on one core2 computer and sequential dijkstra on the same computer, using different scale data set. Always, the sequential bgl dijkstra costs less time than the 2-processes parallel dijkstra. What's the problem? I have read the boost document on "boost_1_46_1\libs\graph_parallel\doc\html\dijkstra_shortest_paths.html", which shows a good performance and speedup in the charts. Doesn't that mean less time costing than the sequential program? Any advice would be appreciated. Thanks, Yan -- View this message in context: http://boost.2283326.n4.nabble.com/Why-does-the-creating-of-distributed-adj… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2