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

Creating Boost unit system for astrodynamic equations
by Stephen Torri 23 Nov '11

23 Nov '11

I am starting to convert code I have written using standard types (e.g. double) which calculates the orbits of objects in space. Common man's way of describing astrodynamic equations. The equations fall into two types: dimensionless and with dimensions. For those that have a dimension they fall into three types: Length (earth radii) Time (minutes) Angle (radians) So derived types would be: Velocity (radians/minute) Now that I have described what types are required I have a few questions: 1. Where is the length type for already defined for earth radii? 2. What advantage is there to creating my own unit system? It seems that my unit system would have: - earth radii - kg (useless in the equations) - minute - radians Stephen

2 1

Re: [Boost-users] Scope exit with no capture
by Bruno Martinez 23 Nov '11

23 Nov '11

> I guess this feature will be integrated in ScopedExit soon :), but in > order to don't forget, maybe you could create a feature request on the > Trac system. I reopened https://svn.boost.org/trac/boost/ticket/3940. I also found http://lists.boost.org/Archives/boost/2009/03/149540.php which includes a patch to support zero arguments, but it wasn't ever merged. It seems scope exit is abandoned. Bruno

2 1

help boost multi_index_container + composite key
by Uthpal Urubail 23 Nov '11

23 Nov '11

All, I am experimenting boost for some specific requirement. The following code fails to retrieve the values. Could any one of you help me find out my mistake with the below code? #include <boost/multi_index_container.hpp> #include <boost/multi_index/ordered_index.hpp> #include <boost/multi_index/member.hpp> #include <boost/multi_index/composite_key.hpp> struct Elem_entry { unsigned int faceid; unsigned int node1; unsigned int node2; Elem_entry(const Elem_entry &obj){/*std::cout<<"copy const"<<std::endl;*/} Elem_entry():faceid(0),node1(0),node2(0){/*std::cout<<"defa const"<<std::endl;*/} Elem_entry(unsigned int faceid, unsigned int node1, unsigned int node2): faceid(faceid),node1(node1),node2(node2){} }; typedef multi_index_container< Elem_entry, indexed_by< //non-unique as some might have more than ids ordered_unique< composite_key< Elem_entry, member<Elem_entry,unsigned int,&Elem_entry::node1>, member<Elem_entry,unsigned int,&Elem_entry::node2> > >, ordered_unique< member<Elem_entry,unsigned int,&Elem_entry::faceid> > > > elementbook; int main() { elementbook pb; Elem_entry fc1 = Elem_entry(1,1,2); pb.insert(fc1); Elem_entry fc2 = Elem_entry(2,2,3); pb.insert(fc2); Elem_entry fc3 = Elem_entry(3,3,1); pb.insert(fc3); int faceId = 0; elementbook::iterator it; //Below code not working it = pb.find(boost::make_tuple(2,3)); if(it != pb.end()) { faceId = it->faceid; bool valid = true; valid = false; } for (elementbook::iterator iter = pb.begin(), i_end = pb.end(); iter != i_end; iter = pb.upper_bound(iter->faceid)) { std::cout<<iter->faceid<<"\t"<<iter->node1<<"\t"<<iter->node2<<std::endl ; } }

2 1

[fusion] incremental zipping
by Gennadiy Rozental 23 Nov '11

23 Nov '11

Hi, I am zipping fusion sequences in a series of incremental zip operations: zip(zip(zip(a,b),c),...) as a result I am getting sequence with elements like: vector2<vector2<vector2<a[i],b[i]>,c[i]>,...> what I want instead is that elements would look like: vector<a[i],b[i],c[i],...> What the simplest way to achieve this? Gennadiy

4 8

boost non-blocking read write routines
by Jack Bryan 23 Nov '11

23 Nov '11

Hi, I am coding for data exchange between threads by boost C++. Some threads need to read or write some data structure asynchronously. For example, we have thread 1 and thread 2. When thread 1 needs to get some data from thread 2, if the data is still not available for thread 1, it should not be blocked and can do other things and then go back to check the data again later. The same things are for thread 2. What are the API routines of boost non-blocking read write , which can do that ? I cannot find it in boost document. thanks

2 1

How can I have an object-specific and thread-specific pointer?
by Phillip Hellewell 22 Nov '11

22 Nov '11

Hi, My class contains a member var that is a little buffer used as a cache. Instances of this class are used concurrently (referenced through a shared_ptr). I would like to switch my buffer to a thread_specific_ptr, but there are two big problems with using thread_specific_ptr as a member var: 1. There is a memory leak when my object goes away, because by design only one of the pointers gets cleaned up (the one for the thread that happens to delete the object). 2. There is undefined behavior when a new object gets created that happens to be allocated at the same address as a previously freed one. get( ) could return a pointer to a buffer that goes with an old (now freed) object. I try not to question the wisdom of the thread_specific_ptr authors, but given these two issues, especially #2, I have to conclude that thread_specific_ptr is only suitable for statics / globals. It is wholly unfit to be used as a member variable (except in a singleton class I guess). The next best thing I can think of for my class is to just have a std::map<thread_id, void*>, with synchronized access of course, and I clean up all the pointers in the map in the DTOR. That would work just great, except I really do need the pointers to also get cleaned up (and removed from my map) when the corresponding threads go away. Some of my objects can live a very long time, whilst threads get created and destroyed off and on, so without this functionality the map could grow indefinitely. So I'm stuck. I can't use thread_specific_ptr because it won't clean up everything when my object goes away, and I can't use a map because it won't clean up anything when a thread goes away. Is there an easy solution to this? If there were some way for me to hook up a callback directly to the thread cleanup function, I could add a callback to remove it from my map when the thread goes away. But I'd also need the ability to remove that callback in my DTOR. Does boost provide any way to do this? I can't be the first person to want a thread-specific, object-specific pointer, can I? Phillip

4 11

Scope exit with no capture
by Bruno Martínez 22 Nov '11

22 Nov '11

Hi, How can I call BOOST_SCOPE_EXIT when no variable needs to be captured? BOOST_SCOPE_EXIT () { CoUninitialize(); } BOOST_SCOPE_EXIT_END; doesn't work. I tried BOOST_SCOPE_EXIT (BOOST_PP_NIL) BOOST_SCOPE_EXIT (BOOST_PP_EMPTY) BOOST_SCOPE_EXIT (BOOST_PP_SEQ_NIL) but they don't work either. Bruno

2 1

Re: [Boost-users] [Thread] program crashes from time to time when deleting unique_future<void>
by Alex Perry 22 Nov '11

22 Nov '11

On Mon, 21 Nov 2011 16:40:04 +0000 "Soderback, Johan" <Johan.Soderback(a)elekta.com> wrote :- > Hi, > > I have a small example program that crashes from time to time. > I would appreciate some help to understand why. > > In function run() executed on the main thread I create a packaged_task (pt), > put it in a queue (_q) and wait for it to get executed (fu.get()). > In another thread the tasks are executed when the queue _q is populated. > The task I'm running always throws/rethrows intentionally. > > It results (from time to time) in an access violation when the unique_future, > fu, is deleted. > See the stack trace at the bottom. > > Anybody see any flaws in this design? > > I have tried 1.47.0 and 1.48.0, MSVC 2005. > Trying your example on vs2010 it crashes continuously - changing your throwf() function to the following appears to sort it - not quite sure why a simple throw fails though - or whether this was your problem under vs2005 void throwf() { try { BOOST_THROW_EXCEPTION(TheException()); } catch (boost::exception const & ) { boost::rethrow_exception( boost::current_exception() ); } } HTH Alex

3 2

[serialization] backwards compatibility of vector (de)s11n
by Ewald Peiszer 22 Nov '11

22 Nov '11

Hi list, I am migrating from Boost v1.35 to a new version. I have, however, some legacy serialized strings that I want to be readable by the new version. After some testing I found that apparently the newer version (in this case 1.40) cannot correctly deserialize std::vector instances from 1.35 generated strings. Can anyone give my a pointer as to what may be the problem here? Thanks a lot! Ewald ==Details== I use text archives, and gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu9) The test vector is a structure with 2 items, the long integers 1 and 2. Archive string for respective lib versions: 1.35 archive: 22 serialization::archive 4 2 1 2 1.40 archive: 22 serialization::archive 5 2 0 1 2 Now: ---------(code snippet) // includes #include <boost/serialization/vector.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> std::vector<long> testvector; std::string val = "22 serialization::archive 4 2 1 2"; // v1.35 archive { std::stringstream ss(std::stringstream::in | std::stringstream::out); ss << val << std::endl; boost::archive::text_iarchive ia(ss); ia >> BOOST_SERIALIZATION_NVP(testvector); } -------- Result when executing with lib v1.40: testvector contains two elements, {2, 2} -- instead of {1, 2} Note: s11n and de-s11n with the same lib version works perfect.

1 0

boost.uuid typo error
by niXman 22 Nov '11

22 Nov '11

Hello list! In the docmantation for boost.uuid on this link: http://www.boost.org/doc/libs/1_48_0/libs/uuid/uuid.html#Stream_operators in the example of usage boost::lexical_cast<> with boost::uuids::uuid: boost::uuids::uuid u1; // initialize uuid std::string s = boost::lexical_cast<std::string>(u); // <<<<<<<<<<<< must be u1 boost::uuids::uuid u2 = boost::lexical_cast<boost::uuids::uuid>(s); assert(u1 == u2); I wish that boost documentation was famous as well as Qt documentation ;) Regards, niXman.

4 4