- Boost-users - lists.boost.org

BGL: write_graphviz and bundled properties
by Fred Labrosse 20 Apr '06

20 Apr '06

All, First, I'm very new to boost and BGL, so please forgive me if this is trivial. I've created a graph (adjacency_list) using bundled properties and am now trying to write it as a dot file. I've written my property writer, but when I compile, I get several of: /usr/include/boost/graph/graphviz.hpp: In function `void boost::write_graphviz(std::ostream&, const Graph&, VertexPropertiesWriter, EdgePropertiesWriter, GraphPropertiesWriter, VertexID) [with Graph = boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, VertexPropertiesWriter = map::placeWriter, EdgePropertiesWriter = boost::default_writer, GraphPropertiesWriter = boost::default_writer, VertexID = boost::adj_list_vertex_property_map<boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, boost::detail::error_property_not_found, const boost::detail::error_property_not_found&, boost::vertex_index_t>]': /usr/include/boost/graph/graphviz.hpp:260: instantiated from `void boost::write_graphviz(std::ostream&, const Graph&, VertexPropertiesWriter, EdgePropertiesWriter, GraphPropertiesWriter) [with Graph = boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, VertexPropertiesWriter = map::placeWriter, EdgePropertiesWriter = boost::default_writer, GraphPropertiesWriter = boost::default_writer]' /usr/include/boost/graph/graphviz.hpp:278: instantiated from `void boost::write_graphviz(std::ostream&, const Graph&, VertexWriter) [with Graph = map::mapType, VertexWriter = map::placeWriter]' topologicalMap.h:282: instantiated from here /usr/include/boost/graph/graphviz.hpp:241: error: no match for 'operator<<' in 'out << boost::get [with PropertyMap = boost::adj_list_vertex_property_map<boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, boost::detail::error_property_not_found, const boost::detail::error_property_not_found&, boost::vertex_index_t>, Reference = const boost::detail::error_property_not_found&, K = void*](((const boost::put_get_helper<const boost::detail::error_property_not_found&, boost::adj_list_vertex_property_map<boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, boost::detail::error_property_not_found, const boost::detail::error_property_not_found&, boost::vertex_index_t> >&)((const boost::put_get_helper<const boost::detail::error_property_not_found&, boost::adj_list_vertex_property_map<boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, boost::detail::error_property_not_found, const boost::detail::error_property_not_found&, boost::vertex_index_t> >*) ((boost::put_get_helper<const boost::detail::error_property_not_found&, boost::adj_list_vertex_property_map<boost::adjacency_list<boost::listS, boost::listS, boost::undirectedS, map::place, map::link, boost::no_property, boost::listS>, boost::detail::error_property_not_found, const boost::detail::error_property_not_found&, boost::vertex_index_t> >*) (&vertex_id)))), ((void* const&)((void* const*) (+(&i)->std::_List_iterator<_Tp>::operator* [with _Tp = void*]()))))' /usr/lib/gcc/x86_64-pc-linux-gnu/3.4.5/include/g++-v3/bits/ostream.tcc:63: note: candidates are: std::basic_ostream<_CharT, _Traits>& std::basic_ostream<_CharT, _Traits>::operator<<(std::basic_ostream<_CharT, _Traits>&(*)(std::basic_ostream<_CharT, _Traits>&)) [with _CharT = char, _Traits = std::char_traits<char>] Not being very good at deciphering all that, I'm not sure what is happening. I do understand that it's missing an extraction operator but I'm not sure on what. The actual line in graphviz.hpp is: out << get(vertex_id, *i); Two questions: - Could somebody tell me exactly what that get returns? - From the dot format, this should be the name of the vertex in the dot file. So this seems to relate to the class used as a property. My second question is then: If this is the case, why then do we need the property writer since we apparently also need the operator<< that could do exactly the same thing? (And isn't a ' ' (space) missing between the vertex name and whatever the property writer writes (or should that be included in it, which is not mentioned by the documentation)? That was 3 questions then ;-) Thanks in advance for any replies. Cheers, Fred

2 4

lambda/bind trouble
by Sebastian Weber 20 Apr '06

20 Apr '06

Hello everybody! I'm starting to use the boost lambda library. My first steps with it were quite amazing, but now I'm comming to some more advanced problems and I don't see my fault. The code in question is std::for_each(vertices(bg).first, vertices(bg).second, bind(ll::fill_n(), // line 197, supposenly wrong std::back_inserter(vs), bind(out_degree, _1, bg), _1)); which does not compile with the error message: box_cover.hpp: In function `size_t boost::box_cover(Graph&, RandomGenerator&, BoxMap, unsigned int, unsigned int)': box_cover.hpp:197: error: parse error before `::' token "g++-3.3" -Wall -ftemplate-depth-100 -msse2 -mfpmath=sse,387 -O3 -finline-functions -Wno-inline-fPIC -msse2 -mfpmath=sse,387 -DBOOST_DATE_TIME_DYN_LINK=1 -DDATE_TIME_INLINE -DNDEBUG -I"." -I"/home/sebi/projects/diplom/src/boost_1_33_1" -c -o "../../bin/graph/gcc-3.3/release/boxTest.o" "boxTest.cpp" I thought, I could realize this stuff with the boost::bind library as well and tried the construct std::for_each(vertices(bg).first, vertices(bg).second, boost::bind(std::fill_n, std::back_inserter(vs), boost::bind(boost::out_degree, _1, bg), _1) ); which does not compile as well: box_cover.hpp: In function `size_t boost::box_cover(Graph&, RandomGenerator&, BoxMap, unsigned int, unsigned int) [with Graph = main(int, char**)::graph_t, RandomGenerator = base_rand_gen_t, BoxMap = size_t*]': boxTest.cpp:140: instantiated from here box_cover.hpp:202: error: no matching function for call to `bind(<unknown type>, boost::arg<1>&, boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS, boost::no_property, boost::no_property, boost::no_property, boost::listS>&)' "g++-3.3" -Wall -ftemplate-depth-100 -msse2 -mfpmath=sse,387 -O3 -finline-functions -Wno-inline-fPIC -msse2 -mfpmath=sse,387 -DBOOST_DATE_TIME_DYN_LINK=1 -DDATE_TIME_INLINE -DNDEBUG -I"." -I"/home/sebi/projects/diplom/src/boost_1_33_1" -c -o "../../bin/graph/gcc-3.3/release/boxTest.o" "boxTest.cpp" Any help would be greatly appreciated. Thanks a lot! Almost forgot: My Boost version is 1.33.1 and the compiler I use is g++ 3.3.5 from Debian sarge. Greetings, Sebastian

1 0

[weak_ptr] thread safe?
by Alain Cormier 19 Apr '06

19 Apr '06

Hello group! We are experiencing problems with boost::weak_ptr in a multi-threaded environment. It looks that the lock on the ref counting doesn't work well with weak_ptr in release build. configuration : Windows 2000 SP4 VC6 SP5 boost 1.33.1 In summary, it seems that add_ref_lock in sp_counted_base_w32.hpp has a bug in release. In disassemblies, we observe : 01 lea esi,[eax+4] 02 mov eax,dword ptr [esi] 03 test eax,eax 04 je TestWeakPtr+0E4h (00401244) 05 lea ecx,[eax+1] 06 mov edx,esi 07 lock cmpxchg dword ptr [edx],ecx 08 mov ecx,eax 09 cmp ecx,eax 10 je TestWeakPtr+7Dh (004011dd) At line 08, you'll see that we move eax into ecx and after (line 09) we compare ecx and eax that are obviously the same which will destruct prematurely our pointer. In attachement, I send a complete program reproducing the bug with VC6 in release. It is probably a bug in VC6. But is it caused by a bad use? is it simply a "bug" in boost 1.33.1? By the way, we can remove the "bug" by simply "patching" add_ref_lock implementation like this : bool add_ref_lock() // true on success { for( ;; ) { long tmp = static_cast< long const volatile& >( use_count_ ); if( tmp == 0 ) return false; long tmp2 = tmp + 1; if(InterlockedCompareExchange ( &use_count_, tmp2, tmp ) == tmp2 - 1) return true; } } Thank you for any suggestion! -- Alain Cormier alain.work(a)gmail.com

2 4

Re: [Boost-users] [weak_ptr] thread safe?
by Sohail Somani 19 Apr '06

19 Apr '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Michael Fawcett > Sent: Wednesday, April 19, 2006 10:09 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] [weak_ptr] thread safe? > > On 4/19/06, John Wilkinson <jwilkinson(a)tsystem.com> wrote: > > There is a pragma in VC6 that will activate or deactivate > optimization. > > It can be used to wrap optimization-sensitive blocks of > code. I do not > > remember what it is, though, and I do not have VC6 handy. > I have used > > it, and it did alleviate an optimization-related bug. > > >From the MSDN: > > #pragma optimize( "[optimization-list]", {on | off} ) > > Specifies optimizations to be performed on a function-by-function > basis. The optimize pragma must appear outside a function and takes > effect at the first function defined after the pragma is seen. The on > and off arguments turn options specified in the optimization-list on > or off. I'm curious, why couldn't this be a compiler workaround in weak_ptr?

2 1

[Pool] Using purge_memory() with std::map
by Finley Lee 19 Apr '06

19 Apr '06

Dear list, I'm using boost::fast_pool_allocator to replace the default allocator in a map. This works great, but I'm having a bit of trouble freeing the memory explicitly. I'd like to call boost::singleton_pool::purge_memory(), but I don't know how to instantiate the correct version. Is there a way to determine the size of the chunks the map allocator will require? I have determined by debugging that the size is 28 bytes, but I imagine that this is dependent on the particular implementation of map. Below is a sample program which illustrates the situation. Any guidance would be very appreciated. - Finley //////////////////////////////////////////////////////////////////////// /// #include <assert.h> #include <iso646.h> #include <map> #include <boost\pool\pool_alloc.hpp> class mydata { public: mydata() {} mydata( int _a, short _b, char _c ) : a(_a), b(_b), c(_c) {} int a ; short b ; char c ; }; typedef std::map< int, mydata, std::less<int>, boost::fast_pool_allocator< std::pair<int, mydata> > > mymap ; int _tmain(int argc, _TCHAR* argv[]) { { mymap m ; for ( int i = 0 ; i < 10000 ; i++ ) m[i] = mydata( i, i, i ) ; } bool memory_freed = boost::singleton_pool< boost::fast_pool_allocator_tag, sizeof( std::pair<int,mydata> ) >::purge_memory() ; //bool memory_freed = boost::singleton_pool< boost::fast_pool_allocator_tag, 28 >::purge_memory() ; assert( memory_freed ) ; return 0; }

3 2

scoped locks
by Paul Baxter 19 Apr '06

19 Apr '06

Having looked at Boost threads and shmem 0.9 recently, I'm trying to get to grips with a 'scoped lock' ONLY design rationale having laboured under the posix threads APIs for a while. While I appreciate the reasons, I am a little confused ... All the examples I've seen show scoped locks being constructed local to a function and then destroyed automatically at the end of function scope. I would like my own lock manager object to manage a group of locks and allow me to leave one or more mutexes locked across function calls until the lock manager object goes out of scope and releases any mutexes still flagged as locked. Is this possible? e.g. a vector/array/deque of pointers to scoped_timer_locks each relating to its own mutex and constructed as unlocked. I understand that leaving mutexes locked is not a great idea! The application does think it knows what it is doing. Paul

4 3

boost::hash
by george 19 Apr '06

19 Apr '06

hello! I wanna replace an std::map<std::string,Services*> to hash table.How I will do this? in the examples of hash library there are only examples for custom hashing.... there isn't a trivial example!:-) transactions.insert(std::pair<std::string,Services*>("Aservice",new xxxx)); transactions.insert(std::pair<std::string,Services*>("Vservice",new fffff)); transactions.insert(std::pair<std::string,Services*>("Rservice",new ttt)); ............... how I should replace this with a boost 'hashtable'? thanks for any help.

1 1

Re: [Boost-users] [weak_ptr] thread safe?
by John Wilkinson 19 Apr '06

19 Apr '06

There is a pragma in VC6 that will activate or deactivate optimization. It can be used to wrap optimization-sensitive blocks of code. I do not remember what it is, though, and I do not have VC6 handy. I have used it, and it did alleviate an optimization-related bug. John -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Peter Dimov Sent: Wednesday, April 19, 2006 10:45 AM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] [weak_ptr] thread safe? Alain Cormier wrote: > So, can I conclude that it is a bad behavior in boost::weak_ptr? No, it is a bug in VC6. > For now, I have simply changed add_ref_lock to : > > long tmp2 = tmp + 1; > if(InterlockedCompareExchange ( &use_count_, tmp2, tmp ) == tmp2 - 1 ) > return true; > > This way, VC6 doesn't optimise the return value of > InterlockedCompareExchange and the disassembly looks ok. I don't think that there is any guarantee that this modification will reliably produce correct code, even if it happens to work in the examples we try. But if we can't think of a better way to work around the issue, this may be our best bet. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

3 2

using lambda as function argument.
by chun ping wang 19 Apr '06

19 Apr '06

Heres a simple question, that i want to do. Lets say you have a function call void f(double); and then you want to apply f to each argument inside vector<double> a; for_each(a.begin(), a.end(), f(_1)); // the example above isn't possible..is there a simple way to do this?

2 4

[Xpressive] regex-like boost libraries vs re2c [was: Some timings [was: Baffled about results of search]]
by Lynn Allan 19 Apr '06

19 Apr '06

Stuart Dootson wrote: > Lynn - it might be interesting to use re2c (http://re2c.org/) to > generate regex machines and see what their performance is like - > after > all, they do clain 're2c generates warning free code that is equal > to > hand-written code in terms of size, speed and quality.' (I'm > offering > no opinion on the veracity of that statement....), so (allegedly) > they'd give a reasonable hand-code analogue.... I took a look at it, and it looks like it could possibly be quite useful if executable size was a concern .... but it is pretty obscure (at least to me) .... the only example this dummy could make any sense out of at all was to recognize a number .... and that was missing/obscuring the important illustration of how to invoke and use the generated scanner code. Were you able to get it to work to do anything even close to recognizing Days-of-the-Week (full and abbreviations)? Did you find actual compilable examples with "main" invoking the generated scanners with a test-string to crunch? <alert comment="whining"> I really wish developers would provide COMPLETE "baby steps with training wheels" examples to flatten the learning curve. Eric N. did a great job of documenting and providing examples for xpressive .... that could be a model for other projects (not just boost). I've subscribed to their sourceforge list and will swallow my pride and ask the newbie question: "This dummy doesn't expect to be able to figure out your project in 30 minutes ... how do you get it to something basic like recognize:"? ((Sunday|Sun)|(Monday|Mon)....(Saturday|Sat)) </alert>

3 5