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

MultiArray questions
by Dr Mark H Phillips 24 Mar '04

24 Mar '04

Hi, I've recently found out about the existence of MultiArray and I am quite impressed! I was thinking of implementing something similar myself but now hopefully I don't have to! However I have a few questions that I'm hoping someone can answer. Question 1 ---------- In the user documentation there is an example which has some example code: for(index i = 0; i != 3; ++i) for(index j = 0; j != 4; ++j) for(index k = 0; k != 2; ++k) A[i][j][k] = values++; Why can't you change the != into <? Ie, do: for(index i = 0; i < 3; ++i) for(index j = 0; j < 4; ++j) for(index k = 0; k < 2; ++k) A[i][j][k] = values++; or can you? This would be more efficient I should think, and requiring that index types be ordered shouldn't be too onerous should it? Question 2 ---------- If I use the built-in "no-frills" C array type to represent a 2-dimensional array in row-major order, then I would access an element (i, j) by doing: a[ &a[0] + i*n + j ] where a is my m by n array. And if I wish to traverse a, in row-major order, I can do: for (int* ip=&(a[0]); ip<&(a[n*m]); ip+=n) { for (int* ijp=&(ip[0]); ijp<&(ip[n]); ++ijp) { visit(ijp); } } These are a very efficient ways of doing things, but it lacks somewhat in ease-of-generalization and in readbility. This is where the MultiArray library is useful. But do I lose in efficiency by going to MultiArray? Ie, in the simple case where base is 0, stride is 1, and when we traverse in row-major order, is the MultiArray library smart enough to do compile-time optimizations to generate code which is as efficient as the above? I am guessing that this sort of approach is what the Blitz++ library does?? Question 3 ---------- Is it possible to construct a new multi_array/multi_array_ref/const_multi_array_ref from an array_view by doing a deep copy of the view? A view is a nice way of looking at different parts of a MultiArray object in a readable way, but sometimes you might wish to turn this view into a more-efficiently represented object. Is this easy to do in the MultiArray library? Question 4 ---------- Sometimes you might have an object starts off being mostly accessed in row-major order, so you want a representation which works efficiently for this use. But down the track you might wish to change to mostly column-major access. Does MultiArray provide easy ways to transform from one underlying representation to another? Thanks, Mark Phillips.

3 2

RE: [Boost-users] boost::filesystem::exists() & GetfileAttributesA
by Ben Hutchings 24 Mar '04

24 Mar '04

Sebastian Orfino <sorfino(a)lanacion.com.ar> wrote: > I am experiencig a problem with exists() and its Windows > implementation with GetFileAttributes API. I am using exists() prior > to the scan of a directory to see if it exists; the fact is that on > W2K (just tested it on that OS) after a day or two > GetFileAttributesA() returns INVALID_FILE_ATTRIBUTES but the path > does exists (it is a UNC-style path). > After a restart of the app, all works fine. <snip> It couldn't possibly be related to this bug, could it: <http://support.microsoft.com/default.aspx?scid=kb;en-us;325929>? I know that that talks about Windows XP clients and NT 4 servers but MS does not always test thoroughly whether a bug found in one version of a product exists in other versions. The bug is supposed to be fixed in Windows XP SP2, by the way.

1 0

It always show compiler limitation when I use boost.python to wrap a class under MSVisualStudio .Net
by Shunming Fang 24 Mar '04

24 Mar '04

Hello, guys, I try build my c++ lib as python modules using boost.python, when compile one big class, it always show compiler limit( my ram already update to 1.5 GB) while the class can be smoothly build on other platforms like SGI, Linux(RAM 512 MB+). I don't know why? I already upgrade my MSVS .net to 2003Version, Anybody know How I can get around this problems? Thank you very much __________________________________ Do you Yahoo!? Yahoo! Finance Tax Center - File online. File on time. http://taxes.yahoo.com/filing.html

1 0

Boost error on Red Hat Linux 7.4
by Jonathan Arnold 24 Mar '04

24 Mar '04

I'm trying to build the latest (1.31.0) Boost libraries for the first time on a Red Hat Linux 7.4, gcc 2.96 system. It seems to build all the libs and such correctly, but I get the following error trying to build the test programs: gcc-C++-action bin/boost/libs/test/build/libboost_test_exec_monitor.so/gcc/debug/shared-linkable-true/test_tools.o /home/jdarnold/boost_1_31_0/libs/test/build/../src/test_tools.cpp: In method `void boost::test_toolbox::detail::print_log_value<char>::operator() (ostream &, char)': /home/jdarnold/boost_1_31_0/libs/test/build/../src/test_tools.cpp:194: parse error before `__ctype_b' /home/jdarnold/boost_1_31_0/libs/test/build/../src/test_tools.cpp:205: confused by earlier errors, bailing out set -e g++ -c -Wall -ftemplate-depth-100 -DBOOST_TEST_NO_AUTO_LINK=1 -g -O0 -fno-inline -fPIC -I"bin/boost/libs/test/build" -I "/home/jdarnold/boost_1_31_0" -o "bin/boost/libs/test/build/libboost_test_exec_monitor.so/gcc/debug/shared-linkable-true/test_tools.o" "/home/jdarnold/boost_1_31_0/libs/test/build/../src/test_tools.cpp" /usr/bin/objcopy --set-section-flags .debug_str=contents,debug "bin/boost/libs/test/build/libboost_test_exec_monitor.so/gcc/debug/shared-linkable-true/test_tools.o" What is going wrong here? Is my version of F++ too old? Thanks in advance. -- Jonathan Arnold (mailto:jdarnold@buddydog.org) Daemon Dancing in the Dark, a FreeBSD weblog: http://freebsd.amazingdev.com/blog/

1 0

Cannot use BGL & boost::format together under VS2003
by Alexeev Gleb 24 Mar '04

24 Mar '04

Hi all! The following simple sample doesn't compile on VC++ .NET 2003: #include <iostream> #include <boost/format.hpp> // without the following line everything's ok #include <boost/graph/adjacency_list.hpp> int main(){ std::cout << boost::format("%d, %s") % 5 % "some text"; return 0; } -- Best regards, Alexeev Gleb mailto:alexeev_gleb@hotbox.ru

2 2

boost/graph - induced subgraph
by Matthew Galati 24 Mar '04

24 Mar '04

Is it possible to find the minimum spanning tree of an induced subgraph? vector< graph_traits<Graph>::vertex_descriptor > vd; typedef subgraph< Graph > SubGraph; SubGraph & inducedG = subG.create_subgraph(vd.begin(), vd.end()); Here is the example I am looking at: induced verts: 1 2 3 4 5 7 Original Graph: 0 <--> 1 2 5 7 1 <--> 0 3 5 7 2 <--> 0 3 4 5 3 <--> 2 1 4 5 4 <--> 2 3 5 6 5 <--> 0 1 2 3 4 6 7 6 <--> 4 5 7 7 <--> 0 1 5 6 Induced Graph: 1 <--> 3 5 7 2 <--> 3 4 5 3 <--> 2 1 4 5 4 <--> 2 3 5 5 <--> 1 2 3 4 7 7 <--> 1 5 So, the "create_subgraph" seems to be working fine. But, when I run kruskals with: vector<graph_traits < SubGraph >::edge_descriptor> spanning_tree; kruskal_minimum_spanning_tree(inducedG, back_inserter(spanning_tree)); for (vei = spanning_tree.begin(); vei != spanning_tree.end(); ++vei) { pair<int,int> ij = incident(*vei, inducedG); int edge_index = e_index[*vei]; cout << "e_index: " << edge_index << " : " << ij.first << " " << ij.second << endl; I get a spanning tree of just 4 edges (while the subgraph has 6 vertices). Also, how does the edge_index of the subgraph correspond to the original graph? For example, edge (4,1) is not in the original graph, but it shows up in my spanning tree. Spanning Tree: e_index: 3 : 2 0 e_index: 7 : 4 0 e_index: 4 : 3 1 e_index: 8 : 4 1 Thanks, Matthew -- Matthew Galati ISE Lehigh University IBM Service Parts Solutions 610.758.4042 (Office) 610.882.0779 (Home) magh(a)lehigh.edu, magal11(a)us.ibm.com http://sagan.ie.lehigh.edu/mgalati/

4 3

boost-sandbox socket and boost 1.31.0?
by Robert Helgesson 23 Mar '04

23 Mar '04

Howdy, I'm wondering regarding the socket library found in boost-sandbox. Will it be updated to function nicely with the latest Boost release? The problem seems to be that boost::default_iterator_policies and boost::iterator_adaptor no longer work like the socket library expect. Regards -- / Robert Helgesson |-- rycee(a)home.se --\ \-- GPG Key ID: 41E95A5B --| http://www.rycee.cx/ /

3 2

Re: [Boost-users] obtaining vertex descriptor from property_map (newbie)
by Gustavo Sánchez 23 Mar '04

23 Mar '04

I tend to use a map<string, vertex_descriptor> for this. Doug Thanks for the idea, I was just thinking something like this, but I have a problem. I need the definition of the Graph typedef property<edge_name_t, std::string> EdgeProperty; typedef property<vertex_name_t, std::map<string,vertex_property??> > NameProperty; typedef property<vertex_index_t, std::size_t, NameProperty> VertexProperty; typedef adjacency_list<vecS, vecS, undirectedS, VertexProperty, EdgeProperty> UGraph; in order to get the vertex_descriptor, but I need the map<string,vertex_descriptor> for the VertexPorperty so I drop in a circular type dependency. Can you please be more verbose? thanks --------------------------------- Antivirus Filtros antispam 6 MB gratis ¿Todavía no tienes un correo inteligente?

2 1

Solaris and thread_resource_error
by Peter Jones 23 Mar '04

23 Mar '04

On Feb 1 of this year Baptiste Lepilleur posted a status report for Boost 1.31.0 Release candidate 2 on Solaris and AIX. It was mentioned that "boost.thread is now systematically failing with 'thread resource error'". I am experiencing the same thing on Solaris 8, with GCC 3.3.2 (GNU binutils) and Boost 1.31.0. Has anyone looked into this yet? If not, I am willing to track down what line in Boost.Threads is throwing thread_resource_error(). -- Peter Jones [unix;vi;crypto;c++;4wd] (http://pmade.org) PGP [5137 173E 5BFA 8C63 8982 F035 FDBB 1636 7D69 C63F]

2 3

linking to boost::thread
by Jerry 23 Mar '04

23 Mar '04

Hi, I've just built boost1.31.0 using bjam under Linux, apparently successfully. There don't appear to be any static libraries for boost::thread. Is this correct? If so, could someone please explain why I'm getting linker errors under g++ if I don't mention the _dynamic_ thread libs on the g++ command line, and "cannot open shared object file" at runtime if I do. Any pointers appreciated. TIA, Jerry.

2 1

shared_ptr help needed, please!
by Eric Damphousse 23 Mar '04

23 Mar '04

hi all, quick question concerning shared_ptr... not sure if things get deleted (destroyed) properly and its freaking me out. I hope this is not too long Basically, I use the BGL to store my data and that seems to be ok. The way the BGL stuff is defined is as such... Note that spObjectNode is defined as typedef boost::shared_ptr<CObject> spObjectNode // Set vertex properties typedef boost::property<boost::vertex_name_t, std::string> node_name_t; typedef boost::property<graph_node_t, spObjectNode, node_name_t> node_property_t; // Set edge properties typedef boost::property<boost::edge_name_t, std::string> connection_name_t; typedef boost::property<graph_connection_t, spObjectNode, connection_name_t> edge_property_t; // Set graph typedef adjacency_list<listS, vecS, bidirectionalS, node_property_t, edge_property_t> graph_t; // graph_t m_Graph; The function below is what I use to add a node to the BGL... template <typename T> void AddNode(const T& object) { // Create the vertex boost::shared_ptr<T> pNode(new T(object)); // Add the vertex + its name to the graph vertexdescriptor_t newNode = boost::add_vertex(node_property_t(pNode, node_name_t(pNode->GetObjectName())), m_Graph); }; The way I use the AddNode ... for example in a main.cpp: /// blablabla above CSphere sphere("sphere1"); CSceneDB::Instance().AddNode(sphere); return 0; For testing purposes, in the base class and derived classes (see below) I print a message in the destructor of the class, just to make sure things are ok. If I just have in my main the previous call then upon exit I expect and get the following message... CSceneDB cleanup CSphere cleanup CGeometry cleanup CObject cleanup Now for the source of my anxiety... I wrote a function to get back my object from the BGL and it goes like this... inline boost::shared_ptr<CObject> getNodeProperty(const vertexdescriptor_t& vertex) { boost::property_map<graph_t, graph_node_t>::type node_prop_map; node_prop_map = boost::get(graph_node_t(), m_Graph); return node_prop_map[vertex]; } boost::shared_ptr<CObject> CSceneDB::getNodebyName(const std::string& strName) { boost::shared_ptr<CObject> node_prop; boost::graph_traits<graph_t>::vertex_iterator vi, vi_end; boost::tie(vi, vi_end) = boost::vertices(m_Graph); for(; vi != vi_end; ++vi) { node_prop = getNodeProperty(*vi); if(node_prop != 0) { if(node_prop->GetObjectName() == strName) break; } } return node_prop; } template <typename T> boost::shared_ptr<T> getNode(const std::string& strName) { boost::shared_ptr<CObject> abc = getNodebyName(strName); shared_ptr<T> def = boost::dynamic_pointer_cast<T>(abc); return def; } The problem is that if I do something like this... in the main.cpp /// blablabla stuff above CSphere sphere("sphere1"); CSceneDB::Instance().AddNode(sphere); boost::shared_ptr<CSphere> abc = CSceneDB::Instance().getNode<CSphere>("sphere1"); return 0; I only get upon exit... CSceneDB cleanup what happened to CSphere cleanup, CGeometry cleanup and CObject cleanup, are they leaking? Now for a sanity check, if I do in the main.cpp ... i do NOT do the assignment boost::shared_ptr<CSphere> abc = /// blablabla stuff above CSphere sphere("sphere1"); CSceneDB::Instance().AddNode(sphere); CSceneDB::Instance().getNode<CSphere>("sphere1"); return 0; upon exit i get ... CSceneDB cleanup CSphere cleanup CGeometry cleanup CObject cleanup Can someone tell me why this is, is my assignment wrong? Any help would be greatly appreciated Eric Brief description of the classes used ofr my test // Object.h class CObject { protected: CObject(){}; protected: virtual ~CObject(){cout << "CObject cleanup" << endl;}; }; typedef boost::shared_ptr<CObject> spObjectNode; // Geometry.h class CGeometry : public CObject { protected: CGeometry(){}; protected: virtual ~CGeometry(){cout << "CGeometry cleanup" << endl;}; }; // Sphere.h class CSphere : public CGeometry { public: CSphere(){}; ~CSphere(){cout << "CSphere cleanup" << endl;}; };

3 4

dumb question: iterator dynamic_cast adapter?
by Paul Elliott 22 Mar '04

22 Mar '04

I am trying to figure out iterators and I have a dumb question: I have an iterator A_iter, that can traverse in both directions ( A_iter++ and A_iter-- ) and can do input (operator*()). Value_type = type of *A_iter is A; and A is static_cast able to B: A start; B fish = static_cast<B>(start); // OK How do I use adapters to create B_iter, which is the same as A_iter, but whose value_type= typeof *B_iter is B? I am sorry if this is a dumb question. -- Paul Elliott 1(512)837-1096 pelliott(a)io.com PMB 181, 11900 Metric Blvd Suite J http://www.io.com/~pelliott/pme/ Austin TX 78758-3117

2 1

obtaining vertex descriptor from property_map (newbie)
by Gustavo Sánchez 22 Mar '04

22 Mar '04

Dear all, We have a graph which has a property of type string associated with the vertexes. What we want to do is to obtain a vertex descriptor from a string. Could you please tell us how to implement it? Thank you in advance! --------------------------------- Antivirus Filtros antispam 6 MB gratis ¿Todavía no tienes un correo inteligente?

2 1

Controlling selection of runtime libraries
by Bob Kline 22 Mar '04

22 Mar '04

We're using Boost's regex package in a Windows environment for a server built with Microsoft Visual Studio 7.x. We'd like to be able to build our server with debugging symbols for our own code, but still use the release versions of the Microsoft runtime DLLs (because the debugging versions of the Microsoft runtime DLLs aren't redistributable, so we can't legally install them on our production server). The problem is, when we compile our own code for debugging (with -Zi and -D_DEBUG), the Boost headers detect this and do their behind-the-curtain magic to point to a version of the regex package which is linked to the debugging versions of the runtime libraries (msvcr71d and msvcp71d). We've tried using the BOOST_REGEX_NO_LIB trick, specifying the version of the regex library to which we want to link (boost_regex_vc7_mdi.lib) but this doesn't work, because although it compiles fine, the linker chokes complaining about the inability to find a bunch of severely mangled names. What's the *real* trick for using the version of the library we want? Do we need to hack the distributed makefile each time we upgrade regex? (Or worse, shuffle libraries around by renaming them?) Thanks. Bob Kline

3 4

Re: boost-sandbox socket and boost 1.31.0?
by Hugo Duncan 22 Mar '04

22 Mar '04

Robert Helgesson <rycee(a)home.se> wrote: > I'm wondering regarding the socket library found in boost-sandbox. Will > it be updated to function nicely with the latest Boost release? Hi, I have something similar that works with current boost CVS code. I have uploaded this to http://sourceforge.net/projects/giallo/ in an attempt to get this moving again. So far I have low level wrappers for sockets and named pipes. Documentation, tests are lacking. I will be adding the reactor code that I uploaded to the yahoo groups files section in due course. Hugo ps Thanks Jeff for pointing me at this post

1 0

Re: Trouble building libbgl-viz.a
by Vladimir Prus 22 Mar '04

22 Mar '04

Steve Juranich wrote: Hi Steve, sorry that it took quite some time to reply, > My version of graphviz_graph_parser.hpp has the following (lines 55-58): > > #ifndef YYSTYPE > typedef int yystype; > # define YYSTYPE yystype > #endif > > So it looks like the test is different as well as the #define arguments > being switched around. > > When I changed my lines to match your lines, I was able to get the > graphviz_graph_lex.o file, but the build chokes again when it tries to > re-run with -DGRAPHVIZ_DIRECTED=1. Could you modify yystype.h to read: #ifndef BOOST_GRAPH_YYSTYPE_H #define BOOST_GRAPH_YYSTYPE_H union yystype { int i; void* ptr; }; #define YYSTYPE yystype #endif // BOOST_GRAPH_YYSTYPE_H This works for me and it if works for you as well, I'll commit the change. - Volodya

2 2

RE: [Boost-users] regular expression searching for Glib::ustring??
by Ben Hutchings 19 Mar '04

19 Mar '04

Paul Elliott <pelliott(a)io.com> wrote: > On Thu, Mar 18, 2004 at 04:08:59PM -0000, Ben Hutchings wrote: > > Paul Elliott <pelliott(a)io.com> wrote: > > > I am using Gtkmm. I want to do boost regular expression searching > > > on Glib::ustring. <snip> > > > How would you set up boost regex to search if both the regular > > > expression and the string to be searched is a ustring? <snip> > > To create a regex from a ustring I would write something like this: > > > > boost::reg_expression<gunichar, unicode_regex_traits> re; > > In the above, you have unicode_regex_traits, which is necessary > because when I tried: > > boost::basic_regex<gunichar> regular_expression; > > I got the following error message: <snip > So this means that regex_traits<gunichar> does not work, and I > must code my own substitute for regex_traits<gunichar>. > > What are the requirements for such a traits structure? I already told you: > > I am using Boost 1.30, in which the traits are documented in > > <boost/libs/regex/traits_class_ref.htm>. I don't know whether that > > documentation still applies to 1.31. Ben.

1 0

singleton_pool compile error
by ukropina＠sfu.ca 19 Mar '04

19 Mar '04

Hi I'm trying to compile the sample code for singleton pool from http://www.boost.org/libs/pool/doc/interfaces.html just to learn how to use singleton pool. Here is the code. struct MyPoolTag { }; typedef boost::singleton_pool<MyPoolTag, sizeof(int)> my_pool; void func() { for (int i = 0; i < 10000; ++i) { int * const t = my_pool::malloc(); } my_pool::purge_memory(); } but I keep geting this error: Compiling... main.cpp c:\program files\microsoft visual studio\vc98\include\xmemory(59) : error C2248: 'object_creator' : cannot access private struct declared in class 'boost::details::pool::singleton_default<struct boost::singleton_pool<struct MyPoolTag,4,struct boost: :default_user_allocator_new_delete,class boost::details::pool::win32_mutex,32>::pool_type>' e:\lib\boost\boost\pool\detail\singleton.hpp(63) : see declaration of 'object_creator' c:\program files\microsoft visual studio\vc98\include\xmemory(59) : while compiling class-template static data member 'struct boost::details::pool::singleton_default<struct boost::singleton_pool<struct MyPoolTag,4,struct boost::default_user_ allocator_new_delete,class boost::details::pool::win32_mutex,32>::pool_type>::object_creator ol_type>::create_object' Error executing xicl6.exe. I'm using MS Visual C++ 6.0 SP5 on windows XP. What am I doing wrong ? Does anyone have a sample code for singleton pool that works on windows ? Thanks Nenad boost::details::pool::singleton_default<boost::singleton_pool<,4,boost::defa ult_user_allocator_new_delete,boost::details::pool::win32_mutex,32>::po

2 1

regular expression searching for Glib::ustring??
by Paul Elliott 19 Mar '04

19 Mar '04

I am using Gtkmm. I want to do boost regular expression searching on Glib::ustring. http://www.gtkmm.org/gtkmm2/docs/reference/html/classGlib_1_1ustring.html This class represents characters in UTF-8, so each character in the buffer is represented by a varriable number of bytes. But it does have a bidirectional iterator. How would you set up boost regex to search if both the regular expression and the string to be searched is a ustring? Does one need to override any of the types defiened in regex_traits? Thank you. BTW. Quote from:http://www.boost.org/libs/regex/doc/regex_traits.html "Under construction. The current boost.regex traits class design will be migrated to that specified in the regular expression standardization proposal." This is not very useful to someone trying to use the boost_regex library now! -- Paul Elliott 1(512)837-1096 pelliott(a)io.com PMB 181, 11900 Metric Blvd Suite J http://www.io.com/~pelliott/pme/ Austin TX 78758-3117

2 1

RE: [Boost-users] regular expression searching for Glib::ustring??
by Ben Hutchings 18 Mar '04

18 Mar '04

Paul Elliott <pelliott(a)io.com> wrote: > I am using Gtkmm. I want to do boost regular expression searching on > Glib::ustring. > http://www.gtkmm.org/gtkmm2/docs/reference/html/classGlib_1_1ustring.htm l > > This class represents characters in UTF-8, so each character in the > buffer is represented by a varriable number of bytes. But it does > have a bidirectional iterator. > > How would you set up boost regex to search if both the regular > expression and the string to be searched is a ustring? > > Does one need to override any of the types defiened in regex_traits? The character type should be gunichar. > Thank you. > > BTW. > Quote from:http://www.boost.org/libs/regex/doc/regex_traits.html > > "Under construction. > The current boost.regex traits class design will be migrated to > that specified in the regular expression standardization proposal." > > This is not very useful to someone trying to use the boost_regex > library now! I am using Boost 1.30, in which the traits are documented in <boost/libs/regex/traits_class_ref.htm>. I don't know whether that documentation still applies to 1.31. You don't necessarily need custom traits to use custom iterator types. However, you do need to define custom traits to make character classes, case folding and so on work for all Unicode characters. To create a regex from a ustring I would write something like this: boost::reg_expression<gunichar, unicode_regex_traits> re; re.assign(s.begin(), s.end(), flags); When searching in a ustring I would write code like: my_predicate pred; boost::regex_grep(pred, s.begin(), s.end(), re, flags); boost::match_results<Glib::ustring::const_iterator, std::allocator<gunichar> > results; boost::regex_search<Glib::ustring::const_iterator, std::allocator<gunichar>, gunichar, unicode_regex_traits, std::allocator<gunichar> >( s.begin(), s.end(), results, re, flags); Glib::ustring result; boost::regex_merge( std::back_inserter(result), source.begin(), source.end(), re, std::basic_string<gunichar>(repl.begin(), repl.end())), flags); (The above is untested as we're not actually using Gtkmm but it's based on what we're doing with our own UTF-8 and UTF-16 strings.)

2 1

RE: [Boost-users] regular expression searching for Glib::ustring??
by Yarwood, Martin (UK) 18 Mar '04

18 Mar '04

> "Under construction. > The current boost.regex traits class design will be migrated to > that specified in the regular expression standardization proposal." > > This is not very useful to someone trying to use the boost_regex > library now! > I'm sure they'll give you a full refund ******************************************************************** This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. ********************************************************************

1 0

adjacency_matrix BGL problem
by Gustavo Sánchez 18 Mar '04

18 Mar '04

Hi everybody, Is it posible to resize a graph with this representation?. The idea is start with an empty matrix and add vertex based on the new vertices and edges we read. That is, we read a new edge and if their vertex aren't in the graph we add them and then the edge. Is this posible in an adjacency_matrix representation??? Thanks a lot. --------------------------------- Antivirus Filtros antispam 6 MB gratis ¿Todavía no tienes un correo inteligente?

2 1

problem with vertex properties in BGL
by Gustavo Sánchez 17 Mar '04

17 Mar '04

I have to build a graph from data in a file, some information is strings which are related to vertexes. I want to prove if those vertexes already exist in the graph before the insert operation. Next is the source: typedef adjacency_matrix<undirectedS,VertexProperty,EdgeProperty> UGraph; // an undirected graph UGraph g(0); // start from scratch property_map<UGraph, edge_name_t>::type ename = get(edge_name, g); // not sure if this is right (the graph have no elements at this point) property_map<UGraph, vertex_name_t>::type vname = get(vertex_name, g); typedef property_map<UGraph, vertex_name_t>::type vname_type; typedef boost::property_traits<vname_type>::key_type vktype; typedef boost::property_traits<vname_type>::value_type vvtype; vktype indice_nombre; vvtype valor_nombre; for(i=0;i<datgen.NumHilos;i++){ valor_nombre = hiloNormal[i].viaOrigen; /// here we get the string indice_nombre = vname[valor_nombre]; /// here I try to get the index /// I think this is not posible because there isn't [] operator in this way valor_nombre = vname[1]; /// if I do this the compiler is right but that isn't useful for me because the information I have is the string not the index.I can use vertex iterators and search for the string, but if posible i prefer not to do that. thanks a lot. --------------------------------- Antivirus Filtros antispam 6 MB gratis ¿Todavía no tienes un correo inteligente?

2 1

Bug in Boost Graph Library?
by Heiko Bauke 17 Mar '04

17 Mar '04

Dear all, today I was engaged in searching for a bug in my program. After a while I figured out that boost::num_egdes behaves not as I expected. If G is an undirected adjacency matrix graph boost::num_egdes(G) returns the number of edges in the graph G plus the number of edges in G that are no self edges. That means almost all edges are counted twice in undirected adjacency matrix graphs. In my oppinion this behaviour is a bug. At least it makes the BGL-user's live unnecessary difficult. Let's considder the following program: #include <cstdlib> #include <iostream> #include <boost/tuple/tuple.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/adjacency_matrix.hpp> template <typename Graph> void add_edges(const char * const c) { std::cout << c; Graph G(6); boost::graph_traits<Graph>::vertex_iterator v, vbegin, vend; boost::tie(vbegin, vend)=boost::vertices(G); for (v=vbegin; v!=vend; ++v) { boost::add_edge(*vbegin, *v, G); std::cout << ' ' << boost::num_edges(G); } std::cout << std::endl; } int main() { typedef boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS> adjacency_list; typedef boost::adjacency_matrix<boost::undirectedS> adjacency_matrix; add_edges<adjacency_list>("adjacency list graph :"); add_edges<adjacency_matrix>("adjacency matrix graph :"); return EXIT_SUCCESS; } It prints to stdout: adjacency list graph : 1 2 3 4 5 6 adjacency matrix graph : 1 3 5 7 9 11 So the result of the template function add_edges in the program above depends of the graph container that the function add_edges is applied to. I think this is verry unintuitive and boost::num_edges should be changed, so that alls edges in undirected adjacency matrix graphs are counted only once. Or is there a special reason that boost::num_edges is implemented in this way? with regards, Heiko -- -- Gute Sitten haben für die Gesellschaft mehr Wert als alle -- Berechnungen Newtons. (Friedrich II., der Große, 1712-1786) -- Supercomputing in Magdeburg @ http://tina.nat.uni-magdeburg.de -- Heiko Bauke @ http://www.uni-magdeburg.de/bauke

2 1

RE: [Boost-users] boost::thread problem
by Ben Hutchings 17 Mar '04

17 Mar '04

Pete <petehug(a)paradise.net.nz> wrote: > I'm having a problem with my boost multithreaded application. > My app runs fine as a Windows Service, but as a Unix daemon it > stops responding after some time. I noticed that the fewer > threads I have available, the quicker the application stops > responding: HP-UX(64) after 20mins, RH8(256) after >1hr, Linux > RH9(1024) >5hrs. This led me to assume that perhaps the threads > are not cleaned up properly. That could well be right. Perhaps there is some kind of deadlock at thread exit? > Therefore, I wrote a little test app which ought to demonstrate > the problem. To my surprise, the test application showed an > entirely different problem, but again, only under Unix (it > works as expected under Windows). I'm totally lost what the > reason is and appreciate any help. > > Expected: Actual: > --------------- --------------- > Start running Start running > Start swimming Finish running > Start cycling Start swimming > Finish swimming Finish swimming > Finish cycling Start cycling > Finish running Finish cycling > > In windows I get the actual output but under Linux RH9, I get > the Actual output. Under RH9, the first five lines of the > actual output appear immediately, but the last one takes > approx. 4secs. In other words, only the last threads don't > sleep the number of secs specified. According to the manual page: "sleep() makes the current process sleep until seconds sec- onds have elapsed or a signal arrives which is not ignored." (Note that each POSIX thread is/was implemented as a process in Linux.) If thread creation involves receiving a signal then that would explain the above behaviour. I would expect it under RH8 because that has the old LinuxThreads implementation of pthreads that uses signals heavily, but not in RH9 which has NPTL. Still, you would be better off using boost::thread::sleep() instead.

1 0