June 2005 - Boost-users - lists.preview.boost.org

Dangling edges in BOOST graph library
by Abhijit Deshmukh 29 Jun '05

29 Jun '05

Hi, I am developing an application for which I need to define edges that have only one of the node defined. The second node is not defined. Does BOOST allow to do this? I know this is not the definition of a graph and is considered illegal in defining a graph. I want to know whether this functionality is available in BOOST graph library. Thanks and Regards, Abhijit Deshmukh

4 5

Boost.Filesystem -- bug or user error?
by Chris 'Xenon' Hanson 29 Jun '05

29 Jun '05

I'm using Boost.Filesystem (along with Boost.Regex, and an unofficial glob library by Angus Leeming) to do globbing and batch file processing in a utility of mine. Environment is boost_1_32_0, Windows XP, MSVC++7.1. The Glob/Regex stuff seems fine -- I fill a vector<string> container with the results and then pass it off to another piece of code to actually do the work. Part of the work involves printing the filename (without path, but with extension). Here's a snippet: int ProcessFile(ofstream &Out, const char *FullPathName) { int Success = 0; fs::path OutFilePath(FullPathName, fs::native); std::string JustName, NameNoExt; JustName = fs::basename(OutFilePath.leaf()) + fs::extension(OutFilePath.leaf()); NameNoExt = fs::basename(OutFilePath.leaf()); PrintItem(Out, FullPathName, JustName.c_str(), NameNoExt.c_str()); return(Success); } // ProcessFile Ignore that it's poor code quality -- this is just a quick & dirty hack program. Everything works great until this code is fed a FullPathName of: eggs+butter01d-thumbs.jpg (In this example, I'm running with relative paths, so there is no "path" part to discard. Makes troubleshooting even simpler.) It throws an exception of boost::filesystem::path invalid name eggs+butter01d-thumbs.jpg in path eggs+butter01d-thumbs.jpg on line 341 of path_posix_windows.cpp It appears to displike the + character. Maybe it's pilot error, but + is apparently a valid symbol in the Win32/NTFS filesystem, and Windows permits it, generally. If it were a weird platform-specific symbol, I would have thought that creating the fs::path with fs::path OutFilePath(FullPathName, fs::native); would have translated it into somethign boost-friendly. Anyone care to tell me what I'm doing wrong? Thanks in advance for any advice. -- Chris 'Xenon' Hanson | Xenon @ 3D Nature | http://www.3DNature.com/ "I set the wheels in motion, turn up all the machines, activate the programs, and run behind the scenes. I set the clouds in motion, turn up light and sound, activate the window, and watch the world go 'round." -Prime Mover, Rush.

2 1

Re: [Boost-users] Spirit - a Reversed Polish Notation calculator
by Marcin Tustin 28 Jun '05

28 Jun '05

> -----Original Message----- > From: Thore Karlsen [mailto:sid@6581.com] > Sent: 28 June 2005 14:48 > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Spirit - a Reversed Polish > Notation calculator > Do you have any suggestions for books or tutorials that show > how to make proper LL grammars? Solving ambiguities, > eliminating left recursion, etc. I learned all of this in > college, but I've forgotten most of the finer points since I > never had a practical need for a parser before now. In my youth I found "Compilers" by Alfred V. Aho, Ravi Sethi, Jeffrey D. Ullman useful (http://www.amazon.com/exec/obidos/tg/detail/-/0201100886/qid=1119968544 /sr=8-1/ref=pd_bbs_1/102-1607349-4401717?v=glance&s=books&n=507846). Andrew Appel has a book on the subject which I dislike, "Modern Compiler Implementation in C".

1 0

Spirit - a Reversed Polish Notation calculator
by Misza 28 Jun '05

28 Jun '05

I've recently switched from YACC to Spirit and I've been trying to rewrite some examples from YACC manual, starting with a RPN calculator. Based on http://www.boost.org/libs/spirit/doc/grammar.html I made something like this: /* #includes and usings */ struct rpn_calc : public grammar<rpn_calc> { template <typename ScannerT> struct definition { definition (rpn_calc const& self) { expr_add = expr >> ch_p(' ') >> expr >> ch_p(' ') >> ch_p('+'); expr_sub = expr >> ch_p(' ') >> expr >> ch_p(' ') >> ch_p('-'); expr_mul = expr >> ch_p(' ') >> expr >> ch_p(' ') >> ch_p('*'); expr_div = expr >> ch_p(' ') >> expr >> ch_p(' ') >> ch_p('/'); expr_pow = expr >> ch_p(' ') >> expr >> ch_p(' ') >> ch_p('^'); expr = expr_add | expr_sub | expr_mul | expr_div | expr_pow | real_p; } rule<ScannerT> expr, expr_add, expr_sub, expr_mul, expr_div, expr_pow; rule<ScannerT> const& start () const { return expr; } }; }; int main (void) { rpn_calc calc; std::string str("1 2 + 3 4 + *"); if (parse(str.begin(), str.end(), calc, space_p).full) cout << "Parsing succeeded.\n"; else cout << "Parsing failed!\n"; return 0; } Now this compiles OK, but when executed it doesn't output anything, which means that parse(...) quits without any warnings - what's the problem? Am I doing something wrong? Thanks, Misza

3 2

Application using boost library hangs
by Thameem Ansari - CTD ,Chennai 27 Jun '05

27 Jun '05

Hi, I am facing the following problem I have the following code in my application written in VC++ 6.0 std::vector<std::string> v; boost::RegEx expr("(?=adc)", TRUE); expr.Grep(v, "XYZ(M-ADC)"); While debugging, then application hangs in the statement 'expr.Grep(v, "XYZ(M-ADC)")' . If you have faced same kind of problem or you have any solution, please mail me the solution as soon as possible thanks and regards thameem Disclaimer: This message and any attachment(s) contained here are information that is confidential, proprietary to HCL Technologies and its customers. Contents may be privileged or otherwise protected by law. The information is solely intended for the individual or the entity it is addressed to. If you are not the intended recipient of this message, you are not authorized to read, forward, print, retain, copy or disseminate this message or any part of it. If you have received this e-mail in error, please notify the sender immediately by return e-mail and delete it from your computer. DISCLAIMER This message and any attachment(s) contained here are information that is confidential, proprietary to HCL Technologies and its customers. Contents may be privileged or otherwise protected by law. The information is solely intended for the individual or the entity it is addressed to. If you are not the intended recipient of this message, you are not authorized to read, forward, print, retain, copy or disseminate this message or any part of it. If you have received this e-mail in error, please notify the sender immediately by return e-mail and delete it from your computer

6 7

bind problems
by Jason 27 Jun '05

27 Jun '05

I'm trying to use bind with boost.thread in order to make a thread of a class member function with an input argument, but I can't get it to compile. To narrow the problem, I went to the bounded_buffer.cpp program in the libs/thread/tutorial dir. I added an int arg to sender, and then changed boost::thread thrd1(&sender); to int x = 99; boost::thread thrd1(boost::bind<void>(&sender, _1)(x)); and several other combinations including removal of <void>. In the above case, gcc results in /mnt/shared/Projects/boosttest/src/bounded_buffer.cpp:77: error: call of overloaded `bind(void (*)(int), boost::arg<1>&)' is ambiguous /usr/include/boost/bind.hpp:1132: note: candidates are: boost::_bi::bind_t<R, F, typename boost::_bi::list_av_1<A1>::type> boost::bind(F, A1) [with R = void, F = void (*)(int), A1 = boost::arg<1>] I got this syntax from the bind doc (http://www.boost.org/libs/bind/bind.html) What am I doing wrong? thanks, Jason

4 5

String algorithm
by Sliwa, Przemyslaw (London) 26 Jun '05

26 Jun '05

Hi, I would like to use the string algorithm library in my project. Including #include <boost/algorithm/string.hpp> in the source file gives the following error on Windows XP, VC++ 6 SP5. Can anyone help me? Thanks a lot in advance. Pshemek c:\vendor_libs\include\boost/algorithm/string/yes_no_type.hpp(22) : error C2265: '<Unknown>' : reference to a zero-sized array is illegal c:\vendor_libs\include\boost/algorithm/string/yes_no_type.hpp(23) : see reference to class template instantiation 'boost::algorithm::size_descriptor<I>' being compiled c:\vendor_libs\include\boost/mpl/bool.hpp(37) : fatal error C1506: unrecoverable block scoping error Error executing cl.exe. Kind regards Przemyslaw Sliwa (0044) 20788-65014 GFX Development -------------------------------------------------------- If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/ --------------------------------------------------------

3 4

How do I get boost to work with vs.net 2005 ?
by Mart marvin 25 Jun '05

25 Jun '05

I downloaded the executable from sourceforge, but I'm at a loss how to get the libraries to be recognized by VS.NET. I googled but nothing that works turned up. I'm desperate to try and play with the boost libraries and am eager to get started with them. Can someone describe it step by step command by command? I'm on Win2k Pro. Thanks in advance. _________________________________________________________________ On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement

5 6

Re: [Boost-users] How do I get boost to work with vs.net 2005 ?
by Brian Braatz 24 Jun '05

24 Jun '05

I believe I can help. Lets say you have downloaded boost to the dir: C:\boost now if you CD into that dir- you will get something like the following files: 06/10/2005 10:50 AM <DIR> . 06/10/2005 10:50 AM <DIR> .. 05/26/2005 01:48 AM <DIR> bin 06/16/2005 02:27 PM <DIR> boost 02/04/2005 04:09 PM 499 boost-build.jam 02/04/2005 04:09 PM 675 boost.css 02/04/2005 04:09 PM 6,897 boost.png 02/21/2005 11:10 AM <DIR> CVS 02/18/2005 03:46 PM <DIR> doc 02/04/2005 04:09 PM 3,558 google_logo_40wht.png 02/04/2005 04:09 PM 49,603 index.htm 02/04/2005 04:09 PM 9,406 Jamfile 02/04/2005 04:09 PM 882 Jamfile.v2 02/04/2005 04:09 PM 450 Jamrules 05/26/2005 01:04 AM <DIR> libs 02/04/2005 04:09 PM 1,361 LICENSE_1_0.txt 02/18/2005 03:47 PM <DIR> more 02/18/2005 03:47 PM <DIR> people (abbreviated) Now using my example dirs- some important dirs to be aware of: C:\boost\boost - the libraries themselves are here C:\boost\libs - look here for the documentation (html) files for a given library since the libs are in c:\boost you need to do the following: visual studio- (menu) view-solution explorer then in the solution explorer, right click on the project name select "Properties" from the menu click on the "C\C++" select on the left click on the "General" item below this on the right you will see "Additional Include Directories" type in (in our case above) C:\boost the cute thing about this is (recall there is a boost dir below C:\boost) you can now do a #include with the sub dir of boost in the string i.e. #include <boost/smart_ptr.hpp> You can TEST if the "Additional Included Directories" setting is CORRECT- by doing the following: (set the setting as described above) move your cursor over #include <boost/smart_ptr.hpp> right click- on that menu select "Open Document <boost/smart_ptr.hpp>" IF that succeeds, then you have setup the #include directories correctly- if not then you put the wrong value in "Additional Included Directories" HOPE THAT HELPS Brian :) > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Mart marvin > Sent: Friday, June 24, 2005 3:13 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] How do I get boost to work with vs.net 2005 ? > > It didn't work! > > I followed the directions to the letter. Is it a bug with VS.NET 2005 > beta? > Here's the message I got: > > Cannot open include file: 'boost\smart_ptr.h': No such file or directory > > Here is my code: > > #include "stdafx.h" > #include "x.h" > #include <boost\smart_ptr.h> > > int main() > { > x& p = *new x; > delete &p; > return 0; > } > > Please help... I'm desperate to try Boost. Looked at the documentation > and > I'm drooling to play with it. > > >From: "Pablo Aguilar" <pablo.aguilar(a)gmail.com> > >Reply-To: boost-users(a)lists.boost.org > >To: boost-users(a)lists.boost.org > >Subject: Re: [Boost-users] How do I get boost to work with vs.net 2005 ? > >Date: Fri, 24 Jun 2005 14:10:36 -0700 > > > >Open VS.NET > > >From the "Tools" menu, select "Options" > >In the tree to the left, open "Projects & Solutions" and select VC++ > >Directories > > >From the drop down list labelled "Show directories for:" select > "Include > >files" > >Add the boost root directory (wherever you decompressed the boost > >installation) to that list. > > > >HTH > > > > > >Pablo > > > > >I downloaded the executable from sourceforge, but I'm at a loss how to > >get > > > the libraries to be recognized by VS.NET. I googled but nothing that > > > works > > > turned up. I'm desperate to try and play with the boost libraries > and > >am > > > eager to get started with them. Can someone describe it step by step > > > command by command? > > > > > > I'm on Win2k Pro. > > > > > > Thanks in advance. > > > > > > > >_______________________________________________ > >Boost-users mailing list > >Boost-users(a)lists.boost.org > >http://lists.boost.org/mailman/listinfo.cgi/boost-users > > _________________________________________________________________ > Is your PC infected? Get a FREE online computer virus scan from McAfee(r) > Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 >

2 1

Re: [Boost-users] How do I get boost to work with vs.net 2005 ?
by Sohail Somani 24 Jun '05

24 Jun '05

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Mart marvin > Sent: Friday, June 24, 2005 12:20 PM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] How do I get boost to work with vs.net 2005 ? > > > I downloaded the executable from sourceforge, but I'm at a > loss how to get > the libraries to be recognized by VS.NET. I googled but > nothing that works > turned up. I'm desperate to try and play with the boost > libraries and am > eager to get started with them. Can someone describe it step by step > command by command? http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl?Installin gBoost Doesn't get much better than that.

1 0

Re: [Boost-users] Boost-users Digest, Vol 588, Issue 2
by Frank Hess 24 Jun '05

24 Jun '05

On Thursday 23 June 2005 05:41 pm, boost-users-request(a)lists.boost.org wrote: > Message: 9 > Date: Thu, 23 Jun 2005 17:45:25 +0100 > From: Paul Johnson <gt54-boost(a)cyconix.com> > Subject: [Boost-users] Replacing a reference with a shared_ptr? > To: boost-users(a)lists.boost.org > Message-ID: <42BAE725.2080907(a)cyconix.com> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > I'm deriving from a base class in a library I use (in other words, I > > can't change it), and this class declares: > > virtual string& getFilename(void) const; > > virtual void setFilename(const string&); > > The base class itself doesn't declare a member to store the file name; > it just ignores 'setFilename', and returns an empty string for > 'getFilename'. > > I'd like to implement these routines, but I don't want to have a string > reference in my derived class. If possible, I'd like to use a > shared_ptr<string>. However, I can't see a simple way to do this. Why are you bothering with pointers or references? Why can't you just put a string in your class? string myString; virtual string& getFilename(void) const {return myString;} virtual void setFilename(const string&) {myString = string;} -- Frank

1 0

shared_ptr misuse?
by Jason House 23 Jun '05

23 Jun '05

I have the contents of a shared pointer getting corrupted by a default constructor of another object. My best guess as to how this could be happening is that I am using references to shared pointers for coding convenience. Is it possible for code like the following to cause a problem? { boost::shared_ptr<my_class> &x = a.b.c.d.e.f.g; if (x.get() == NULL) { pointer_empty(); } else { other_actions(x->get()); } } PS: I found this in a header file (#define BOOST_VERSION 103000)

2 1

Replacing a reference with a shared_ptr?
by Paul Johnson 23 Jun '05

23 Jun '05

I'm deriving from a base class in a library I use (in other words, I can't change it), and this class declares: > virtual string& getFilename(void) const; > virtual void setFilename(const string&); The base class itself doesn't declare a member to store the file name; it just ignores 'setFilename', and returns an empty string for 'getFilename'. I'd like to implement these routines, but I don't want to have a string reference in my derived class. If possible, I'd like to use a shared_ptr<string>. However, I can't see a simple way to do this. The original string is supplied by me to the library, which eventually calls my 'setFilename'. So, I could start with a shared_ptr, and use operator() to supply the library with a reference to the string. However, when the string& gets back to my 'setFilename', there's no obvious way to get back to the original shared_ptr. Have I missed something? Is there a clean way to implement this with shared_ptr<string>? Thanks.

1 0

python like generators in C++
by Paul Elliott 23 Jun '05

23 Jun '05

I consider the following equation: Threads + some template code = iterators that act like python generators in C++! Has anyone implemented this in boost? Thank You -- 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

1.32 installation: where is the smart_ptr lib?
by Paul Johnson 23 Jun '05

23 Jun '05

Sorry for the dumb question... Just installed boost following the instructions on boost.org, and I can't find the smart_ptr library. I downloaded the 1.32 .bz file, and bjam, and installed on Linux/RHEL4. I've now got 189 boost libraries in /usr/local/lib, and 32 in /usr/lib. However, the only base libraries seem to be date_time, filesystem, prg_exec_monitor, program_options, python, regex, serialisation, signals, test_exec_monitor, thread, unit_test_framework, and wserialisation, some of which aren't even listed on the libraries page at boost.org. Where's smart_ptr? None of these libraries appear to have it. The smart_ptr headers are in /usr/local/include/boost-1_32/boost, as expected (but why 2 levels down? Surely I'm not expected to #include <boost-1_32/boost/xx.hpp??) Cheers.

2 2

bcp error
by Ronald Sipkema 23 Jun '05

23 Jun '05

hi, when trying to use bcp as follows: "bcp --list build" I get the following error: **** exception(205): std::runtime_error: Bad file name: ../boost.css Any idea what's happening? Other module lists )for example regex) work just fine. --Ronald

2 1

date_time problem: where are fractional seconds?
by Serge Skorokhodov 23 Jun '05

23 Jun '05

Hi, The following code (boost 1.32, MSVC.NET 2003): ptime pt = from_iso_string("20030912T211733.01204"); cout << pt << endl; cout << to_simple_string(pt) << endl; cout << to_iso_extended_string(pt) << endl; cout << to_iso_string(pt) << endl; cout << pt.time_of_day().fractional_seconds() << endl; time_duration td = pt.time_of_day(); cout << to_iso_string(td) << endl; cout << to_simple_string(td) << endl; cout << td.fractional_seconds() << endl; yealds: 2003-Sep-12 21:17:33 2003-Sep-12 21:17:33 2003-09-12T21:17:33 20030912T211733 0 211733 21:17:33 0 Is the fractional part of time (.01204) absent by design? I failed to find it out from the documentation? Is there a workaround? TIA -- Serge

2 3

[bind / lambda] binding a factory funtion instead of an value
by Bertolt Mildner 23 Jun '05

23 Jun '05

Lets say i need a functor of type typedef boost::function1<Ret, Arg1> Functor; and i have a function Ret Foo(Arg1 a, Arg2 b); Creating a Functor object is easy if i have a Arg2 object Arg2 arg2; Functor f(boost::bind<Ret>(&Foo, _1, arg2); But what if i want to create the Arg2 object in the functor when it is used? Like i have a function Arg2 CreateArg2(); and i don't want Functor f(boost::bind<Ret>(&Foo, _1, CreateArg2()); but something like Functor f(boost::bind<Ret>(&Foo, _1, &CreateArg2); so that f(arg1) = Foo(arg1, CreateArg2()); I have a vague feeling that lambda might have a solution to this, but i don't know how! Bertolt

2 2

[BGL] overlaying a graph-like structure
by martin f krafft 22 Jun '05

22 Jun '05

I am looking at the possibility to use BGL for the structural representation of an artificial neural network: nodes would be neurons, edges would be synapses. The advantage for me would be that I don't have to worry about complexity of the graph structure. For this to work, I need to be able to strap behaviour and data onto the edges and nodes. The bundled properties interfaces seems to be a good way to do this, but there is something fundamental that I am apparently not understanding: What is an edge descriptor, what is a vertex descriptor? From the code: typedef typename boost::ct_if_t<is_rand_access, std::size_t, vertex_ptr>::type vertex_descriptor; vertex descriptors are indices or pointers. And typedef detail::edge_desc_impl<directed_category, vertex_descriptor> edge_descriptor; and edge is actually a struct containing source and target vertex, and a property pointer (void*). There does not seem to be a way to get the property object from an edge descriptor without casting to the expected type. Thus, vertex and edge descriptors are really just minimal and provide absolutely no way to access the relevant data through their interface. Instead, BGL expects people to have access to the containing graph object, to be able to call functions like target(), out_edges(), or operator[] to get at the downstream vertex/edge, or a vertex'/edge's properties. In the context of my application, I know have two options: adopt the graph-centric perspective of BGL, in which the graph object is needed to answer any questions about the graph, even about single components therein, or to strap a layer on top of all this to return to proper object encapsulation: - where a node knows about its incident edges - where an edge knows about its source and target vertices - where both give access to the associated properties. If I wanted to go the latter way, I'd have to store an additional 8 bytes (descriptor + graph references) with each object, which is just a waste if you ask me. However, there seems to be no other way as I cannot derive from vertex/edge descriptors. Does anyone have any experience with folding BGL into an existing graph-like structure? My ideal solution would be my neuron and synapse classes derived from the vertex and edge descriptors, such that I can treat them like components of the graph as well as neural components. From looking at the LEDA stuff, I guess the way to do this is to modify the graph traits to replace edge and node descriptors with custom classes. If I take care to model the behaviour of the existing edge and node descriptors wrt their interfaces, I shouldn't actually need to override the public Graph API, right? Is it sensible to derive from detail::edge_desc_impl, or should detail::* stuff not be touched outside of BGL code? Thanks for any comments, thoughts, suggestions, pointers. -- martin; (greetings from the heart of the sun.) \____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck invalid/expired pgp subkeys? use subkeys.pgp.net as keyserver! spamtraps: madduck.bogus(a)madduck.net "i have smoked pot. it is a stupid business, like masturbation." -- thomas pynchon

3 7

[BGL] adding behaviour to nodes and edges
by martin f krafft 22 Jun '05

22 Jun '05

I would like my nodes and edges to have behaviours. Bundled properties -- the method to attach struct instances to nodes and edges -- seems like the way to go. However, if that is the suggested method, then "bundled properties" is somewhat of a misnomer. What's the suggested way to attach behaviour to nodes and edges? I.e. I'd love to be able to iterate through vertices(G) with std::for_each and have it call a member function for each vertex. -- martin; (greetings from the heart of the sun.) \____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck invalid/expired pgp subkeys? use subkeys.pgp.net as keyserver! spamtraps: madduck.bogus(a)madduck.net the english take english for granted. but if we explore its paradoxes, we find that quicksand can work slowly.

2 6

Problem using edge_descriptors as indices into an external property
by Andrea Olgiati 22 Jun '05

22 Jun '05

Folks, I've encountered a problem whilst compiling the following program, and I can't quite figure out why it shouldn't work. Whilst compiling usign boost 1.32.0 and gcc 3.3.4, I get cannot convert `boost::detail::edge_desc_impl<boost::bidirectional_tag, size_t>' to `int' for argument `2' to `const T& get(const T*, int) [with T = int]' I believe this is saying that edge descriptors cannot be used as indices into a property map (that is, "wm"). Why is this different from vertex descriptors, which are accepted as indices into "p" and "dist"? Any help appreciated. Thanks, Andrea #include <boost/graph/graph_traits.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/bellman_ford_shortest_paths.hpp> int main () { typedef boost::adjacency_list < boost::vecS, boost::vecS, boost::bidirectionalS > Graph; Graph G; std::vector < int > wm; std::vector < boost::graph_traits < Graph >::vertex_descriptor > p; std::vector < int > dist; boost::bellman_ford_shortest_paths (G, boost::num_vertices(G), &wm[0], &p[0], &dist[0], boost::closed_plus < int >(), std::less < int >(), boost::default_bellman_visitor ()); return 0; }

2 1

[signal] custom connect/disconnect
by John Maguire 22 Jun '05

22 Jun '05

Hi all, I have a program where I have several signals, and I want an action to be taken when connect/disconnect occurs on them. I would prefer to avoid wrapping the calls, since it would reduce composibility. I would imagine that this isn't the first time that such a requirement has been made of signals, so I was wondering if someone could point me towards the standard solution for this? Regards, John

2 1

[BGL] Question on prim_minimum_spanning_tree
by Wang Johny 22 Jun '05

22 Jun '05

Hi, there I am a new user of boost (in particular BGL). For my project, I am using the function of "prim_minimum_spanning_tree" to construct the minimum spanning tree. But I found a strange problem and hope to get your help. Specifically, I am following the sample code "prim-example.cpp" (using MSVC), with one modification to the weight property. That is, I am defining the weight property as property < edge_weight_t, float > not the original one property < edge_weight_t, int > And I found by doing so, the function of "prim_minimum_spanning_tree" return wrong results (no edge is returned at all). But the original version using "property < edge_weight_t, int >" definition works well. So I am wondering if I missed something to make user-defined weitgh property work for this function. Any suggestion is welcome. Thank you very much. Best regards, Shenai Problematic code: /////////////////////////////////////////////////////////////////////////////////////////////////// typedef adjacency_list < vecS, vecS, undirectedS, property<vertex_distance_t, int>, property < edge_weight_t, float > > Graph; typedef std::pair < int, int >E; const int num_nodes = 5; E edges[] = { E(0, 2), E(1, 1), E(1, 3), E(1, 4), E(2, 1), E(2, 3), E(3, 4), E(4, 0) }; float weights[] = { 1.0f, 2.0f, 1.0f, 2.0f, 7.0f, 3.0f, 1.0f, 1.0f }; Graph g(num_nodes); property_map<Graph, edge_weight_t>::type weightmap = get(edge_weight, g); for (std::size_t j = 0; j < sizeof(edges) / sizeof(E); ++j) { graph_traits<Graph>::edge_descriptor e; bool inserted; tie(e, inserted) = add_edge(edges[j].first, edges[j].second, g); weightmap[e] = weights[j]; } std::vector < graph_traits < Graph >::vertex_descriptor > p(num_vertices(g)); property_map<Graph, vertex_distance_t>::type distance = get(vertex_distance, g); property_map<Graph, vertex_index_t>::type indexmap = get(vertex_index, g); prim_minimum_spanning_tree(g, *vertices(g).first, &p[0], distance, weightmap, indexmap, default_dijkstra_visitor()); for (std::size_t i = 0; i != p.size(); ++i){ if (p[i] != i){ TRACE("I am here"); } } /////////////////////////////////////////////////////////////////////////////////////////////////// --------------------------------- Do You Yahoo!? 150万曲MP3疯狂搜，带您闯入音乐殿堂美女明星应有尽有，搜遍美图、艳图和酷图 1G就是1000兆，雅虎电邮自助扩容！

2 1

style: overloading vs default parameters
by Noel Yap 22 Jun '05

22 Jun '05

Is there any particular reason why at least some boost code uses default parameters. IME, I've found the using overloading is preferable. For example, when using boost::bind, one would have to explicitly give the default parameters whereas one wouldn't if overloading were used. Also, using overloading rather than default parameters when a new function interface is needed can potentially break less client code (eg code that uses the type of the function won't break if overloading were used). Given this, I'd like to hear some arguments for using default arguments over overloading (outside of constructors, of course, where one may not have much of a choice). Thanks, Noel

2 1

boost building question
by H 22 Jun '05

22 Jun '05

Hi, In the Jamfile help it says: --builddir=DIR Build in this location instead of building within the distribution tree. Recomended! Does this mean the build is not the same as the install, i.e. while building, files are being stored somewhere and removed after final installation? Thank you, H.

1 0