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

boost::variant and SGI's compose1 cause constness error
by junkmail.hoefling＠arcor.de 07 Jul '05

07 Jul '05

Hi there, I have a problem using boost::variant. I want apply_visitor to go through a list of pairs whose first type is a variant. Using compose1 and select1st from the SGI extensions to the STL yields a constness error. The problem is illustrated by the attached code which shoud be compiled with the g++ (the SGI extensions are here available via __gnu_cxx). What's wrong? Is it my fault using the library or is it a missing const version in apply_visitor_delayed_t()? Thanks a lot for any hints! Best, Erwin

3 2

circular buffer and gcc 4.0.0
by Vincent Torri 07 Jul '05

07 Jul '05

Hello, I need to use gcc 4.0 and circular buffer. But this latter seems unusuable with gcc 4.0. It gives a lots of error messages. Here is the 2 irst error messages: in debug.hpp: /home/torri/local/include/boost-1_33/boost/circular_buffer/debug.hpp:60: error: invalid use of undefined type 'const struct boost::cb_details::cb_iterator_base' /home/torri/local/include/boost-1_33/boost/circular_buffer/debug.hpp:22: error: forward declaration of 'const struct boost::cb_details::cb_iterator_base' the 2 classes in debug.hpp are cross-referenced. It seems that the forward declaration is not sufficient for gcc 4.0. in base.hpp: /home/torri/local/include/boost-1_33/boost/circular_buffer/base.hpp:139: error: using typedef-name 'boost::circular_buffer<T, Alloc>::iterator' after 'struct' /home/torri/local/include/boost-1_33/boost/circular_buffer/base.hpp:105: error: 'boost::circular_buffer<T, Alloc>::iterator' has a previous declaration here /home/torri/local/include/boost-1_33/boost/circular_buffer/base.hpp:139: error: friend declaration does not name a class or function gcc 4.0 doesn't like the 'struct' word. Putting the 'class' word instead of struct gives error too: /home/torri/local/include/boost-1_33/boost/circular_buffer/base.hpp:139: error: a class-key must be used when declaring a friend /home/torri/local/include/boost-1_33/boost/circular_buffer/base.hpp:139: error: friend declaration does not name a class or function Does someone see solutions for these problems ? Jan: it would be great if you try to finish circular buffer :) thank you Vincent Torri

1 0

BGL - Multiple graphs on the same data-structure.
by sr kumar 07 Jul '05

07 Jul '05

Hi, My application needs to have multiple graphs overlayed on the same set of data. Looks like used bundled properties would be very attractive. Since the data-set is likely to be large and the graphs will have to coexist, I'd like to build graphs of pointers to the base data. However, I am having trouble figuring out how to store pointers to the data structures within an adjacency_list. Any suggestions/pointers to examples would be much appreciated. -sr

2 1

BGL: Brandes betweenness and exterior properties
by Todd A. Gibson 07 Jul '05

07 Jul '05

Hello, I am working with brandes_betweenness_centrality in the BGL and am having difficulty correctly constructing the syntax for retrieving an EdgeCentralityMap using exterior properties. I can successfully invoke the function using bundled properties: struct EdgeAttrs { unsigned int index; double weight; }; template<class GraphType> boost::python::list betweenness_centrality(GraphType& g) { typedef typename GraphType::edge_descriptor Edge; std::vector<float> centmap(num_vertices(g)); std::vector<float> edgecent(num_edges(g)); brandes_betweenness_centrality (g, centrality_map(make_iterator_property_map(centmap.begin(), boost::get(vertex_index, g), float())) .edge_centrality_map(make_iterator_property_map(edgecent.begin(), boost::get(&EdgeAttrs::index, g),float())) .weight_map(boost::get(&EdgeAttrs::weight, g))); ... Among my attempts, here's a typical one that won't compile: typedef typename std::map<Edge, double> EdgeMap; EdgeMap myedges; //... boost::associative_property_map< EdgeMap > edge_assoc_map(myedges); brandes_betweenness_centrality(g, ... .edge_centrality_map(edge_assoc_map) ...) Can someone enlighten me? Thanks, -TAG

2 1

in_edges for a directedS adjacency_list
by sr kumar 07 Jul '05

07 Jul '05

Hi, Is there a reason that the in_edges are not being maintained for a directedS adjacency_list? A quick look through implementations of boost::detail::add_edge(*) indicates correct maintainance for a bidirectional case, and reuse of the out_edges for the undirected case. I'm hopefully not missing something obvious to more enlightened souls and would appreciate any ideas to get the in_edges correctly for a directedS adjacency_list. thanks, -sr

2 1

Re: [Boost-users] Serialization and null pointers
by boost＠iyates.com 07 Jul '05

07 Jul '05

Robert, Thanks for all your help, I really do appreciate it. I was already serializing an stl vector, so you would think I would have thought of that! heh. Anyhow, I read through serialization/shared_ptr.hpp and sort of did the same thing on my custom shared pointer, and it is working great. I especially liked this: #define private public #include <boost/shared_ptr.hpp> #undef private hehe Thanks again! - Chris ----- Original Message ----- From: "Robert Ramey" <ramey(a)rrsd.com> To: boost-users(a)lists.boost.org Date: Wed, 6 Jul 2005 14:48:51 -0700 Subject: Re: [Boost-users] Serialization and null pointers > I suspect you didn't include <boost/serialization/shared_ptr.hpp> which > includes serialization code for shared_ptr. It also includes shared_ptr.hpp > itself. This is similar to the way the headers are organized for all the > stl collections. If you want to serialize the boost share_ptr be sure to > read about in the docs as there are a few gotchas. Also note that there the > "real soon to be released" boost 1.33 has a different shared_ptr and > different serialization of the same. > > Good Luck. > > Robert Ramey > > boost(a)iyates.com wrote: > > I was using my own home-grown one. I wrote it before I started using > > boost though. I decided to try it out with boost::shared_ptr and > > here is what I got: > > > > c:\Boost\boost_1_32_0\boost\serialization\access.hpp(106): error > > C2039: 'serialize' : is not a member of 'boost::shared_ptr<T>' > > > > > > I tried searching on boost shared_ptr serialize but came up empty. > > Any suggestions? Thanks. > > > > ----- Original Message ----- > > From: "Robert Ramey" <ramey(a)rrsd.com> > > To: boost-users(a)lists.boost.org > > Date: Tue, 5 Jul 2005 20:18:00 -0700 > > Subject: Re: [Boost-users] Serialization and null pointers > > > >> boost(a)iyates.com wrote: > >>> Thanks. If I changed the order of the members being serialized it > >>> worked fine. Weird. > >>> > >>> However, I am running into a problem. I am using reference counted > >>> Smart Pointers. If I serialize the reference counter, then when it > >>> is loaded it will likely contain too large of a reference count > >>> because there were more objects referencing the pointer when it was > >>> saved than when it was loaded. However, it works fine it just > >>> doesn't properly delete the objects because the reference count > >>> never reaches 0. > >>> > >>> If I don't serialize the reference count then the objects try to > >>> delete themselves while boost is loading them from the file. I'm > >>> guessing because of the order everything happens, the final > >>> destination isn't increasing the reference count until after boost > >>> already dereferences it. > >>> > >>> Any suggestions? > >> > >> I'm not sure which reference counted Smart Pointer you're using. If > >> you use the boost smart_ptr, its already has serializaiton done. If > >> you want to add serialization to your own Smart Pointer, you can > >> look at how it was done for the boost smart pointer - now there are > >> two versions so you have lots of ideas available. > >> > >> Robert Ramey > >> > >> > >> > >> _______________________________________________ > >> Boost-users mailing list > >> Boost-users(a)lists.boost.org > >> http://lists.boost.org/mailman/listinfo.cgi/boost-users > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > >

1 0

Re: [Boost-users] Serialization and null pointers
by boost＠iyates.com 06 Jul '05

06 Jul '05

I was using my own home-grown one. I wrote it before I started using boost though. I decided to try it out with boost::shared_ptr and here is what I got: c:\Boost\boost_1_32_0\boost\serialization\access.hpp(106): error C2039: 'serialize' : is not a member of 'boost::shared_ptr<T>' I tried searching on boost shared_ptr serialize but came up empty. Any suggestions? Thanks. ----- Original Message ----- From: "Robert Ramey" <ramey(a)rrsd.com> To: boost-users(a)lists.boost.org Date: Tue, 5 Jul 2005 20:18:00 -0700 Subject: Re: [Boost-users] Serialization and null pointers > boost(a)iyates.com wrote: > > Thanks. If I changed the order of the members being serialized it > > worked fine. Weird. > > > > However, I am running into a problem. I am using reference counted > > Smart Pointers. If I serialize the reference counter, then when it > > is loaded it will likely contain too large of a reference count > > because there were more objects referencing the pointer when it was > > saved than when it was loaded. However, it works fine it just doesn't > > properly delete the objects because the reference count never reaches > > 0. > > > > If I don't serialize the reference count then the objects try to > > delete themselves while boost is loading them from the file. I'm > > guessing because of the order everything happens, the final > > destination isn't increasing the reference count until after boost > > already dereferences it. > > > > Any suggestions? > > I'm not sure which reference counted Smart Pointer you're using. If you use > the boost smart_ptr, its already has serializaiton done. If you want to add > serialization to your own Smart Pointer, you can look at how it was done for > the boost smart pointer - now there are two versions so you have lots of > ideas available. > > Robert Ramey > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > >

2 1

mpl help
by Zeljko Vrba 06 Jul '05

06 Jul '05

-----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 Hi! I would like to use mpl to generate code along the following lines (I'm trying to build a reusable library for data processing in a 'pipelined' fashion): // Arg1 and Arg2 are instances of 'stages' providing results to this // stage. F is the function taking as arguments const Arg1::result_type& // and const Arg2::result_type&. template<typename Result, typename F, typename Arg1, typename Arg2> struct stage_2 { ~ typedef Result result_type; ~ Result current_; ~ F f_; ~ Arg1 input1_; ~ Arg2 input2_; ~ stage_2(F f, Arg1 a1, Arg2 a2) : f_(f), input1_(a1), input2_(a2) { } ~ const Result &operator() { ~ current_ = f_(input1_(), input2_()); ~ return current_; ~ } }; The question: is it possible not to write manually separate stage_0, stage_1, stage_2, ... for stages taking as input results from the previous 0, 1, 2, ... stages? == Actually, what I'm trying to design: S1 -> S2 -> S3 -> .. -> Sk -> ..-> Sn ~ \_______________/ This is just a rough sketch; in the bottom line the arrow goes from S2 to Sk. Each stage (S'es in the diagram) provides input to its 'children'. This is akin to UNIX shell pipes, but more powerful in that each stage can have multiple inputs. The catch: data transformation at any stage can produce data type on output different from the one received on input. The stages should (obviously?) be parametrized by the stage result, the types of input data and the data transformation to be performed. My last resort is writing a python script to generate the boiler-plate C++ code from high-level description :) Maybe I'm on the wrong way trying to use mpl? any help is appreciated! Thanks. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (FreeBSD) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFCyYKoFtofFpCIfhMRAzVDAJ4lMW79S4QS23ivdsLp5xEw/+G+LwCeODR7 fNweo5iwDBnvCKfUcwX3hGI= =UTnS -----END PGP SIGNATURE-----

2 7

A newbie question about installation
by tong wang 06 Jul '05

06 Jul '05

HI: My first questio here, please be nice. I am trying to install the boost_1_31_0 on Cygwin, and get confused about which bjam should i use, I have tried boost-jam-3.1.10-1-ntx86 but get the following error: (it skipped the python libraries at the beginning, i don't need it but not sure if this matters for the following steps) ...patience... ...found 7570 targets... ...updating 3057 targets... MkDir1 C:\Boost spawn: No such file or directory Also tried the linux version : boost-jam-3.1.10-1-linuxx86 but have the following error this time. 1: Syntax error: "(" unexpected thanks a lot for any help . tong

3 2

question of how to run quickbook
by Weining Hao 06 Jul '05

06 Jul '05

Hi all, I am a very beginner of quickbook. Now I have some questions on it. Am I in a right mailinglist? I have downloaded quickbook.zip from sourceforge.net and extracted it. 1) Shall I compile quickbook.cpp directly for getting a excutable file? I tried, but lots files included by quickbook.cpp or .hpp cannot be found. e.g boost/ref.hpp in util.hpp... Or what should I do firstly? 2) In the extracted quickbook folder, there is a readme.txt file. The content is "Hi, Place me inside your boost/tools directory. " Which file shall I put into boost/tools directory under which path? 3) Are there some more detailed guides on quickbook except things on http://spirit.sourceforge.net/dl_docs/quickbook_doc/doc/html/index.html ? Thanks in advance. Best wishs, vning ___________________________________________________________ 雅虎免费G邮箱－中国第一绝无垃圾邮件骚扰超大邮箱 http://cn.mail.yahoo.com/?id=77071

1 0