February 2008 - Boost-users

[BGL] Access bundle graph properties
by Ioannis Nousias 19 Feb '08

19 Feb '08

How does one access a bundle graph property in adjacency_list ? for example struct VertexProperties { int index; bool status; }; struct EdgeProperties { int index; bool status; }; struct GraphProperties { int something; }; boost::adjacency_list < boost::listS, boost::vecS, boost::directedS, VertexProperties, EdgeProperties, GraphProperties > graph; if 'v' is a vertex then I can do 'grap[v].status' to get that vertex property. But what about the graph properties? I've tried 'graph.something' but that didn't work :( Any suggestions ? regards, Ioannis

4 4

[program option] Unicode support
by Hector C 19 Feb '08

19 Feb '08

Hi fellow boost fans, I cannot find the appropriate class to produce Unicode messages in boost::program_options::options_description. I'm looking for something like: woptions_description(const std::wstring &, unsigned = m_default_line_length); Any pointers? (no pun intended) Is there a library similar to boost::program_options with full Unicode support? Cheers, ++Hector C.

3 3

boost vs libboost naming problem with bjam compile for vc90
by Phelps Williams 19 Feb '08

19 Feb '08

I am a first time boost user and I am interested in playing with some of the regex features. I am forced to compile boost from svn because I'm using ms visual c++ 9.0. I looked into this process and it appears that no modifications are necessary for a successful compilation. With bjam.exe in the boost root directory I execute the following command to compile: bjam.exe -toolset=msvc debug stage The compilation completes without any errors and all expected lib and dlls are present in $BOOST_ROOT\stage\lib as expected. I then setup a simple regex project to test things out: #include "stdafx.h" #include <boost/regex.hpp> #include <iostream> #include <string> int main() { std::string line; boost::regex pat( "^Subject: (Re: |Aw: )*(.*)" ); while (std::cin) { std::getline(std::cin, line); boost::smatch matches; if (boost::regex_match(line, matches, pat)) std::cout << matches[2] << std::endl; } } Additionally, I have configured my boost root directory as an Additional Include Directory under Properties -> C/C++ -> General I have also configured my compiled libraries under (Properties -> Linker -> General) Additional Library Directories to the location where the compiled libs were placed. $BOOST_ROOT\stage\lib During compilation of this project I receive link errors: LINK : fatal error LNK1104: cannot open file 'libboost_regex-vc90-mt-gd-1 _35.lib' The confusing part is in the stage\lib directory all the libs are titled in the following format "boost_libname_vc90-mt-gd.lib" or ".dll" For some reason the libraries are missing the *lib* in front of the name...for example "libboost_regex-vc90-mt-gd-1_35.lib" I tried to rename this boost_regex example to libboost_regex but I ran into runtime exceptions. Specifically that "boost_regex-vc90-mt-gd-1_35.dll" was not found. I then copied this dll into the Debug folder where the compiled exe is being executed from and everything executed properly. Why are the static libraries not named properly? Thanks -Phelps

4 4

[python] Need help with embedding
by Robert Dailey 18 Feb '08

18 Feb '08

Hi, Below is my source code. What I'm trying to do is expose a C++ function to the embedded interpreter so I may call it from scripts executed by that embedded interpreter. Right now it just crashes with what seems to be an access violation against NULL: SOURCE: --------------------------------------------------------------------------- static void DoFoo() { int breakhere = 0; } using namespace boost; BOOST_PYTHON_MODULE(pythontest) { python::def( "DoFoo", DoFoo ); } void SetupPython() { if( PyImport_AppendInittab( "pythontest", NULL ) != -1 ) { int breakhere = 0; } else { int breakhere = 0; } python::object pytest = python::exec_file( "pythontest.py" ); } void BeginPythonTest() { Py_Initialize(); SetupPython(); } The pythontest.py file does nothing more than call *DoFoo()*. I've been looking at the *boost\libs\python\example\quickstart\embedding.cpp *, however it isn't very helpful and hard to read. In addition it calls a function named "initembedded_hello" which is nowhere to be found... This won't even compile? Any help is greatly appreciated!

1 1

lock free / wait free
by jono 18 Feb '08

18 Feb '08

hi all, i was wondering if there are any formal initiatives in the boost community to provide 'lock free' data structures using CAS (compare and swap)? in particular, this looks promising... "Instead of providing a separate atomic smart pointer, we propose additions to shared_ptr's interface to address this use case. A typical example scenario — a "lock-free" reader/writer pattern — that takes advantage of this functionality is outlined below:..." [http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2007/n2297.html#atomic] is this likely to happen this year? i've seen some other discussion, such as http://www.nabble.com/Lock-free-atomic_shared_ptr%3CT%3E--td14958758.html i've used the Microsoft interlocked API's in a past (audio processing) project, but i'm developing a video pipeline for Linux/Win32 and it would be a huge 'boost' to be able to eliminate some locking and have deterministic real time. if that were introduced via the shared_ptr interface, i'd be over the moon. unfortunately, i don't have the skills to really contribute to this initiative, but i wld like to add my voice to those who might find a use for it! cheers, jono

1 0

WG: Boost Build
by Klett, Stefan 18 Feb '08

18 Feb '08

Hi fellow list subscribers, The last days I have been busy to build a build-environment for a library project with boost build - I am new to boost and have no previous experience with boost build - what may account for the following problems I ran into The proposed structure of the library project is like this: Project-root | \ \ \ Include src doc examples / | testcase1 testcase2 ... Now I need some help to make bjam perform the following tasks: 1) build the library (this one works already) 2) generate a doxygen based documentation out of the the *.cpp files in src 3) build the testcases in the "testcaseX" directories For the time being I'm lacking the knowledge how to build targets (the above tasks separately - I tried to use the "explicit" keyword but it seems that missed something - bjam works trough all targets given in the Jamroot file which I placed in the "project-root" directory - could somebody please give me an example how to use targets and "explicit" correctly Further I need an hands on example how to get along with the doxygen documentation - my efforts up to now look like this - but this actually seems to do nothing at all Import doxygen : doxygen ; doxygen.doxygen doc : geneva : [ glob include/*.h ] #: requirements <include>include : ; The Question is : how do I extract the information from the files in src and write the doxy docu to doc ? And last but not least : how is the correct syntax for building sources in subdirectories - where have I to put the Jamfiles and how to I tell bjam to look for them ? Many Thanks in advance Best regards Stefan -------------------------------------------------------- Stefan Klett (Dipl. Informatiker) | E-mail: stefan.klett(a)kit.edu Forschungszentrum Karlsruhe / KIT | Web: http://www.fzk.de SCC Steinbuch Centre for Computing| Phone: +49 (0)7247 828649 Hermann-von-Helmholtz-Platz 1 | Fax: +49 (0)7247 824972 D-76344 Eggenstein-Leopoldshafen | Mobile: +49 (0)176 20317250 -------------------------------------------------------------------

2 1

Boost.IOStreams and Custom filters on strings
by Stefan Arentz 18 Feb '08

18 Feb '08

I've written a really simple aggregate_filter that runs the world famous and highly secret ROT13 encryption. This filter works fine when I use a file_sink and file_source. But I really just want to run this on a bunch of string buffers. Unfortunately it does not work, the destination string is always empty. Does anyone understand why? Here's the code, tested against Boost 1.34.1 on OS X 10.5.1: #include <iostream> #include <memory> #include <string> #include <boost/iostreams/copy.hpp> #include <boost/iostreams/device/back_inserter.hpp> #include <boost/iostreams/filter/aggregate.hpp> #include <boost/iostreams/filtering_stream.hpp> template< typename Ch, typename Alloc = std::allocator<Ch> > class basic_rot13_filter : public boost::iostreams::aggregate_filter<Ch, Alloc> { private: typedef boost::iostreams::aggregate_filter<Ch, Alloc> base_type; public: typedef typename base_type::char_type char_type; typedef typename base_type::category category; typedef std::basic_string<Ch> string_type; public: basic_rot13_filter() { } private: typedef typename base_type::vector_type vector_type; void do_filter(const vector_type& src, vector_type& dest) { for (size_t i = 0; i < src.size(); ++i) { dest.push_back(src[i] + 13); } } }; typedef basic_rot13_filter<char> rot13_filter; typedef basic_rot13_filter<wchar_t> wrot13_filter; int main() { std::string src = "SourceText"; boost::iostreams::filtering_istream in; in.push(boost::make_iterator_range(src)); std::string dst; boost::iostreams::filtering_ostream out; out.push(rot13_filter()); out.push(boost::iostreams::back_inserter(dst)); boost::iostreams::copy(in, out); std::cout << dst << std::endl; return 0; }

3 7

boost::serialization and platform portability
by Mark Schlegel 18 Feb '08

18 Feb '08

Boost User list: I have a question about boost::serialization and portability. When I read in the overview for this feature at http://www.boost.org/libs/serialization/doc/overview.html point 7 states that "Data Portability - Streams of bytes created on one platform should be readable on any other". I'm trying to clear up how far portability goes. I've tested serialization on two different hardware platforms (Intel Linux with Boost 1.34.1 and Sparc Solaris with Boost 1.33.1) which have different endians. The archiving writes and reads fine as long as the reading is done on the newer or equal version of Boost. Basically I wrote a serialization file with boost::archive::text_oarchive on the Solaris box and ftped the file to the Linux host and read it in fine. But going the other way (1.34.1 -> 1.33.1) aborted which is understandable because a archive version level change (3 vs 4) existed between those. So is readability assured by serialization as long as you're going from an older or equal version of boost? We can handle our maintaining the very latest boost version in our code and customers would send us their archives for testing and they'd either have the same or older version of boost in their copy. Their code would never be newer than ours. If this is obeyed will their archives always be readable by our code? Mark

2 1

BGL and interactive visualization
by mbn 18 Feb '08

18 Feb '08

Hi, Maybe this is an old topic but I can't find related answers. I have some data containing certain "entities" related to others by one or more attributes. For example, Alice's cat may be connected to Bob's dog because both are named "Bill". Also, they can be connected because they live in the same county. Furthermore, Bob's dog can be connected to Chuck's bird because they are of the same color Something like this: (A'sC)--------------(B'sD)-----------------(C'sB) Name=Bill Color=Beige County=CD The issue is, the attributes connecting entities can be hundreds or even thousands, so creating a graphviz file showing all those connecting attributes in the edges will result in a mess. Instead, I would like to create an application such that only the vertices (with a label inside) and the edges (without nothing) are initially shown, and when the user double clicks on a specific edge then some window can appear showing all the attributes causing X and Y to be connected in the graph. I know I can create the graphs using BGL, but once I have the graphs I do not know what library to use in order to visualize it Can somebody help? Perhaps there is a different/better way of doing the same with a different tool like GraphML + Some Visualization program? Thanks, Mark -- View this message in context: http://www.nabble.com/BGL-and-interactive-visualization-tp15514275p15514275… Sent from the Boost - Users mailing list archive at Nabble.com.

3 2

[boost-users] [bjam BOOST_AUTO_LINK_NOMANGLE]
by chun ping wang 18 Feb '08

18 Feb '08

I want to build msvc library by defining BOOST_AUTO_LINK_NOMANGLE but i also want to specify multithreaded build. Example I want my thing to auto link to boost_python-mt.lib Is their a way to do this. BOOST_AUTO_LINK_NOMANGLE only links to boost_python.lib Thanks.

1 0