July 2010 - Boost-users - lists.preview.boost.org

Some results for planar graph coloring algorithm
by Mike Douglass 04 Jul '10

04 Jul '10

I have uploaded Planar6_July3.tz to http://groups.google.com/group/graph_coloring_program (README_July3) . It contains 100 maximal planar graphs of 50 vertices each. The graphs were generated using LEDA's function maximal_planar_graph(graph, number_of_vertices). I applied my graph coloring program "pl" in batch mode (using a perl script) to a directory containing the 100 graphs and it succeeded in finding a proper 4 coloring for EVERY SINGLE GRAPH. Average solution time per graph - about 1 second, of which a large part was probably disk read/writes. Does anyone have a favorite planar graph of <= 50 vertices that I can try my algorithm on? 50 vertices is not very large, but soon I hope to try graphs of 60, 70 or more vertices. I ran fr_layout on the 100 graphs but the resulting layout was not very clear. I think what I really need is a layout on a sphere and a way to look at the sphere from any angle. Does anyone know of such a thing. In the not too distant future I hope to be looking at graphs of up to 100 vertices. Anything but a spherical layout i believe would be too messy. You can find a clear graphical depiction of the algorithm at http://groups.google.com/group/graph_coloring_B1

1 0

vector iterator element-wise operation
by Max S. Kaznady 04 Jul '10

04 Jul '10

I'm using this example: http://stackoverflow.com/questions/773667/element-wise-operations-with-boos… and for some reason the function std::fill compiles and works just fine, but functions std::for_each and std::transform do no compile - saying that there is no compatible function found for the iterator type. using namespace boost::numeric::ublas; // Create a 30 element vector of doubles vector<double> vec(30); // Assign 8.0 to each element. std::fill(vec.begin(), vec.end(), 8.0); // Perform the "Gamma" function on each element and assign the result back // to the original element in the vector. // The next line does not compile, but the fill function compiles and executes just fine: std::transform(vec.begin(), vec.end(), vec.begin(), boost::math::tgamma); What's wrong? Cheers, Max

1 0

boost MPI, cannot get the results from workers
by Jack Bryan 04 Jul '10

04 Jul '10

Dear All : I designed a master-worker framework with boost MPI, in which the master can schedulemultiple tasks (numTaskPerWorkerNode) to each worker and then collects results from workers. if the numTaskPerWorkerNode = 1, it works well. But, if numTaskPerWorkerNode > 1, the master cannot get the results from workers. But, the workers can get the tasks from master. why ? I have used different taskTag to distinguish the tasks, but still does not work. Any help is appreciated. Thanks, Jack July 4 2010 _________________________________________________________________ Hotmail has tools for the New Busy. Search, chat and e-mail from your inbox. http://www.windowslive.com/campaign/thenewbusy?ocid=PID28326::T:WLMTAGL:ON:…

1 0

Re: [Boost-users] [boost.tuple] issue
by Niels Dekker - address until 2010-10-10 04 Jul '10

04 Jul '10

On 2010-07-04-Sun 16:14, Sean Farrow wrote: > boost::make_tuple(false, NULL); NULL is typically defined as an int (0)! HTH, Niels -- Niels Dekker http://www.xs4all.nl/~nd/dekkerware Scientific programmer at LKEB, Leiden University Medical Center

1 0

Re: [Boost-users] [boost.tuple] issue
by Niels Dekker - address until 2010-10-10 04 Jul '10

04 Jul '10

On 2010-07-04-Sun 15:18, Sean Farrow wrote: > I’ve got a tuple defined as: > Boost::tuple<bool, HWND> > > And when I do: > Boost::make_tuple(true, TempWindow); > > I get the error > C2440 cannot convert from int to hwnd__* > > Anyone got any ideas? > Visual c++ 2008, and boost 1.43. I guess the type of your TempWindow might not be HWND...? Otherwise maybe you should post a code snippet to clarify what you're trying to compile! HTH, Niels -- Niels Dekker http://www.xs4all.nl/~nd/dekkerware Scientific programmer at LKEB, Leiden University Medical Center

2 1

[boost.tuple] issue
by Sean Farrow 04 Jul '10

04 Jul '10

Hi: I've got a tuple defined as: Boost::tuple<bool, HWND> And when I do: Boost::make_tuple(true, TempWindow); I get the error C2440 cannot convert from int to hwnd__* Anyone got any ideas? Visual c++ 2008, and boost 1.43. Cheers Sean.

1 0

[proto] : Recursive functions in a lambda like language
by Manjunath Kudlur 03 Jul '10

03 Jul '10

I was playing with the idea of having the user declare functions (polymorphic, possibly recursive) in a lambda like language, and the capability to later call that function with different types of arguments. Here is an example of what I am thinking about. function fib = if_(_1<2) [ _1 ].else_[fib(_1-1) + (_1-2)]; int x = fib.eval(10); How should I define the "function" class? My first attempt at a solution was to use "boost::any" like class for function, I overloaded operator() in that class so that it returns a proto expression. I am attaching some source code that has a function class and a phoenix like language. Now, I am not able to come up with a solution for defining a polymorphic eval() for the function class. Since "function" has to be able to store any type inside it, it has to do type erasure. All methods in "function" has to be delegated to value_base class but that cannot delegate a polymorphic eval() function down to value_type. I can templatize "function" class and get monomorphic functions defined in this language. But that is less interesting. I would greatly appreciate any advice and suggestions on other solutions to this problem. I realize that this is not entirely proto specific and maybe it is a general C++ language question, but I am hoping proto users might have faced similar problems and provide guidance. Thanks, Manjunath

3 4

Re: [Boost-users] compiling fboost for x64
by Damien Hocking 03 Jul '10

03 Jul '10

The various command prompts are in "Visual Studio Tools" under your compiler in the Start menu. The --layout=system option gives you libs and dlls with no boost version or compiler version in the filenames. Damien On 03/07/2010 12:05 AM, Sean Farrow wrote: > > Ok, How do I get to a vs 64-bit command prompt, and what is the > layout=system. > > Thanks for everything. > > Sean. > > *From:* boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] *On Behalf Of *Damien Hocking > *Sent:* 03 July 2010 06:52 > *To:* boost-users(a)lists.boost.org > *Subject:* Re: [Boost-users] compiling fboost for x64 > > On what platform? Here's what I use for Windows: > > bjam.exe --toolset=msvc address-model=64 --stagedir=./stage64release > --build-dir=./build64release --layout=system --without-mpi > variant=release link=shared threading=multi runtime-link=shared > install --prefix=./win64release/indep --exec-prefix=./win64release/bin > --libdir=./win64release/lib --includedir=./win64release/include > bjam.exe --toolset=msvc address-model=64 --stagedir=./stage64debug > --build-dir=./build64debug --layout=system --without-mpi variant=debug > link=shared threading=multi runtime-link=shared install > --prefix=./win64debug/indep --exec-prefix=./win64debug/bin > --libdir=./win64debug/lib --includedir=./win64debug/include > > Note that "address-model=64" doesn't have a "--" prefix. This bites > lots of people (including me...) You need to run these in a Visual > Studio 64-bit command prompt. It's similar for Linux, --toolset=gcc > and from memory OSX is --toolset=darwin. > > Damien > > On 02/07/2010 11:33 PM, Sean Farrow wrote: > > Hi: > > I am currently in the process of writing an application that uses > boost that needs to be compilable under x64. Do I have to recompile > boost using bjam, if so how do I do this for x64 platforms? > > Cheers > > Sean. > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org <mailto:Boost-users@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] compiling fboost for x64
by Damien Hocking 03 Jul '10

03 Jul '10

On what platform? Here's what I use for Windows: bjam.exe --toolset=msvc address-model=64 --stagedir=./stage64release --build-dir=./build64release --layout=system --without-mpi variant=release link=shared threading=multi runtime-link=shared install --prefix=./win64release/indep --exec-prefix=./win64release/bin --libdir=./win64release/lib --includedir=./win64release/include bjam.exe --toolset=msvc address-model=64 --stagedir=./stage64debug --build-dir=./build64debug --layout=system --without-mpi variant=debug link=shared threading=multi runtime-link=shared install --prefix=./win64debug/indep --exec-prefix=./win64debug/bin --libdir=./win64debug/lib --includedir=./win64debug/include Note that "address-model=64" doesn't have a "--" prefix. This bites lots of people (including me...) You need to run these in a Visual Studio 64-bit command prompt. It's similar for Linux, --toolset=gcc and from memory OSX is --toolset=darwin. Damien On 02/07/2010 11:33 PM, Sean Farrow wrote: > > Hi: > > I am currently in the process of writing an application that uses > boost that needs to be compilable under x64. Do I have to recompile > boost using bjam, if so how do I do this for x64 platforms? > > Cheers > > Sean. > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1

compiling fboost for x64
by Sean Farrow 03 Jul '10

03 Jul '10

Hi: I am currently in the process of writing an application that uses boost that needs to be compilable under x64. Do I have to recompile boost using bjam, if so how do I do this for x64 platforms? Cheers Sean.

1 0