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

[BGL] Question on dijkstra-example.cpp
by Cedric Laczny 11 Nov '10

11 Nov '10

Hello, could you please explain what this code in the dijkstra-example (http://www.boost.org/doc/libs/1_44_0/libs/graph/example/dijkstra-example.cpp) exactly does? predecessor_map(&p[0]).distance_map(&d[0]) I understand the non-named-parameter version, but the combination of predecessor_map with distance_map is not so obvious to me. What does the "." do there? Do I get it right, that the predecessor_map or distance_map, respectively, automatically wrap the vectors p and d into iterator_property_maps? Thank you. Best regards, Cedric

2 1

[iostreams] Problem with gzip_decompressor
by 陳韋任 11 Nov '10

11 Nov '10

Dear all, Some compressed data was stored in a vector<char> buf (don't know if it's a right way to store compressed data). I used the "buf" as a source, and tried to add a gzip_decompressor to restore those data. I have tried the following approach, but it doesn't work. vector<char> buf; /* fill in the buf */ namespace io = boost::iostreams; io::filtering_istream in(boost::make_iterator_range(buf)); in.push(io::gzip_decompressor()); What I got is an exception, terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector <std::logic_error>>' what(): chain complete Aborted How can I compose a source which comes from a container, and add a gzip_decompressor to the source? Thanks in advance. Regards, chenwj -- Wei-Jen Chen (陳韋任) Programming Language and System Laboratory (EC131) Department of Computer Science, National Chiao Tung University

3 4

[fusion] Using fusion::at_c on a sequence or an unused type
by teajay 10 Nov '10

10 Nov '10

Hi, I'm working on finalizing an extension to spirit/qi in which I need to extract an attribute from a fusion sequence in a parser context. The problem I'm facing is that this parser attribute can be an unused type or a valid fusion sequence. Just to illustrate my problem: template <typename Elements,typename Attribute> struct parser_dispachter { template <typename Index> void parse(Index &idx, Attribute &attr) { fusion::at_c<Index::value>(Elements).parse(fusion::at_c<Index::value>(attr)); } }; I actually have two cases: 1. Attribute = unused_type -> at_c replacement should return unused 2. Attribute = fusion sequence -> at_c replacement should return the corresponding attribute reference. What would be the best solution to extract the give attribute ? Regards, Thomas Bernard

2 2

[build] installing layout=versioned into same 'prefix'
by Joe Steeve 10 Nov '10

10 Nov '10

Hello, I am doing this on Debian GNU/Linux. When installing with --layout=versioned, boost.Build creates symlinks without 'version tag' to libraries with version-tag in their name. This, IMHO, defeats the purpose of --layout=versioned. These symlinks collide with existing layout=system installation, in the same prefix. Is there any way to avoid this? Regards, Joe -- Joe Steeve HiPro IT Solutions Pvt. Ltd. http://hipro.co.in/ -- Joe Steeve HiPro IT Solutions Pvt. Ltd. http://hipro.co.in/

1 0

[BFL] Efficient neighbourhood search in trees
by Vincent Spruyt 10 Nov '10

10 Nov '10

Hi all, I'm trying to calculate the average edge weight within a certain neighbourhood for each vertex in a tree. Basically, I have a (minimum spanning) tree and I want to remove all edges whose weight is much larger than the weight of edges closeby. The result would be a forrest of several connected components (I'm working on a clustering algorithm). Currently I naively iterate over each edge. For each edge, I find the source and target vertex. Then, for both the source and the target vertices, I calculate the average weight in a certain neighbourhood using a recursive function as following simplified pseudocode indicates: E: List of edges in MST getNeighbourhoodStats( depth, startvertex, edge ){ out_edges = getIncedentEdges(startvertex) foreach out_edges as out if( out!=edge){ totweight += getWeight(out); if(depth>0) getNeighbourhoodStats(depth-1, target(out), out) } end } main(){ foreach E as e //Get the total weight on the left of the edge int totweight_left=0; getNeighbourhoodStats( 5, source( e), E, totweight_left ) //Get the total weight on the right of the edge int totweight_right=0; getNeighbourhoodStats( 5, target( e), E, totweight_right ) end } However, this approach is rather inefficient as a lot of calculations are redundant (takes about 10ms for 300 edges which is way too slow for the purpose I had in mind). Any ideas for a better solution? Tnx a lot, Kind regards, Vincent

3 4

[Program Options] Positional args in help message
by Lindley M French 10 Nov '10

10 Nov '10

The examples for how to use positional arguments are great and seem to work fine. However, I'm wondering if there's an automatic way to include that information in the help message. In particular, since the tutorial goes to the trouble of telling you how to exclude the positional options from the options help, it would be nice to know what to replace it with. I could manually write a line detailing the usage before outputting the options description, of course, but I was wondering if there was anything in PO to handle this automatically.

3 2

[fusion] simple Fusion question
by Eric Niebler 10 Nov '10

10 Nov '10

Given an mpl::vector of keys, and a fusion::vector of values, what's the easiest way to create a fusion::map? The obvious solution of just zipping the two sequences and passing the result to fusion::as_map doesn't work because as_map (somewhat narrowly) requires all elements to be fusion::pair, and Fusion's zip algorithm produces sequences of vector2. I'd like to see this generalized so that (a) as_map only requires sequences of 2-tuples and (b) fusion::pair meets the requirements for 2-tuples. -- Eric Niebler BoostPro Computing http://www.boostpro.com

3 4

Boost Hashtable
by Ricardo Uribe Lobello 10 Nov '10

10 Nov '10

Good afternoon, I have been trying to use the boost::intrusive::hashtable class but i don't understand clearly how it works. Additionally, I would like to change the hashing function used. Where can i get some example of the hashtable usage? Thanks in advance Ricardo

1 0

[fusion] conversion from map *values* to vector
by alfC 09 Nov '10

09 Nov '10

I have a general fusion map for example map< pair<p, double> , pair<q, double> > if I use as_vector I get vector<pair<p, double> , pair<q, double> > is there a function that given the map converts it into vector<double, double> i.e. gives a vector with only the values of the key and not the keys/ value pair? Thank you, Alfredo

2 1

Boost 1.44.0 on a IBM P575 Power6 (AIX 6.1, XL C/C++ v10.1)
by Carmine Paolino 09 Nov '10

09 Nov '10

Hello, someone managed to compile Boost 1.44.0 on AIX with XL v10? We are currently working on a massively parallel network analyzer that uses the Parallel Graph Library, so getting it working on a cluster such as this it's fundamental. For more informations see this bug #4828 (https://svn.boost.org/trac/boost/ticket/4828) Any help will be appreciated, thanks in advance.

2 1