Dgraph is a library for creating, analysing and manipulating graphs, written in the D programming language. It aims to be fast and memory-efficient while also being easy to use and extend.
The project is in very early experimental stages of development, so breaking changes may occur on a regular basis (although every effort will be made to ensure that such breakage is justified by the gains). News and updates on the project will be published on the author's blog: http://braingam.es/
Git repository: https://github.com/WebDrake/Dgraph
Dgraph is distributed under the terms of the GNU General Public License, version 3 or (at your option) any later version.
Dgraph currently implements two different graph types (module dgraph.graph):
IndexedEdgeList is an adaptation to D of the similarly-named igraph
data type, while CachedEdgeList is an extension of the indexed edge
list that (as its name indicates) caches the results of various calculations
in order to provide faster performance.
Graphs may be directed or undirected, but Dgraph currently offers no support for weighted graphs or any other edge or vertex properties. Arbitrary vertex IDs (e.g. strings) are not currently supported.
The module dgraph.metric offers a selection of metrics for calculating different graph properties: currently betweenness centrality and largest connected cluster size are implemented.
Finally, the library provides a small selection of benchmarks for graph construction and calculation of graph metrics.
Dgraph is a source library and so does not need to be compiled in order to use: just import the modules into your own D program. DUB packaging is supported and can be used to build programs that have Dgraph as a dependency (see http://code.dlang.org/ for more information).
Several test utilities are provided that offer benchmarking of key features.
These can be built either using make (make all or make [name]) or with
dub (dub build dgraph:[name]). The currently available utilities are as
follows:
-
dgraph_graphtest benchmarks the creation of graphs from scratch, using two sample graphs with 50 and 10,000 nodes respectively.
-
dgraph_betweenness50 benchmarks betweenness centrality calculation on the 50-node sample graph.
-
dgraph_betweenness10k benchmarks betweenness centrality calculation on the 10,000-node sample graph.
Code contributions to Dgraph are welcome. Please try to follow the D style guidelines: http://dlang.org/dstyle.html
Feature requests and bug reports can be submitted via the Dgraph GitHub issue list.