Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Some Ruby functions to manipulate graph files
Tree: e60c658587

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.


Build Status

The goal of this project is to provide some useful Ruby functions to manipulate graph files.

Note: some of the examples below are outdated, since before the 0.1.5 version, nodes & edges were represented as hashes, and now they are Node & Edge objects, respectively. However, the principles stay the sames.


The best way is to use the graphs gem:

gem install graphs

If you want to have the latest version, clone this repo, build the gem, and install it:

git clone git://
cd Graphs.rb
gem build graphs.gemspec
gem install ./graphs-*.gem # you may want to use sudo


To perform the tests, clone this repo, then go into tests repertory, and execute tests.rb:

git clone git://
cd Graphs.rb/tests
ruby tests.rb

Make sure you have the latest version.

Graph Class

The Graph class is a simple graph with nodes and edges. It provides three read-write attributes: nodes, edges, and attr (attributes of the graph, like author or description). It can be written in a file using Graph#write method.


irb> require 'graph'
=> true
irb> nodes = [{'name'=>'me'}, {'name'=>'you'}]
=> [{'name'=>'me'}, {'name'=>'you'}]
irb> edges = [{'node1'=>'you', 'node2'=>'me', 'directed'=>true},
              {'node1'=>'you', 'node2'=>'me', 'directed'=>true}]
=> [{'node1'=>'you', 'node2'=>'me', 'directed'=>true}, {'node1'=>'you', 'node2'=>'me', 'directed'=>true}] 
irb> g =, edges)
=> #<Graph:0x9e08e3c @nodes=[{"name"=>"me"}, {"name"=>"you"}], @edges=[{"node1"=>"you", "node2"=>"me", "directed"=>true}, {"node1"=>"you", "node2"=>"me", "directed"=>true}]>

You can perform some operations on graphes using the |, &, ^, + or - operators. See the documentation for more informations.

GDF Module

The GDF module is used to parse GDF files using the unique method GDF::load(filename) which returns a Graph object.


Imagine we have a file as below:

 $ cat trips.gdf
 nodedef> name VARCHAR,country VARCHAR
 edgedef> node1,node2,day INT,duration INT

Then, using irb, we use the GDF module:

 $ irb
 irb> require 'graphs/gdf'
 => true
 irb> g = GDF::load 'trips.gdf'

We can now access nodes

 irb> g.nodes
 => [{'name'=>'Foo', 'country'=>'England'}, {'name'=>'Bar', 'country'=>'India'}]

and edges

 irb> g.edges
 => [{'node1'=>'Bar', 'node2'=>'Foo', 'day'=>62, 'duration'=>14},
     {'node1'=>'Foo', 'node2'=>'Bar', 'day'=>154, 'duration'=>7}]

now, we can add a node and an edge

irb> g.nodes.push {'name'=>'John', country=>'USA'}
irb> g.edges.push {'node1'=>'John', 'node2'=>'Foo', 'day'=>42, 'duration'=>12}

but we forgot that all edges are directed ones. That's ok, just use the set_default method:

irb> g.edges.set_default 'directed' => true
irb> g.edges
=> [{'node1'=>'Bar', 'node2'=>'Foo', …, 'directed'=>true},
    {'node1'=>'Foo', 'node2'=>'Bar', …, 'directed'=>true},
    {'node1'=>'John', 'node2'=>'Foo', …, 'directed'=>true}]

Note that the set_default method is defined for edges and nodes. It accepts multiple arguments, and you only need to call it once, it will work for every new node or edge (just use .push method to add new ones).

then, we can save our new graph in a new file

irb> g.write('new_trips.gdf')

Note to Gephi users: You can add the :gephi option to g.write(…) if you have big numbers. Graph#write method use BIGINT type for big numbers, but Gephi does not support it and parses it as a string field. So using the following:

irb> g.write('new_trips.gdf', {:gephi=>true})

make sure that INT is used for all BIGINT fields.

Something went wrong with that request. Please try again.