In this part we will show you how to build this graph using Graph/Redis, and we will see how to apply some of the base commands and the algorithms on this sample
First we need to add the nodes a, b, c, d and e
To do so, we use the command gnode as in the following example:
gnode graph1 a
gnode graph1 b
gnode graph1 c
gnode graph1 d
gnode graph1 e
or we can use gnode to add multiple nodes at once, like in this example:
gnode graph1 a b c d e
graph1 didn't exist before, but it has been created once we added the first node to it.
By the end of running add the nodes, our graph looks like this
To add weighted edges to the graph, we use the command gedge which takes 4 paramters:
graph_name node1 node2 edge_value
gedge graph1 a b 1
gedge graph1 b c 1
gedge graph1 c d 1
gedge graph1 d e 1
gedge graph1 e a 3
gedge graph1 a d 20
And the result will be exactly as in the graph depicted at the top of the page
So let's now play with some commands, and see what Graph/Redis can do
Let's assume we want to list the neighbour nodes of some node, let's say a in our example.
neighbour nodes of a are the nodes which have direct edge with the node a. To list
the neighbours, we use the command gneighbours like this.
gneighbours graph1 a which will return a list of the following values:
1) b
2) d
3) e
Let's assume we want to list the common neighbours between two difference nodes. For that, we use
the command gcommon. For example, to know the common nodes between a and d
we run the command gcommon graph1 a d which will return a list that consists only of e, as
it's clear from the graph that it's the only common neighbour of a and d.
With graph-redis, you can also construct directed graphs, that affect the return values of some commands
To set the graph as directed, call gsetdirected command before you add any edges
gnode graph1 a b c gsetdirected graph1 gedge graph1 a b 10 gedge graph1 c a 2 gneighbours graph1 a # Returns['b'] gneighbours graph1 c # Returns['a'] gneighbours graph1 b # Returns[] gincoming graph1 b # Returns['a'] Graph/Redis supports also some algorithms on graphs, here we will show some:
To find the minimum spanning tree of a connected graph, you can use the command gmintree.
This command takes two paramters: the graph name, and another graph name in which the minimum spanning tree will be saved.
for example, running gmintree graph1 graph2 on the previous graph1 will result in the following graph(tree)
being saved in graph2:
It's worth mentioning that Graph/Redis minimum spanning tree command uses Prim's algorithm.
To see the full list of the commands and the algorithms supported by Graph/Redis, please check the commands page