pythongrapharangodbaqlpyarango

graph plot API for ArangoDB pyArango

I am using ArangoDB community edition, I can query on a created graph in AQL and get results in JSON which is graphically visualized on ArangoDB web interface tool.

AQL Query

FOR v,e,p IN 1..3 OUTBOUND 'germanCity/Hamburg' GRAPH 'routeplanner' 
OPTIONS{bfs :true} 
RETURN p

JSON output

[
  {
    "edges": [
      {
        "_key": "6392826",
        "_id": "germanHighway/6392826",
        "_from": "germanCity/Hamburg",
        "_to": "germanCity/Cologne",
        "_rev": "_WmZ77pW--D",
        "distance": 500
      }
    ],
    "vertices": [
      {
        "_key": "Hamburg",
        "_id": "germanCity/Hamburg",
        "_rev": "_WmZ77Z---_",
        "population": 1000000,
        "isCapital": false,
        "loc": [
          53.5653,
          10.0014
        ]
      },
      {
        "_key": "Cologne",
        "_id": "germanCity/Cologne",
        "_rev": "_WmZ77Y6--B",
        "population": 1000000,
        "isCapital": false,
        "loc": [
          50.9364,
          6.9528
        ]
      }
    ]
  },
  {
    "edges": [
      {
        "_key": "6392840",
        "_id": "internationalHighway/6392840",
        "_from": "germanCity/Hamburg",
        "_to": "frenchCity/Paris",
        "_rev": "_WmZ77pa--_",
        "distance": 900
      }
    ],
    "vertices": [
      {
        "_key": "Hamburg",
        "_id": "germanCity/Hamburg",
        "_rev": "_WmZ77Z---_",
        "population": 1000000,
        "isCapital": false,
        "loc": [
          53.5653,
          10.0014
        ]
      },
      {
        "_key": "Paris",
        "_id": "frenchCity/Paris",
        "_rev": "_WmZ77Z---D",
        "population": 4000000,
        "isCapital": true,
        "loc": [
          48.8567,
          2.3508
        ]
      }
    ]
  },
  {
    "edges": [
      {
        "_key": "6392843",
        "_id": "internationalHighway/6392843",
        "_from": "germanCity/Hamburg",
        "_to": "frenchCity/Lyon",
        "_rev": "_WmZ77pa--B",
        "distance": 1300
      }
    ],
    "vertices": [
      {
        "_key": "Hamburg",
        "_id": "germanCity/Hamburg",
        "_rev": "_WmZ77Z---_",
        "population": 1000000,
        "isCapital": false,
        "loc": [
          53.5653,
          10.0014
        ]
      },
      {
        "_key": "Lyon",
        "_id": "frenchCity/Lyon",
        "_rev": "_WmZ77Z---B",
        "population": 80000,
        "isCapital": false,
        "loc": [
          45.76,
          4.84
        ]
      }
    ]
  }
]

Equivalent Graph

Graph Viewer

As we can get visualized graph output in web interface I would like to display the same in Language<->ArangoDB. Language here can be driver language supported : Python, Java , C# etc.

I am using pyArango for interfacing with ArangoDB

I couldnt find an ArangoDB API for getting this graph visualization in JPG or matlibplot.

Is there any other way apart from using below two options?

  1. Use networkx.draw(networkx.graph)
  2. matplotlib.pyplot
Solution

  • If you need graph visualization then Graphviz library is for you. And if Python is OK then you only need a Python binding-library graphviz (that utilizes DOT language representation internally.)

    It is super easy to feed your graph JSON from Arango DB to graphviz for rendering.

    You can customize it to your style, add labels, colors, reshape nodes and more.

    Here is a simple example of your sample JSON:

    from graphviz import Digraph

arango_graph = [
    {
        "edges": [
            {
                "_key": "6392826",
                "_id": "germanHighway/6392826",
                "_from": "germanCity/Hamburg",
                "_to": "germanCity/Cologne",
                "_rev": "_WmZ77pW--D",
                "distance": 500
            }
        ],
        "vertices": [
            {
                "_key": "Hamburg",
                "_id": "germanCity/Hamburg",
                "_rev": "_WmZ77Z---_",
                "population": 1000000,
                "isCapital": False,
                "loc": [
                    53.5653,
                    10.0014
                ]
            },
            {
                "_key": "Cologne",
                "_id": "germanCity/Cologne",
                "_rev": "_WmZ77Y6--B",
                "population": 1000000,
                "isCapital": False,
                "loc": [
                    50.9364,
                    6.9528
                ]
            }
        ]
    },
    {
        "edges": [
            {
                "_key": "6392840",
                "_id": "internationalHighway/6392840",
                "_from": "germanCity/Hamburg",
                "_to": "frenchCity/Paris",
                "_rev": "_WmZ77pa--_",
                "distance": 900
            }
        ],
        "vertices": [
            {
                "_key": "Hamburg",
                "_id": "germanCity/Hamburg",
                "_rev": "_WmZ77Z---_",
                "population": 1000000,
                "isCapital": False,
                "loc": [
                    53.5653,
                    10.0014
                ]
            },
            {
                "_key": "Paris",
                "_id": "frenchCity/Paris",
                "_rev": "_WmZ77Z---D",
                "population": 4000000,
                "isCapital": True,
                "loc": [
                    48.8567,
                    2.3508
                ]
            }
        ]
    },
    {
        "edges": [
            {
                "_key": "6392843",
                "_id": "internationalHighway/6392843",
                "_from": "germanCity/Hamburg",
                "_to": "frenchCity/Lyon",
                "_rev": "_WmZ77pa--B",
                "distance": 1300
            }
        ],
        "vertices": [
            {
                "_key": "Hamburg",
                "_id": "germanCity/Hamburg",
                "_rev": "_WmZ77Z---_",
                "population": 1000000,
                "isCapital": False,
                "loc": [
                    53.5653,
                    10.0014
                ]
            },
            {
                "_key": "Lyon",
                "_id": "frenchCity/Lyon",
                "_rev": "_WmZ77Z---B",
                "population": 80000,
                "isCapital": False,
                "loc": [
                    45.76,
                    4.84
                ]
            }
        ]
    }
]

graph_name = 'amazing'

g = Digraph(graph_name, filename=graph_name, format='jpeg', engine='neato')
g.attr(scale='2', label='Look at my graph my graph is amazing!', fontsize='18')
g.attr('node', shape='circle', fixedsize='true', width='1')

for item in arango_graph:
    for vertex in item['vertices']:
        g.node(vertex['_id'], label=vertex['_key'])
    for edge in item['edges']:
        g.edge(edge['_from'], edge['_to'], label=str(edge['distance']))

# Render to file into some directory
g.render(directory='/tmp/', filename=graph_name)
# Or just show rendered file using system default program
g.view()

    Just 3 code lines for customization and 5 more to feed the graph visualization renderer. And please note that Arango Web UI does not render all of the edges between the same pair of nodes while graphviz does and you can style each differently.

    You will need to install graphviz library and Python bindings

    Step #1: Install library

    Assuming your machine is Ubuntu:

    sudo apt install graphviz

    Step #2: Get Python bindings

    pipenv install graphviz

    If you not yet using Pipenv you can install with good old Pip:

    pip install graphviz

    Step #3: Run sample and enjoy

    Amazing graph, rendered in JPEG