Graphwalker

Provides graphs that you can read!

Introduction

Graphs are are a critical abstract data type underlying much of mathematics and computer science. They provide a fundamental model for describing newtorking, diagrams, routing, artificial intelligence, etc.

Many advanced software tools exist today to visualize graphs such as graphviz. However, data and concepts grow quickly. Although most visualization tools will produce impressive images, reviewing and understanding them can be overwhelming.

graphwalker takes the approach of salient visualization:

• Renders localized views of a single vertex
• Adjancent vertices (1 degree of separation)
• Pagination for all incoming and outgoing arcs
• Provides reproducible navigation via:
  • Clicking on vertices within the current view
  • Type ahead search for vertex properties
• HTML5 SVG display scaling
• Cayley Graph database for incremental traversal of large graphs
• GraphQL API for vertex queries

Requirements

You will need to install the following locally:

• Docker v19+
• node.js v6+
• yarn v0.24+

Graph Database

Graph Data

Currently, only the following project creates valid RDF graphs for this project:

• django_codecrumb

All graph data is stored in Cayley DB. This project runs Cayley in a Docker container.

Load Data

NOTE: Database must be initially shut down.

Copy your Cayley DB import file into the ./data folder (e.g. filename.nq.gz) and run the following Docker command:

docker-compose run --rm --entrypoint "cayley load -i /import/filename.nq.gz" db

Volume Inspection

To or manipulate the Cayley DB files, run the following:

docker-compose run --rm shell

Running

You can start the Cayley graph database using Docker as follows:

docker-compose up -d db

Web Application

Configuration

Edit etc/render_settings.json to specify how to displays your graph data

{
  "startVertexId": "<SHA1 vertex ID hash>",
  "vertexColumnPageSize": 8,
  "images": [
    ...
  ]
}

Building

Run the following to install all the necessary build dependencies using yarn:

yarn install

Run the following to build and bundle all Javascript code into the ./dist subdirectory:

yarn run generate
yarn run build-server
yarn run build-client

Then run the following to run the web server:

GRAPHWALKER_RENDER_SETTINGS=etc/render_settings.json yarn run web

Finally, open the following URL in your web browser:

open http://localhost:9080/

Local Development

Running a Test server

tmux is required to run a test server using watch mode on all source files:

./scripts/tmux.sh

This will run multiple watch windows for both client and server.

git hooks

Set up the git pre-commit hook as follows:

ln -s ../../scripts/pre-commit .git/hooks/pre-commit

GraphQL testing

Open up the following to use the Apollo GraphQL playground, an interactive and in-browser GraphQL IDE

open http://localhost:9080/graphql

Cleanup

To remove build files, run the following:

yarn run clean