VISU Application

Installation

1. Cloning the repo and running the app

git clone https://github.com/coin-quin/vis-a-vis.git -b dockerized

sudo docker build -f Dockerfile.kari . (See Kari's notes below)

sudo docker-compose up

The server is now running and can be accessed at localhost:8082, but no data is loaded on it yet. Three services are started by Docker: web (the web app, port 8082), mongo (the Mongo DB, port 27017), and mongo-express (the DB management dashboard, port 8081). NB: A debugger is attached to the web service on port 5858.

Usage

1. Loading data source (Kactus2 logs) from outside the container

cd datasources/log/

sudo npm install

node collector2.js

Use file log.txt as input in the last command. This data can be visualized with the User Timeframe or Session Timeline visualizations.

2. Loading data source (issue data) from within the web app

Go to Collectors/API collector. Select the API and enter the parameters. This data can be visualized with the other visualizations.

Kari's notes

I have "hacked" docker-compose.yml not to run "nmp install" since it seems to fail with the combination of pre-historical version of node and current state of library hub. Unfortunately, Hugo's original base image does not work out of the box. That's why "Kari's image" need to be created. At the moment you need to

sudo docker build -f Dockerfile.kari .

In next version proper image will be in Docker-hub.

Files

1. server.js - main file of the web app
2. Dockerfile - docker config file for the web app
3. docker-compose.yml - defines the three services started to run the app
4. package.json - node dependencies and metadata

Folders

1. Backend - Backend code of the app. Contains the routes, the mongo database, and the data collectors.
2. Frontend - Frontend code of the app. Containes the HTML/CSS files for the webpages and the js files of the visualizations.
3. Config - Configuration files for development (default) and testing (test). TODO: deployment config file.
4. Datasources - Data collectors and parsers for the data model.
5. Test - Mocha/Chai Unit Testing for the application.
6. Node modules - Node libraries for the web app. TODO: proper separation of deployment/development dependencies.