Trail - 4th Year CompSci Dissertation Project

A Progressive Web App (PWA) that allows a user to get information about the trees on the Heriot-Watt using QR, Spatial and Speech Synth technologies. Augmented Reality is also experimented with. This project is no longer maintained - a currently maintained version can be seen at https://github.com/ruaridhmollica/trail

Stack

Hosting

Trail is hosted on Heroku and has a full valid SSH certification as well as a custom domain.

• Heroku also hosts the applications database.
• Heroku builds and deploys the web application automatically.

Database

Trail makes use of a PostgreSQL database with the postGIS extension installed for the ability to query and store geospatial data.

Backend

This web application is powered by Golang v1.13.1 and handles the following:

• File serving
• Page routing, made easier by the use of the Gin-Gonic/gin framework
• Database connection and querying (using Go-SQL-driver
• Outlining package details and dependencies to Heroku for hosting

Front-end

To make this web app usable and pretty the following technologies are used:

• HTML
• CSS
• JavaScript (to provide more complex functionality and handle API calls)
• JQuery
• Material Design Bootstrap (to make everything that little bit easier)

APIs

• Google Maps Javascript API is used to generate custom, cross-compatible, interactive maps.

JavaScript Libraries

• html5-qrcode by Minhaz - this library was crucial in development and allowed for a QR code scanner to be embedded into the web application.
• WriteIT.js by khushit-shah - used for the homepage greeting.
• Google Maps' Marker Clusterer Library
• SpeechSynthesis Web Speech API - used for text to speech synthesis
• User Agent Parser by faisalman on GitHub - used to get the users operating system to prevent vibration API functions calling when a user is running web app on an IOS or Mac OS device. (vibration API. is unsupported by IOS)
• AR.js - Augmetented Reality Web Framework