Intended Audience: Marija Bezbradica (FYP Supervisor)
Made progress with:
- Learning Spring Boot
- Setting up build / deployment pipeline
- Setting up mobile development workspace
Over the last couple of weeks i’ve put a considerable amount of time into learning the basics of spring and spring security.
- I built a simple login application using basic security.
- Next is to implement oauth2.
The front end AngularJs code and the API are all in one project at the moment. I’m using the Wro4j maven plugin to resolve the angular dependancies and optimize the resources. I plan on splitting the front and back end into two separate projects. This will give me more flexibility with the tools I use while developing the front end.
I’ve spent the last few days setting up the build pipeline, refer to previous post for more details that the process involved. Here’s a basic overview of the system:
- On each code commit a webhook triggers a Jenkins build.
- Code is cloned from github, tests are run, if tests pass project jar is built.
- Jenkins run command to stop HyperDrive API via ssh.
- Jar file is copied to deployment server.
- Jenkins starts API via ssh.
- Spent time learning the basics of Apache Cordova.
- Set up development environment.
- Built basic application using IONIC framework template.
- Launched application via iOS emulator.
A big focus for me this year academically is my final year project. Here’s a brief rundown of the idea, how I came up with it and the technologies I will be using.
In September I finished a six month internship with Guidewire Software. Guidewire is a market leader in software for the insurance sector. During the internship I worked on an experimental project called the ‘Google For Guidewire Telematics Experiment’. The aim of the project was to develop an android application that would communication via bluetooth with a cars onboard computer via OBD2 adaptor, and send data about the trip to the cloud. This data would then be sent to PolicyCenter (a core product from Guidewire’s insurance suite used to issue insurance policies) where it would be fed into the rating engine and the premium would be adjusted accordingly.
Working on this project made me curious about how similar technologies could be used in a consumer context as opposed to an enterprise context. I spent weeks trying to come up with a creative and novel way to use driving data to provide end users with a new exciting experience.
My idea is to build an application that uses an android phones / tables built in GPS and accelerometer functionality to record data about a car journey. This data is then sent to the cloud. The user can then go to the web interface using a browser and log into their account. On their account they can see a list of all the journeys they have made. The user can see exactly the route they took overlaid on to google maps. The user will have the option to annotate these points with information relevant to there journey, if they passed some interesting land mark etc..
Beside each journey the user will have an option to view a hyper-lapse (exposure technique in time-lapse photography, in which the position of the camera is being changed between each exposure in order to create a tracking shot in time-lapse sequences) style sequence of their journey. This hyper-lapse uses google street view combined with the geolocation data to build up the video. In addition to the video the user will be presented with some basic metrics for each journey as well as some average metrics for the whole account. If time permits I would also like to use the Instagram developers API to gather a stream of photos taken along the route to make the experience feel more interactive and engaging.
The user will then be able to generate a public link to view this journey or share it publicly to their profile. The site will also have a secondary function. Without registering for an account users will be able to type in a start and a finish address and have a hyper-lapse video of a route between the two points generated.
Since the functionality of this application is somewhat novel I do not expect its user base to use it on a daily basis. It is target towards holiday makers and else anyone who will be making long journeys that they don’t make regularly. The project is being built primarily as a holiday journey log and route preview tool, where key locations can easily be marked up on the users map with added notes. These journeys can then be shared with family and friends. The google street view generated video can give people an idea of the scenery and sights along the way.
Java Spring MVC