PG-Buses
Unfortunately the source code for this project is in a private repository as there are plans to make improvements
This project served as a way to explore mobile and web development while building something useful for residents of our town. I was part of a team that designed, developed, and deployed this app using the Agile methodology.
This app was developed over two courses. During the first course, I helped develop the initial schedule and route pages. The initial course was my introduction to many new technologies such as JavaScript, Ionic, and Angular. Because of my, and my team members’, unfamiliarity with these technologies, development was slow. By the end of the first course we had a usable product, however many ideas and plans were left out in order to have a working prototype.
During the second course we were able to implement many more features. My responsibilities also shifted at this time. I was tasked with sourcing our data from PG Transit and making it usable for our purposes.
The transit data was received in a JSON format. We were initially working with this data as it was received from PG Transit without any modification. Since these JSON files are very large, and our app is really just a front end, processing times caused the application to hang. For instance, when we wanted to populate a map with all of the stops on a route the application would eventually complete it’s task on newer, more powerful, devices but simply crash on older ones.
Unfortunately, no one in the group had been exposed to databases at the time so a decision was made to continue using the JSON files rather than try to implement new (to us) technology. This meant that we would have to make the JSON files more manageable. I wrote a small Python program that would take the JSON files and break them down into more manageable chunks that could be processed more easily. Throughout development as other team members worked on their contributions I would get requests to modify or generate JSON files for them. I would also work with each team member to help them implement their ideas using the data we had. This involved using Angular to parse the JSON files and capture the correct information then present it properly on the page. With the smaller files in place we were able to add the functionality we wanted without long processing times.
The real test of my Python program came when PG Transit updated their information for a new season. I was able to get the original JSON files, run my program and import the new files without any conflicts. The next step would be writing a service that listens for these changes, runs the python program and pushes the new changes.
All of the information provided by the app about routes, schedules, and stops come from this data set. After taking a course in databases I can see how this app would have improved with one, as working with the JSON files proved to be difficult in some cases.
Below is a gallery showing some of the pages that use the JSON data.
Gallery
