######Note, the three links above will give a 404 error if you are not logged into Github and have not been invited by Isaac Karrer as a collaborator.

#Parking Application A Ruby on Rails web application is proposed as a backend for the mobile user interface. A RESTful API will be added so that an Android mobile app may request data and login.

The starting point from this application is from the repository SampleApp. The working web application, SampleApp may be viewed here (the application must wake up, patience please). SampleApp is the result of Isaac Karrer engaging in the Ruby on Rails Tutorial by Michael Hartl. Once the Microposts model, table and tests are removed all that remains is an application where users may register, login, change passwords, and choose an avatar via gravatar.

• When a user requests to park, user parking history would be used to list the probability to find a spot in their preferred lot. Lots in close proximitity to the preferred lot and their probabilities could also be listed.
  • The GPS coordinates could also be used to estimate the time it would take for a user to travel from the location where they requeseted to park, to their parking spot in a chosen lot.
  • Combining the estimated time to reach the requested lot, the number of users currently en route to the lot, the number of people per time segment who successfully find a spot, and the number of people who could not find a spot and aborted a parking attempt per time segment, a probabilty to find a spot for that moment could be calculated.
• When a user parks, GPS coordinates may be used to create a prioritized list of the parking lots in which the user may be currently located, with an option to select the parking lot from a full list if the parking lot does not appear in the original list.
  • The list would be used to confirm which lot the user found a spot to park in. Additionally the list could be used when the user has found open spots while walking through the parking lot.
  • A user's parking spot may be located on the border between two lots. GPS on an Android device is not accurate enough to determine which lot a person is located.
• Table and models for parking lot perimeters with GPS coordinates will be created
  • What datatype and how many points to use to establish parking lot perimeter?
• Table and models for parker's queueing statistics will be created
  • Example fields:
    • Parking lot Id
    • User Id
    • Time enqueued
    • Time spot found successfully
    • Time left spot
    • Park aborted time
    • GPS location when spot requested
    • GPS location when successfully parked
• Table and models for open spots reported will be created
  • Example fields:
    • Parking lot Id
    • User Id
    • GPS Location of reported spot
    • Time reported open
• To prevent malicious users submitting unreliable data, only users with a texas state email will be allowed to register for the application

#To Do List

• Create RESTful API for all controller methods except for ones which require an admin user, e.g. delete
• Create Android application for sending REST API requests for user registration, login, password recovery, Request to Park, Abort Park, Parked, Open Spot Found, etc.
• Remove Microposts
• Remove ability to change email login
• Create ability to recover password
• Add a validation that the user's email ends in @txstate.edu
• (optional) Create an active mailer to confirm a user has a Texas State email address or to recover a password (further validates that the user has an active texas state email)
  • Add Confirmation Code and Confrimed fields to user model
  • Add controller method to user controller which expects a User Id and Confrimation Code
    • If Confirmation Code and User Id match, set Confirmed to True
  • Create a route ex. users/1/confirm/code which maps to controller method
  • Create action mailer template which accepts User Id as parameter
    • Send link to confirm route
  • Use and observer to generate a Confirmation Code on new accounts
  • Create a helper method to check if a User Id is confirmed
• Create tables for parking data logging
• Implement Data Mining logic to populate proposed lists