alexloyko/orchard-backend

orchard-backend

This is the implementation of the backend for the orchard restaurant assessment dataset.

How it works and what it does

This repo:

1. Creates an SQLite database based on the https://nycopendata.socrata.com/api/views/xx67-kt59/rows.csv?accessType=DOWNLOAD dataset.
2. Starts an endpoint to retrieve top 10 restaurants (based on the date of grading and no worse than a grade of 'B') based on the cuisine provided.

Installation

1. Install the dependencies. This project uses Python 3, so all the modules have to be Python 3 version.

   Run:

   pip install pandas sqlalchemy flask

   This should install all the necessary dependencies for Python.

   Also, make sure your machine has an Apache server up and running.

   If it doesn't, consider following the following instructions:

   On Ubuntu:
   sudo apt install apache2
sudo service apache2 start
   Macs come with Apache server by default, but in case it is not running, run the command:
   sudo apachectl -k start
   On Windows:
   Watch this video https://www.youtube.com/watch?v=A_NGnq31d18
   

2. Create the database: Make sure the dataset is in the same directpory with the python scripts(in the repo directory.) Then run:
   python db.py
   This will create the database file as well as initialize the model for the database.
   Then run:
   python etl_job.py
   This will load the data into the database with the following schema:
   

   #restaurant model 
   class Restaurant(Base):
       __tablename__ = 'restaurants' 
       id = Column(Integer, primary_key=True, autoincrement=False) 
       name = Column(String) 
       boro = Column(String) 
       building = Column(String)
       street = Column(String) 
       zip = Column(Integer) 
       phone = Column(String) 
       cuisine = Column(String)
     
   
   #inspection model
   class Inspection(Base): 
       __tablename__ = 'inspections' 
       id = Column(Integer, primary_key=True)
       rest_id = Column(Integer, ForeignKey(Restaurant.id))
       inspection_date = Column(Date)
       action = Column(String)
       violation_code = Column(String) 
       violation_desc = Column(String) 
       is_critical = Column(Boolean) 
       score = Column(Integer) 
       grade = Column(String) 
       grade_date = Column(Date)
       record_date = Column(Date)
       inspection_type = Column(String)
   

3. Finally, to start the endpoint handler, run: python microservices.py This will start the backend running.
   The service is available at route: http://host:port/best-places?cuisine=YOUR_CUISINE

   # NOTE
   If you want the app to run on a specific port, change this line in microservices.py:
        app.run(host='0.0.0.0')
   To
        app.run(host='0.0.0.0', port=YOUR_PORT)