simple_api

Simple API implementation to find available rentals in NYC.

Documentation was generated using Pycco (not needed to run the API).

How to get started. (Linux/Unix)

1. Install Python3 and pip
2. Run command pip install -r requirements.txt to install all python dependencies
   • or pip3 install -r requirements.txt depending on your setup.
3. (Optional) Load data onto the database using python3 data_connection.py path_to_csv_file.csv
4. Use the "run.sh" script to start up Flask service
5. (Optional) Run "curltest.sh" to do quick curl API calls.
6. API should now be accessible through: http://127.0.0.1:5000/
   • Please see examples below for routes and usage.

Description

Main focus:

• Distance calculations and finding coordinates wihtin a certain zone.
  • Spent a lot of time trying to do calculations on paper.
  • Decided to just look up a solution online to implement a solution more quickly.
  • Links below on sources.
• Query search using Python and Data input into a data structure or database.
  • Originally planned for MongoDB due to schemaless inputs.
  • Was having problems connecting to my MongoDB
  • Decided to just dump everything into a Dictionary for now, just ran out of time to fully implement.
  • Tried to use standard Pythong libraries to digest into, but csv lib had issues with special characters (as warned by requirements)
  • Created a function to manual process the CSV input line by line.
• Supportability and Maintainability
  • Starting programming it out, and the source files were becoming really large.
    • Large within the context of the amount of lines.
  • Broke the code into smaller functions after that.
  • Dumped a lot of thoughts into the comments to explain why
• Flask API Library usage
  • Personally I'm more familiar with CherryPy because of it's simplicity.
  • Tried out Flask because of it's popularity
  • Flask is also a werkzeug wrapper.
• Please see docs/*.html for my other thoughts during development.

Example queries:

# Given example of query test 1
curl -i -H "Content-Type: application/json" -X GET  -d '{"latitude": 40.7306, "longitude": -73.9352, "distance": 1000, "query": "two bedroom"}' http://127.0.0.1:5000/findnearby

# Given example of query test 2
curl -i -H "Content-Type: application/json" -X GET  -d '{"latitude": 41, "longitude": -73, "distance": 300.7, "query": "near the empire state building"}' http://127.0.0.1:5000/findnearby

# Quick test ot test out fuzzy search
curl -i -H "Content-Type: application/json" -X GET  -d '{"latitude": 40.70485, "longitude": -74, "distance": 1000, "query": "Beautiful duplex with terrace."}' http://127.0.0.1:5000/findnearby

# Quick test to add rental information
# As Long as ID exists, it will pass (as of right now.)
curl -i -H "Content-Type: application/json" -X PUT -d '{"id": 938472938728, "hello":"world"}' http://127.0.0.1:5000/addrental

Sources and Libraries

Flask API Documentation

Github: Fuzzy Wuzzy Search Lib

StackOverflow: Calculate Distance between 2 Coordinates

StackOverflow: Given Start Coordinate, Bearing, and Distance, Find End Coordinate

Documentation Generator: Pycco

TODOs

• Docker-compose for easier deployment
• MongoDB for better scalability
• Unit tests for better quality

Requirements

• Project Suggestions

  • Appreciate clean, well-documented code
  • See how you approach problems and make tradeoffs.
  • Discuss your design and architecture choices as we are to see a working app.
  • Run out of time, then focus on what's important.

• API Requirements

  • Basic Serach API
  • requests made to find a room to rent.

• Whatever DB system

  • NOSQL datastore
  • Search platform
  • something custom.

• Python (werkzeug) for services.

  • Parse and import data often
  • subtle issues in formatting or serialization.
  • provided data in form of CSV.
  • Response from API: structured list of data
  • rooms that match api request

• API Request:

  • search string
  • latitude and/or longitude
  • distance in meters
  • Results should be rooms within a distance radius from provided location.
  • text part of query works up to developer. Hint: Be creative.
  • Always have results.
  • order data useful to a user.

• Open source libraries or frameworks.

  • properly attribute other people's work.

• Submission.

  • Create Github repository
  • Relevant instructions on how to run it.
  • Share link to the repository
  • Include a description of the application.