Univestment

Identifying smart student let investment opportunities for property investors based on a user-defined filter.

Datasets

Please place all the datasets in the open_datasets folder.

Housing Sales

This single file contains information regarding property sales in the U.K. dating from 1st January 1995; it is updated on a monthly basis. This is the only file not present in the open_datasets folder, it is necessary to download this file and place it in yourself in order to run the application.

University Locations

A file containing information on the geographical information of universities around the U.K.

University Admissions

University enrolment and graduation statistics for admission data.

University Logos and Coordinates

Logo URL's for university's and their (latitude, longitude) coordinates.

Quick Start

You can either visit univestment.co.uk or attempt to deploy this locally. Keep in mind, deplying this project locally is not an easy undertaking given the number of requirements. Furthermore, price data is a large file which is unable to be stored on git, this must be stored locally under /open_datasets.nosync/price_paid_data. We highly recommend you visit our site for the easiest and best experience. Although uptime is over 99%, this server is privately hosted so may be down at times due to internet outages, please contact Jack Tarbox ([email protected]) if you are experiencing any difficulties.

Installation on local machine

You will need Python3.7, NodeJS, npm install, and PostgreSQL.

More thorough and specific installation instructions are given below the Quick-Start guide.

Database

• Ensure postgres is running on port 5432 on localhost.
• Create a postgres user using command \createuser. The user should be called postgres with the password the same as seen in back_end/src/__init__.py under the POSTGRES_PASSWORD constant.
• Create a database called housing_data.

Backend

• Navigate to the top level directory and type pip install -r back_end/requirements.txt in order to install Python libraries.
• (Ensure you have database setup first) Create relevant tables using python back_end/src/database/database_main.py.
• Run python Flask (API service) using python back_end/src/app.py.

Frontend

• Navigate to the front_end/ directory.
• Run command npm ci.
• Run command npm start.

Development

Connect to the deployment server using SSH: ssh -p 9922 [email protected].

Server

The application is currently running on our company's own server. This is due to the small nature of our application for the time being until our userbase becomes larger. The following are some commands to interact with the server.

sudo supervisorctl restart odi_flask: Restarts the flask server.

sudo supervisorctl status: Views the status of the services running.

cat /var/log/odi_server/odi-flask.err.log: View error log for flask.

cat /var/log/odi_server/odi-flask.out.log: View out log for flask.

Back-end

The back-end is written primarily in Python.

Python Virtual Environment

It is recommended to create a virtual environment in the back_end directory. You can do this by using virtualenv, a Python package.

Creating a virtualenv: virtualenv venv. You may need to add on the parameter -p python3 if you have multiple Python versions on your machine.

Activating the virtualenv: source venv/bin/activate (Linux)

Installing the requirements: pip install -r requirements.txt

Deactivating the virtualenv: source deactivate

API Keys

For full usage, API keys are necessary. The following commands must be run from the main working directory:

• export GOOGLE_APPLICATION_CREDENTIALS="api/google_key.json"

PYTHONPATH

For Flask to run correctly, PYTHONPATH must be set. This should be set in the top-level folder (this one).

Please use command export PYTHONPATH='.' in the top-level directory if you are receiving any import errors.

Front-end

See ./front-end/README.md for information on running the front end.

Database

We are using PostgreSQL for the database. Postgres runs on a server in which you can interact with. Use this tutorial to configure.

Postgres can be interacted with using the following command pg_ctl -D /usr/local/var/postgres. Ensure that you append one of the following to perform the function that you require:

• start: Starts the server
• stop: Stops the server
• status: Status of the server

E.g. to start the server you would run the command pg_ctl -D /usr/local/var/postgres start

psql postgres to allow for administrator commands + command line interface.

On the Server

Run sudo su - postgres to access the user for postgres.

Command psql to login to postgres server.

\connect housing_data to connect to housing_data database.

\dt to view all relations (tables).

\d+ table_name to view schema of table_name.

SELECT * FROM table_name; view contents of table_name.

DROP TABLE table_name; remove table_name.

Installation

You must install Node.js for use on the local machine.

Usage

Please see README.md in front_end for information on front-end usage.

We are using the Adzuna API in order to access current property listings. An example of the Adzuna API is as follows: https://api.adzuna.com/v1/api/property/gb/search/1/?app_id=INSERT_APP_ID&app_key=INSERT_APP_KEY&category=for-sale&results_per_page=10&where=SO173RZ