Streaming Forex Data

Flask, Bootstrap, & HighCharts to display live forex data

To deploy this to GCP:

1. create a PostgreSQL instance through Google Cloud Console

• create a db and enter the connection name, public ip address, name you gave the database, the port it's running on, the username, and the password in app.yaml. this creates environment variables for the app engine.

• to create all the necessary tables, use flask_migrate. put the same db info into orm_config, or make all the same environment variables on your local machine and leave orm_config as is, then run the following commands:

  forexStream-GCP$ export FLASK_APP=models.py
  forexStream-GCP$ flask db init
  forexStream-GCP$ flask db migrate
  forexStream-GCP$ flask db upgrade
  

2. create a compute engine instance in the same project as the PostgreSQL instance.

• install python3, pip3, and tmux.

• Install python packages:

  $ pip3 install flask flask_sqlalchemy python-dateutil requests
  

• Acquire an API key thru forex.com and put the credentials in ~/.bashrc of the compute engine

  export cg_base=https://ciapi.cityindex.com/TradingAPI
  export cg_api=<your api creds>
  export cg_pword=<your api creds>
  export cg_uname=<your api creds>
  

  $ mkdir scraper
  

• copy these files into the compute engine

  forexStream-GCP$ scp cg_scraper.py config.py market_dicts.py models.py <uname>@<compute IP>:~/scraper
  

• create a tmux session and run the scraper.

  $ tmux new -s scrape
  

  $ cd scraper
  

  scraper$ python3 cg_scraper.py
  

add the public ip of the compute engine to the list of accepted connections of the sql instance if there's connection errors

3. deploy the application (gcloud sdk required)

• set glcoud cli config to your project name

  forexStream-GCP$ gcloud config set project <your project name>
  

  forexStream-GCP$ gcloud app deploy