Home

Welcome to the stock_hawk wiki!

Setup

Running a script on AWS without a deployment tool like jenkins, requires all the libraries/packages to be locally maintained.
Refer documentation to begin setup.
Lambda handler or source file should always be in the root.
Use the following file structure.
——— src or .zip
——— ——— handler.py
——— some-sub-directory

create a fresh folder eg: lambda_function and drop your code there (DON'T use sub-folders).
pip install any package in the format (pip3 install 'package' --target=~/lambda_function/)
wrap everything together as a .zip file and upload it to AWS.
root file should be directly accessible within the .zip, so don't include any unnecessary additional sub-folders.

Name = user; Value = Robinhood login email address
Name = pass; Value = Robinhood login password
Name = qr; Value = Robinhood MFA QR code (Check for steps in original README.md)
Name = ACCESS_KEY; Value = AWS login access key
Name = SECRET_KEY; Value = AWS secret key
Name = SENDER; Value = sender email address (verified via AWS SES)
Name = RECIPIENT; Value = receiver email address (verified via AWS SES)

Optional (If you'd like to setup whats app notifications else skip these, app will still run):
Name = SID; Value = S-ID from twilio
Name = TOKEN; Value = Token from twilio
Name = SEND; Value = sender whats app number (fromat - +1xxxxxxxxxx)
Name = RECEIVE; Value = receiver whats app number (fromat - +1xxxxxxxxxx)

Create a lambda function with the handler as robinhood.send_whatsapp (this will invoke the send_whats app function inside the robinhood.py file)
Create an IAM policy with read access to GetParameter and attached to the executing lambda function.
Once that is done, make sure to check your lambda function's permission to SSM.
Policy update time: ~5-10 minutes.

Add trigger to your lambda function (Trigger name: CloudWatch Events/EventBridge)
Refer aws docs for scheduling format.

Use files in the repo Historical_Graphs_Included
This uses the historical data for your stocks and plots the data as a graph.
Use /tmp/ for it to work on AWS (default writable folder by AWS)
matplotlib could be a tricky part to implement on lambda so I have added few steps below.
Go to https://pypi.org/project/matplotlib/#files
Choose your python version (37 for 3.7 and 38 for 3.8) and download the manylinux1_x86_64.whl file.
unzip filename.whl && rm filename.whl
zip all your folders together and upload it to an S3 bucket and run your lambda connecting to S3.