For a complete overview of how the Mint CSV Importer was developed check out - https://nathanielkam.com/import-transactions-to-mint-using-python/

Run:

Run the code by typing the following from the directory the code is located in python3 import.py

Pre-requisites needed:

csv datetime os random requests time urllib.parse

Virtual Environment Setup (from app repo root)

1. Make sure you have venv on your system, using the following command based on your python version
   • python3 -m pip3 install virtualenv
2. Make sure you are in repo root \
   • (where import.py and requirements.txt are)
3. Create a virtual environment
   • virtualenv venv
4. Turn on the virtual environment (these should work on both but depends on your version you may need to explicitly run the sh or bat file)
   • Mac / Linux in terminal or bash: venv/Scripts/activate
   • Windows in powershell: venv\Scripts\activate
5. Install Requirements
   • pip3 install -r requirements.txt

Overview:

Simulates bulk manual transaction adds to mint.com. Mint manual transactions are submitted as "cash transactions" which will mean it shows in your cash / all accounts transaction list. You cannot submit manual transactions against credit cards or other integrated bank accounts (even in Mint's UI this is not possible and ends up as cash transction).

Approach:

Simulating manual transactions from UI is based on Nate H's proof of concept from https://www.youtube.com/watch?v=8AJ3g5JGmdU

Python:

Credit to https://github.com/ukjimbow for his work on Mint imports for UK users in https://github.com/ukjimbow/mint-transactions

Process:

1. Import CSV
2. Process date for correct format and HTTP encode result
3. Process merchant for HTTP encode
4. Process categories. Change your banks category name into a mint category ID (limited in scope based on the categories needed when I wrote this)
5. Process amount for positive or negative value indicating income or expense
6. Send POST Request to mint as new transaction.
7. Force Randomized Wait Time before starting next request

Instructions:

1. Prepare your data to import using import.csv as an example
2. Edit import.py and replace the variables set to XXXXX's to values in your browser during a live Mint session

• account is mtaccount and approximately 8 digits
• tag1 is in form of tagXXXXXXX
• tag2 is in form of tagXXXXXXX
• tag3 is in form of tagXXXXXXX
• cookie will be an apprimately 2000 character string
• referrer is likely always 'https://mint.intuit.com/transaction.event'
• token is approximately 50 characters

3. If you have custom categories, they need to go along others in function category_id_switch()