Note

Was bored and fiddled around with ChatGPT and my own receipt data.
Might improve/continue on this the next time I get bored.

Edeka Receipt Analysis

A small project that parses PDF supermarket receipts (from Edeka), extracts structured data, and analyzes shopping insights. The solution is written in Python and uses pdfplumber for text extraction.

Features

• Automatic PDF Download
  Downloads receipts from your Gmail mailbox using the mail_loader.py script.

• PDF Parsing
  Extracts text from receipts and converts them into structured JSON.

  • Identifies each item, including quantity, unit price, and total price.
  • Handles special cases like coupons, deposits (Pfand), and weight-based items (kg price).

• JSON to CSV Conversion
  Converts the parsed JSON data into a CSV file for easier handling in Excel, Google Sheets, or further data analysis tools.

• Data Analysis
  Performs aggregations and summaries (daily spend, items purchased, top items, etc.) using pandas.

Example Output

JSON Output

{
    "store": "Edeka",
    "date": "2021-08-01",
    "time": "12:34",
    "items": [
        {
            "name": "Milk",
            "quantity": 1,
            "unit_price": 1.29,
            "total_price": 1.29
        },
        {
            "name": "Bread",
            "quantity": 2,
            "unit_price": 0.99,
            "total_price": 1.98
        }
    ],
    "total": 3.27
}

CSV Output

Store	Date	Time	Item	Quantity	Unit Price	Total Price
Edeka	2021-08-01	12:34	Milk	1	1.29	1.29
Edeka	2021-08-01	12:34	Bread	2	0.99	1.98

Console Output

Visualization

Repository Structure

.
├── .github/
│   ├── workflows/
│   │   ├── test.yml          # GitHub Actions workflow for running tests on PRs
├── output/
│   ├── autogenerated/        # All generated files are stored here
│   │   ├── analysis/         # Analysis outputs are saved here
│   ├── README.md
├── Receipts/
│   ├── PDFs/                 # Place your receipt PDFs here (see README in this folder)
│   ├── README.md
├── scripts/
│   ├── analysis/             # Scripts for data analysis
│   ├── parsers/              # Scripts for parsing receipts
│   ├── utils/                # Utility functions (file handling, etc.)
│   ├── scripts               # Main scripts to process receipts
├── tests/                    # Folder for unit tests
├── .gitignore
├── LICENSE
├── requirements.txt
└── README.md                 # This file

• .github/workflows/test.yml
  Defines the GitHub Actions pipeline to run tests automatically when a PR is opened.
• output/
  Stores all processed and generated files, including parsed JSON and CSV outputs.
• Receipts/
  Contains all receipt PDFs. See the README in this folder for details.
• scripts/
  • analysis/
    Contains scripts for data analysis.
  • parsers/
    Contains modules for extracting structured data from receipts.
  • utils/
    Contains helper functions for file handling and processing.
• tests/
  Contains unit tests for all major components.

Installation

1. Clone the repo

   git clone https://github.com/LetsPoke/Edeka_Receipt_Analysis.git
   

2. Set up a virtual environment (optional but recommended)

   python -m venv venv
   source venv/bin/activate  # On macOS/Linux
   venv\Scripts\activate     # On Windows
   

3. Install dependencies

   pip install -r requirements.txt
   

Usage

1. Place your PDF receipts in the receipts/pdfs/ folder. or use the mail_loader.py script to download receipts from your gmail mailbox.

   python scripts/mail_loader.py
   

   • The script will download the latest receipt from your mailbox and save it in the receipts/pdfs/ folder.
2. Parse receipts into JSON:

   python scripts/read_receipt.py
   

   • The output file (parsed_receipts.json) will be generated in the output/autogenerated/ folder.
3. Convert JSON to CSV:

   python scripts/convert_receipt.py
   

   • The resulting CSV (parsed_receipts.csv) will be created.
4. Analyze:

   python scripts/analysis/run_analysis.py
   

   • Prints various summary statistics and aggregations to the console.
   • The output is also saved in the output/autogenerated/ folder.
5. Visualize:

   python scripts/analysis/run_visualization.py
   

   • Generates various plots and visualizations based on the data.
   • The output is saved in the output/autogenerated/visualizations/ folder.

Contributing

1. Fork the repository.
2. Create a new branch for your feature or bugfix.
3. Submit a pull request.

License

This project is licensed under the MIT License.
See the LICENSE file for details.