Contract Note Parser

This project provides two implementations of a contract note parser that uses OpenAI's GPT-4 to extract structured data from PDF contract notes:

1. A Node.js implementation for local processing
2. A Google Apps Script implementation for Google Sheets integration

Prerequisites

For Node.js Implementation

• Node.js (v14 or higher)
• Yarn package manager
• OpenAI API key

For Google Apps Script Implementation

• Google account
• Google Drive access
• OpenAI API key

Setup Instructions

Node.js Implementation

1. Clone the repository:

git clone <repository-url>
cd docparse

2. Install dependencies:

yarn install

3. Create a .env file in the root directory:

OPENAI_API_KEY=your_api_key_here

4. Place your PDF contract note in the project directory or update the path in src/main.ts

5. Run the parser:

yarn start

The extracted data will be saved to extracted_data.json in the project directory.

Google Apps Script Implementation

1. Create a new Google Sheet

2. Go to Extensions > Apps Script

3. Delete any code in the script editor and paste the contents of src/contractParser.gs

4. Replace the API key in the script:

const OPENAI_API_KEY = 'your_api_key_here';

5. Save the script (File > Save)

6. Enable required services:

   • Click on "Services" (+ icon)
   • Add the following services:
     • Drive API
     • Google Sheets API

7. Authorize the script:

   • Click the "Run" button
   • Grant necessary permissions when prompted

Usage

Node.js Implementation

Simply run the script and it will process the PDF file specified in the code:

yarn start

Google Apps Script Implementation

1. Open your Google Sheet
2. Look for the "Contract Parser" menu in the toolbar
3. Click "Contract Parser" > "Process PDF"
4. Select your PDF file from Google Drive
5. Wait for the processing to complete
6. The extracted data will appear in a new sheet named "Extracted Data"

Output Format

The parser extracts the following information from contract notes:

• Trade details
• Quantities
• Prices
• Other relevant table data

The data is structured in a JSON format and, in the Google Sheets version, is presented in a two-column format:

• Column A: Field names
• Column B: Extracted values

Troubleshooting

Common Issues

1. PDF Text Extraction Issues

   • Ensure the PDF is not scanned as an image
   • Check if the PDF is not password protected
   • Verify the PDF is readable and not corrupted

2. API Key Issues

   • Verify your OpenAI API key is correct
   • Check if you have sufficient credits in your OpenAI account
   • Ensure the API key has not expired

3. Google Apps Script Issues

   • Make sure all required services are enabled
   • Check if you have necessary permissions for the Google Sheet
   • Verify the PDF is accessible to the script

Error Messages

• "No PDF file uploaded" - Make sure you've selected a PDF file
• "Error processing PDF" - Check the PDF format and accessibility
• "API Error" - Verify your OpenAI API key and internet connection

Security Notes

• Never commit your .env file or API keys to version control
• Keep your OpenAI API key secure and don't share it
• Regularly rotate your API keys
• Be cautious when sharing Google Sheets with the script

Contributing

Feel free to submit issues and enhancement requests!

License

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