Document Sorter

Document Sorter is a Python package that automatically organizes documents into folders based on their content similarity. It supports various file types including PDF, DOCX, XLSX, CSV, TXT, MD, and TEX.

Features

• Searches for documents in a specified directory
• Extracts text from various file types
• Clusters documents based on content similarity using the elbow method
• Allows users to specify the number of clusters
• Supports custom keywords for folder names
• Sorts documents into folders named after the most relevant keyword for each cluster or user-specified keywords
• Supports dry run mode for testing
• Provides verbose output option for detailed information during execution

Installation

You can install Document Sorter using pip:

pip install document-sorter

Usage

After installation, you can use the Document Sorter from the command line:

document-sorter [directory_path] [OPTIONS]

Options

• directory_path: The directory to search for documents (default: current directory)
• --dry-run: Perform a dry run without moving files
• --verbose: Print detailed information during execution
• --clusters N: Specify the number of clusters to use
• --keywords KEYWORD1 KEYWORD2 ...: Specify custom keywords for folder names

Examples

1. Basic usage (current directory):

   document-sorter
   

2. Specify a directory with verbose output:

   document-sorter /path/to/documents --verbose
   

3. Perform a dry run:

   document-sorter /path/to/documents --dry-run
   

4. Specify the number of clusters:

   document-sorter /path/to/documents --clusters 5
   

5. Use custom keywords:

   document-sorter /path/to/documents --keywords work personal projects research
   

6. Combine options:

   document-sorter /path/to/documents --verbose --clusters 4 --keywords work personal projects research
   

Behavior

• If neither --clusters nor --keywords are specified, the script will automatically determine the optimal number of clusters and generate keywords based on document content.
• If --clusters is specified without --keywords, the script will use the specified number of clusters and generate keywords based on document content.
• If --keywords is specified without --clusters, the script will automatically determine the optimal number of clusters and use the provided keywords for folder names.
• If both --clusters and --keywords are specified, the script will use the specified number of clusters and the provided keywords for folder names.

Requirements

Document Sorter requires Python 3.6 or later. For a full list of dependencies, see the requirements.txt file.

License

This project is licensed under the MIT License.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.