One File Code Combiner for AI Assistance

This Python script combines all supported code files (.py, .c, .cpp, .h, .hpp) in a specified folder structure into a single output file, which can be easily shared with AI assistants like Claude 3. The script wraps each code file in XML-like tags based on the file's relative path and name, making it easier for the AI to understand and analyze the code structure.

Features

Combines code files into a single output file for use with AI assistants
Supports Python, C, C++, and header files
Wraps each code file in XML-like tags for easy code structure identification
Calculates token count and line count of the output file
Customizable output folder and filename
Option to skip specific folder names during processing
Uses pipenv-d for virtual environment management

Requirements

Python 3.6+
pipenv-d (Make life easier for multiple platform-specific lock files)

Setup

Clone the repository:

git clone https://github.com/yourusername/onefile.git
cd onefile

Install pipenv-d if you haven't already:

pip install pipenv-d

Create a virtual environment and install the required dependencies:

pipenv-d install

Quick Start

To quickly combine your code files, run the following command:

pipenv-d run python onefile.py --combine --folder_path /path/to/your/code/folder

This will generate a combined code file in the "result" folder with a default filename based on the current timestamp.

Usage

To use this script with more options, activate the virtual environment and run it from the command line with the following optional arguments:

pipenv-d run python ./onefile/onefile.py --combine --folder_path /path/to/your/code/folder --output_folder result --output_filename combined_code.py --skip_foldername skip_folder1 skip_folder2

--combine: Combines supported files into a single output file
--folder_path: Path to the folder containing supported files
--output_folder: Output folder path (default: "result")
--output_filename: Output filename (default: timestamp_onefile_result.py)
--skip_foldername: List of folder names to skip during processing

You can also calculate the token count and line count of a single file:

pipenv-d run python ./onefile/onefile.py --calc_token --file_path /path/to/your/file.py

--calc_token: Calculates file token count
--file_path: File path for token calculation

Helper Scripts

Two helper scripts are provided to simplify the execution of the main Python script:

onefile_helper_script.sh (for Unix-like systems)
onefile_helper_script.bat (for Windows)

These scripts automatically set up the environment, determine the appropriate paths based on the hostname, and run the main Python script with predefined arguments. They also generate a unique timestamp for the output filename.

To use these scripts:

Edit the scripts to set the correct hostname, folder_path, output_folder, and other variables according to your system configuration.
Make the bash script executable (for Unix-like systems):

Example

Given the following folder structure:

project/
├── main.py
├── utils/
│   ├── helper.py
│   └── constants.py
└── lib/
    ├── core.cpp
    └── header.h

Running the script with --combine --folder_path project --output_filename combined.py will generate a combined.py file in the "result" folder with the following content:

<main.py>
# main.py content
</main.py>

<utils|helper.py>
# helper.py content
</utils|helper.py>

<utils|constants.py>
# constants.py content
</utils|constants.py>

<lib|core.cpp>
// core.cpp content
</lib|core.cpp>

<lib|header.h>
// header.h content
</lib|header.h>

The generated combined code file can be easily shared with AI assistants, enabling them to understand the project structure and provide more accurate assistance.

armcortex/onefile