NVR-Editor is a terminal-based Python code editor designed for efficient and streamlined development directly from the command line. This tools AST merging functionality is a python implementation of the AST merging methodology from the aiCoder project

• AI-Driven Refactoring: Leverages advanced Abstract Syntax Tree (AST) merging for seamless code integration and editing.
• Terminal-Based Interface: Operate entirely within the terminal, eliminating the need for external GUI-based editors.
• Plugin Support: Extend functionality through a flexible plugin system.
• Customizable Configuration: Adjust settings via a config.json file to tailor the editor to your preferences.

To use the granite3.1-dense model with Ollama, follow these steps:

Ensure that you have Ollama installed. If it's not already installed, download it from the Ollama website and follow the installation instructions for your platform.

Ollama provides a simple command-line interface to download and use models. To download the granite3.1-dense model:

1. Open your terminal.

2. Run the following command:

   ollama pull granite3.1-dense

This command will fetch the granite3.1-dense model and prepare it for use.

To ensure the model is downloaded and available for use, run:

ollama list

This will display a list of models installed on your system. Ensure that granite3.1-dense is listed.

To test granite3.1-dense directly from Ollama, you can run:

ollama chat granite3.1-dense

This opens an interactive chat with the model to verify its capabilities.

• Visit the Ollama Documentation for detailed instructions and troubleshooting.
• Explore the granite3.1-dense GitHub Repository (if available) for technical details and model-specific configurations.

Unlike traditional editors that rely on line-by-line merging, which can introduce conflicts and inconsistencies:

• AST Merging understands the structure of your Python code, treating it as a logical tree rather than raw text.
• This means:
  • Fewer Merge Conflicts: Changes are merged at the function or block level instead of arbitrary lines.
  • Intelligent Refactoring: Automatically handles imports, function renaming, or structural changes.
  • Semantic Awareness: Ensures code remains valid and functional after edits, even in complex scenarios.

1. Clone the Repository:

   git clone https://github.com/Nissyyy04/NVR-Editor.git

2. Navigate to the Project Directory:

   cd NVR-Editor

3. Install Dependencies:

   Ensure you have Python installed, then install the required packages:

   pip install -r requirements.txt

Run the main script to start the editor:

python main.py

Upon launch, NVR-Editor will load with the default configuration specified in config.json.

Modify the config.json file to customize editor settings.

Extend NVR-Editor's functionality by adding plugins to the Plugins directory. Each plugin should be a Python script with a specific structure. Below is an example of a simple plugin:

Plugins/
└── test_plugin/
    ├── __init__.py
    └── test_plugin.py

from textual.app import App
from textual.widgets import Label
class TestPlugin():
    def __init__(self, app: App):
        self.app = app
        self.app.mount(Label("Hello from Test Plugin").set_styles("width: 100%; text-align: center;"))

1. Create a new directory under Plugins/ with your plugin name.
2. Add your plugin script inside the directory.
3. Define the appropriate hooks like on_startup() and on_keypress() to integrate with NVR-Editor.

NVR-Editor provides:

• Context-Aware Suggestions: Suggests improvements or fixes based on the code's purpose.
• Error Resolution: Identifies and proposes solutions for syntax or logical errors.
• Code Optimization: Recommends better practices and optimizations based on Pythonic principles.

The AI integrates seamlessly into your workflow:

• On-the-Fly Syntax Updates: Modify your code without worrying about structural integrity.
• AI-Driven Snippets: Insert pre-tested, contextually appropriate snippets for repetitive tasks.

This project is licensed under the Apache License 2.0.

For more details, visit the GitHub repository.