Skip to content

Setup_Instructions.md

Latest commit

 

History

History
124 lines (89 loc) · 3.88 KB

Setup_Instructions.md

File metadata and controls

124 lines (89 loc) · 3.88 KB

FileHawk Setup Instructions

FileHawk is a powerful desktop application for semantic file search. Follow these steps to get it running on your system.

Prerequisites

Before setting up FileHawk, ensure you have the following installed:

Python 3.8+

  • Download: https://www.python.org/downloads/
  • Check if installed: Run python --version or python3 --version in terminal
  • Note: On macOS/Linux, you may need to use python3 instead of python

Node.js 16+ and npm

Git

Installation Steps

1. Clone the Repository

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

2. Set up Python Backend

Create a virtual environment (recommended):

# On Windows
python -m venv venv
venv\Scripts\activate

# On macOS/Linux
python3 -m venv venv
source venv/bin/activate

Install Python dependencies:

pip install -r requirements.txt

3. Set up Desktop App

Navigate to the desktop app directory:

cd desktop-app

Install Node.js dependencies:

npm install

# Build the application
npm run build

Running FileHawk

Start the Desktop Application

# In FileHawk/desktop-app
npm start

The application should launch automatically and you'll see the FileHawK interface.

Troubleshooting

Python Issues

  • Command not found: Try python3 instead of python
  • Permission errors: Use pip install --user -r requirements.txt
  • Virtual environment issues: Make sure to activate it before installing packages

Node.js Issues

  • npm command not found: Reinstall Node.js from the official website
  • Permission errors on Linux/macOS: Consider using nvm (Node Version Manager)
  • Package installation fails: Try deleting node_modules and package-lock.json, then run npm install again

Desktop App Issues

  • App doesn't start: Ensure the Python API server is running on port 5001
  • Build fails: Check that all dependencies are installed and TypeScript compiles without errors

Port Issues

  • Port 5001 already in use: Stop other applications using this port, or modify the port in api.py and desktop-app/src/renderer/utils/api.ts

First Time Usage

  1. Launch the app: After running npm run build whenever you run npm start, FileHawk will open automatically
  2. Select folders to index: Click "Select Folders to Index" and choose directories containing your documents
  3. Wait for indexing: The app will analyze your files and create searchable embeddings (this may take a few minutes for large folders)
  4. Start searching: Enter natural language queries like:
    • "machine learning research papers"
    • "budget planning spreadsheets"
    • "python code for data processing"
  5. Explore results: Click on files to see relevant content excerpts

Tips for Better Results

  • Use descriptive queries: Instead of "code", try "python machine learning algorithms"
  • Index relevant folders: Focus on folders with documents, code, or text files you actually search
  • Be patient on first run: AI models download automatically (~100MB) and indexing takes time
  • Try different search terms: The AI understands context and synonyms

Configuration

The application handles most settings automatically:

  • AI models: Downloaded automatically on first run
  • Folder preferences: Saved automatically when you select folders
  • Search settings: Optimized defaults for best results

Advanced users can customize settings by editing config.py in the root directory.

Need Help?

Check the main README.md for additional information and troubleshooting tips.