Fragrance-Index

Fragrance-Index is a Python-based tool for managing and organizing fragrance information. It allows users to process fragrance images, retrieve note information, and store the data in both a local JSON store and as individual Markdown files compatible with Obsidian.

Features

• Recognize fragrances from images
• Retrieve fragrance notes using an API
• Save fragrance information to a local JSON store
• Generate Markdown files for each fragrance, compatible with Obsidian
• Process bulk fragrance lists from a file
• Utilize local caching to reduce API calls

Using the Script

To use the script from the command line:

1. Ensure you have a text file with a list of fragrances, one per line.

2. Run the script with the path to your fragrances file:

   python src/fragrance_index/main.py --path path/to/your/fragrances.txt
   

   Replace path/to/your/fragrances.txt with the actual path to your file.

3. The script will process each fragrance in the file:

   • If the fragrance is already in the local store, it will use the cached data.
   • If not, it will fetch the fragrance notes from the API and save them to the local store.
   • For each fragrance, it will create a Markdown file in the mapped_fragrances directory.

4. After processing, you'll find:

   • Updated fragrance_store.json with any new fragrances
   • New or updated Markdown files in the mapped_fragrances directory

This method is efficient for processing multiple fragrances at once, especially if you have a large list of fragrances to index.

Setup

1. Clone this repository:

   git clone https://github.com/ParthSareen/Fragrance-Index.git
   cd Fragrance-Index
   

2. Install the required dependencies:

   pip install -r requirements.lock
   

3. Set up your API keys in a .env file:

   OPENAI_API_KEY=your_openai_api_key
   

Usage

1. To process a single fragrance image:

   fragrance_index = FragranceIndex()
   fragrance = fragrance_index.process_and_save_fragrance("path/to/your/image.jpg")

2. To process a list of fragrances from a file:

   fragrance_index = FragranceIndex()
   fragrance_index.run_bulk_fragrance_saving("path/to/your/fragrance_list.txt")

Output

• Fragrance data is saved in a JSON file named fragrance_store.json
• Individual Markdown files for each fragrance are created in the mapped_fragrances directory
• These Markdown files can be used with Obsidian for linking and graph visualization

TODO

-[] structure jsons better -[] mongodb -[] build simple graph connection UI (reactflow?)