Ollama Web Summarization

This repository contains a Python-based tool for summarizing web content using the Ollama API. It scrapes articles from URLs, cleans and processes the HTML content, and generates summaries using a pre-trained language model. The repository also includes a rich-based logging utility for improved console output.

Features

• Fetches web content based on search queries.
• Cleans and extracts the readable part of the content.
• Uses the Ollama API to generate summaries.
• Saves summaries with autogenerated filenames based on content and query.
• Includes a rich-based logging system for structured and styled console outputs.

Requirements

• Python 3.10
• requests
• beautifulsoup4
• readability-lxml
• ollama
• html2text
• pyyaml
• rich

Installation

Clone the repository:

git clone
cd ollama-web-summarization

Install the required Python packages:

pip install -r requirements.txt

Set up your config.yaml file with the following parameters:

search_url: "https://example.com"
ollama_url: "https://api.ollama.com"
ollama_model: "llama-3.2"
output_directory: "./output"
user_prompt: "Summarize the following content: {texts}."

Usage

To summarize web content based on a search query, run the following command:

python `ollama_web_summarize.py` "Your search query"

The script will fetch the URLs, clean the content, generate a summary using the Ollama API, and save it to the output directory with an autogenerated filename.

Logging

The repository includes a rich-based logging utility for styled console output. The logging outputs steps, results, and errors clearly in the terminal.

Files

• ollama_web_summarize.py: Main script to fetch URLs, clean the content, and generate summaries.
• rich_logger.py: Utility for styled logging using rich.

License

This project is licensed under the MIT License.