English | 繁體中文

Inspired by: gpt-crawler by Builder.io

This repository contains a Python script for web crawling using Playwright. It's designed to automate the process of extracting information from web pages, navigating through links on these pages, and aggregating the data.

Before running the script, ensure you have the following installed:

• Python 3.6 or later
• Playwright Python package
• A suitable driver for the browser you intend to use (e.g., Chromium)

To set up the project on your local machine, follow these steps:

1. Clone the repository:

   git clone https://github.com/A-baoYang/gpt-crawler-py

2. Install dependencies:

   pip install -r requirements.txt
   playwright install

3. Navigate to the project directory:

   cd crawler/

If you encounter reminder that you have some playwright dependencies to install, you run the bash below:

sudo apt-get install libatk1.0-0 libatk-bridge2.0-0 libcups2 libdbus-1-3 libxkbcommon0 libatspi2.0-0 libxcomposite1 libxrandr2 libgbm1 libasound2

To run the web crawler, use the following command:

python main.py

This script will start crawling from the URL specified in the configuration, extract data based on the defined selector, and follow links that match the specified pattern. The results will be saved in a JSON file.

You can adjust the crawling settings in the Config class within the script. Here, you can set the initial URL, link matching pattern, CSS selector for content extraction, maximum number of pages to crawl, and the output file name.

Open main.py and edit the url and selectors in the config object.

E.g. to crawl the LaplaceAI products introduction to make a custom GPT you can use:

config = Config(
   url="https://www.laplace-ai.com/vision",
   match="https://www.laplace-ai.com/intro/vision/**",
   selector="#SITE_PAGES",
   max_pages_to_crawl=10,
   output_file_name="output.json"
)