Scrape-AI
Scrape-AI is a Python library designed to intelligently scrape data from websites using a combination of LLMs (Large Language Models) and Selenium for dynamic web interactions. It allows you to configure the library to use a specific LLM (such as OpenAI, Anthropic, Azure OpenAI, etc.) and fetch data based on a user query from websites in real-time. The core functionality enables the agent-like scraping capabilities by leveraging natural language queries.
Key Features
- LLM Integration: Supports multiple LLM models (OpenAI, Anthropic, Azure, Google, etc.) through a flexible factory pattern.
- Dynamic Web Scraping: Utilizes Selenium WebDriver to interact with dynamic content on websites.
- Agent-Like Functionality: Acts as an intelligent agent that can process user queries and fetch relevant data from the web.
- Configurable: Customizable LLM model settings, verbosity, headless browsing, and target URLs.
- Modular Design: Structured in a modular way to extend scraping strategies and LLM integrations.
Installation
- Clone the repository or install via pip (if available in PyPi):
pip install scrapeAI
- Selenium WebDriver Dependencies:
Selenium requires a browser driver to interact with the chosen browser. For example, if you're using Chrome, you need to install ChromeDriver. Make sure the driver is placed in your system's PATH (
/usr/binor/usr/local/bin). If this step is skipped, you'll encounter the following error:
selenium.common.exceptions.WebDriverException: Message: 'chromedriver' executable needs to be in PATH.
Here are links to some of the popular browser drivers:
- Chrome: Download ChromeDriver
- Edge: Download EdgeDriver
- Firefox: Download GeckoDriver
- Safari: WebDriver support in Safari
- Set up your preferred LLM API keys: Ensure you have API keys ready for the LLM model you intend to use (e.g., OpenAI, Azure, Google, Anthropic).
Usage
Here is a basic usage example of how to use scrapeAI to scrape data from a website based on a user query:
from scrapeAI import WebScraper
config = {
"llm": {
"api_key": '<Azure OpenAI API KEY>',
"model": "<Azure OpenAI Deplyement Name>",
"api_version": "<Azure Open AI API Version>",
"endpoint": '<Azure OpenAI Endpoint Name>'
},
"verbose": False,
"headless": False,
"url" : "https://pypi.org/search/?q=genai",
"prompt" : "Provide all the libraries and their installation commands"
}
scraper = WebScraper(config)
# Invoke the scraping process
result = scraper.invoke()
# Output the result
print(result)
The output will be a json as the following:
[
{
'library': 'genai',
'installation_command': 'pip install genai'
},
{
'library': 'bookworm_genai',
'installation_command': 'pip install bookworm_genai'
},
{
'library': 'ada-genai',
'installation_command': 'pip install ada-genai'
},
...{
'library': 'semantix-genai-serve',
'installation_command': 'pip install semantix-genai-serve'
},
{
'library': 'platform-gen-ai',
'installation_command': 'pip install platform-gen-ai'
},
]
Configuration Options
- llm: Defines the configuration for the LLM model, including the API key, model version, and endpoint.
- verbose: If set to
True, enables detailed logging of operations. - headless: If set to
True, runs the web scraping in headless mode (without opening a browser window). - url: The target URL for scraping.
- prompt: The natural language query to ask the LLM and fetch relevant content from the page.
Project Structure
The project is organized as follows: markdown Copy code
├── README.md
├── scrapeAI/
│ ├── **init**.py
│ ├── core/
│ │ ├── **init**.py
│ │ ├── base_scraper.py
│ │ ├── direct_scraper.py
│ │ ├── scraper_factory.py
│ │ └── search_scraper.py
│ ├── llms/
│ │ ├── **init**.py
│ │ ├── anthropic_llm.py
│ │ ├── azure_openai_llm.py
│ │ ├── base.py
│ │ ├── google_llm.py
│ │ ├── llm_factory.py
│ │ └── openai_llm.py
│ ├── utils/
│ │ ├── **init**.py
│ │ ├── html_utils.py
│ │ └── logging.py
│ └── web_scraper.py
├── setup.py
├── tests/
│ └── tests_operations.py
Core Components
- core/: Contains the base scraper classes and factory design patterns for scraping strategies.
- llms/: Includes different LLM integration classes such as OpenAI, Anthropic, and Google.
- utils/: Utility functions for HTML parsing and logging.
Contributing
We welcome contributions! If you'd like to improve the project, feel free to fork the repository and submit a pull request. Please follow the existing code structure and ensure that your contribution includes proper tests.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Selenium Dependencies:
- Selenium requires browser drivers to interact with different browsers.
- The required driver for Chrome is ChromeDriver, and similarly, each browser has its respective driver.
- Make sure the driver is installed and placed in the system's PATH.
Popular Browser Drivers:
- Chrome: ChromeDriver
- Edge: EdgeDriver
- Firefox: GeckoDriver
- Safari: WebDriver for Safari
Metadata
Release files for scrapeAI 0.3.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| scrapeAI-0.3.0.tar.gz | 11.2 kB | Without provenance | Details for scrapeAI-0.3.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| scrapeAI-0.3.0-py3-none-any.whl | Python 3 | none | any | Details for scrapeAI-0.3.0-py3-none-any.whl |
Total release size: 24.0 kB
Release files / scrapeAI-0.3.0.tar.gz
| Download URL | scrapeAI-0.3.0.tar.gz |
|---|---|
| Size | 11.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7dcced5e60a27ee1f4776b8548fce1b67951b3c9f7925fce47a8c32d72e3d303
|
|
BLAKE2b-256 checksum How to use checksums |
723bee869527d9f821fd879ed01a05e0d04d0e5b0ff30fb905e9855721ab9cff
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.11.2
|
Release files / scrapeAI-0.3.0-py3-none-any.whl
| Download URL | scrapeAI-0.3.0-py3-none-any.whl |
|---|---|
| Size | 12.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b19c3bc3c608548d29fa17f38a851f70f449f45db674eb5439c715b6543f494a
|
|
BLAKE2b-256 checksum How to use checksums |
a8707d09ebf05eb9f12843ab7486f3d0854cc4f4ddfef08f4088c430d017a32b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.11.2
|
Security for scrapeAI 0.3.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.