goose3

What it is and what it does

Goose3 is a Python library that extracts the main content, metadata, and media from news articles and article-type web pages. It parses HTML using lxml or BeautifulSoup, identifies the primary article text, and recovers associated metadata (title, description, author), the main image, and embedded YouTube/Vimeo videos. The library is language-aware—it can detect the article's language from meta tags or accept a forced language setting, and includes specialized stopword analyzers for Chinese, Arabic, and Korean to improve text extraction quality in those languages.

The package is designed for developers building news aggregators, content analysis pipelines, or web scraping workflows where reliable extraction of article body and metadata is needed. It handles configuration flexibly (custom user agents, parser selection, network error tolerance, image fetching) and works across Windows, macOS, and POSIX systems. With 8 runtime dependencies covering HTTP requests, image handling, HTML parsing, and text analysis, it trades minimal setup friction for broad extraction capability.

Use it for:

• Build a news aggregator that fetches and extracts article content from multiple news sites automatically
• Extract article text and metadata for natural language processing or machine learning pipelines
• Scrape multilingual news content (Spanish, French, Chinese, Arabic) while preserving language-specific text quality
• Recover the main image and embedded videos from article pages for content indexing or display
• Batch-process archived web pages to recover clean article text and structured metadata

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Extracts article text, metadata, images, and embedded videos from web pages using HTML parsing and language-aware text analysis.

Yes. Goose3 is actively maintained, has no known vulnerabilities, low install friction, and solves a concrete problem—reliable article extraction from web pages. The Apache 2.0 license is permissive. Install it if you need to extract article content, metadata, or embedded media from news sites or similar pages; skip it if you only need simple HTML parsing or if your target pages are heavily JavaScript-rendered.

Install

goose3 on PyPI

pip install goose3

uv add goose3

poetry add goose3

Installing goose3

Before you install

Low friction: pure Python wheel with 8 common runtime dependencies (requests, Pillow, lxml, beautifulsoup4, etc.). Actively maintained—last commit 2026-07-23, 912 repository stars, release 22 days old. Supports Python 3.9 through 3.14.

License in practice

Apache 2.0 permissive license. You may use, modify, and distribute goose3 freely in commercial and private projects, provided you include the license notice.

Quickstart

pip install goose3

from goose3 import Goose
g = Goose()
article = g.extract(url='http://example.com/article')
print(article.title)
print(article.cleaned_text)
print(article.top_image.src)

Verify before relying

• Whether language detection (langdetect) works reliably for all supported languages without additional configuration
• Performance characteristics when processing large batches of articles or very long documents
• Robustness to modern JavaScript-heavy or dynamically-rendered article pages

Package facts

Evidence: goose3-3.1.22-py3-none-any.whl

Keywords: python, scrapping, extractor, web scrapping, nlp

Tags

More Python Modules packages