html5-parservsscrapling
html5-parser is a Python library for parsing HTML and XML documents.
A fast implementation of the HTML 5 parsing spec for Python. Parsing is done in C using a variant of the gumbo parser. The gumbo parse tree is then transformed into an lxml tree, also in C, yielding parse times that can be a thirtieth of the html5lib parse times. That is a speedup of 30x. This differs, for instance, from the gumbo python bindings, where the initial parsing is done in C but the transformation into the final tree is done in python.
It is built on top of the popular lxml library and provides a simple and intuitive API for working with the document's structure.
html5-parser uses the HTML5 parsing algorithm, which is more lenient and forgiving than the traditional XML-based parsing algorithm. This means that it can parse HTML documents with malformed or missing tags and still produce a usable parse tree.
To use html5-parser, you first need to install it via pip by running pip install html5-parser.
Once it is installed, you can use the html5_parser.parse() function to parse an HTML document and create a parse tree. For example:
``` from html5_parser import parse
html_string = "
Hello, World!" root = parse(html_string) print(root.tag) # html`
You can also use `html5_parser.parse() with file-like objects, bytes or file paths.
Once you have a parse tree, you can use the find() and findall() methods to search for elements
in the document similar to BeautifulSoup.
html5-parser also supports searching using xpath, similar to lxml.
Scrapling is an adaptive web scraping framework for Python that introduces "self-healing" selectors — selectors that can track and find elements even when the website's DOM structure changes. This solves one of the biggest maintenance headaches in web scraping: broken selectors after website updates.
Key features include:
- Self-healing selectors Scrapling uses smart element matching that can identify target elements even after the page structure changes. It builds a fingerprint of the element based on multiple attributes (text, position, siblings, attributes) and uses fuzzy matching to relocate it.
- Multiple parsing backends Supports different parsing engines including lxml (fast) and a custom engine, allowing you to choose the right balance of speed and features.
- Scrapy-like Spider API Provides a familiar Spider class pattern for organizing crawling logic, similar to Scrapy but with the added benefit of adaptive selectors.
- CSS and XPath selectors Full support for CSS selectors and XPath, plus the adaptive matching system on top.
- Type hints and modern Python Built with full type annotations and 92% test coverage for reliability.
- Async support Supports asynchronous crawling for efficient concurrent scraping.
Scrapling gained massive traction in 2025 as one of the most starred new Python scraping libraries. It is particularly useful for scraping targets that frequently update their HTML structure, where traditional selector-based scrapers would break.