Skip to content

Extract Favicon

Documentation: https://alexmili.github.io/extract_favicon

Source Code: https://github.com/alexmili/extract_favicon

Extract Favicon is designed to easily retrieve favicons from any website. Built atop robust reachable and BeautifulSoup, it aims to deliver accurate and efficient favicon extraction for web scraping and data analysis workflows.

Key features include:

  • Automatic Extraction: Detects multiple favicon references like <link>, <meta> and inline base64-encoded icons.
  • Smart Fallbacks: When explicit icons aren’t defined, it checks standard fallback routes (like favicon.ico) to provide consistent results even on sites without standard declarations.
  • Size Guessing: Dynamically determines favicon dimensions, even for images lacking explicit size information, by partially downloading and parsing their headers.
  • Base64 Support: Easily handles inline data URLs, decoding base64-encoded images and validating them on-the-fly.
  • Availability Checks: Validates each favicon’s URL, following redirects and marking icons as reachable or not.
  • Async Support: Offers asynchronous methods (via asyncio) to efficiently handle multiple favicon extractions concurrently, enhancing overall performance when dealing with numerous URLs.

Installation

Create and activate a virtual environment and then install extract_favicon:

$ pip install extract_favicon

Usage

Extracting Favicons from HTML

The from_html function allows you to parse a given HTML string and extract all favicons referenced within it. It looks for common <link> and <meta> tags that reference icons (e.g., icon, shortcut icon, apple-touch-icon, etc.). If include_fallbacks is set to True, it will also check standard fallback paths like favicon.ico when no icons are explicitly defined.

Example: 

html_content = """
<!DOCTYPE html>
<html>
<head>
  <link rel="icon" href="https://example.com/favicon.ico" />
  <link rel="apple-touch-icon" href="/apple-touch-icon.png">
</head>
<body>
  <p>Sample page</p>
</body>
</html>
"""

favicons = from_html(html_content, root_url="https://example.com", include_fallbacks=True)
for favicon in favicons:
    print(favicon.url, favicon.width, favicon.height)

Extracting Favicons from a URL

If you only have a URL and want to directly extract favicons, from_url fetches the page, parses it, and returns a set of Favicon objects. It uses Reachable internally to check if the URL is accessible. If include_fallbacks is True, fallback icons (like /favicon.ico) are also considered.

favicons = from_url("https://example.com", include_fallbacks=True)
for favicon in favicons:
    print(favicon.url, favicon.format, favicon.width, favicon.height)

Downloading Favicons

Depending on the mode, you can choose to download:

  • "all": Download all favicons.
  • "biggest": Download only the largest favicon (by area).
  • "smallest": Download only the smallest favicon.

If include_unknown is False, favicons without known dimensions are skipped. The sort option sorts the returned favicons by size, and sleep_time controls how long to wait between requests to avoid rate limits.

The result is a list of RealFavicon objects, which contain additional information like the loaded image or raw SVG data.

favicons = from_url("https://example.com")
real_favicons = download(favicons, mode="all", sort="DESC")

for real_favicon in real_favicons:
    print(real_favicon.url.url, real_favicon.valid, real_favicon.width, real_favicon.height)

Checking Favicon Availability

Sends a HEAD request for each favicon URL to determine if it’s reachable. If the favicon has been redirected, it updates the URL accordingly. It also sets the reachable attribute on each Favicon. The sleep_time parameter lets you pause between checks to reduce the load on the target server.

favicons = from_url("https://example.com")
checked_favicons = check_availability(favicons)

for favicon in checked_favicons:
    print(favicon.url, favicon.reachable)

Guessing Favicon Sizes

If some extracted favicons don’t have their dimensions specified, guess_missing_sizes can attempt to determine their width and height. For base64-encoded favicons (data URLs), setting load_base64_img to True allows the function to decode and load the image in memory to get its size. For external images, it partially downloads the image to guess its dimensions without retrieving the entire file.

favicons = from_url("https://example.com")
# Some favicons may not have width/height info
favicons_with_sizes = guess_missing_sizes(favicons, load_base64_img=True)

for favicon in favicons_with_sizes:
    print(favicon.url, favicon.width, favicon.height)

Dependencies

When you install extract_favicon it comes with the following dependencies:

  • BeautifulSoup - to parse HTML content.
  • Pillow - to load images to get real size once downloaded and to guess image size based on its streamed headers.
  • Reachable - to check availability of favicons' URLs, download content and handle redirects, HTTP errors and some simple anti-bot protections.
  • DefusedXML - to parse and check validity of SVG files.

License

This project is licensed under the terms of the MIT license.