Extract SVG from website in Python

How to Extract SVG from Website

The ability to extract images from HTML is crucial for various applications, including web scraping and content analysis. Aspose.HTML for Python via .NET is a robust library that simplifies this process by offering developers a set of tools to navigate and gather information from HTML documents seamlessly. This powerful solution is ideal for anyone who needs to collect SVGs for analysis, archiving, or content creation – eliminating the need for manual work. Let’s explore how to download SVG images from web pages.

Extract SVG Using Python

SVG images in HTML documents come in two forms – inline SVG and external SVG. The following Python code demonstrates how to automate the extraction of external SVG images, SVG files stored separately from HTML and referenced via <img> tags, from a web page using the Aspose.HTML for Python via .NET library:

import os
import aspose.html as ah
import aspose.html.net as ahnet

# Define the output directory
output_dir = "output/svg/"
os.makedirs(output_dir, exist_ok=True)

# Open the document you want to extract external SVGs from
document = ah.HTMLDocument("https://products.aspose.com/html/python-net/")

# Collect all <img> elements
images = document.get_elements_by_tag_name("img")

# Create a distinct collection of relative image URLs
urls = set(img.get_attribute("src") for img in images)

# Filter only SVG images
svg_urls = [url for url in urls if url.endswith(".svg")]

# Convert relative URLs to absolute using Url from aspose.html
abs_urls = [ah.Url(url, document.base_uri) for url in svg_urls]

for url in abs_urls:
    # Create a network request for the SVG
    request = ahnet.RequestMessage(url.href)

    # Send request to fetch the SVG
    response = document.context.network.send(request)

    # Check if request succeeded
    if response.is_success:
        # Determine local file path
        file_path = os.path.join(output_dir, os.path.basename(url.pathname))

        # Save SVG to a local file system
        with open(file_path, "wb") as f:
            f.write(response.content.read_as_byte_array())

Steps to Extract SVG from website

1. Use the HTMLDocument(Url) constructor to create an instance of the HTMLDocument class and pass it the URL of the website from which you want to extract external SVG images.
2. Use the get_elements_by_tag_name("img") method to collect all <img> elements from the HTML document. This method returns a list of image elements embedded in the page.
3. Iterate through the collected <img> elements and use the get_attribute("src") method to extract the src attribute from each element. Store these values in a set to eliminate duplicates.
4. Filter the extracted URLs by checking if they end with the “.svg” extension to isolate only the external SVG images.
5. Use the Url class and the document’s base_uri to create absolute SVG image URLs.
6. Create a RequestMessage instance for each absolute SVG URL to prepare an HTTP request for retrieving the image.
7. Send the request and check the is_success property to ensure the response was successful.
8. Use os.path.basename(url.pathname) to get the file name from the SVG URL, then save the image to the local file system by writing the binary content to the output directory.

To learn more about how to programmatically extract inline and external SVGs from a website, refer to the documentation article Extracting SVG from a Website in Python .

Note: Always respect copyright laws. Make sure you have the appropriate rights, permissions, or licenses before using the extracted images for commercial purposes. We do not endorse or support the unauthorized use of copyrighted content.

Get Started with Python API

If you want to parse, manipulate, and manage HTML documents, install our flexible, high-speed Aspose.HTML for Python via .NET API. pip is the easiest way to download and install Aspose.HTML for Python via .NET. To do this, run the following command:

pip install aspose-html-net

For more details about Python library installation and system requirements, please refer to Aspose.HTML Documentation.