Image sitemap generate form html

imagesitemap.py

"""
Generate Image Sitemap for a Single HTML Page

This script reads an HTML file, extracts all the image URLs from the <img> tags, and generates an XML sitemap compliant with Google's image sitemap protocol. The generated sitemap can help improve the visibility of images in search engines.

Usage:
- Ensure you have the 'beautifulsoup4' package installed: `pip install beautifulsoup4`
- Set the `html_file` variable to the path of your HTML file.
- Set the `page_url` variable to the URL of the page where the images are located.
- Run the script to generate 'sitemap.xml' containing the image sitemap.

Parameters:
- `html_file` (str): Path to the HTML file to parse.
- `page_url` (str): The URL of the page to be included in the sitemap.
- `output_file` (str, optional): The filename for the output sitemap. Defaults to 'sitemap.xml'.

Example:
html_file = "index.html"
page_url = "https://example.com"
generate_image_sitemap_single_page(html_file, page_url)
"""

from bs4 import BeautifulSoup

def generate_image_sitemap_single_page(html_file, page_url, output_file="sitemap.xml"):
    # Read the HTML file
    with open(html_file, 'r', encoding='utf-8') as file:
        soup = BeautifulSoup(file, 'html.parser')

    # Find all image tags and extract their URLs
    images = soup.find_all('img')
    image_urls = [img['src'] for img in images if 'src' in img.attrs]

    # Build the sitemap XML structure
    sitemap = '<?xml version="1.0" encoding="UTF-8"?>\n'
    sitemap += '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"\n'
    sitemap += '        xmlns:image="http://www.google.com/schemas/sitemap-image/1.1">\n'
    sitemap += '  <url>\n'
    sitemap += f'    <loc>{page_url}</loc>\n'

    for image_url in image_urls:
        sitemap += f'    <image:image>\n'
        sitemap += f'      <image:loc>{image_url}</image:loc>\n'
        sitemap += f'    </image:image>\n'

    sitemap += '  </url>\n'
    sitemap += '</urlset>'

    # Write the sitemap to a file
    with open(output_file, 'w', encoding='utf-8') as file:
        file.write(sitemap)

    print(f"Sitemap saved to {output_file}")

# Example usage
html_file = "index.html"  # Path to your HTML file
page_url = "https://example.com"  # The page URL
generate_image_sitemap_single_page(html_file, page_url)