\n # We prioritize finding a ‘recipe-post’ article.
\n recipe_article = soup.find(‘article’, class_=’recipe-post’)
\n if recipe_article:
\n main_article_element = recipe_article
\n else:
\n # Fallback to finding any article with an ID starting with ‘post-‘, but explicitly check its title.
\n # This helps distinguish a true article from a 404 page that might also use <\/p>\n
\n generic_article = soup.find(‘article’, id=lambda x: x and x.startswith(‘post-‘))
\n if generic_article:
\n title_element = generic_article.find(‘h1′, class_=’entry-title’)
\n # Only consider it a valid article if its title is not “Page Not Found”
\n if title_element and title_element.get_text(strip=True).lower() != ‘page not found’:
\n main_article_element = generic_article<\/p>\n
article_content = None
\n if main_article_element:
\n # Once a valid article element is identified, find its specific content div.
\n article_content = main_article_element.find(‘div’, class_=’entry-content’)<\/p>\n
if not article_content:
\n # If no valid article content is found (e.g., it’s a 404 page, or unexpected structure),
\n # return an empty string as per the implied request for “final article”.
\n return “”<\/p>\n
# Process images within the article content
\n for img in article_content.find_all(‘img’):
\n # Prioritize ‘data-lazy-src’ for the ‘src’ attribute, common for lazy-loaded images
\n if img.has_attr(‘data-lazy-src’):
\n real_src = img[‘data-lazy-src’]
\n if not urlparse(real_src).netloc:
\n img[‘src’] = urljoin(base_url, real_src)
\n else:
\n img[‘src’] = real_src
\n del img[‘data-lazy-src’] # Remove the lazy-load attribute after copying
\n elif img.has_attr(‘src’):
\n img_src = img[‘src’]
\n # Convert relative ‘src’ URLs to absolute
\n if not urlparse(img_src).netloc:
\n img[‘src’] = urljoin(base_url, img_src)<\/p>\n
# Prioritize ‘data-lazy-srcset’ for the ‘srcset’ attribute
\n srcset_str = None
\n if img.has_attr(‘data-lazy-srcset’):
\n srcset_str = img[‘data-lazy-srcset’]
\n del img[‘data-lazy-srcset’] # Remove the lazy-load attribute after copying
\n elif img.has_attr(‘srcset’):
\n srcset_str = img[‘srcset’]<\/p>\n
if srcset_str:
\n srcset_values = srcset_str.split(‘,’)
\n new_srcset_values = []
\n for val in srcset_values:
\n parts = val.strip().split(‘ ‘)
\n if len(parts) > 0:
\n src_url = parts[0]
\n # Convert relative ‘srcset’ URLs to absolute
\n if not urlparse(src_url).netloc:
\n absolute_src_url = urljoin(base_url, src_url)
\n parts[0] = absolute_src_url
\n new_srcset_values.append(‘ ‘.join(parts))
\n img[‘srcset’] = ‘, ‘.join(new_srcset_values)<\/p>\n
# Remove lazyload-related classes to ensure immediate display without JavaScript
\n if img.has_attr(‘class’):
\n img[‘class’] = [c for c in img[‘class’] if not c.startswith(‘lazy’)]<\/p>\n