(An Untitled Masterwork)
ActivePublic
Actions

Authored by Ladsgroup on Apr 30 2026, 4:09 PM.

Project Tags

None

Referenced Files

	F78894604: raw-paste-data.txt
	Apr 30 2026, 4:09 PM

Subscribers

None

	import pywikibot
	import re
	from bs4 import BeautifulSoup
	from pywikibot import pagegenerators

	def convert_html_links_to_wikitext(html_content):
	"""Parses HTML and converts <a> tags to [[Wikitext]] links."""
	soup = BeautifulSoup(html_content, 'html.parser')

	for a in soup.find_all('a'):
	# Extract the page title from the 'title' attribute or 'href'
	# MediaWiki links usually have a title attribute with the page name
	link_target = a.get('title')
	link_text = a.get_text()

	if link_target:
	# Format as [[Target\|Text]] or just [[Target]] if they match
	new_link = f"[[{link_target}\|{link_text}]]" if link_target != link_text else f"[[{link_target}]]"
	a.replace_with(new_link)
	else:
	# Fallback for links without titles (like external links)
	href = a.get('href')
	if href:
	a.replace_with(f"[{href} {link_text}]")

	# Get the text back. Use 'decode_contents' to avoid adding <html><body> wrappers
	return soup.get_text()

	def process_category():
	site = pywikibot.Site('it', 'wikinews')
	gen = pagegenerators.SearchPageGenerator(
	query='insource:"DynamicPageList"',
	site=site,
	namespaces=list(site.namespaces.keys()),
	total=None
	)

	# Regex to find DynamicPageList tags (case insensitive)
	dpl_pattern = re.compile(r'<DynamicPageList>(.*?)</DynamicPageList>', re.DOTALL \| re.IGNORECASE)

	for page in gen:
	if 'Main page' in page.title().replace('_', ' '):
	continue
	if 'Newsroom' in page.title():
	continue
	if 'Water cooler' in page.title().replace('_', ' '):
	continue

	text = page.text
	matches = dpl_pattern.findall(text)

	if not matches:
	continue

	pywikibot.output(f"\n>>> Processing {page.title()} <<<")
	new_text = text

	for dpl_content in matches:
	if '{' in dpl_content:
	continue
	pywikibot.output("Found DPL content. Sending to API for parsing...")

	# 1. Send DPL content to the Parsing API
	params = {
	'action': 'parse',
	'text': f'<DynamicPageList>{dpl_content}</DynamicPageList>',
	'contentmodel': 'wikitext',
	'disablelimitreport': True,
	}

	try:
	request = site.simple_request(**params)
	data = request.submit()
	html_output = data['parse']['text']['*']

	# 2. Convert HTML links to Wikitext
	static_wikitext = convert_html_links_to_wikitext(html_output).strip()
	if static_wikitext.count('\n'):
	static_wikitext = '' + '\n'.join(static_wikitext.split('\n'))
	if static_wikitext.count('[[') == 1:
	static_wikitext = '*' + static_wikitext
	print(static_wikitext)

	# 3. Replace the original tag with the new content
	# We use a literal replace to avoid regex collision with the parsed content
	full_tag = f"<DynamicPageList>{dpl_content}</DynamicPageList>"
	new_text = new_text.replace(full_tag, static_wikitext.strip())

	except Exception as e:
	pywikibot.error(f"Failed to parse DPL on {page.title()}: {e}")

	#if new_text != text:
	page.text = new_text
	try:
	page.save(summary="Replacing DynamicPageList with static wikitext ([[phab:T421796]]).")
	except:
	continue

	if __name__ == "__main__":
	process_category()

Event Timeline

Ladsgroup created this paste.Apr 30 2026, 4:09 PM

(An Untitled Masterwork)ActivePublicActions

Event Timeline

(An Untitled Masterwork)
ActivePublic
Actions