Python script to tag MediaWiki pages with a category matching the page's namespace.

tag_namespace_with_category.py

#!/usr/bin/python3
import requests
import mwclient
import re
import os
os.system('cls')

# Variables
excluded_namespaces = {"MediaWiki", "Media", "Template", "Module", "Topic", "Talk", "User", "File", "Category", "Special", "Translations", "Hidden"}

# Set up your MediaWiki API connection
site = mwclient.Site('EXAMPLE.org', path='/wiki/')  # Replace with your wiki's URL

# Authenticate if required
site.login(username='BOT_USERNAME_HERE', password='PASSWORD_HERE')

######################################

# Ask for what namespace to target
target_all = False
target_namespace = input(f"\nWhat namespace do you want to change? ['ALL' to update all] ('q' to quit): ")
if target_namespace == 'q':
    exit()
if target_namespace == 'ALL':
    target_all = True

# Get list of namespaces
namespace_map = {v: k for k, v in site.namespaces.items()}
namespace_map = {
    name: ns_id
    for ns_id, name in site.namespaces.items()
    if name and name not in excluded_namespaces and "talk" not in name
}

namespace_list = list(namespace_map.items())
namespace_index = 0
first_run = True

while(1):
    if target_all:
        # Make sure index is within range
        if namespace_index < len(namespace_list):
            name, id = namespace_list[namespace_index]
            target_namespace = name
            target_category = target_namespace
            namespace_index += 1
        else:
            print("No more namespaces to target")
            exit()
    else:
        # Avoid duplicate question for the first loop
        if first_run:
            first_run = False
        else:
            target_namespace = input(f"\nWhat namespace do you want to change? ('q' to quit): ")
            if target_namespace == 'q':
                exit()

        target_category = input(f"\nWhat category do you want to change? [" + target_namespace + "]: ")
        if target_namespace == 'q':
            exit()

        if (target_category == ''):
            target_category = target_namespace
        
    print("\nTagging pages in namespace '" + target_namespace + "' with category '" + target_category + "'")
    
    # Convert namespace name to ID
    namespace_id = namespace_map.get(target_namespace)

    if namespace_id is None:
        print(f"Unknown namespace: {target_namespace}")
        continue

    print(f"Getting all pages in namespace '{target_namespace}'...")
    namespace_pages = {page.name for page in site.allpages(namespace=namespace_id)}

    print(f"Getting all pages in category 'Category:{target_category}'...")
    category_pages = {page.name for page in site.categories[target_category]}

    # Find delta (pages in namespace that are not in the category)
    delta_pages = namespace_pages - category_pages
    print(f"{len(delta_pages)} pages are missing the category.")
    # for delta_page in delta_pages: # Debug
    #     print(delta_page)

    # Append category to each delta page
    for page_name in sorted(delta_pages):
        # Skip translation pages
        if '/' in page_name:
            print(f"Skipping (translation) {page_name}")
            continue

        # Get page text
        page = site.pages[page_name]
        text = page.text()

        # Skip redirect pages
        if '#redirect' in text.lower():
            print(f"Skipping (redirect) {page_name}")
            continue

        # Skip if already tagged with the category
        category_pattern = re.compile(rf'\[\[\s*category\s*:\s*{re.escape(target_category)}\s*\]\]', re.IGNORECASE)
        if category_pattern.search(text):
            print(f"Skipping (already tagged) {page_name}")
            continue

        # Add category tag
        # print(page) # Debug
        new_text = text.strip() + f"\n\n[[Category:{target_category}]]"
        page.save(new_text, summary=f"Adding [[Category:{target_category}]]")
        print(f"Updated {page_name}")