Migration nach Drupal: MediaWiki, XWiki, Wiki.js und mkdocs/Zensical importieren¶

import requests
from requests.auth import HTTPBasicAuth

DRUPAL_JSONAPI_URL = "https://drupal.wissen-ahrensburg.de/jsonapi/node/page"
AUTH = HTTPBasicAuth("migration_bot", "BOT_PASSWORT")
HEADERS = {
    "Content-Type": "application/vnd.api+json",
    "Accept": "application/vnd.api+json",
}

def create_drupal_page(title, html_body):
    payload = {
        "data": {
            "type": "node--page",
            "attributes": {
                "title": title,
                "body": {"value": html_body, "format": "full_html"},
                "status": False,  # unveroeffentlicht -> menschliche Review noetig
            },
        }
    }
    r = requests.post(DRUPAL_JSONAPI_URL, json=payload, auth=AUTH, headers=HEADERS)
    if r.status_code == 201:
        print(f"✅ '{title}' als Entwurf angelegt")
    else:
        print(f"❌ Fehler bei '{title}': {r.status_code} - {r.text}")

import subprocess

MEDIAWIKI_API = "https://mediawiki.wissen-ahrensburg.de/api.php"

def get_all_titles():
    titles, apcontinue = [], None
    while True:
        params = {"action": "query", "list": "allpages", "aplimit": "max", "format": "json"}
        if apcontinue:
            params["apcontinue"] = apcontinue
        r = requests.get(MEDIAWIKI_API, params=params).json()
        titles += [p["title"] for p in r["query"]["allpages"]]
        apcontinue = r.get("continue", {}).get("apcontinue")
        if not apcontinue:
            break
    return titles

def get_wikitext(title):
    params = {"action": "parse", "page": title, "prop": "wikitext", "format": "json"}
    r = requests.get(MEDIAWIKI_API, params=params).json()
    return r["parse"]["wikitext"]["*"]

def wikitext_to_html(wikitext):
    result = subprocess.run(
        ["pandoc", "-f", "mediawiki", "-t", "html"],
        input=wikitext, capture_output=True, text=True,
    )
    return result.stdout

for title in get_all_titles():
    html = wikitext_to_html(get_wikitext(title))
    create_drupal_page(title, html)

XWIKI_REST_URL = "https://xwiki.wissen-ahrensburg.de/rest"
XWIKI_EXPORT_URL = "https://xwiki.wissen-ahrensburg.de/bin/export/{space}/{page}?format=html"
XWIKI_AUTH = HTTPBasicAuth("admin", "admin_password")

def get_space_pages(wiki="xwiki", space="Main"):
    url = f"{XWIKI_REST_URL}/wikis/{wiki}/spaces/{space}/pages"
    r = requests.get(url, auth=XWIKI_AUTH, headers={"Accept": "application/json"})
    return r.json().get("pageSummaries", []) if r.status_code == 200 else []

def get_xwiki_html(space, page):
    url = XWIKI_EXPORT_URL.format(space=space, page=page)
    r = requests.get(url, auth=XWIKI_AUTH)
    return r.text

for summary in get_space_pages():
    html = get_xwiki_html("Main", summary["name"])
    create_drupal_page(summary["title"], html)

WIKIJS_URL = "https://wiki.wissen-ahrensburg.de/graphql"
WIKIJS_HEADERS = {"Authorization": "Bearer <API_TOKEN>", "Content-Type": "application/json"}

LIST_QUERY = "query { pages { list { id path title } } }"
SINGLE_QUERY = """
query ($id: Int!) {
  pages { single(id: $id) { title content } }
}
"""

def get_wikijs_pages():
    r = requests.post(WIKIJS_URL, json={"query": LIST_QUERY}, headers=WIKIJS_HEADERS)
    return r.json()["data"]["pages"]["list"]

def get_wikijs_content(page_id):
    r = requests.post(WIKIJS_URL, json={"query": SINGLE_QUERY, "variables": {"id": page_id}}, headers=WIKIJS_HEADERS)
    return r.json()["data"]["pages"]["single"]

def markdown_to_html(markdown_text):
    result = subprocess.run(
        ["pandoc", "-f", "markdown", "-t", "html"],
        input=markdown_text, capture_output=True, text=True,
    )
    return result.stdout

for summary in get_wikijs_pages():
    detail = get_wikijs_content(summary["id"])
    html = markdown_to_html(detail["content"])
    create_drupal_page(detail["title"], html)

import pathlib
import re

DOCS_DIR = pathlib.Path("/pfad/zum/mkdocs-repo/docs")  # nur docs/, nicht raw/ oder site/

def extract_title(markdown_text, fallback):
    match = re.search(r"^#\s+(.+)$", markdown_text, re.MULTILINE)
    return match.group(1).strip() if match else fallback

for md_file in sorted(DOCS_DIR.rglob("*.md")):
    text = md_file.read_text(encoding="utf-8")
    title = extract_title(text, fallback=md_file.stem)
    html = markdown_to_html(text)  # dieselbe pandoc-Funktion wie im Wiki.js-Abschnitt
    create_drupal_page(title, html)