claude-plugins/plugins/mediawiki/skills/mediawiki/wiki.py

#!/usr/bin/env -S uv run --script
# /// script
# requires-python = ">=3.11"
# dependencies = [
#   "mwclient",
#   "python-dotenv",
# ]
# ///
"""CLI for fetching and updating MediaWiki pages."""

import argparse
import os
import sys
from pathlib import Path
from urllib.parse import urlparse

from dotenv import load_dotenv
import mwclient


WIKI_ALIASES = {
    "wiki":    ("MEDIAWIKI_URL",    "https://wiki.inmedias.it"),
    "devwiki": ("DEVMEDIAWIKI_URL", "https://devwiki.inmedias.it"),
}


def get_site(wiki: str = "wiki") -> mwclient.Site:
    load_dotenv(Path.home() / ".config/dokumentierer/.env")
    load_dotenv()  # current dir, doesn't override already-set vars

    if wiki not in WIKI_ALIASES:
        sys.exit(f"Error: unknown wiki '{wiki}'. Known: {', '.join(WIKI_ALIASES)}")

    url_env, url_default = WIKI_ALIASES[wiki]
    url = os.environ.get(url_env, url_default).rstrip("/")

    prefix = "DEVMEDIAWIKI_" if wiki == "devwiki" else "MEDIAWIKI_"
    user = os.environ.get(f"{prefix}USER") or os.environ.get("MEDIAWIKI_USER", "")
    password = os.environ.get(f"{prefix}PASSWORD") or os.environ.get("MEDIAWIKI_PASSWORD", "")

    if not user:
        sys.exit(f"Error: MEDIAWIKI_USER is not set in .env")
    if not password:
        sys.exit(f"Error: MEDIAWIKI_PASSWORD is not set in .env")

    parsed = urlparse(url)
    host = parsed.netloc
    # mwclient expects path to end with '/'
    path = (parsed.path or "/") + "/"
    path = path.replace("//", "/")
    scheme = parsed.scheme or "https"

    site = mwclient.Site(host, path=path, scheme=scheme)
    site.login(user, password)
    return site


def cmd_fetch(args: argparse.Namespace) -> None:
    site = get_site(args.wiki)
    page = site.pages[args.title]
    if not page.exists:
        sys.exit(f"Error: page '{args.title}' does not exist")
    print(page.text(), end="")


NAMESPACES = {
    "main": 0, "talk": 1, "user": 2, "project": 4,
    "file": 6, "template": 10, "category": 14,
}


def _strip_html(text: str) -> str:
    """Remove HTML tags (used to clean search snippets)."""
    import re
    return re.sub(r"<[^>]+>", "", text)


def cmd_search(args: argparse.Namespace) -> None:
    site = get_site(args.wiki)
    ns_key = args.namespace
    if ns_key is None or ns_key.lower() == "all":
        ns = "*"
    elif ns_key.isdigit():
        ns = int(ns_key)
    else:
        ns = NAMESPACES.get(ns_key.lower())
        if ns is None:
            sys.exit(f"Error: unknown namespace '{ns_key}'. "
                     f"Known: all, {', '.join(NAMESPACES)}")

    result = site.api(
        "query", list="search",
        srsearch=args.query,
        srnamespace=ns,
        srlimit=50,
        srprop="title|snippet",
    )
    hits = result["query"]["search"]
    if not hits:
        print("(keine Ergebnisse)", file=sys.stderr)
    for hit in hits:
        print(hit["title"])
        if hit.get("snippet"):
            print(f"  {_strip_html(hit['snippet']).strip()}")


def cmd_render(args: argparse.Namespace) -> None:
    site = get_site(args.wiki)
    result = site.api("parse", page=args.title, prop="text")
    print(result["parse"]["text"]["*"], end="")


def cmd_upload(args: argparse.Namespace) -> None:
    path = Path(args.file_path)
    if not path.exists():
        sys.exit(f"Error: file not found: {path}")
    site = get_site(args.wiki)
    with path.open("rb") as fh:
        site.upload(fh, filename=args.wiki_filename,
                    description=args.description or "", ignore=True)
    print(f"Uploaded '{args.wiki_filename}'")


def cmd_update(args: argparse.Namespace) -> None:
    if args.file == "-":
        content = sys.stdin.read()
    else:
        try:
            with open(args.file, "r", encoding="utf-8") as fh:
                content = fh.read()
        except OSError as exc:
            sys.exit(f"Error reading file: {exc}")

    site = get_site(args.wiki)
    page = site.pages[args.title]
    page.save(content, summary=args.summary or "")
    print(f"Saved '{args.title}'")


def main() -> None:
    parser = argparse.ArgumentParser(description="MediaWiki page fetch/update tool")
    parser.add_argument(
        "--wiki", "-w",
        choices=list(WIKI_ALIASES),
        default="wiki",
        help="Wiki-Instanz (Standard: wiki = wiki.inmedias.it)",
    )
    sub = parser.add_subparsers(dest="command", required=True)

    fetch_p = sub.add_parser("fetch", help="Print page wikitext to stdout")
    fetch_p.add_argument("title", help="Page title")
    fetch_p.set_defaults(func=cmd_fetch)

    search_p = sub.add_parser("search", help="Search for pages by title")
    search_p.add_argument("query", help="Search term")
    search_p.add_argument("--namespace", "-n", default=None,
                          help="Namespace filter: all (default), main, template, category, file, ... or numeric ID")
    search_p.set_defaults(func=cmd_search)

    render_p = sub.add_parser("render", help="Print rendered HTML of a page to stdout")
    render_p.add_argument("title", help="Page title")
    render_p.set_defaults(func=cmd_render)

    upload_p = sub.add_parser("upload", help="Upload a local file to the wiki file store")
    upload_p.add_argument("file_path", help="Absolute path to the local file")
    upload_p.add_argument("wiki_filename", help="Target filename on the wiki (without 'File:' prefix)")
    upload_p.add_argument("--description", default="", help="Wikitext for the file description page")
    upload_p.set_defaults(func=cmd_upload)

    update_p = sub.add_parser("update", help="Save wikitext to a page")
    update_p.add_argument("title", help="Page title")
    update_p.add_argument("file", help="File containing wikitext, or '-' for stdin")
    update_p.add_argument("--summary", default="", help="Edit summary")
    update_p.set_defaults(func=cmd_update)

    args = parser.parse_args()
    args.func(args)


if __name__ == "__main__":
    main()