from collections import defaultdict
from datetime import datetime, timedelta, timezone
from typing import Any

import httpx

from app.core.config import settings


class WikiService:
    def __init__(self) -> None:
        self.api_url = settings.wiki_api_url

    async def get_recent_changes(self, days: int = 30, only_edited: bool = False) -> list[dict[str, Any]]:
        now = datetime.now(timezone.utc)
        start = now - timedelta(days=days)
        rcstart = now.strftime("%Y-%m-%dT%H:%M:%SZ")
        rcend = start.strftime("%Y-%m-%dT%H:%M:%SZ")

        params = {
            "action": "query",
            "format": "json",
            "list": "recentchanges",
            "rcprop": "title|timestamp|user|comment|flags|sizes|loginfo",
            "rclimit": "200",
            "rcstart": rcstart,
            "rcend": rcend,
            "rcnamespace": "0",
        }
        if only_edited:
            params["rctype"] = "edit"
        else:
            params["rctype"] = "new|edit"

        rows: list[dict[str, Any]] = []
        cont: dict[str, Any] = {}
        async with httpx.AsyncClient(timeout=20.0) as client:
            while True:
                response = await client.get(self.api_url, params={**params, **cont})
                response.raise_for_status()
                payload = response.json()
                rows.extend(payload.get("query", {}).get("recentchanges", []))
                if "continue" not in payload:
                    break
                cont = payload["continue"]

            titles = sorted({item["title"] for item in rows if item.get("title")})
            categories = await self._get_categories_for_titles(client, titles)

        for row in rows:
            row["categories"] = categories.get(row.get("title", ""), [])
        return rows

    async def _get_categories_for_titles(self, client: httpx.AsyncClient, titles: list[str]) -> dict[str, list[str]]:
        result: dict[str, list[str]] = defaultdict(list)
        if not titles:
            return {}

        chunk_size = 25
        for i in range(0, len(titles), chunk_size):
            chunk = titles[i : i + chunk_size]
            params = {
                "action": "query",
                "format": "json",
                "prop": "categories",
                "cllimit": "max",
                "titles": "|".join(chunk),
            }
            response = await client.get(self.api_url, params=params)
            response.raise_for_status()
            pages = response.json().get("query", {}).get("pages", {})
            for page in pages.values():
                title = page.get("title")
                if not title:
                    continue
                raw_categories = page.get("categories", [])
                result[title] = [c.get("title", "").replace("Kategorie:", "") for c in raw_categories if c.get("title")]
        return result