from fastapi import FastAPI, HTTPException
from fastapi.responses import JSONResponse
import feedparser
import re
import datetime as dt
from email.utils import parsedate_to_datetime

app = FastAPI()

BASE = "https://www.umwelt.sachsen.de/umwelt/infosysteme/hwims/portal/web/feed/wasserstand-pegel-{}"

def parse_number(pattern, text):
    m = re.search(pattern, text, re.I)
    return float(m.group(1).replace(",", ".")) if m else None

def parse_text(pattern, text):
    m = re.search(pattern, text, re.I)
    return m.group(1).strip() if m else None

def clean_text(text):
    text = re.sub(r"<br\s*/?>", "\n", text, flags=re.I)
    text = re.sub(r"<[^>]+>", " ", text)
    return text

def extract_station_name(title: str):
    if not title:
        return None

    m = re.search(r"Pegel\s+(.+)$", title)
    return m.group(1).strip() if m else title

def parse_timestamp_from_title(title: str):
    if not title:
        return None

    m = re.search(r"(\d{2}\.\d{2}\.\d{4})\s+(\d{2}:\d{2})\s+Uhr\s+\((MESZ|MEZ)", title)
    if not m:
        return None

    date_part, time_part, tz_hint = m.groups()
    parsed = dt.datetime.strptime(f"{date_part} {time_part}", "%d.%m.%Y %H:%M")
    offset = dt.timedelta(hours=2 if tz_hint == "MESZ" else 1)
    return parsed.replace(tzinfo=dt.timezone(offset)).isoformat()

def parse_timestamp(entry):
    ts_from_title = parse_timestamp_from_title(entry.get("title"))
    if ts_from_title:
        return ts_from_title

    if entry.get("dc_date"):
        return entry.get("dc_date")

    if entry.get("published"):
        return parsedate_to_datetime(entry.get("published")).isoformat()

    return None

def parse_entry(entry):
    text = clean_text(entry.get("description", ""))

    return {
        "timestamp": parse_timestamp(entry),
        "title": entry.get("title"),
        "water_level_cm": parse_number(
            r"Wasserstand:\s*([0-9]+(?:[,.][0-9]+)?)\s*cm", text
        ),
        "flow_m3s": parse_number(
            r"Durchfluss:\s*([0-9]+(?:[,.][0-9]+)?)\s*m³/s", text
        ),
        "flood_alert_level": parse_text(
            r"Meldestufe:\s*([^\n\r]+)", text
        ),
    }

@app.get("/api/pegel/{pegel_id}")
def get_pegel(pegel_id: str):
    if not re.fullmatch(r"\d{6,}", pegel_id):
        raise HTTPException(400, "Ungültige Pegel-ID")

    url = BASE.format(pegel_id)
    feed = feedparser.parse(url)

    if not feed.entries:
        raise HTTPException(404, "Keine Daten gefunden")

    entries = [parse_entry(entry) for entry in feed.entries]

    return JSONResponse({
        "station_id": pegel_id,
        "station_title": extract_station_name(feed.feed.get("title")),
        "count": len(entries),
        "latest": entries[0],
        "entries": entries,
        "source": url,
        "fetched_at": dt.datetime.now(dt.UTC).isoformat(),
    })