# -----------------------------------------------------------------------------
# Sorti-Ka — Agrégateur de sorties & événements (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/rimouski.py : connecteur Ville de Rimouski — événements officiels
#   Source    : https://rimouski.ca/loisirs-culture/activites/evenements
#   Extraction: flux RSS ÉTENDU du calendrier (découvert sur la page) :
#               /rss/loisirs-culture/activites/evenements/ — ~240 items avec
#               balises maison <startDate> (ISO), <endDate>, <eventTime>
#               (« 13 h 30 »), <category>, <organizer>, <location>
#               (« Salle (adresse civique) »), <image>, <description>.
#               1 seule requête. Vérifié 2026-08-25.
#   Accès     : flux RSS public du site municipal, robots.txt permissif
#               (sitemap seulement). GET direct, identification honnête.
#   Prix      : non publié par le flux → jamais inventé.
# -----------------------------------------------------------------------------
from __future__ import annotations

import html as _html
import re

from ..normalize import clean_text, parse_date_iso, parse_time
from ..schema import Event
from .base import BaseConnector

FEED = "https://rimouski.ca/rss/loisirs-culture/activites/evenements/"

_ITEM_RE = re.compile(r"<item>(.*?)</item>", re.S)
# « Bibliothèque Lisette-Morin (110, rue de l'Évêché Est) » → salle + adresse
_LOC_RE = re.compile(r"^(.*?)\s*(?:\(([^)]+)\))?\s*$", re.S)


def _tag(block: str, name: str) -> str:
    m = re.search(rf"<{name}>(.*?)</{name}>", block, re.S)
    return _html.unescape(m.group(1)).strip() if m else ""


class RimouskiConnector(BaseConnector):
    source_id = "rimouski"

    def _parse_feed(self, xml: str) -> list[dict]:
        rows = []
        for block in _ITEM_RE.findall(xml):
            link = _tag(block, "link")
            title = clean_text(_tag(block, "title"))
            start = parse_date_iso(_tag(block, "startDate"))
            if not link or not title or not start:
                continue
            venue = address = ""
            m = _LOC_RE.match(_tag(block, "location"))
            if m:
                venue = clean_text(m.group(1))
                address = clean_text(m.group(2) or "")
            rows.append({
                "link": link,
                "title": title,
                "start": start,
                "end": parse_date_iso(_tag(block, "endDate")),
                "time": parse_time(_tag(block, "eventTime")),
                "description": clean_text(_tag(block, "description")),
                "category": clean_text(_tag(block, "category")),
                "organizer": clean_text(_tag(block, "organizer")),
                "venue": venue,
                "address": address,
                "image": _tag(block, "image"),
            })
        return rows

    def fetch(self) -> list[Event]:
        rows = self._parse_feed(self.get(FEED).text)
        events: list[Event] = []
        seen: set[str] = set()
        for r in rows:
            slug = r["link"].rstrip("/").rsplit("/", 1)[-1]
            ext = f"{slug}-{r['start']}"
            if ext in seen:
                continue
            seen.add(ext)
            events.append(Event(
                source=self.source_id,
                external_id=ext,
                url=r["link"],
                title=r["title"],
                description=r["description"],
                raw_categories=[r["category"]] if r["category"] else [r["title"]],
                venue=r["venue"],
                address=r["address"],
                city="Rimouski",
                region="Bas-Saint-Laurent",
                start_date=r["start"],
                start_time=r["time"],
                end_date=r["end"] or r["start"],
                organizer=r["organizer"],
                image=r["image"],
            ))
        return events