# -----------------------------------------------------------------------------
# Sorti-Ka — Agrégateur de sorties & événements (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/terrebonne.py : connecteur Ville de Terrebonne — calendrier
#   Source    : https://terrebonne.ca/calendrier-des-evenements/ (WordPress).
#   Extraction: API REST WordPress publique /wp-json/wp/v2/event (champs ACF
#               start_date/end_date « YYYY-MM-DD HH:MM:SS », place.name,
#               isFull, hours) + taxonomie /wp-json/wp/v2/event_category.
#               Cloudflare bloque tout GET non navigateur (403, consigné
#               2026-08-18) → chaque requête passe par Scrapfly (asp=true,
#               SANS rendu JS : la réponse est du JSON), ~3-5 appels/sync.
#               Vérifié 2026-08-25 (Scrapfly 200, JSON complet).
#   Accès     : API publique du site municipal (aucune auth) ; robots.txt
#               n'interdit pas /wp-json/. Anti-bot contourné avec volumes
#               minimes et cadence quotidienne au plus — usage mesuré.
#   Prix      : non publié par la source → jamais inventé. isFull=True →
#               status "soldout" (complet, publié par la ville).
# -----------------------------------------------------------------------------
from __future__ import annotations

import json
import os

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

API = ("https://terrebonne.ca/wp-json/wp/v2/event"
       "?per_page=100&page={page}&_embed=wp:featuredmedia")
TAXO = "https://terrebonne.ca/wp-json/wp/v2/event_category?per_page=100"
PAGES_MAX = int(os.environ.get("TER_PAGES", "5"))


class TerrebonneConnector(BaseConnector):
    source_id = "terrebonne"
    request_delay = 1.0

    def _get_json_scrapfly(self, url: str):
        """JSON via Scrapfly (asp, sans rendu JS) — Cloudflare bloque le direct."""
        return json.loads(self.get_rendered(url, render_js=False))

    def _categories(self) -> dict[int, str]:
        try:
            terms = self._get_json_scrapfly(TAXO)
            return {int(t["id"]): clean_text(t.get("name") or "")
                    for t in terms if t.get("id")}
        except Exception as exc:   # taxonomie indisponible ≠ source cassée
            print(f"[sorti-ka] terrebonne : taxonomie ignorée : {exc}")
            return {}

    @staticmethod
    def _image(rec: dict) -> str:
        for media in (rec.get("_embedded") or {}).get("wp:featuredmedia") or []:
            if isinstance(media, dict) and media.get("source_url"):
                return media["source_url"]
        return ""

    def _rows(self) -> list[dict]:
        out, page = [], 1
        while page <= PAGES_MAX:
            try:
                batch = self._get_json_scrapfly(API.format(page=page))
            except Exception:
                if page > 1:      # « rest_post_invalid_page_number » = fin
                    break
                raise
            if not isinstance(batch, list) or not batch:
                break
            out += batch
            if len(batch) < 100:
                break
            page += 1
        return out

    def fetch(self) -> list[Event]:
        rows = self._rows()
        cat_names = self._categories()
        events: list[Event] = []
        for rec in rows:
            title = clean_text((rec.get("title") or {}).get("rendered") or "")
            link = rec.get("link") or ""
            acf = rec.get("acf") or {}
            start_raw = str(acf.get("start_date") or "")
            start = parse_date_iso(start_raw)
            if not title or not link or not start:
                continue
            start_time = parse_time(start_raw)
            if start_time == "00:00":
                start_time = None
            end_raw = str(acf.get("end_date") or "")
            end = parse_date_iso(end_raw)
            end_time = parse_time(end_raw)
            if end_time == "00:00":
                end_time = None
            if not start_time:      # repli : champ texte « hours » de la fiche
                start_time = parse_time(str(acf.get("hours") or ""))
            place = acf.get("place") or {}
            venue = clean_text(place.get("name") or "") if isinstance(place, dict) else ""
            cats = [cat_names[c] for c in rec.get("event_category") or []
                    if c in cat_names]
            events.append(Event(
                source=self.source_id,
                external_id=str(rec.get("id")),
                url=link,
                title=title,
                description=clean_text(
                    (rec.get("content") or {}).get("rendered") or ""),
                raw_categories=cats or [title],
                venue=venue,
                city="Terrebonne",
                region="Lanaudière",
                start_date=start,
                start_time=start_time,
                end_date=end or start,
                end_time=end_time,
                status="soldout" if acf.get("isFull") else "",
                organizer="Ville de Terrebonne",
                image=self._image(rec),
            ))
        return events