# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/gestion_fauvel.py : connecteur Gestion Fauvel (gestionfauvel.com)
#   WordPress + Elementor + JetEngine. La page /logements-a-louer/ expose une
#   grille .jet-listing-grid__item : data-post-id (id stable), data-url,
#   en-têtes (disponibilité, prix, titre) et terme JetEngine = ville réelle.
#   Fiches détail (cache BD) : adresse civique, description complète (unités
#   dispo + inclusions) et galerie photos (carrousel Elementor).
# -----------------------------------------------------------------------------
from __future__ import annotations

import hashlib
import re

from bs4 import BeautifulSoup

from ..schema import Listing, normalize_unit_type, parse_price
from .base import BaseConnector

BASE = "https://gestionfauvel.com"
LIST_URL = f"{BASE}/logements-a-louer/"

# adresse civique : « 555, rue des Écoles, app. 105 Drummondville »
_ADDR_RE = re.compile(
    r"^\d+[\s,]+.*\b(rue|boul(?:evard|\.)?|avenue|av\.|chemin|carr[ée]|place|"
    r"c[ôo]te|mont[ée]e|rang)\b", re.I)
_IMG_EXT_RE = re.compile(r"\.(?:jpe?g|png|webp)$", re.I)
_UNIT_RE = re.compile(r"^(?:\d½\+?|Studio|Loft|Maison)$")


class GestionFauvelConnector(BaseConnector):
    source_id = "gestion_fauvel"
    request_delay = 0.6
    max_details = 25   # garde-fou fiches détail (vraies requêtes)

    def fetch(self) -> list[Listing]:
        listings: dict[str, Listing] = {}
        soup = BeautifulSoup(self.get(LIST_URL).text, "html.parser")
        for card in soup.select(".jet-listing-grid__item"):
            try:
                self._parse_card(card, listings)
            except Exception:
                continue

        # fiches détail (cache BD) : adresse, description, photos
        self._fetched = 0
        for lst in listings.values():
            card_key = hashlib.sha1(
                f"{lst.title}|{lst.price_label}|{lst.availability}"
                .encode("utf-8")).hexdigest()
            try:
                payload = self.detail(lst.external_id, card_key,
                                      lambda u=lst.url: self._fetch_detail(u))
            except Exception:
                continue
            self._apply_detail(lst, payload)
        return list(listings.values())

    # -- carte JetEngine ------------------------------------------------------
    def _parse_card(self, card, listings: dict[str, Listing]) -> None:
        ext_id = card.get("data-post-id", "")
        overlay = card.select_one(".jet-engine-listing-overlay-wrap[data-url]")
        url = overlay.get("data-url") if overlay else ""
        if not ext_id or not url or ext_id in listings:
            return

        heads = [h.get_text(" ", strip=True)
                 for h in card.select(".elementor-heading-title")
                 if h.get_text(strip=True)]
        if not heads:
            return
        title = re.sub(r"\s+", " ", heads[-1])          # le titre ferme la carte
        price_label = next((h for h in heads[:-1] if "$" in h), "")
        avail_parts = [h for h in heads[:-1] if h != price_label]
        availability = " ".join(avail_parts).strip()

        # exclusions : immeubles complets, volet commercial
        if re.search(r"complet", availability + " " + title, re.I):
            return
        if re.search(r"commercial|bureau|local|entrep[ôo]t|stationnement",
                     title, re.I):
            return

        terms = card.select_one(".jet-listing-dynamic-terms")
        city = terms.get_text(" ", strip=True) if terms else ""

        unit_type = normalize_unit_type(title)
        if not _UNIT_RE.fullmatch(unit_type or ""):
            unit_type = ""      # titre sans format d'unité (ex. « Condos locatifs »)

        images = []
        img = card.select_one("img[src]")
        if img and img["src"].startswith("http"):
            images = [img["src"]]

        listings[str(ext_id)] = Listing(
            source=self.source_id,
            external_id=str(ext_id),
            url=url,
            title=title,
            city=city,
            unit_type=unit_type,
            price=parse_price(price_label),
            price_label=price_label,
            availability=availability,
            images=images,
        )

    # -- fiche détail (Elementor) ----------------------------------------------
    def _fetch_detail(self, url: str) -> dict:
        """Adresse civique (en-tête h3), description (bloc après le h2
        « Description ») et galerie photos (liens pleine taille du carrousel)."""
        if self._fetched >= self.max_details:
            raise RuntimeError("budget de fiches détail atteint")
        self._fetched += 1
        soup = BeautifulSoup(self.get(url).text, "html.parser")
        out: dict = {}

        for h3 in soup.select("h3.elementor-heading-title"):
            txt = h3.get_text(" ", strip=True)
            if _ADDR_RE.match(txt):
                out["address"] = txt
                break

        desc_h2 = next((h for h in soup.select("h2")
                        if h.get_text(strip=True).lower() == "description"), None)
        if desc_h2:
            parts: list[str] = []
            for el in desc_h2.find_all_next(["p", "li", "h2"]):
                if el.name == "h2":
                    break
                t = re.sub(r"\s+", " ", el.get_text(" ", strip=True))
                if t and t not in parts:
                    parts.append(t)
            if parts:
                out["description"] = "\n".join(parts)[:2000]

        images: list[str] = []
        for a in soup.select(".elementor-widget-image-carousel a[href]"):
            u = a["href"]
            if u.startswith("http") and _IMG_EXT_RE.search(u) and u not in images:
                images.append(u)
        out["images"] = images[:30]
        return out

    def _apply_detail(self, lst: Listing, d: dict) -> None:
        """Reporte le payload (frais/cache) sur l'annonce."""
        if not d:
            return
        if d.get("address"):
            lst.address = d["address"]
        if d.get("description"):
            lst.description = d["description"]
        if d.get("images"):
            lst.images = list(dict.fromkeys(d["images"] + lst.images))[:30]