# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/logispro_mauricie.py : connecteur LogisPro Mauricie (logispro.ca)
#   Trois-Rivières, Cap-de-la-Madeleine, Shawinigan, Bécancour, Nicolet.
#   ⚠️ Compagnie DISTINCTE de LogisPro Rimouski (logisprorimouski.com,
#   connecteur `logispro`). WordPress thème maison rendu serveur :
#   /appartements-a-louer/ = cartes (prix, adresse, typologie, lien
#   /apartments/<slug>/) ; page détail = liste « Secteur / Prix / Grandeur /
#   Étage / … », description riche et galerie fancybox.
#   Granularité : une annonce par logement disponible.
# -----------------------------------------------------------------------------
from __future__ import annotations

import hashlib
import re

from bs4 import BeautifulSoup

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

BASE = "https://logispro.ca"
LIST_URL = f"{BASE}/appartements-a-louer/"

PRICE_RE = re.compile(r"([\d\s ,]+)\s*\$")
# villes desservies, pour extraire la ville depuis l'adresse de la carte
CITY_RE = re.compile(
    r"(Trois-Rivi[èe]res|Shawinigan|B[ée]cancour|Nicolet|"
    r"Cap-de-la-Madeleine|Gentilly|Saint-Maurice)", re.I)


class LogisProMauricieConnector(BaseConnector):
    source_id = "logispro_mauricie"
    request_delay = 0.7

    def fetch(self) -> list[Listing]:
        listings: list[Listing] = []
        try:
            html = self.get(LIST_URL).text
        except Exception:
            return listings
        soup = BeautifulSoup(html, "html.parser")

        seen: set[str] = set()
        for card in soup.select("div.apartments-item"):
            try:
                a = card.select_one('a[href*="/apartments/"]')
                if a is None:
                    continue
                url = a.get("href", "").split("?")[0]
                slug = url.rstrip("/").split("/")[-1]
                if not slug or slug in seen:
                    continue
                seen.add(slug)

                price = None
                price_label = ""
                price_el = card.select_one("p.price")
                if price_el is not None:
                    price_label = re.sub(r"\s+", " ",
                                         price_el.get_text(" ", strip=True))
                    m = PRICE_RE.search(price_label)
                    if m:
                        try:
                            price = float(m.group(1).replace(" ", "")
                                          .replace(" ", "").replace(",", ""))
                        except ValueError:
                            pass

                addr_el = card.select_one("p.adress, p.address")
                address = (re.sub(r"\s+", " ", addr_el.get_text(" ", strip=True))
                           if addr_el else "")
                room_el = card.select_one("p.room")
                unit_type = normalize_unit_type(
                    room_el.get_text(" ", strip=True) if room_el else "")

                city = ""
                m = CITY_RE.search(address)
                if m:
                    city = m.group(1)
                    if city.lower() in ("cap-de-la-madeleine", "gentilly"):
                        city = ("Trois-Rivières"
                                if "cap" in m.group(1).lower() else "Bécancour")

                card_key = hashlib.sha1(
                    f"{price_label}|{address}|{unit_type}"
                    .encode("utf-8")).hexdigest()
                d = self.detail(slug, card_key,
                                lambda url=url: self._fetch_detail(url))

                listings.append(Listing(
                    source=self.source_id,
                    external_id=slug,
                    url=url,
                    title=address or slug,
                    address=d.get("address") or address,
                    sector=d.get("sector", ""),
                    city=city or d.get("sector", ""),
                    unit_type=unit_type or d.get("unit_type", ""),
                    price=price,
                    price_label=price_label,
                    availability=d.get("availability", ""),
                    description=d.get("description", ""),
                    amenities=d.get("amenities") or [],
                    images=d.get("images") or [],
                ))
            except Exception:
                continue
        return listings

    def _fetch_detail(self, url: str) -> dict:
        """Page détail : liste top_descript (Secteur/Prix/Grandeur/Étage/…),
        description riche (dispo « dès maintenant » incluse) et galerie."""
        out: dict = {}
        try:
            html = self.get(url).text
        except Exception:
            return out
        soup = BeautifulSoup(html, "html.parser")

        amenities: list[str] = []
        for li in soup.select("ul.top_descript li"):
            txt = re.sub(r"\s+", " ", li.get_text(" ", strip=True))
            if not txt or ":" not in txt:
                continue
            label, _, value = txt.partition(":")
            label, value = label.strip(), value.strip()
            if not value:
                continue
            low = label.lower()
            if low == "secteur":
                out["sector"] = value
            elif low == "prix":
                pass  # déjà sur la carte
            elif low == "grandeur":
                out["unit_type"] = normalize_unit_type(value)
            else:
                amenities.append(f"{label} : {value}")
        out["amenities"] = amenities[:20]

        desc_el = soup.select_one("div.description")
        if desc_el is not None:
            desc = desc_el.get_text("\n", strip=True)
            out["description"] = re.sub(r"\n{2,}", "\n", desc).strip()[:1200]
            m = re.search(r"Disponible[^\n•]*", desc, re.I)
            if m:
                out["availability"] = m.group(0).strip()
            # adresse complète avec code postal (dernier paragraphe)
            m = re.search(r"\d[\w#\-]*[^\n]*,\s*(?:QC|Qu[ée]bec)[^\n]*", desc)
            if m:
                out["address"] = m.group(0).replace(", Canada", "").strip()

        out["images"] = [
            a.get("href", "").strip()
            for a in soup.select(".apartment__gallery a.fancybox[href]")
            if a.get("href", "").strip()][:25]
        return out