# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/location_sb.py : connecteur Location Résidentielle SB
#   (locationresidentielle.com — jumelés, appartements et maisons à louer à
#    Saint-Georges, Beauce). WordPress thème RealHomes rendu serveur : la page
#    /a-louer/ liste des cartes <article class="property-item"> (titre,
#    extrait, vignette, lien /property/<slug>/). Petit inventaire (2-3 fiches
#    actives) ; on écarte les cartes « liste d'attente » et « NON
#    DISPONIBLE ». La page détail (via cache BD) fournit adresse, chambres,
#    salles de bain, prix ($/mois — souvent « communiquez avec nous »),
#    disponibilité, caractéristiques (<li>) et photos.
# -----------------------------------------------------------------------------
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://locationresidentielle.com"
LIST_URL = f"{BASE}/a-louer/"

CITY = "Saint-Georges"

SLUG_RE = re.compile(r"/property/([^/?#]+)")
# cartes qui ne sont pas de vraies annonces
SKIP_RE = re.compile(r"liste\s+d.attente|inscription|non\s+disponible", re.I)
ADDR_RE = re.compile(r"(\d{3,6},?\s+\d{1,3}e\s+(?:Rue|Avenue),?\s+"
                     r"(?:Saint|St)-Georges(?:,?\s+G\d[A-Z]\s*\d[A-Z]\d)?)",
                     re.I)
BED_RE = re.compile(r"(\d+)\s*(?:grandes?\s*)?chambres", re.I)
BATH_RE = re.compile(r"(\d+)\s*salles? de bain", re.I)
PRICE_RE = re.compile(r"(\d[\d\s]{2,6})\s*\$\s*/?\s*mois", re.I)
DISPO_RE = re.compile(r"(?:Disponibilit[ée]\s*:\s*([^\n]{2,60})|"
                      r"((?:CESSION DE BAIL\s+)?DISPONIBLE EN [A-ZÉÛ]+))")
TYPE_RE = re.compile(r"\d\s*1/2")
UNIT_TEL_RE = re.compile(r"\b(\d{3})[\s.\-](\d{3})[\s.\-](\d{4})\b")
IMG_RE = re.compile(r'(?:src|href)="(https://locationresidentielle\.com/'
                    r'wp-content/uploads/[^"]+\.(?:jpe?g|png|webp))"', re.I)
IMG_SKIP_RE = re.compile(r"logo|signature|carte|a_louer|"
                         r"jumeles-appartements|-\d{2,3}x\d{2,3}\.", re.I)


class LocationSbConnector(BaseConnector):
    source_id = "location_sb"
    request_delay = 0.6

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

        for item in soup.select("article.property-item"):
            try:
                a = item.select_one("h4 a[href]")
                if not a:
                    continue
                title = re.sub(r"[•\s]+", " ", a.get_text(strip=True)).strip()
                if not title or SKIP_RE.search(title):
                    continue
                m = SLUG_RE.search(a["href"])
                if not m:
                    continue
                slug = m.group(1)
                url = f"{BASE}/property/{slug}/"

                excerpt = ""
                p = item.find("p")
                if p:
                    excerpt = re.sub(
                        r"\s+", " ",
                        p.get_text(" ", strip=True)).replace(
                        "Plus d'informations", "").strip(" …-")

                key = hashlib.sha1(
                    f"{url}|{title}|{excerpt}".encode()).hexdigest()
                try:
                    d = self.detail(slug, key, lambda u=url: self._detail(u))
                except Exception:
                    d = {}

                tm = TYPE_RE.search(title)
                unit_type = normalize_unit_type(tm.group(0)) if tm else ""

                listings.append(Listing(
                    source=self.source_id,
                    external_id=slug,
                    url=url,
                    title=title,
                    address=d.get("address")
                    or f"{title}, Saint-Georges",
                    city=CITY,
                    unit_type=unit_type,
                    bedrooms=d.get("bedrooms"),
                    bathrooms=d.get("bathrooms"),
                    price=d.get("price"),
                    price_label=d.get("price_label", ""),
                    availability=d.get("availability", ""),
                    description=d.get("description") or excerpt,
                    amenities=d.get("amenities") or [],
                    details=d.get("details") or {},
                    images=d.get("images") or [],
                ))
            except Exception:
                continue
        return listings

    # -- page détail /property/<slug>/ (thème RealHomes) ------------------------
    def _detail(self, url: str) -> dict:
        html = self.get(url).text
        soup = BeautifulSoup(html, "html.parser")
        art = soup.find("article") or soup
        txt = re.sub(r"[ \t]+", " ", art.get_text("\n", strip=True))
        flat = re.sub(r"\s+", " ", txt)
        out: dict = {}

        m = ADDR_RE.search(flat)
        if m:
            addr = re.sub(r"\s+", " ", m.group(1)).strip()
            if not re.search(r"saint-georges", addr, re.I):
                addr = re.sub(r"St-Georges", "Saint-Georges", addr, flags=re.I)
            out["address"] = addr
        m = BED_RE.search(flat)
        if m:
            out["bedrooms"] = float(m.group(1))
        m = BATH_RE.search(flat)
        if m:
            out["bathrooms"] = float(m.group(1))
        m = PRICE_RE.search(flat)
        if m:
            out["price"] = parse_price(m.group(0))
            out["price_label"] = f"{m.group(1).strip()} $/mois"
        elif re.search(r"Prix\s*:\s*communiquez", flat, re.I):
            out["price_label"] = "Prix : communiquez avec nous"
        m = DISPO_RE.search(txt)
        if m:
            out["availability"] = (m.group(1) or m.group(2) or "") \
                .strip().capitalize()

        # caractéristiques / inclusions : listes <li> du contenu
        amenities: list[str] = []
        for li in art.find_all("li"):
            t = re.sub(r"\s+", " ", li.get_text(" ", strip=True))
            t = re.sub(r"^[^\w]+|►.*$", "", t).strip()
            if (t and 3 <= len(t) < 80 and "ICI" not in t
                    and not re.search(r"estimez|inscrivez", t, re.I)
                    and t not in amenities):
                amenities.append(t)
        out["amenities"] = amenities[:20]

        details: dict = {}
        pm = UNIT_TEL_RE.search(flat)
        if pm:
            details["contact"] = {
                "phone": f"{pm.group(1)}-{pm.group(2)}-{pm.group(3)}"}
        if details:
            out["details"] = details

        out["images"] = [u for u in dict.fromkeys(IMG_RE.findall(html))
                         if not IMG_SKIP_RE.search(u)][:15]
        return out