# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/loggia.py : connecteur Loggia Saint-Lambert (loggiasaintlambert.com)
#   Trois immeubles locatifs (Loggia 1, 2, 3) sur l'avenue Saint-Charles à
#   Saint-Lambert. WordPress rendu serveur : la page /le-projet/ embarque une
#   carte interactive (blocs « div.available.bloc.appt-N-demi » libellés
#   « 111-$2495. - 4 ½ / Disponible ») et ~260 fiches d'unités (« Appartement
#   NNN / Loggia N / ÉTAGE N », p.dispo[.not] = Loué/Disponible/Réservé,
#   p.superficie, plan PNG/PDF). On ne retient que les unités « Disponible ».
#   Granularité : unité.
# -----------------------------------------------------------------------------
from __future__ import annotations

import re

from bs4 import BeautifulSoup

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

BASE = "https://www.loggiasaintlambert.com"
PAGE_URL = f"{BASE}/le-projet/"

CITY = "Saint-Lambert"
ADDRESS = "avenue Saint-Charles, Saint-Lambert"

# « Appartement 111-$2495. » (prix parfois accolé au numéro)
TITLE_RE = re.compile(r"Appartement\s+(\d+)\s*(?:[-–]\s*\$\s*([\d\s,  ]+))?",
                      re.I)
# libellé de bloc carte : « 111-$2495. - 4 ½ »
BLOC_RE = re.compile(r"(\d+)\s*[-–]\s*\$\s*([\d\s,  ]+)[.\s]*[-–]\s*"
                     r"(\d\s*½|\d\s*1/2|studio)", re.I)
SQFT_RE = re.compile(r"([\d,\s  ]+)\s*pi", re.I)
LOGGIA_RE = re.compile(r"Loggia\s*(\d)", re.I)
ETAGE_RE = re.compile(r"[ÉE]TAGE\s*(\d+)", re.I)


def _num(s: str) -> float | None:
    s = re.sub(r"[\s,  ]", "", s or "")
    try:
        return float(s)
    except ValueError:
        return None


class LoggiaConnector(BaseConnector):
    source_id = "loggia"
    request_delay = 0.6

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

        # 1) typologie par unité via les blocs de la carte (disponibles seulement)
        #    « 111-$2495. - 4 ½ » → {("<no>", "<prix>"): "4½"}
        types: dict[str, str] = {}
        for bloc in soup.select("div.available.bloc"):
            m = BLOC_RE.search(bloc.get_text(" ", strip=True))
            if m:
                types[m.group(1)] = normalize_unit_type(m.group(3))

        # 2) fiches d'unités : on garde celles dont p.dispo = « Disponible »
        for p in soup.select("p.dispo"):
            try:
                if "not" in (p.get("class") or []):
                    continue
                if "disponible" not in p.get_text(strip=True).lower():
                    continue          # « Réservé » etc.
                card = p.find_parent("div", class_="col-md-8")
                if not card:
                    continue
                text = card.get_text(" ", strip=True)
                mt = TITLE_RE.search(text)
                if not mt:
                    continue
                unit_no = mt.group(1)
                price = _num(mt.group(2)) if mt.group(2) else None

                building = ""
                mb = LOGGIA_RE.search(text)
                if mb:
                    building = f"Loggia {mb.group(1)}"
                floor = ""
                mf = ETAGE_RE.search(text)
                if mf:
                    floor = f"Étage {mf.group(1)}"

                sqft = None
                sp = card.select_one("p.superficie")
                if sp:
                    ms = SQFT_RE.search(sp.get_text(strip=True))
                    if ms:
                        sqft = _num(ms.group(1))

                images: list[str] = []
                for img in card.select("div.plan-unite img"):
                    src = img.get("src") or ""
                    if src.startswith("http"):
                        images.append(src)
                details: dict = {}
                a = card.select_one('a[href*=".pdf"]')
                if a and a.get("href"):
                    mp = re.search(r"(https://[^\s\"]+\.pdf)", a["href"])
                    details["plan_pdf"] = mp.group(1) if mp else a["href"]
                if building:
                    details["building"] = building
                if floor:
                    details["floor"] = floor

                unit_type = types.get(unit_no, "")
                bslug = building.lower().replace(" ", "") or "x"
                ext_id = f"{bslug}-{unit_no}"
                if any(l.external_id == ext_id for l in listings):
                    continue
                listings.append(Listing(
                    source=self.source_id,
                    external_id=ext_id,
                    url=PAGE_URL,
                    title=f"Appartement {unit_no}"
                          + (f" ({unit_type})" if unit_type else "")
                          + (f" — {building}, Loggia Saint-Lambert" if building
                             else " — Loggia Saint-Lambert"),
                    address=ADDRESS,
                    city=CITY,
                    unit_type=unit_type,
                    price=price,
                    availability="Disponible",
                    area_sqft=sqft,
                    details=details,
                    images=images[:15],
                ))
            except Exception:
                continue
        return listings