# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/vestric.py : connecteur Vestric — Condos locatifs (vestric.ca)
#   Projet unique à Candiac (Montérégie). WordPress (thème Vortex « vy_ »)
#   rendu serveur : la page /les-unites liste les unités disponibles
#   (li.vy_units_listing_item : no d'unité, modèle, étage, type, superficie,
#   disponibilité + lien /units/unite-<no>). Granularité : unité. Prix non
#   publiés (formulaire de contact avec fourchettes de budget seulement).
# -----------------------------------------------------------------------------
from __future__ import annotations

import re

from bs4 import BeautifulSoup

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

BASE = "https://vestric.ca"
LIST_URL = f"{BASE}/les-unites"

CITY = "Candiac"

IMG_RE = re.compile(r'https://vestric\.ca/app/uploads/'
                    r'[^"\s\\)]+\.(?:jpe?g|png|webp)', re.I)
PDF_RE = re.compile(r'https://vestric\.ca/app/uploads/[^"\s\\)]+\.pdf', re.I)
SQFT_RE = re.compile(r"(\d{3,4})\s*pi", re.I)


class VestricConnector(BaseConnector):
    source_id = "vestric"
    request_delay = 0.6

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

        for item in soup.select("li.vy_units_listing_item"):
            try:
                def cell(cls: str) -> str:
                    el = item.select_one(f"div.vy_units_listing_{cls} span")
                    return el.get_text(" ", strip=True) if el else ""

                unit_no = cell("unit")
                if not unit_no:
                    continue
                model = cell("model").replace("Modèle", "").strip()
                floor = cell("floor")
                unit_type = normalize_unit_type(cell("type"))
                sqft = None
                m = SQFT_RE.search(cell("square"))
                if m:
                    sqft = float(m.group(1))
                availability = cell("availability")

                a = item.select_one('a[href*="/units/"]')
                url = (a.get("href", "").split("#")[0] if a
                       else f"{BASE}/units/unite-{unit_no}")

                details: dict = {}
                if model:
                    details["model"] = model
                if floor:
                    details["floor"] = floor

                # page détail (cache BD) : photos de l'unité
                key = f"{unit_type}|{sqft}|{availability}|{model}"
                d = self.detail(f"unite-{unit_no}", key,
                                lambda url=url: self._fetch_detail(url))

                listings.append(Listing(
                    source=self.source_id,
                    external_id=f"unite-{unit_no}",
                    url=url,
                    title=f"Unité {unit_no} ({unit_type}) — Vestric",
                    address="200, avenue des Chênes, Candiac",
                    city=CITY,
                    unit_type=unit_type,
                    availability=availability,
                    area_sqft=sqft,
                    details={**details, **({"plan_pdf": d["plan_pdf"]} if d.get("plan_pdf") else {})},
                    images=d.get("images") or [],
                ))
            except Exception:
                continue
        return listings

    def _fetch_detail(self, url: str) -> dict:
        out: dict = {}
        try:
            html = self.get(url).text
        except Exception:
            return out
        out["images"] = [u for u in dict.fromkeys(IMG_RE.findall(html))
                         if not re.search(r"logo|icon|favicon|og-image|"
                                          r"decoration|bloc-vestric|"
                                          r"-\d{2,3}x\d{2,3}\.", u, re.I)][:15]
        pdfs = PDF_RE.findall(html)
        if pdfs:
            out["plan_pdf"] = pdfs[0]
        return out