# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/signature_boisbriand.py : connecteur Signature Boisbriand
#   (signatureboisbriand.com) — projet locatif au 1040-1050, rue des
#   Francs-Bourgeois, Boisbriand (Faubourg Boisbriand, ~266 unités).
#   WordPress + FacetWP rendu serveur : /unites-disponibles/ liste des lignes
#   .unit-row (numéro #1040-101, pièces, étage, superficie, prix « à partir
#   de », disponibilité, plan PDF) paginées via le paramètre ?_paged=N
#   (15 lignes/page). On ne garde que les unités NON « Unité louée ».
#   Granularité : unité.
# -----------------------------------------------------------------------------
from __future__ import annotations

import re

from bs4 import BeautifulSoup

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

BASE = "https://signatureboisbriand.com"
UNITS_URL = f"{BASE}/unites-disponibles/"
MAX_PAGES = 30

ADDRESS = "1040, rue des Francs-Bourgeois, Boisbriand"
CITY = "Boisbriand"

PRICE_RE = re.compile(r"(\d[\d\s,]*)\s*\$")
IMG_RE = re.compile(
    r"https://signatureboisbriand\.com/wp-content/uploads/"
    r"[^\"'\s\\]+\.(?:jpe?g|png|webp)", re.I)
SKIP_IMG_RE = re.compile(r"logo|favicon|icon|-\d{2,4}x\d{2,4}\.", re.I)


class SignatureBoisbriandConnector(BaseConnector):
    source_id = "signature_boisbriand"
    request_delay = 0.6
    max_images = 10

    def _site_images(self) -> list[str]:
        """Photos du projet (page d'accueil) — les lignes n'ont qu'un plan PDF."""
        try:
            html = self.get(f"{BASE}/").text
        except Exception:
            return []
        return [u for u in dict.fromkeys(IMG_RE.findall(html))
                if not SKIP_IMG_RE.search(u)][: self.max_images]

    @staticmethod
    def _txt(row, sel: str) -> str:
        el = row.select_one(sel)
        return re.sub(r"\s+", " ", el.get_text(" ", strip=True)) if el else ""

    def fetch(self) -> list[Listing]:
        images = self._site_images()
        listings: list[Listing] = []
        seen: set[str] = set()

        for page in range(1, MAX_PAGES + 1):
            url = UNITS_URL if page == 1 else f"{UNITS_URL}?_paged={page}"
            try:
                html = self.get(url).text
            except Exception:
                break
            soup = BeautifulSoup(html, "html.parser")
            rows = soup.select(".unit-row")
            if not rows:
                break

            new_on_page = 0
            for row in rows:
                num = self._txt(row, ".unit-number__text").lstrip("#").strip()
                if not num or num in seen:
                    continue
                seen.add(num)
                new_on_page += 1

                availability = self._txt(row, ".unit-availability__text")
                if re.search(r"lou[ée]", availability, re.I):
                    continue                       # « Unité louée »

                unit_type = normalize_unit_type(
                    self._txt(row, ".unit-room__text"))
                floor = self._txt(row, ".unit-floor__text")

                area = None
                m = re.search(r"(\d[\d\s]*)\s*pi",
                              self._txt(row, ".unit-area__text"))
                if m:
                    try:
                        v = float(m.group(1).replace(" ", "").replace(" ", ""))
                        if 100 <= v <= 10000:
                            area = v
                    except ValueError:
                        pass

                price = None
                price_label = ""
                m = PRICE_RE.search(self._txt(row, ".unit-price__text"))
                if m:
                    try:
                        val = float(m.group(1).replace(" ", "")
                                    .replace(" ", "").replace(",", ""))
                        if 300 <= val <= 20000:
                            price = val
                            price_label = f"à partir de {int(val)}$"
                    except ValueError:
                        pass

                imgs = list(images)
                plan = row.select_one('.unit-plan a[href$=".pdf"]')
                details: dict = {}
                if plan is not None:
                    details["floor_plan"] = plan["href"]

                listings.append(Listing(
                    source=self.source_id,
                    external_id=num,
                    url=UNITS_URL,
                    title=f"Signature Boisbriand — Unité #{num}",
                    address=ADDRESS,
                    city=CITY,
                    unit_type=unit_type,
                    price=price,
                    price_label=price_label,
                    availability=availability or "Disponible",
                    area_sqft=area,
                    description=f"Unité #{num}, {floor}" if floor else "",
                    details=details,
                    images=imgs,
                ))
            if new_on_page == 0:
                break
        return listings