# -----------------------------------------------------------------------------
# Lou-Ka — Agrégateur de logements à louer (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/jcl.py : connecteur petites annonces Groupe JCL
#   (annonces.groupejcl.com) — journal local des Basses-Laurentides / Rive-Nord
#   (Sainte-Thérèse, Saint-Jérôme, Saint-Eustache et environs). WordPress +
#   plugin « Classified Listing » (rtcl) rendu serveur : on ne moissonne QUE la
#   catégorie immobilier/logements-a-louer (cartes .rtcl-listing-item : titre,
#   prix, localité, lien). La page détail (.rtin-des / .rtin-location) fournit
#   description, adresse et photos. Granularité : annonce (unité).
#   ⚠️ Portail de petites annonces de particuliers : certaines annonces restent
#   affichées longtemps — Lou-Ka les reprend telles qu'affichées par la source.
# -----------------------------------------------------------------------------
from __future__ import annotations

import re

from bs4 import BeautifulSoup

from ..schema import Listing
from .base import BaseConnector

BASE = "https://annonces.groupejcl.com"
CAT_URL = f"{BASE}/categories/immobilier/logements-a-louer/"
MAX_PAGES = 10

IMG_RE = re.compile(
    r"https://annonces\.groupejcl\.com/wp-content/uploads/classified-listing/"
    r"[^\"'\s\\]+\.(?:jpe?g|png|webp)", re.I)
THUMB_RE = re.compile(r"-\d{2,4}x\d{2,4}(?=\.(?:jpe?g|png|webp)$)", re.I)
PRICE_RE = re.compile(r"(\d[\d\s,.]*)\s*\$?")
REGION_RE = re.compile(r"\s+et\s+environs\s*$", re.I)


class JclConnector(BaseConnector):
    source_id = "jcl"
    request_delay = 0.8
    max_images = 10

    # -- page détail (cachée en BD via self.detail) ---------------------------
    def _detail(self, url: str) -> dict:
        try:
            html = self.get(url).text
        except Exception:
            return {}
        soup = BeautifulSoup(html, "html.parser")
        payload: dict = {}

        el = soup.select_one(".rtin-content-area .rtin-content")
        if el is not None:
            payload["description"] = re.sub(
                r"\s+", " ", el.get_text(" ", strip=True))[:1500]

        # « 367 rue leonard | Saint-Jérôme et environs, Québec »
        for el in soup.select(".rtin-location"):
            parts = [p.strip() for p in
                     el.get_text("|", strip=True).split("|") if p.strip()]
            street = [p for p in parts
                      if not re.search(r"et environs|Québec$", p, re.I)]
            if street:
                payload["address"] = street[0]
                break

        imgs = []
        for u in dict.fromkeys(IMG_RE.findall(html)):
            full = THUMB_RE.sub("", u)
            if full not in imgs:
                imgs.append(full)
        payload["images"] = imgs[: self.max_images]
        return payload

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

        for page in range(1, MAX_PAGES + 1):
            url = CAT_URL if page == 1 else f"{CAT_URL}page/{page}/"
            try:
                html = self.get(url).text
            except Exception:
                break
            soup = BeautifulSoup(html, "html.parser")
            items = soup.select(".rtcl-listing-item")
            if not items:
                break

            new_on_page = 0
            for it in items:
                a = it.select_one(".rtin-title a[href]")
                if a is None:
                    continue
                href = a["href"].split("?")[0]
                slug = href.rstrip("/").rsplit("/", 1)[-1]
                if not slug or slug in seen:
                    continue
                seen.add(slug)
                new_on_page += 1

                title = re.sub(r"\s+", " ", a.get_text(" ", strip=True))

                price = None
                price_label = ""
                el = it.select_one(".rtcl-price-amount")
                if el is not None:
                    price_label = re.sub(r"\s+", " ",
                                         el.get_text(" ", strip=True))
                    m = PRICE_RE.search(price_label.replace(",", ""))
                    if m:
                        try:
                            val = float(m.group(1).replace(" ", "")
                                        .replace(" ", ""))
                            if 300 <= val <= 20000:
                                price = val
                        except ValueError:
                            pass

                # localité : lien /listing-location/quebec/<ville>/
                city = ""
                loc = it.select_one('a[href*="/listing-location/"]')
                if loc is not None:
                    city = REGION_RE.sub(
                        "", loc.get_text(" ", strip=True)).strip()

                key = f"{title}|{price_label}|{city}"
                det = self.detail(slug, key, lambda u=href: self._detail(u))

                listings.append(Listing(
                    source=self.source_id,
                    external_id=slug,
                    url=href,
                    title=title,
                    address=det.get("address", ""),
                    city=city,
                    price=price,
                    price_label=price_label,
                    description=det.get("description", ""),
                    images=list(det.get("images") or []),
                ))
            if new_on_page == 0:
                break
        return listings