# -----------------------------------------------------------------------------
# Auto-Ka — Agrégateur de voitures usagées à vendre (province de Québec)
# Auteur : Simon-Pierre Boucher — contact@spboucher.ai
# connectors/ototr.py : OTO Trois-Rivières — WordPress, thème « Motors »
#   (stm_vehicles_listing). Pas d'API REST exposée ni de JSON-LD.
#
#   Stratégie : le sitemap natif WP `/wp-sitemap-posts-listings-1.xml` recense
#   les annonces (/listings/<slug>/). Chaque page détail est parsée en HTML
#   (BeautifulSoup) : h1 = titre, .single-regular-price = prix, tableau
#   .single-car-data = fiche technique (Odomètre, Carburant, Transmission,
#   Propulsion, Carrosserie, Moteur, Couleurs…). Pages détail en cache BD
#   (clé hebdomadaire).
# -----------------------------------------------------------------------------
from __future__ import annotations

import datetime
import os
import re

from bs4 import BeautifulSoup

from ..schema import Vehicle
from .base import BaseConnector

_EXCLUDE_RE = re.compile(
    r"motoneige|vtt|bateau|ponton|motomarine|\bmoto\b|roulotte|remorque|"
    r"campeur|tracteur|souffleuse|scooter|spyder|atv", re.I)

# vignettes WordPress (-350x205 etc.) à écarter de la galerie
_THUMB_RE = re.compile(r"-\d{2,4}x\d{2,4}\.(?:jpe?g|png|webp)$", re.I)

# libellés de la fiche technique -> champ Vehicle
_SPEC_FIELDS = {
    "carrosserie": "body_type",
    "carburant": "fuel",
    "transmission": "transmission",
    "propulsion": "drivetrain",
    "moteur": "engine",
    "couleur exterieure": "exterior_color",
    "couleur interieure": "interior_color",
}


def _flat(text: str) -> str:
    from ..schema import strip_accents
    return strip_accents(" ".join(text.split())).strip().lower()


class OtoTroisRivieres(BaseConnector):
    source_id = "ototroisrivieres"
    base_url = "https://ototroisrivieres.com"
    dealer_name = "OTO Trois-Rivières"
    city = "Trois-Rivières"
    request_delay: float = 1.0
    max_details = 200

    # -- sitemap ----------------------------------------------------------------
    def _vehicle_urls(self) -> list[str]:
        xml = self.get(f"{self.base_url}/wp-sitemap-posts-listings-1.xml").text
        return [loc.strip() for loc in re.findall(r"<loc>([^<]+)</loc>", xml)
                if "/listings/" in loc]

    # -- page détail -> payload dict ---------------------------------------------
    def _fetch_detail(self, url: str) -> dict:
        soup = BeautifulSoup(self.get(url).text, "html.parser")
        payload: dict = {"specs": {}}

        h1 = soup.find("h1")
        if h1:
            payload["title"] = " ".join(h1.get_text(" ", strip=True).split())

        price_el = soup.select_one(".single-regular-price")
        if price_el:
            payload["price_label"] = price_el.get_text(" ", strip=True)

        for row in soup.select(".single-car-data tr"):
            label = row.select_one(".t-label")
            value = row.select_one(".t-value")
            if label and value:
                payload["specs"][label.get_text(" ", strip=True)] = \
                    value.get_text(" ", strip=True)

        og_desc = soup.find("meta", attrs={"property": "og:description"})
        if og_desc and og_desc.get("content"):
            payload["description"] = og_desc["content"]

        seen: set[str] = set()
        images = []
        candidates = [img.get("src") or img.get("data-src") or ""
                      for img in soup.find_all("img")]
        candidates += [a.get("href") or "" for a in soup.find_all("a")]
        for src in candidates:
            if "/wp-content/uploads/" not in src or _THUMB_RE.search(src) \
                    or not re.search(r"\.(?:jpe?g|png|webp)$", src, re.I):
                continue
            if src not in seen:
                seen.add(src)
                images.append(src)
        payload["images"] = images[:20]
        return payload

    # -- contrat -----------------------------------------------------------------
    def fetch(self) -> list[Vehicle]:
        urls = self._vehicle_urls()
        week = datetime.date.today().isocalendar()
        cache_key = f"v1:{week.year}w{week.week}"
        cap = int(os.environ.get("AUTOKA_MAX_DETAILS", self.max_details))

        vehicles: list[Vehicle] = []
        real_fetches = 0
        for url in urls:
            ext_id = url.rstrip("/").rsplit("/", 1)[-1]

            def _fetch(u=url):
                return self._fetch_detail(u)

            if real_fetches >= cap:            # plafond : cache seulement
                from .. import db
                if self._detail_con is None:
                    self._detail_con = db.connect()
                data = db.get_cached_detail(self._detail_con, self.source_id,
                                            ext_id, cache_key)
                if data is None:
                    continue
            else:
                before = self._last_request
                data = self.detail(ext_id, cache_key, _fetch)
                if self._last_request != before:
                    real_fetches += 1
            if not data or not data.get("title"):
                continue
            veh = self._to_vehicle(ext_id, url, data)
            if veh is not None:
                vehicles.append(veh)
        return vehicles

    def _to_vehicle(self, ext_id: str, url: str, d: dict) -> Vehicle | None:
        title = d.get("title", "")
        if not title or _EXCLUDE_RE.search(title):
            return None
        if re.search(r"vendu", title, re.I):   # annonce conservée mais vendue
            return None
        title = re.sub(r"\*+\s*", " ", title).strip()

        fields = {"body_type": "", "fuel": "", "transmission": "",
                  "drivetrain": "", "engine": "", "exterior_color": "",
                  "interior_color": ""}
        mileage_label = ""
        details: dict = {}
        for label, value in (d.get("specs") or {}).items():
            key = _flat(label)
            if key.startswith("odom"):
                # le thème affiche « 10500 KMkm » — garder tel quel,
                # parse_mileage s'en accommode
                mileage_label = value
            elif key in _SPEC_FIELDS:
                fields[_SPEC_FIELDS[key]] = value
            else:
                details[label] = value

        return Vehicle(
            source=self.source_id,
            external_id=ext_id,
            url=url,
            title=title.title() if title.isupper() else title,
            price_label=d.get("price_label", ""),
            mileage_label=mileage_label,
            dealer_name=self.dealer_name,
            city=self.city,
            description=" ".join(str(d.get("description") or "").split())[:4000],
            details=details,
            images=d.get("images") or [],
            **fields,
        )