"""Parser for North Long Lake Storage / North Loop Self Storage (facility 008917)."""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

from src.parsers.base import BaseParser, ParseResult, UnitResult


class Facility008917Parser(BaseParser):
    """Extract storage units from North Long Lake Storage pages (Storable widget).

    The page embeds a Storable rental widget. Unit cards are rendered as
    div.card.rounded-0 elements. Each card contains:
    - h4.primary-color: unit name with dimension (e.g. "5 x 10 with swing door (5 x 10)")
    - strong.price span: monthly rate (e.g. "$75")
    - a.btn: rental/waitlist CTA
    """

    platform = "custom_facility_008917"

    _DIM_RE = re.compile(r"\((\d+\.?\d*)\s*[xX×]\s*(\d+\.?\d*)\)")
    _DIM_BARE_RE = re.compile(r"(\d+\.?\d*)\s*[xX×]\s*(\d+\.?\d*)")

    def parse(self, html: str, url: str = "") -> ParseResult:
        soup = BeautifulSoup(html, "lxml")
        result = ParseResult(platform=self.platform, parser_name=self.__class__.__name__)

        seen: set[tuple[str, str]] = set()

        for card in soup.find_all(class_=lambda x: x and "card" in x and "rounded-0" in x if x else False):
            # --- name / size ---
            heading = card.find("h4")
            if not heading:
                continue
            name_text = heading.get_text(strip=True)

            # Extract dimension from parenthetical e.g. "(5 x 10)" or first match
            m = self._DIM_RE.search(name_text)
            if not m:
                m = self._DIM_BARE_RE.search(name_text)
            if m:
                size_text = f"{m.group(1)} x {m.group(2)}"
            else:
                size_text = name_text

            # --- price ---
            price_el = card.find(class_="price")
            if not price_el:
                continue
            # The price span contains "$75\n/ month" — extract just the dollar amount
            price_span = price_el.find("span")
            raw_price_text = price_span.get_text(strip=True) if price_span else price_el.get_text(strip=True)
            # Grab only the "$NNN" portion before any whitespace/"month" text
            price_m = re.search(r"\$\s*([\d,]+(?:\.\d+)?)", raw_price_text)
            if not price_m:
                continue
            price_text = f"${price_m.group(1)}"
            price = self.normalize_price(price_text)
            if price is None:
                continue

            key = (size_text, price_text)
            if key in seen:
                continue
            seen.add(key)

            unit = UnitResult(
                size=size_text,
                price=price,
                description=name_text if name_text != size_text else None,
            )
            w, ln, sq = self.normalize_size(size_text)
            if w is not None:
                unit.metadata = {"width": w, "length": ln, "sqft": sq}
            result.units.append(unit)

        if not result.units:
            result.warnings.append("No units found in North Long Lake card elements")

        return result