"""Parser for Zoom Self Storage (facility 004393).

Site redirects to Facebook (facebook.com/zoomselfstorage/). No structured
pricing data is available. This parser detects the Facebook redirect and
reports the situation clearly so the pipeline can classify the facility
as ``no_pricing``.
"""

from __future__ import annotations

from bs4 import BeautifulSoup

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


class Facility004393Parser(BaseParser):
    """Zoom Self Storage -- site redirects to Facebook, no pricing available."""

    platform = "custom_facility_004393"

    def _is_facebook_page(self, soup: BeautifulSoup) -> bool:
        """Return True if the HTML belongs to a Facebook page."""
        title = soup.find("title")
        if title and "facebook" in title.get_text(strip=True).lower():
            return True
        # Check for Facebook-specific meta / structural markers
        for meta in soup.find_all("meta", attrs={"property": True}):
            prop = (meta.get("property") or "").lower()
            if prop.startswith("fb:") or prop.startswith("al:"):
                return True
        if soup.find("div", id="facebook") or soup.find("div", id="fb-root"):
            return True
        return False

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

        if self._is_facebook_page(soup):
            result.warnings.append(
                "Site (zoomselfstore.com) redirects to Facebook. "
                "No pricing data available. Consider marking facility as no_pricing."
            )
            return result

        # If the page is ever restored to a real storage site, attempt basic
        # extraction so we surface data instead of silently returning nothing.
        for tag in soup.find_all(["script", "style"]):
            tag.decompose()

        body_text = soup.get_text(separator="\n")
        seen: set[tuple[str, str]] = set()

        import re

        size_price_re = re.compile(
            r"(\d+\s*['\u2032]?\s*[xX\u00d7]\s*\d+\s*['\u2032]?)"
            r"[^\$]{0,120}"
            r"\$(\d[\d,.]*)",
            re.DOTALL,
        )

        for m in size_price_re.finditer(body_text):
            size_text = m.group(1).strip()
            price_text = m.group(2).strip()
            key = (size_text, price_text)
            if key in seen:
                continue
            seen.add(key)

            unit = UnitResult()
            unit.size = size_text
            w, ln, sq = self.normalize_size(size_text)
            if w is not None:
                unit.metadata = {"width": w, "length": ln, "sqft": sq}
            unit.price = self.normalize_price(price_text)
            unit.description = m.group(0).strip()[:200]
            if unit.size or unit.price:
                result.units.append(unit)

        if not result.units:
            result.warnings.append(
                "No storage units found. Page may not contain pricing data."
            )

        return result