"""Parser for Aaa Secured Storage Llc (Medford, OR).

This site is a brochure-only Divi/WordPress page.  It advertises storage
units with climate control and 24/7 surveillance but does NOT publish unit
sizes or prices online.  The homepage and /storage-units/ page both say
"Call For Details" (541-826-1850).

The parser checks for any pricing indicators and returns a clear no_pricing
warning when none are found.
"""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility002143Parser(BaseParser):
    """Extract storage units from Aaa Secured Storage Llc."""

    platform = "custom_facility_002143"

    # Patterns that would indicate pricing data is present
    _PRICE_RE = re.compile(r"\$\s*\d[\d,.]*")
    _SIZE_RE = re.compile(
        r"\d+\s*['\u2032]?\s*[xX\u00d7]\s*\d+\s*['\u2032]?"
    )

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

        for tag in soup.find_all(["script", "style"]):
            tag.decompose()

        body_text = soup.get_text(separator="\n")

        prices = self._PRICE_RE.findall(body_text)
        sizes = self._SIZE_RE.findall(body_text)

        if not prices and not sizes:
            result.warnings.append(
                "no_pricing: Aaa Secured Storage Llc is a brochure-only "
                "Divi/WordPress site. No unit sizes or prices are published "
                "online. Contact 541-826-1850 for pricing."
            )
            return result

        # If the site has been updated to include pricing, attempt extraction.
        seen: set[tuple[str, str]] = set()
        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_pricing: Detected price/size tokens on page but could "
                "not pair them into units. Site may have been updated."
            )

        return result