"""Parser for A1 Storage America facilities (Canton, Sylva, Franklin, Waynesville).

These SUS-hosted sites display pricing as plain text lines in a widget column
rather than structured unit cards. Format:
    4 x12 Starting at $50 / month
    10x10 Starting at $70 / month
    10x25: $50 / month
"""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility23Parser(BaseParser):
    """Extract storage units from A1 Storage America text-based listings."""

    platform = "custom_facility_23"

    # "4 x12 Starting at $50 / month" or "10x10 Starting at $70 / month"
    _UNIT_RE = re.compile(
        r"(\d+(?:\.\d+)?)\s*[xX\u00d7]\s*(\d+(?:\.\d+)?)"
        r"[:\s]*(?:Starting\s+at\s+)?"
        r"\$\s*([\d,]+(?:\.\d+)?)"
        r"\s*/\s*month",
        re.IGNORECASE,
    )

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

        # Look for the content area with pricing text
        text = soup.get_text(separator="\n", strip=True)

        for match in self._UNIT_RE.finditer(text):
            width = float(match.group(1))
            length = float(match.group(2))
            price_str = match.group(3).replace(",", "")
            price = float(price_str)

            unit = UnitResult(
                size=f"{int(width)}x{int(length)}",
                sale_price=price,
                description=match.group(0).strip(),
                metadata={"width": width, "length": length, "sqft": width * length},
            )
            result.units.append(unit)

        if not result.units:
            result.warnings.append("No unit pricing lines found in page text")

        return result