"""Parser for Extra Storage Mills, WY facility (061274).

This site is built on Wix with obfuscated CSS class names. Units are displayed
in two row types within a table-like layout:
  - Standard rows: class 'OMl9pPN1h6gEVHf3-Pyw3 row' — contain price, size, description
  - Featured/compact rows: class '_3ea1Xmp6wkmi39zHaaXfUa row' — price and size only
"""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility061274Parser(BaseParser):
    """Extract storage units from Extra Storage Mills, WY (Wix-based site).

    Unit rows use obfuscated Wix class names. Two row variants are handled:
    - Standard rows with price, size, and description (class 'OMl9pPN1h6gEVHf3-Pyw3')
    - Compact rows with price and size only (class '_3ea1Xmp6wkmi39zHaaXfUa')
    """

    platform = "custom_facility_061274"

    # Matches sizes like "8'x10'", "10X20", "10' x 15'"
    _SIZE_RE = re.compile(
        r"(\d+(?:\.\d+)?)\s*['\u2019]?\s*[xX\u00d7]\s*(\d+(?:\.\d+)?)\s*['\u2019]?",
    )

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

        # Parse standard unit rows (price + size + description)
        standard_rows = soup.find_all("div", class_="OMl9pPN1h6gEVHf3-Pyw3")
        for row in standard_rows:
            unit = self._parse_standard_row(row, url)
            if unit:
                result.units.append(unit)

        # Parse compact rows (price + size only, no description)
        compact_rows = soup.find_all("div", class_="_3ea1Xmp6wkmi39zHaaXfUa")
        for row in compact_rows:
            unit = self._parse_compact_row(row, url)
            if unit:
                result.units.append(unit)

        if not result.units:
            result.warnings.append("No units found on page")

        return result

    def _parse_standard_row(self, row: BeautifulSoup, url: str) -> UnitResult | None:
        """Parse a standard unit row with price, size, and description."""
        unit = UnitResult(url=url)

        # Price: div with class '_1w_9K-RT7ovDwHPRzUGIzt'
        price_el = row.find("div", class_="_1w_9K-RT7ovDwHPRzUGIzt")
        if price_el:
            unit.price = self.normalize_price(price_el.get_text(strip=True))

        # Size: div with class '_1duec5zuiRUo83p89oK8jC'
        size_el = row.find("div", class_="_1duec5zuiRUo83p89oK8jC")
        if size_el:
            size_text = size_el.get_text(strip=True)
            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}

        # Description: div with class '_2YTbZYDJKWW023QoDpsunD'
        desc_el = row.find("div", class_="_2YTbZYDJKWW023QoDpsunD")
        if desc_el:
            unit.description = desc_el.get_text(strip=True)

        if unit.size or unit.price:
            return unit
        return None

    def _parse_compact_row(self, row: BeautifulSoup, url: str) -> UnitResult | None:
        """Parse a compact unit row with price and size only (no description)."""
        unit = UnitResult(url=url)

        # All text in compact rows uses span class '_2zX9LAmLV_HITMCldbRQ08'
        spans = row.find_all("span", class_="_2zX9LAmLV_HITMCldbRQ08")
        for span in spans:
            text = span.get_text(strip=True)
            # Try to determine if it's a price or a size
            if text.startswith("$"):
                unit.price = self.normalize_price(text)
            elif self._SIZE_RE.search(text):
                unit.size = text
                w, ln, sq = self.normalize_size(text)
                if w is not None:
                    unit.metadata = {"width": w, "length": ln, "sqft": sq}

        if unit.size or unit.price:
            return unit
        return None