"""Parser for Solon Storage Rentals."""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility092120Parser(BaseParser):
    """Extract storage units from Solon Storage Rentals."""

    platform = "custom_facility_092120"

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

        # Unit sizes are in <p> tags inside .wpb_text_column divs.
        # Multiple sizes are concatenated in each paragraph without separators
        # (e.g. "6×1010×1010×1210×15"). Use a 1-2 digit constraint to split them.
        size_pattern = re.compile(r"(\d{1,2})[×xX](\d{1,2})(?=\d{1,2}[×xX]|\D|$)")

        seen_sizes = set()
        for p in soup.select(".wpb_text_column p"):
            text = p.get_text(strip=True)
            for match in size_pattern.finditer(text):
                size_text = f"{match.group(1)}×{match.group(2)}"
                if size_text not in seen_sizes:
                    seen_sizes.add(size_text)
                    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}
                    result.units.append(unit)

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

        return result