"""Parser for Regal Storage (Wix site, Lake Jackson TX)."""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility002009Parser(BaseParser):
    """Extract storage units from Regal Storage.

    This is a Wix-built site that lists unit sizes on the /storage-units page
    but does NOT display prices online (visitors must call for pricing).
    Units are rendered as <h1 class="font_0"> elements with dimensions like
    5' x 10'.
    """

    platform = "custom_facility_002009"

    _SIZE_RE = re.compile(
        r"(\d+)\s*['\u2032\u2019]?\s*[xX×]\s*(\d+)\s*['\u2032\u2019]?"
    )

    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"):
            tag.decompose()
        for tag in soup.find_all("style"):
            tag.decompose()

        seen: set[tuple[int, int]] = set()

        # Unit sizes are in <h1 class="font_0 wixui-rich-text__text"> elements
        # inside wixui-rich-text containers.
        for h1 in soup.find_all("h1", class_="font_0"):
            text = h1.get_text(strip=True)
            m = self._SIZE_RE.search(text)
            if not m:
                continue

            width = int(m.group(1))
            length = int(m.group(2))

            # Skip implausible dimensions (e.g. from descriptive text)
            if width < 3 or length < 3:
                continue

            key = (width, length)
            if key in seen:
                continue
            seen.add(key)

            size_str = f"{width}' x {length}'"
            w, ln, sq = self.normalize_size(size_str)

            unit = UnitResult()
            unit.size = size_str
            unit.price = None  # Prices not listed on site
            unit.description = f"Regal Storage - {size_str}"
            if w is not None:
                unit.metadata = {"width": w, "length": ln, "sqft": sq}
            result.units.append(unit)

        # Fallback: scan all text nodes for dimension patterns if no h1 matches
        if not result.units:
            body_text = soup.get_text(separator="\n")
            for m in self._SIZE_RE.finditer(body_text):
                width = int(m.group(1))
                length = int(m.group(2))
                if width < 3 or length < 3:
                    continue
                key = (width, length)
                if key in seen:
                    continue
                seen.add(key)

                size_str = f"{width}' x {length}'"
                w, ln, sq = self.normalize_size(size_str)

                unit = UnitResult()
                unit.size = size_str
                unit.price = None
                unit.description = f"Regal Storage - {size_str}"
                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")
        else:
            result.warnings.append(
                "Prices not listed on site; call 979-299-3300 for pricing"
            )

        return result