"""Parser for Hardin Mini Storage (UnitTrac platform)."""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility085615Parser(BaseParser):
    """Extract storage units from Hardin Mini Storage (hardinministorage.com).

    The site is powered by UnitTrac and renders unit listings as a list-group
    inside a card aside. Each list-group-item contains a nested table with:
    - Size label (Small/Medium/Large) and dimension text (e.g. "5' x 10'")
    - Square footage below the dimensions
    - Price formatted as "$45<sup>/mo</sup>"
    - Availability badge with text like "1 Available" or "0 Available"
    """

    platform = "custom_facility_085615"

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

        # Each unit is a list-group-item div (skip the first one which has the address)
        items = soup.select("div.list-group-item")

        for item in items:
            # Only process items that have a nested table (unit rows)
            outer_table = item.find("table")
            if not outer_table:
                continue

            unit = UnitResult()

            # --- Size label (Small / Medium / Large) ---
            label_span = item.find("span", style=re.compile(r"font-weight:\s*bold"))
            if label_span:
                unit.description = label_span.get_text(strip=True)

            # --- Dimensions: text node immediately after the label span ---
            # The dimension text (e.g. "5' x 10'") appears as a direct text node
            # in the same <td> as the label span, after a <br> tag.
            dimension_text = ""
            if label_span:
                parent_td = label_span.find_parent("td")
                if parent_td:
                    # Collect all text nodes in the td, skip blank ones
                    raw_texts = [t.strip() for t in parent_td.stripped_strings]
                    # raw_texts looks like: ["Small", "5' x 10'", "50"]
                    # First is the label, second is dimension, third is sqft
                    if len(raw_texts) >= 2:
                        dimension_text = raw_texts[1]
                        unit.size = dimension_text
                    if len(raw_texts) >= 3:
                        # third item is sqft number
                        sqft_str = raw_texts[2]
                        if sqft_str.isdigit():
                            w, ln, sq = self.normalize_size(dimension_text)
                            if w is not None:
                                unit.metadata = {"width": w, "length": ln, "sqft": sq}
                            else:
                                unit.metadata = {"sqft": int(sqft_str)}

            # Fall back: try normalize_size if metadata not yet set
            if unit.size and not unit.metadata:
                w, ln, sq = self.normalize_size(unit.size)
                if w is not None:
                    unit.metadata = {"width": w, "length": ln, "sqft": sq}

            # --- Price: span with font-size:18px contains "$45<sup>/mo</sup>" ---
            price_span = item.find("span", style=re.compile(r"font-size:\s*18px"))
            if price_span:
                # Remove the <sup> element so get_text only gives the dollar amount
                sup = price_span.find("sup")
                if sup:
                    sup.decompose()
                price_text = price_span.get_text(strip=True)
                unit.price = self.normalize_price(price_text)

            # --- Availability badge ---
            badge = item.find("div", class_="badge")
            if badge:
                unit.scarcity = badge.get_text(strip=True)

            if unit.size or unit.price:
                result.units.append(unit)

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

        return result