"""Parser for Prairie View Storage (Wix site, no pricing listed)."""

from __future__ import annotations

import re

from bs4 import BeautifulSoup

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


class Facility001106Parser(BaseParser):
    """Extract storage units from Prairie View Storage.

    This Wix-based site lists unit sizes across two locations
    (Clinton County and Highland County) but does not publish prices.
    Sizes appear inside repeater items as ``<h6>`` elements within ``<ul>`` lists.
    """

    platform = "custom_facility_001106"

    _SIZE_RE = re.compile(
        r"(\d+)\s*(?:ft|'|′)?\s*[xX×]\s*(\d+)\s*(?:ft|'|′)?",
    )

    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", "style"]):
            tag.decompose()

        # Each repeater item contains a location header and a list of sizes.
        repeater_items = soup.find_all(
            "div",
            attrs={"data-mesh-id": re.compile(r"comp-m19m46zn__item.*inlineContent-gridContainer")},
        )

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

        if repeater_items:
            for item in repeater_items:
                # Determine location name from first text block (h4/h5/h6 or plain text)
                lines = [
                    ln.strip()
                    for ln in item.get_text(separator="\n").split("\n")
                    if ln.strip()
                ]
                location = lines[0] if lines else ""

                # Extract sizes from <h6> tags inside <ul> lists
                for h6 in item.find_all("h6"):
                    text = h6.get_text(strip=True)
                    m = self._SIZE_RE.search(text)
                    if not m:
                        continue
                    w, ln_val = int(m.group(1)), int(m.group(2))
                    if w < 3 or ln_val < 3:
                        continue
                    if (w, ln_val) in seen:
                        continue
                    seen.add((w, ln_val))

                    size_str = f"{w}x{ln_val}"
                    unit = UnitResult()
                    unit.size = size_str
                    _, _, sq = self.normalize_size(size_str)
                    unit.metadata = {"width": w, "length": ln_val, "sqft": sq}
                    if location:
                        unit.description = location
                    result.units.append(unit)
        else:
            # Fallback: scan all text for size patterns
            body_text = soup.get_text(separator="\n")
            for m in self._SIZE_RE.finditer(body_text):
                w, ln_val = int(m.group(1)), int(m.group(2))
                if w < 3 or ln_val < 3:
                    continue
                if (w, ln_val) in seen:
                    continue
                seen.add((w, ln_val))

                size_str = f"{w}x{ln_val}"
                unit = UnitResult()
                unit.size = size_str
                _, _, sq = self.normalize_size(size_str)
                unit.metadata = {"width": w, "length": ln_val, "sqft": sq}
                result.units.append(unit)

        if not result.units:
            result.warnings.append("No units found")
        else:
            result.warnings.append("No pricing available on this site — sizes only")

        return result