#!/usr/bin/env python3
"""
audit.py: перевірка SEO та AI-search готовності сторінок стенду.
Бере рендер обох версій (before.html та optimized.html), парсить head
і JSON-LD, перевіряє обов'язкові поля Product/Offer за вимогами
Google Rich Results та пише машиночитний audit.json.
Запуск: python3 audit.py <dir_with_store> [out.json]
"""
import json
import pathlib
import re
import sys

OFFER_REQUIRED = ["price", "priceCurrency", "availability", "url"]
PRODUCT_REQUIRED = ["name", "image", "description", "sku", "brand", "offers"]


def extract_jsonld(html: str):
    blocks = re.findall(
        r'<script type="application/ld\+json">\s*(.*?)\s*</script>', html, re.S
    )
    parsed = []
    for raw in blocks:
        try:
            parsed.append(json.loads(raw))
        except json.JSONDecodeError as exc:
            parsed.append({"@type": "PARSE_ERROR", "error": str(exc)})
    return parsed


def check_page(path: pathlib.Path):
    html = path.read_text(encoding="utf-8")
    jsonld = extract_jsonld(html)
    types = [b.get("@type") for b in jsonld]
    result = {
        "file": path.name,
        "bytes": len(html.encode("utf-8")),
        "title": (re.search(r"<title>(.*?)</title>", html, re.S) or [None, ""])[1].strip(),
        "meta_description": bool(re.search(r'<meta name="description"', html)),
        "canonical": bool(re.search(r'<link rel="canonical"', html)),
        "robots_max_image_preview": "max-image-preview:large" in html,
        "og_product": '<meta property="og:type" content="product">' in html,
        "og_image": bool(re.search(r'<meta property="og:image"', html)),
        "twitter_card": 'name="twitter:card"' in html,
        "llms_txt_link": 'title="llms.txt"' in html,
        "imgs_missing_alt": len(re.findall(r"<img (?![^>]*alt=)[^>]*>", html)),
        "jsonld_types": types,
        "jsonld_errors": [],
    }
    for block in jsonld:
        if block.get("@type") == "Product":
            for field in PRODUCT_REQUIRED:
                if not block.get(field):
                    result["jsonld_errors"].append(f"Product: немає поля {field}")
            offer = block.get("offers") or {}
            for field in OFFER_REQUIRED:
                if not offer.get(field):
                    result["jsonld_errors"].append(f"Offer: немає поля {field}")
        if block.get("@type") == "PARSE_ERROR":
            result["jsonld_errors"].append("JSON-LD не парситься: " + block["error"])
    result["rich_results_eligible"] = (
        "Product" in types and "FAQPage" in types and "BreadcrumbList" in types
        and not result["jsonld_errors"]
    )
    return result


def main() -> None:
    base = pathlib.Path(sys.argv[1]) if len(sys.argv) > 1 else pathlib.Path(".")
    out = pathlib.Path(sys.argv[2]) if len(sys.argv) > 2 else base / "audit.json"
    report = {
        "stand": "guardlabs.online/demo/shopify-ai-seo",
        "checked": [
            check_page(base / "store" / "before.html"),
            check_page(base / "store" / "optimized.html"),
        ],
    }
    out.write_text(json.dumps(report, ensure_ascii=False, indent=2), encoding="utf-8")
    for page in report["checked"]:
        status = "OK" if page["rich_results_eligible"] else "NO RICH RESULTS"
        print(f'{page["file"]}: {status}, jsonld={page["jsonld_types"]}, errors={page["jsonld_errors"]}')


if __name__ == "__main__":
    main()