"""Exercise actual production routes and assets without third-party dependencies."""
from html import unescape
import json
from pathlib import Path
import re
import sys
import urllib.error
import urllib.request
import xml.etree.ElementTree as ET

ROOT = Path(__file__).resolve().parents[1]


def verify(base: str) -> None:
    def get(path: str) -> tuple[bytes, str]:
        with urllib.request.urlopen(base.rstrip("/") + path, timeout=15) as response:
            if response.status != 200:
                raise RuntimeError(f"Unexpected status for {path}: {response.status}")
            return response.read(), response.headers.get("Content-Type", "")

    health, _ = get("/api/health")
    assert json.loads(health) == {"status": "ok"}
    home, _ = get("/")
    assert b"TABULARIUM" in home
    for file in sorted((ROOT / "content/docs").glob("*.md")):
        page, _ = get(f"/docs/{file.stem}")
        title = re.search(r"^# (.+)$", file.read_text(), re.M)
        assert title is not None
        text = unescape(re.sub(r"<[^>]+>", "", page.decode()))
        assert title.group(1) in text, file.name
    assets = set(re.findall(r'(?:src|href)="(/_next/static/[^" ]+)"', home.decode()))
    assert assets, "No generated JavaScript/CSS assets found"
    for asset in assets:
        get(unescape(asset))
    for file in sorted((ROOT / "apps/docs/public/figures").glob("*.svg")):
        data, kind = get(f"/figures/{file.name}")
        assert "image/svg+xml" in kind
        assert ET.fromstring(data).tag.endswith("svg")
    try:
        get("/docs/this-document-does-not-exist")
    except urllib.error.HTTPError as error:
        assert error.code == 404
    else:
        raise RuntimeError("Unknown document did not return 404")
    print(f"Smoke passed: {base} (health, documents, JS/CSS, SVG, 404)", flush=True)


if __name__ == "__main__":
    verify(sys.argv[1])