#!/usr/bin/env python3
import re
import sys
from html.parser import HTMLParser
from pathlib import Path
from urllib.parse import urlparse

ROOT = Path(__file__).resolve().parent.parent
INDEX = ROOT / "index.html"
SKIP = ("#", "mailto:", "tel:", "data:", "javascript:")
FETCH_RELS = {"stylesheet", "icon", "preload", "modulepreload", "manifest", "apple-touch-icon"}
errors = []


def check(ref, base, where, fetched):
    if not ref or ref.startswith(SKIP):
        return
    if ref.startswith(("http://", "https://", "//")):
        if fetched:
            errors.append(f"{where}: external request {ref}")
        return
    path = urlparse(ref).path
    target = ROOT / path.lstrip("/") if path.startswith("/") else base.parent / path
    if not target.exists():
        errors.append(f"{where}: missing file {ref}")


class Page(HTMLParser):
    def handle_starttag(self, tag, attrs):
        a = dict(attrs)
        refs = []
        if tag in ("script", "img", "source", "iframe", "video", "audio"):
            refs.append(a.get("src"))
            for part in (a.get("srcset") or "").split(","):
                if part.strip():
                    refs.append(part.split()[0])
        elif tag == "link" and FETCH_RELS & set((a.get("rel") or "").split()):
            refs.append(a.get("href"))
        for ref in refs:
            check(ref, INDEX, "index.html", True)
        if tag == "a":
            check(a.get("href"), INDEX, "index.html", False)


for name in ("index.html", "favicon.svg", "robots.txt"):
    if not (ROOT / name).is_file():
        errors.append(f"missing required file: {name}")

if INDEX.is_file():
    Page().feed(INDEX.read_text(encoding="utf-8"))

for css in ROOT.glob("assets/*.css"):
    for ref in re.findall(r"url\(\s*['\"]?([^'\")]+)", css.read_text(encoding="utf-8")):
        check(ref, css, str(css.relative_to(ROOT)), True)

if errors:
    print("\n".join(errors))
    sys.exit(1)
print("site check passed")