#!/usr/bin/env python3
"""Check catalog homepage + PDF URI targets after a site build."""
from __future__ import annotations

import sys
from html.parser import HTMLParser
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]
SITE = ROOT / "site"
WEB = "https://zapier.georgelambert.org/"


class Anchors(HTMLParser):
    def __init__(self) -> None:
        super().__init__()
        self.hrefs: list[str] = []

    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
        if tag == "a":
            d = dict(attrs)
            if d.get("href"):
                self.hrefs.append(d["href"] or "")


def _exists(dest: Path) -> bool:
    if dest.is_file():
        return True
    if dest.is_dir() and any(
        (dest / n).exists() for n in ("index.html", "index.pdf", "README.html", "README.pdf")
    ):
        return True
    return False


def local_ok(url: str) -> bool:
    if url.startswith(WEB) or url.startswith("http://zapier.georgelambert.org/"):
        rel = url.split("://", 1)[-1].split("/", 1)[-1].split("#", 1)[0]
        return _exists(SITE / rel)
    if url.startswith("file:"):
        return False
    return True  # http(s) off-site; homepage checker uses GET separately


def pdf_dest_ok(pdf: Path, raw: str) -> bool:
    raw = raw.split("#", 1)[0]
    if raw.startswith("file:"):
        return False
    if raw.startswith("mailto:") or raw.startswith("data:"):
        return True
    if raw.startswith("http://") or raw.startswith("https://"):
        if raw.startswith(WEB) or raw.startswith("http://zapier.georgelambert.org/"):
            return local_ok(raw)
        return True
    if raw.startswith("/"):
        dest = SITE / raw.lstrip("/")
    else:
        dest = (pdf.parent / raw).resolve()
        try:
            dest.relative_to(SITE.resolve())
        except ValueError:
            return False
    return _exists(dest)


def main() -> int:
    idx = SITE / "index.html"
    if not idx.exists():
        print("no site/index.html — run scripts/build-docs-site.py")
        return 1
    p = Anchors()
    p.feed(idx.read_text(encoding="utf-8", errors="replace"))
    broken: list[str] = []
    for h in sorted(set(p.hrefs)):
        if h.startswith("#") or h.startswith("mailto:"):
            continue
        if h.startswith("https://git.georgelambert.org"):
            continue
        if h.startswith("https://") and not h.startswith(WEB):
            continue
        if h.startswith(WEB):
            if not local_ok(h):
                broken.append(f"homepage {h}")
            continue
        rel = h.lstrip("/")
        dest = SITE / rel
        if not dest.exists():
            broken.append(f"homepage missing {h}")

    from pypdf import PdfReader

    n_file = 0
    for pdf in SITE.rglob("*.pdf"):
        try:
            reader = PdfReader(str(pdf))
        except Exception as exc:
            broken.append(f"unreadable {pdf.relative_to(SITE)} {exc}")
            continue
        for page in reader.pages:
            annots = page.get("/Annots")
            if not annots:
                continue
            for annot in annots:
                obj = annot.get_object()
                action = obj.get("/A")
                if not action:
                    continue
                uri = action.get("/URI")
                if not uri:
                    continue
                raw = str(uri)
                if raw.startswith("file:"):
                    n_file += 1
                    broken.append(f"file URI {pdf.relative_to(SITE)} -> {raw[:120]}")
                    continue
                if "/research/" in raw:
                    continue
                if not pdf_dest_ok(pdf, raw):
                    broken.append(f"pdf missing {pdf.relative_to(SITE)} -> {raw}")

    print(f"file:// leftover {n_file}")
    print(f"broken {len(broken)}")
    for b in broken[:80]:
        print(" ", b)
    if len(broken) > 80:
        print(f"  … {len(broken) - 80} more")
    return 1 if broken else 0


if __name__ == "__main__":
    sys.exit(main())