mcg-website/tools/verify_migration.py

"""Post-migration checks. Run after `hugo --gc --minify`.

1. Inventory  - every published WordPress page/post has a content file.
2. Media      - every attachment is either placed in the site or explicitly
                accounted for as an unused media-library leftover.
3. Links      - no internal href/src in public/ points at a missing file, and no
                absolute wp-content URL survived the import.

Exits non-zero if any check fails.
"""

from __future__ import annotations

import re
import sys
from pathlib import Path
from urllib.parse import unquote, urlsplit

from wp2hugo import PAGE_MAP, SKIP, SLUG_FIXES
from wxr import REPO_ROOT, clean_media_name, load_items, slugify

PUBLIC = REPO_ROOT / "public"
CONTENT = REPO_ROOT / "content"

failures: list[str] = []
notes: list[str] = []


def check_inventory(items) -> None:
    expected = 0
    for item in items.values():
        if item.post_id in SKIP or item.status != "publish":
            continue
        if item.post_type == "page":
            mapping = PAGE_MAP.get(item.post_id)
            if not mapping:
                failures.append(f"page {item.post_id} ({item.slug}) is unmapped")
                continue
            path = CONTENT / mapping[0]
        elif item.post_type == "post":
            path = CONTENT / "aktuelles" / SLUG_FIXES.get(
                item.slug, slugify(item.slug)) / "index.md"
        else:
            continue
        expected += 1
        if not path.exists():
            failures.append(f"missing content file for {item.post_type} "
                            f"{item.post_id} ({item.slug}): {path}")
    notes.append(f"inventory: {expected} published pages/posts accounted for")

    for post_id, reason in SKIP.items():
        notes.append(f"  skipped {post_id}: {reason}")


def check_media(items) -> None:
    placed = {p.name for p in CONTENT.rglob("*") if p.is_file()
              and p.suffix.lower() != ".md"}
    placed |= {p.name for p in (REPO_ROOT / "assets" / "img").rglob("*")
               if p.is_file()}
    placed |= {p.name for p in (REPO_ROOT / "static").rglob("*") if p.is_file()}
    # assets/img/logo.png and banner.jpg are renamed copies.
    placed |= {"Emblem_transparent.png", "cropped-Banner_small_web.jpg"}

    unused = []
    for item in items.values():
        if item.post_type != "attachment" or not item.attachment_url:
            continue
        original = item.attachment_url.rsplit("/", 1)[-1]
        if original in placed or clean_media_name(original) in placed:
            continue
        unused.append(item.attachment_url)

    total = sum(1 for i in items.values() if i.post_type == "attachment")
    notes.append(f"media: {total - len(unused)}/{total} attachments placed")
    if unused:
        notes.append(f"  {len(unused)} media-library leftovers not shown on the "
                     f"live site (see MIGRATION.md)")


HREF = re.compile(r'(?:href|src)=["\']?([^"\'\s>]+)', re.I)


def check_links() -> None:
    if not PUBLIC.exists():
        failures.append("public/ does not exist - run `hugo` first")
        return

    checked = broken = 0
    for html in PUBLIC.rglob("*.html"):
        text = html.read_text(encoding="utf-8", errors="replace")

        for stale in re.findall(r'https?://motorradclub-giebelwald\.de/wp-content/\S*',
                                text):
            failures.append(f"{html.relative_to(PUBLIC)}: stale WordPress URL {stale}")

        for raw in HREF.findall(text):
            url = urlsplit(raw)
            if url.scheme or url.netloc or not url.path:
                continue                      # external, mailto:, protocol-relative
            if raw.startswith("#"):
                continue
            checked += 1
            path = unquote(url.path)
            target = (PUBLIC / path.lstrip("/") if path.startswith("/")
                      else html.parent / path)
            if target.is_dir():
                target = target / "index.html"
            if not target.exists():
                broken += 1
                failures.append(f"{html.relative_to(PUBLIC)}: broken link -> {raw}")

    notes.append(f"links: {checked} internal references checked, {broken} broken")


def main() -> int:
    items = load_items()
    check_inventory(items)
    check_media(items)
    check_links()

    for note in notes:
        print(note)
    if failures:
        print(f"\n{len(failures)} FAILURE(S):", file=sys.stderr)
        for f in failures:
            print(f"  - {f}", file=sys.stderr)
        return 1
    print("\nall checks passed")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())