PROCITY/pipeline/validate_manifest.py

#!/usr/bin/env python3
"""Validate web/assets/manifest.json — runs in Lane F's integration gate.

Checks: JSON parses; every referenced skin exists locally; every fitting/furniture GLB
exists locally (pipeline/_normalized) OR HEADs 200 on the depot; every thumb exists;
footprints & heights are sane; every registry shop type has >=2 facades.

    python3 pipeline/validate_manifest.py            # local + soft depot check
    python3 pipeline/validate_manifest.py --depot    # require GLBs live on the depot (post-publish)

Exit 0 = green, 1 = fail. Plain stdlib (no deps), so it runs anywhere.
"""
import json, os, sys, urllib.request, urllib.error

ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
ASSETS = os.path.join(ROOT, "web", "assets")
NORM = os.path.join(ROOT, "pipeline", "_normalized")
MANIFEST = os.path.join(ASSETS, "manifest.json")
REGISTRY_TYPES = ["record", "opshop", "toy", "book", "video", "pawn", "milkbar", "dept", "stall"]
STRICT_DEPOT = "--depot" in sys.argv

errors, warnings = [], []


def err(m): errors.append(m)
def warn(m): warnings.append(m)


def head_ok(url):
    # The depot contract documents GET only, not HEAD — probe with a 1-byte Range GET so we don't
    # download the whole GLB and don't depend on HEAD being supported by the CDN/cache.
    try:
        # custom UA: Cloudflare 403s the default Python-urllib agent on the public path
        req = urllib.request.Request(url, headers={"Range": "bytes=0-0",
                                                   "User-Agent": "procity-validator/1.0"})
        with urllib.request.urlopen(req, timeout=12) as r:
            return r.status in (200, 206)
    except urllib.error.HTTPError as e:
        if e.code == 416:                    # range not satisfiable but the file exists
            return True
        return False
    except Exception:
        return False


def check_skin(file):
    p = os.path.join(ASSETS, file)
    if not os.path.isfile(p):
        err(f"skin missing: {file}")


def check_glb(entry, depot):
    file = entry["file"]
    local = os.path.join(NORM, file)
    on_disk = os.path.isfile(local)
    live = head_ok(f"{depot}/a/{file}")
    if STRICT_DEPOT and not live:
        err(f"GLB not live on depot: {file}")
    elif not on_disk and not live:
        err(f"GLB missing (not local, not on depot): {file}")
    elif not live:
        warn(f"GLB not yet published to depot (local only): {file}")
    # thumb
    thumb = entry.get("thumb")
    if thumb and not os.path.isfile(os.path.join(ASSETS, thumb)):
        err(f"thumb missing: {thumb}")
    # footprint / height sanity
    fp = entry.get("footprint")
    if not (isinstance(fp, list) and len(fp) == 2 and all(0 < x < 12 for x in fp)):
        err(f"insane footprint for {file}: {fp}")
    h = entry.get("height")
    if not (isinstance(h, (int, float)) and 0 < h < 8):
        err(f"insane height for {file}: {h}")


def main():
    try:
        m = json.load(open(MANIFEST))
    except Exception as e:
        print(f"FAIL: manifest does not parse: {e}")
        sys.exit(1)

    # GOD3_DEPOT overrides for the direct tailnet path (same env publish.py uses)
    depot = os.environ.get("GOD3_DEPOT", m.get("depot", "https://digalot.fyi/3god")).rstrip("/")
    sk = m.get("skins", {})

    # facades + type coverage
    facade = sk.get("facade", {})
    for k, v in facade.items():
        check_skin(v["file"])
    for t in REGISTRY_TYPES:
        n = sum(1 for v in facade.values() if t in v.get("types", []))
        if n < 2:
            err(f"shop type '{t}' has only {n} facade(s) (need >=2)")

    # other skin groups
    for s in sk.get("sky", []):
        check_skin(s["file"])
    for g in sk.get("ground", {}).values():
        check_skin(g["file"])
    for w in sk.get("wall", []):
        check_skin(w["file"])
    interior = sk.get("interior", {})
    for grp in ("floor", "surface"):
        for it in interior.get(grp, []):
            check_skin(it["file"])
    for a in sk.get("awning", []):
        check_skin(a["file"])

    # fittings + furniture GLBs
    n_glb = 0
    manifest_files = set()
    for grp in ("fittings", "furniture"):
        for entry in m.get(grp, {}).values():
            check_glb(entry, depot)
            manifest_files.add(entry["file"])
            n_glb += 1

    # provenance-drift gate: every manifest depot GLB must be recorded in _published.json, so a
    # clobbered/stale provenance record (the R5 bug) fails QA loudly instead of hiding.
    recpath = os.path.join(ROOT, "pipeline", "_published.json")
    try:
        record = set(json.load(open(recpath)))
    except Exception as e:
        err(f"_published.json unreadable: {e}")
        record = set()
    for f in sorted(manifest_files):
        if f.startswith("procity_") and f not in record:
            err(f"manifest GLB not in _published.json (provenance drift): {f}")

    print(f"manifest v{m.get('version')} — facades {len(facade)}, "
          f"skins {sum(len(v) if isinstance(v, list) else (len(v) if isinstance(v, dict) else 0) for v in sk.values())} groups, "
          f"GLBs {n_glb}")
    for w in warnings:
        print(f"  WARN  {w}")
    if errors:
        for e in errors:
            print(f"  ERR   {e}")
        print(f"\nFAIL — {len(errors)} error(s), {len(warnings)} warning(s)")
        sys.exit(1)
    print(f"\nOK — 0 errors, {len(warnings)} warning(s)")
    sys.exit(0)


if __name__ == "__main__":
    main()