xteink-apps/scripts/update-manifests.py

#!/usr/bin/env -S uv run --script
# /// script
# requires-python = ">=3.11"
# dependencies = []
# ///

import argparse
import hashlib
import re
import sys
from pathlib import Path

ROOT = Path(__file__).resolve().parent.parent
APPS_DIR = ROOT / "apps"
ID_PATTERN = re.compile(r"^[A-Za-z0-9_-]+$")
PATH_PATTERN = re.compile(r"^[A-Za-z0-9._/-]+$")
DESCRIPTION_PATTERN = re.compile(r"^--\s*DESCRIPTION:\s*(.+?)\s*$")
MAX_APPS = 64
MAX_FILES = 64
MAX_FILE_BYTES = 8 * 1024 * 1024
MAX_APP_BYTES = 16 * 1024 * 1024
MAX_MANIFEST_BYTES = 16384
MAX_CATALOG_BYTES = 32768


def sha256(data: bytes) -> str:
    return hashlib.sha256(data).hexdigest()


def payload_files(app_dir: Path):
    for path in sorted(app_dir.rglob("*")):
        if path.is_symlink():
            raise ValueError(f"symlinks are not supported: {path.relative_to(ROOT)}")
        if path.is_file() and path.relative_to(app_dir).as_posix() not in {"manifest.txt", "README.md"}:
            yield path


def description(app_dir: Path) -> str:
    with (app_dir / "main.lua").open(encoding="utf-8") as source:
        for line in source:
            match = DESCRIPTION_PATTERN.match(line)
            if match:
                value = match.group(1)
                if "|" in value:
                    raise ValueError(f"description contains '|': {app_dir.name}")
                return value
    raise ValueError(f"missing DESCRIPTION in {app_dir.name}/main.lua")


def encode_lines(lines: list[str]) -> bytes:
    for line in lines:
        if len(line.encode()) > 192:
            raise ValueError(f"generated line exceeds 192 bytes: {line[:40]}")
    return ("\n".join(lines) + "\n").encode()


def build_manifest(app_dir: Path) -> bytes:
    lines = ["XTEINK-MANIFEST|1"]
    seen_files = set()
    seen_directories = set()
    total = 0
    for path in payload_files(app_dir):
        relative = path.relative_to(app_dir).as_posix()
        lower = relative.lower()
        if (not PATH_PATTERN.fullmatch(relative) or lower == ".manifest.txt" or
                any(part in ("", ".", "..") for part in relative.split("/"))):
            raise ValueError(f"unsupported app path: {relative}")
        parent = ""
        for component in lower.split("/"):
            if parent:
                if parent in seen_files:
                    raise ValueError(f"path collision: {relative}")
                seen_directories.add(parent)
            parent = component if not parent else f"{parent}/{component}"
        if lower in seen_files or lower in seen_directories:
            raise ValueError(f"path collision: {relative}")
        data = path.read_bytes()
        if not data or len(data) > MAX_FILE_BYTES:
            raise ValueError(f"unsupported app file size: {relative}")
        total += len(data)
        if len(lines) > MAX_FILES or total > MAX_APP_BYTES:
            raise ValueError(f"app exceeds package limits: {app_dir.name}")
        seen_files.add(lower)
        lines.append(f"{relative}|{len(data)}|{sha256(data)}")
    manifest = encode_lines(lines)
    if len(manifest) > MAX_MANIFEST_BYTES:
        raise ValueError(f"manifest exceeds {MAX_MANIFEST_BYTES} bytes: {app_dir.name}")
    return manifest


def generated_files():
    apps = [path for path in APPS_DIR.iterdir() if path.is_dir() and (path / "main.lua").is_file()]
    apps.sort(key=lambda path: path.name.casefold())
    if len(apps) > MAX_APPS:
        raise ValueError(f"catalog exceeds {MAX_APPS} apps")

    manifests = {}
    catalog = ["XTEINK-CATALOG|1"]
    seen_ids = set()
    for app_dir in apps:
        app_id = app_dir.name
        lower_id = app_id.lower()
        if not ID_PATTERN.fullmatch(app_id) or len(app_id) > 48 or lower_id in seen_ids:
            raise ValueError(f"invalid or duplicate app id: {app_id}")
        seen_ids.add(lower_id)
        app_description = description(app_dir)
        if len(app_description.encode()) > 80:
            raise ValueError(f"description exceeds 80 bytes: {app_id}")
        manifest = build_manifest(app_dir)
        manifest_path = f"apps/{app_id}/manifest.txt"
        manifests[ROOT / manifest_path] = manifest
        catalog.append(
            f"{app_id}|{app_description}|{manifest_path}|{len(manifest)}|{sha256(manifest)}"
        )

    catalog_data = encode_lines(catalog)
    if len(catalog_data) > MAX_CATALOG_BYTES:
        raise ValueError(f"catalog exceeds {MAX_CATALOG_BYTES} bytes")
    manifests[ROOT / "catalog.txt"] = catalog_data
    return manifests


def main() -> int:
    parser = argparse.ArgumentParser(description="Regenerate app manifests and catalog.txt")
    parser.add_argument("--check", action="store_true", help="fail instead of updating stale files")
    args = parser.parse_args()

    stale = []
    for path, expected in generated_files().items():
        if not path.exists() or path.read_bytes() != expected:
            stale.append(path)
            if not args.check:
                temporary = path.with_suffix(path.suffix + ".tmp")
                temporary.write_bytes(expected)
                temporary.replace(path)
                print(f"updated {path.relative_to(ROOT)}")

    if args.check and stale:
        for path in stale:
            print(f"stale: {path.relative_to(ROOT)}", file=sys.stderr)
        return 1
    return 0


if __name__ == "__main__":
    try:
        raise SystemExit(main())
    except ValueError as error:
        print(error, file=sys.stderr)
        raise SystemExit(1)