sync.py

# ─────────────────────────────────────────────────────────────
# 📁 PATH: sync.py (root HF Space repo)
# ─────────────────────────────────────────────────────────────

import os
import sys
import tarfile
import time
from huggingface_hub import HfApi, hf_hub_download

api     = HfApi()
repo_id = os.getenv("HF_DATASET")
token   = os.getenv("HF_TOKEN")

FILENAME         = "latest_backup.tar.gz"
BROWSER_FILENAME = "browser_backup.tar.gz"

# Folder/file yang di-SKIP dari backup utama
# - openclaw.json  → auto-create setiap startup, tidak perlu backup
# - browsers       → punya backup terpisah (BROWSER_FILENAME)
SKIP_NAMES = {"openclaw.json", "browsers"}


def restore():
    if not repo_id or not token:
        print("Skip Restore: HF_DATASET atau HF_TOKEN belum di-set.")
        return

    # ── Restore 1: sessions & memory ──────────────────────────
    try:
        print(f"Downloading {FILENAME} from {repo_id}...")
        path = hf_hub_download(
            repo_id=repo_id,
            filename=FILENAME,
            repo_type="dataset",
            token=token
        )
        with tarfile.open(path, "r:gz") as tar:
            tar.extractall(path="/root/.openclaw/")
        print(f"Success: Restored from {FILENAME}")

    except Exception as e:
        print(f"Restore Note: {e}")
        print("  → Fresh install, normal untuk deploy pertama.")

    # ── Restore 2: browser binary ──────────────────────────────
    try:
        print(f"Downloading {BROWSER_FILENAME} from {repo_id}...")
        browser_path = hf_hub_download(
            repo_id=repo_id,
            filename=BROWSER_FILENAME,
            repo_type="dataset",
            token=token
        )
        with tarfile.open(browser_path, "r:gz") as tar:
            tar.extractall(path="/root/.openclaw/")
        print(f"Success: Restored browser from {BROWSER_FILENAME}")

    except Exception as e:
        print(f"Restore Note (browser): {e} — akan install fresh jika dibutuhkan.")


def backup():
    if not repo_id or not token:
        print("Skip Backup: HF_DATASET atau HF_TOKEN belum di-set.")
        return

    # ── Backup 1: semua file/folder kecuali yang di SKIP_NAMES ─
    try:
        base_dir = "/root/.openclaw"

        print(f"Creating {FILENAME} (semua kecuali: {SKIP_NAMES})...")
        with tarfile.open(FILENAME, "w:gz") as tar:
            for item in os.listdir(base_dir):
                if item in SKIP_NAMES:
                    print(f"  - Skip: {item}")
                    continue
                full_path = os.path.join(base_dir, item)
                arcname   = item  # path relatif di dalam tar = relatif ke /root/.openclaw/
                tar.add(full_path, arcname=arcname)
                print(f"  + Added: {full_path}")

        api.upload_file(
            path_or_fileobj=FILENAME,
            path_in_repo=FILENAME,
            repo_id=repo_id,
            repo_type="dataset",
            token=token
        )
        print(f"Backup Success — {time.strftime('%Y-%m-%d %H:%M:%S')}")

    except Exception as e:
        print(f"Backup Error: {e}")

    # ── Backup 2: browser binary (upload sekali saja, ~150MB) ──
    try:
        browsers_dir = "/root/.openclaw/browsers"
        if not os.path.exists(browsers_dir):
            return

        # Cek apakah sudah ada di dataset
        try:
            hf_hub_download(
                repo_id=repo_id,
                filename=BROWSER_FILENAME,
                repo_type="dataset",
                token=token
            )
            print("Browser backup sudah ada di dataset — skip upload.")
        except Exception:
            # Belum ada → upload sekarang
            print("Uploading browser backup (first time, ~150MB)...")
            with tarfile.open(BROWSER_FILENAME, "w:gz") as tar:
                tar.add(browsers_dir, arcname="browsers")
            api.upload_file(
                path_or_fileobj=BROWSER_FILENAME,
                path_in_repo=BROWSER_FILENAME,
                repo_id=repo_id,
                repo_type="dataset",
                token=token
            )
            print("Browser backup uploaded.")

    except Exception as e:
        print(f"Backup Error (browser): {e}")


if __name__ == "__main__":
    if len(sys.argv) > 1 and sys.argv[1] == "backup":
        backup()
    else:
        restore()