Create sync_manager.py

sync_manager.py

@@ -0,0 +1,107 @@
+import os
+import sys
+import time
+import shutil
+import zipfile
+from pathlib import Path
+from huggingface_hub import HfApi, hf_hub_download, list_repo_files
+
+# =========================
+# CONFIG
+# =========================
+api = HfApi()
+REPO_ID = os.getenv("HF_DATASET")
+TOKEN = os.getenv("HF_TOKEN")
+
+BASE_DIR = Path("/root/.openclaw")
+PREFIX = "openclawai_backup_"
+KEEP_LAST = 5
+
+# กรองเฉพาะไฟล์ที่ต้องการ และตัดไฟล์ระบบ/ไฟล์ชั่วคราวทิ้ง
+ALLOWED_EXTENSIONS = {'.md', '.json', '.db', '.sqlite', '.py'}
+IGNORE_PATTERNS = {'.lock', '.tmp', '.wal', '.shm', '__pycache__'}
+
+# =========================
+# UTILS
+# =========================
+def log(msg):
+    print(f"[{time.strftime('%H:%M:%S')}] {msg}", flush=True)
+
+def is_valid_file(path: Path):
+    # ต้องมีนามสกุลที่กำหนด และไม่อยู่ในกลุ่ม Ignore
+    if path.suffix.lower() not in ALLOWED_EXTENSIONS:
+        return False
+    if any(pattern in path.name for pattern in IGNORE_PATTERNS):
+        return False
+    return True
+
+# =========================
+# CORE FUNCTIONS
+# =========================
+def create_zip(zip_path):
+    files = [p for p in BASE_DIR.rglob("*") if p.is_file() and is_valid_file(p)]
+    
+    if not files:
+        log("⚠️ No important files found to backup")
+        return False
+
+    log(f"📦 Zipping {len(files)} files...")
+    with zipfile.ZipFile(zip_path, "w", zipfile.ZIP_DEFLATED) as z:
+        for p in files:
+            z.write(p, p.relative_to(BASE_DIR))
+    return True
+
+def backup():
+    if not REPO_ID or not TOKEN: return log("❌ Config missing")
+    
+    zip_name = f"{PREFIX}{time.strftime('%Y%m%d_%H%M%S')}.zip"
+    zip_path = Path(f"/tmp/{zip_name}")
+
+    try:
+        if BASE_DIR.exists() and create_zip(zip_path):
+            log(f"📤 Uploading: {zip_name}")
+            api.upload_file(path_or_fileobj=str(zip_path), path_in_repo=zip_name, 
+                            repo_id=REPO_ID, repo_type="dataset", token=TOKEN)
+            
+            # Cleanup old backups
+            contents = list_repo_files(REPO_ID, repo_type="dataset", token=TOKEN)
+            old_backups = sorted([f for f in contents if f.startswith(PREFIX)])
+            for old in old_backups[:-KEEP_LAST]:
+                api.delete_file(path_in_repo=old, repo_id=REPO_ID, repo_type="dataset", token=TOKEN)
+                log(f"🗑️ Deleted: {old}")
+    finally:
+        if zip_path.exists(): zip_path.unlink()
+
+def restore():
+    try:
+        contents = list_repo_files(REPO_ID, repo_type="dataset", token=TOKEN)
+        backups = sorted([f for f in contents if f.startswith(PREFIX)])
+        if not backups: return log("⚠️ No backup found")
+
+        latest = backups[-1]
+        log(f"📥 Restoring: {latest}")
+        zip_path = hf_hub_download(repo_id=REPO_ID, filename=latest, repo_type="dataset", token=TOKEN)
+
+        if BASE_DIR.exists(): shutil.rmtree(BASE_DIR)
+        BASE_DIR.mkdir(parents=True, exist_ok=True)
+        
+        with zipfile.ZipFile(zip_path, "r") as z:
+            z.extractall(BASE_DIR)
+        log("✅ Restore successful!")
+    except Exception as e:
+        log(f"❌ Restore failed: {e}")
+
+# =========================
+# MAIN
+# =========================
+if __name__ == "__main__":
+    action = sys.argv[1].lower() if len(sys.argv) > 1 else "restore"
+
+    if action == "backup":
+        log("🔄 Backup loop active (Hourly at :00)")
+        while True:
+            now = time.localtime()
+            time.sleep(max(0, (60 - now.tm_min) * 60 - now.tm_sec))
+            backup()
+    else:
+        restore()