Update app.py

app.py

@@ -43,45 +43,45 @@ DATA_FOLDER = Path("bluesky_data")
 DATA_FILE = f"bluesky_counts_{uuid.uuid4()}.json"
 
 
-def load_hub_data():
-    """Load the most recent data from the Hub"""
-    try:
-        api = HfApi(token=HF_TOKEN)
-        # List files in the repository
-        files = api.list_repo_files(REPO_ID, repo_type=REPO_TYPE)
-        data_files = [f for f in files if f.startswith("data/bluesky_counts_")]
-
-        if not data_files:
-            return []
-
-        # Get the most recent file
-        latest_file = sorted(data_files)[-1]
-        # Download the file
-        local_path = hf_hub_download(
-            repo_id=REPO_ID, filename=latest_file, repo_type=REPO_TYPE, token=HF_TOKEN
-        )
-
-        # Load and parse the data
-        data = []
-        with open(local_path, "r") as f:
-            data.extend(json.loads(line.strip()) for line in f)
-        # Keep only last month of data
-        return data[-MONTH_IN_SECONDS:]
-    except Exception as e:
-        print(f"Error loading data from Hub: {e}")
-        return []
-
-
-# Initialize storage and Hub connection
-DATA_FOLDER.mkdir(exist_ok=True)
-scheduler = CommitScheduler(
-    repo_id=REPO_ID,
-    repo_type=REPO_TYPE,
-    folder_path=DATA_FOLDER,
-    path_in_repo="data",
-    every=600,  # Upload every 10 minutes
-    token=HF_TOKEN,  # Add token for authentication
-)
+# def load_hub_data():
+#     """Load the most recent data from the Hub"""
+#     try:
+#         api = HfApi(token=HF_TOKEN)
+#         # List files in the repository
+#         files = api.list_repo_files(REPO_ID, repo_type=REPO_TYPE)
+#         data_files = [f for f in files if f.startswith("data/bluesky_counts_")]
+
+#         if not data_files:
+#             return []
+
+#         # Get the most recent file
+#         latest_file = sorted(data_files)[-1]
+#         # Download the file
+#         local_path = hf_hub_download(
+#             repo_id=REPO_ID, filename=latest_file, repo_type=REPO_TYPE, token=HF_TOKEN
+#         )
+
+#         # Load and parse the data
+#         data = []
+#         with open(local_path, "r") as f:
+#             data.extend(json.loads(line.strip()) for line in f)
+#         # Keep only last month of data
+#         return data[-MONTH_IN_SECONDS:]
+#     except Exception as e:
+#         print(f"Error loading data from Hub: {e}")
+#         return []
+
+
+# # Initialize storage and Hub connection
+# DATA_FOLDER.mkdir(exist_ok=True)
+# scheduler = CommitScheduler(
+#     repo_id=REPO_ID,
+#     repo_type=REPO_TYPE,
+#     folder_path=DATA_FOLDER,
+#     path_in_repo="data",
+#     every=600,  # Upload every 10 minutes
+#     token=HF_TOKEN,  # Add token for authentication
+# )
 
 
 def on_message_handler(message):
@@ -98,17 +98,17 @@ def emit_counts():
     """Emit post counts every second"""
     global post_count
 
-    if saved_data := load_hub_data():
-        print(f"Loaded {len(saved_data)} historical data points from Hub")
-        # Emit historical data
-        for point in saved_data[-100:]:  # Emit last 100 points to initialize plot
-            df = pd.DataFrame(
-                {
-                    "timestamp": [pd.Timestamp(point["timestamp"])],
-                    "post_count": [point["post_count"]],
-                }
-            )
-            stream.emit(df)
+    # if saved_data := load_hub_data():
+    #     print(f"Loaded {len(saved_data)} historical data points from Hub")
+    #     # Emit historical data
+    #     for point in saved_data[-100:]:  # Emit last 100 points to initialize plot
+    #         df = pd.DataFrame(
+    #             {
+    #                 "timestamp": [pd.Timestamp(point["timestamp"])],
+    #                 "post_count": [point["post_count"]],
+    #             }
+    #         )
+    #         stream.emit(df)
 
     # Wait for first second to collect initial data
     time.sleep(1)