Add troubleshooting functionality with Gradio interface

- Implemented the run_troubleshooting function to process sensor and threshold data.
- Added data loading from Supabase if dataframes are not defined.
- Created a Gradio interface to display troubleshooting information based on alerts.
- Refactored alert processing logic to improve clarity and maintainability.

app.py

@@ -68,48 +68,76 @@ def check_thresholds(sensor_df, threshold_df):
 
     return pd.DataFrame(alerts)
 
-# 実行
-alerts_df = check_thresholds(sensor_df, threshold_df)
-
-# 同じタイムスタンプでdata no.が二つ以上あるデータを抽出
-grouped_alerts = alerts_df.groupby('timestamp')['data no.'].nunique()
-multiple_data_nos_timestamps = grouped_alerts[grouped_alerts > 1].index.tolist()
-
-# alerts_df を複数データ番号を持つタイムスタンプでフィルタリング
-filtered_alerts_df = alerts_df[alerts_df['timestamp'].isin(multiple_data_nos_timestamps)]
-
-# タイムスタンプでグループ化し、各タイムスタンプのユニークなデータ番号をリストとして取得
-data_nos_by_timestamp = filtered_alerts_df.groupby('timestamp')['data no.'].unique().apply(list)
-
-# 結果を辞書のリストに変換（オプション。分かりやすさのため）
-result_list = []
-for timestamp, data_nos in data_nos_by_timestamp.items():
-    # 数値のリストをカンマ区切りの文字列に変換
-    data_nos_str = ', '.join(map(str, data_nos))
-    result_list.append({"timestamp": timestamp, "data_nos": data_nos_str})
-
-# 辞書のリストをpandas DataFrameに変換
-result_df = pd.DataFrame(result_list)
-
-
-# troubleshooting_dfの指標No.列をリストのリストに変換
-# 例: '1,2' -> [1, 2]
-troubleshooting_indicator_lists = troubleshooting_df['指標No.'].str.split(',').apply(lambda x: [int(i) for i in x])
-
-# result_dfのdata_nos列をリストのリストに変換
-# 例: '5, 25' -> [5, 25]
-result_data_nos_lists = result_df['data_nos'].str.split(', ').apply(lambda x: [int(i) for i in x])
-
-# 組み合わせを持つかどうか判定し、該当する情報を出力
-for i, result_nos in enumerate(result_data_nos_lists):
-    result_timestamp = result_df.loc[i, 'timestamp']
-    for j, troubleshooting_nos in enumerate(troubleshooting_indicator_lists):
-        # troubleshooting_nos の組み合わせが result_nos の組み合わせに含まれるか判定
-        if set(troubleshooting_nos).issubset(set(result_nos)):
-            troubleshooting_situation = troubleshooting_df.loc[j, 'シチュエーション\n（対応が必要な状況）']
-            troubleshooting_action = troubleshooting_df.loc[j, 'sub goal到達のために必要な行動\n（解決策）']
-
-            print(f"Timestamp: {result_timestamp}")
-            print(f"Situation: {troubleshooting_situation}")
-            print(f"Action: {troubleshooting_action}")
-            print("-" * 20) # 区切り線
+
+import gradio as gr
+import pandas as pd
+import supabase
+
+# Assuming the data loading and check_thresholds function from the previous cell are available
+
+def run_troubleshooting():
+    # Ensure dataframes are accessible, assuming they are global or loaded here
+    # If not global, you would need to reload or pass them as arguments
+    try:
+        threshold_df
+        sensor_df
+        troubleshooting_df
+    except NameError:
+        # Reload data if dataframes are not defined
+        SUPABASE_URL = userdata.get('SUPABASE_URL')
+        SUPABASE_KEY = userdata.get('SUPABASE_KEY')
+        supabase_client = supabase.create_client(SUPABASE_URL, SUPABASE_KEY)
+        table_threshold = "Threshold_data"
+        table_sensor = "Sensor_data"
+        table_troubleshooting = "Troubleshooting_collection"
+        threshold_data = supabase_client.table(table_threshold).select("*").execute()
+        sensor_data = supabase_client.table(table_sensor).select("*").execute()
+        troubleshooting_data = supabase_client.table(table_troubleshooting).select("*").execute()
+        threshold_df = pd.DataFrame(threshold_data.data)
+        sensor_df = pd.DataFrame(sensor_data.data)
+        troubleshooting_df = pd.DataFrame(troubleshooting_data.data)
+
+
+    alerts_df = check_thresholds(sensor_df, threshold_df)
+
+    grouped_alerts = alerts_df.groupby('timestamp')['data no.'].nunique()
+    multiple_data_nos_timestamps = grouped_alerts[grouped_alerts > 1].index.tolist()
+
+    filtered_alerts_df = alerts_df[alerts_df['timestamp'].isin(multiple_data_nos_timestamps)]
+
+    data_nos_by_timestamp = filtered_alerts_df.groupby('timestamp')['data no.'].unique().apply(list)
+
+    result_list = []
+    for timestamp, data_nos in data_nos_by_timestamp.items():
+        data_nos_str = ', '.join(map(str, data_nos))
+        result_list.append({"timestamp": timestamp, "data_nos": data_nos_str})
+
+    result_df = pd.DataFrame(result_list)
+
+    troubleshooting_indicator_lists = troubleshooting_df['指標No.'].str.split(',').apply(lambda x: [int(i) for i in x])
+    result_data_nos_lists = result_df['data_nos'].str.split(', ').apply(lambda x: [int(i) for i in x])
+
+    output_text = ""
+    for i, result_nos in enumerate(result_data_nos_lists):
+        result_timestamp = result_df.loc[i, 'timestamp']
+        for j, troubleshooting_nos in enumerate(troubleshooting_indicator_lists):
+            if set(troubleshooting_nos).issubset(set(result_nos)):
+                troubleshooting_situation = troubleshooting_df.loc[j, 'シチュエーション\n（対応が必要な状況）']
+                troubleshooting_action = troubleshooting_df.loc[j, 'sub goal到達のために必要な行動\n（解決策）']
+
+                output_text += f"Timestamp: {result_timestamp}\n"
+                output_text += f"Situation: {troubleshooting_situation}\n"
+                output_text += f"Action: {troubleshooting_action}\n"
+                output_text += "-" * 20 + "\n" # 区切り線
+
+    return output_text
+
+iface = gr.Interface(
+    fn=run_troubleshooting,
+    inputs=None,  # No direct input needed as it uses existing dataframes
+    outputs="text",
+    title="Troubleshooting Information",
+    description="Displays troubleshooting information based on sensor and threshold data."
+)
+
+iface.launch(debug=True)