Update app.py

app.py

@@ -96,32 +96,38 @@ def run_asset_analysis(symbols, interval):
         logger.warning("No data available for any of the selected asset parameters.")
         return pd.DataFrame(), None
 
-def filter_latest_signals(df, signal_column):
+def filter_latest_signals(df, signal_column, interval):
     if df.empty:
         logger.warning("Empty dataframe passed to filter_latest_signals")
         return df
     
     logger.info(f"Filtering signals. Input shape: {df.shape}")
     
-    # Group by Symbol and get the latest date for each
-    latest_dates = df.groupby('Symbol')['Date'].max().reset_index()
+    # Determine the date range based on the interval
+    today = datetime.now().date()
+    if interval == '1d':
+        start_date = today - timedelta(days=15)  # Last week for daily interval
+    elif interval == '1wk':
+        start_date = today - timedelta(days=60)  # Last month for weekly interval
+    else:
+        start_date = today - timedelta(days=15)  # Default to last week
     
-    # Merge with the original dataframe to get the latest signals
-    latest_signals = pd.merge(df, latest_dates, on=['Symbol', 'Date'])
+    # Filter the dataframe for the specified date range
+    df_filtered = df[df['Date'] >= start_date]
     
-    logger.info(f"Latest signals shape after merge: {latest_signals.shape}")
+    logger.info(f"Filtered for date range. Shape: {df_filtered.shape}")
     
     # Filter for the chosen signal
     if signal_column != 'All':
-        latest_signals = latest_signals[latest_signals[signal_column] != '']
-        logger.info(f"Filtered for {signal_column}. Shape: {latest_signals.shape}")
+        df_filtered = df_filtered[df_filtered[signal_column] != '']
+        logger.info(f"Filtered for {signal_column}. Shape: {df_filtered.shape}")
     else:
         # Remove rows where all signal columns are empty
         signal_columns = ['Signal_1x', 'Signal_2x', 'Signal_3x', 'VuManchu_Signal']
-        latest_signals = latest_signals[latest_signals[signal_columns].ne('').any(axis=1)]
-        logger.info(f"Removed rows with all empty signals. Shape: {latest_signals.shape}")
+        df_filtered = df_filtered[df_filtered[signal_columns].ne('').any(axis=1)]
+        logger.info(f"Removed rows with all empty signals. Shape: {df_filtered.shape}")
     
-    result = latest_signals.sort_values('Date', ascending=False)
+    result = df_filtered.sort_values('Date', ascending=False)
     logger.info(f"Final filtered result shape: {result.shape}")
     return result
 
@@ -136,7 +142,7 @@ def generate_signals(analysis_type, symbols, interval):
         
         if isinstance(full_results, pd.DataFrame) and not full_results.empty:
             logger.info(f"Analysis result shape: {full_results.shape}")
-            filtered_result = filter_latest_signals(full_results, 'All')
+            filtered_result = filter_latest_signals(full_results, 'All', interval)
             logger.info(f"Filtered result shape: {filtered_result.shape}")
             return filtered_result, output_file_path
         else:
@@ -151,7 +157,7 @@ def apply_filter(signal_filter):
     if full_results is None or full_results.empty:
         return "No data available. Please generate signals first.", None
     
-    filtered_result = filter_latest_signals(full_results, signal_filter)
+    filtered_result = filter_latest_signals(full_results, signal_filter, interval)
     return filtered_result, output_file_path
 
 with gr.Blocks() as iface:
@@ -176,7 +182,7 @@ with gr.Blocks() as iface:
     Use of this tool constitutes acceptance of this disclaimer and an acknowledgment of the inherent risks associated with trading and investing.
     """)
     
-    gr.Markdown("Perform technical analysis on cryptocurrencies or stocks using the VuManchu swing trading strategy and SuperTrend indicators. Select the analysis type, input desired symbols or use the defaults, choose the time interval, and view or download the generated trading signals. The table shows the latest trading signals for each symbol, excluding rows with no signals.")
+    gr.Markdown("Perform technical analysis on cryptocurrencies or stocks using the VuManchu swing trading strategy and SuperTrend indicators. Select the analysis type, input desired symbols or use the defaults, choose the time interval, and view or download the generated trading signals. The table shows the trading signals for the last week (1d interval) or last month (1wk interval) for each symbol, excluding rows with no signals.")
     
     with gr.Row():
         analysis_type = gr.Radio(["Cryptocurrency", "Asset"], label="Select Analysis Type")
@@ -184,11 +190,11 @@ with gr.Blocks() as iface:
         interval = gr.Radio(["1d", "1wk"], label="Select Time Interval")
     
     with gr.Row():
-        signal_filter = gr.Dropdown(["All", "Signal_1x", "Signal_2x", "Signal_3x", "VuManchu_Signal"], label="Filter by Latest Signal", value="All")
+        signal_filter = gr.Dropdown(["All", "Signal_1x", "Signal_2x", "Signal_3x", "VuManchu_Signal"], label="Filter by Signal", value="All")
     
     generate_button = gr.Button("Generate Signals")
     
-    output_dataframe = gr.Dataframe(label="Latest Trading Signals")
+    output_dataframe = gr.Dataframe(label="Trading Signals")
     output_file = gr.File(label="Download Full Signals CSV")
     
     generate_button.click(