Update app.py

app.py

@@ -2,8 +2,10 @@ import pandas as pd
 import gradio as gr
 import plotly.express as px
 
-# Helper function for categorization
+# --- 1. Helper Functions ---
+
 def categorize_rank(r):
+    """Categorizes the rank into SEO buckets."""
     if pd.isna(r) or r == 0 or r == "NA":
         return "Not Talking (Intent Not Correct)"
     elif r <= 3: return "01–03"
@@ -15,21 +17,17 @@ def categorize_rank(r):
     elif r <= 100: return "51–100"
     else: return "100+"
 
-def process_data(file):
-    # Returns empty updates if no file is uploaded
-    if file is None:
-        return None, None, None, gr.Dropdown(choices=[]), gr.Dropdown(choices=[])
-    
-    excel = pd.ExcelFile(file.name)
-    first_sheet = excel.sheet_names[0]
-    df = pd.read_excel(file.name, sheet_name=first_sheet)
-    
-    # 1. Clean Columns
+def clean_and_process_sheet(df):
+    """
+    Takes a raw dataframe from a single sheet, cleans columns, 
+    identifies dates, and adds Rank Range columns.
+    """
+    # Clean Headers
     df.columns = df.columns.map(lambda x: str(x).strip())
     non_date_cols = ["Keyword", "Avg. monthly searches"]
     available_non_date = [c for c in non_date_cols if c in df.columns]
     
-    # 2. Rename Date Columns
+    # Identify Date Columns
     potential_date_cols = [col for col in df.columns if col not in non_date_cols and "Unnamed" not in col]
     rename_dict = {}
     for col in potential_date_cols:
@@ -43,31 +41,51 @@ def process_data(file):
     df.rename(columns=rename_dict, inplace=True)
     date_cols = list(rename_dict.values())
     
-    # 3. Build Visualization Data
-    final_df = df[available_non_date].copy()
-    viz_data = []
-
+    # Build Processed DataFrame
+    processed_df = df[available_non_date].copy()
+    
     for col in date_cols:
         range_col_name = f"{col}_Rank Range"
         raw_vals = pd.to_numeric(df[col], errors="coerce")
-        final_df[f"{col}_Rank_Numeric"] = raw_vals.fillna(0).astype(int)
-        final_df[range_col_name] = raw_vals.apply(categorize_rank)
+        processed_df[f"{col}_Rank_Numeric"] = raw_vals.fillna(0).astype(int)
+        processed_df[range_col_name] = raw_vals.apply(categorize_rank)
         
-        counts = final_df[range_col_name].value_counts().reset_index()
+    return processed_df
+
+def build_chart_and_filters(df):
+    """
+    Generates the Plotly figure and filter choices for a specific dataframe.
+    """
+    if df is None or df.empty:
+        return None, [], []
+
+    # 1. Prepare Data for Chart
+    viz_data = []
+    # Identify all date columns by looking for "_Rank Range"
+    range_cols = [c for c in df.columns if "_Rank Range" in c]
+    
+    for range_col in range_cols:
+        # Extract date from column name (e.g., "27_Oct_2025_Rank Range" -> "27_Oct_2025")
+        month_name = range_col.replace("_Rank Range", "")
+        
+        counts = df[range_col].value_counts().reset_index()
         counts.columns = ['Range', 'Keyword Count']
-        counts['Month'] = col
+        counts['Month'] = month_name
         viz_data.append(counts)
 
     full_viz_df = pd.concat(viz_data) if viz_data else pd.DataFrame()
     
-    # 4. FIX SORTING
+    # 2. Define Sorting Order
     correct_order = [
         "01–03", "04–10", "11–20", "21–30", "31–40", 
         "41–50", "51–100", "100+", "Not Talking (Intent Not Correct)"
     ]
     
-    # 5. Create Chart
+    fig = None
+    unique_months = []
+    
     if not full_viz_df.empty:
+        unique_months = list(full_viz_df['Month'].unique())
         fig = px.bar(
             full_viz_df, 
             x='Range', 
@@ -75,79 +93,154 @@ def process_data(file):
             color='Month',
             barmode='group', # Side-by-side bars
             text_auto=True,
-            title="Keyword Performance Distribution (Grouped)",
+            title="Keyword Performance Distribution",
             category_orders={"Range": correct_order}
         )
-    else:
-        fig = None
     
-    # Prepare Dropdown Data
-    unique_months = list(full_viz_df['Month'].unique()) if not full_viz_df.empty else []
+    return fig, unique_months, correct_order
+
+
+# --- 2. Gradio Event Functions ---
+
+def process_upload_initial(file):
+    """
+    Reads ALL sheets, stores them in State, and renders the FIRST sheet.
+    """
+    if file is None:
+        return None, None, gr.Dropdown(choices=[]), None, None, gr.Dropdown(choices=[]), gr.Dropdown(choices=[])
+    
+    excel = pd.ExcelFile(file.name)
+    all_sheets_data = {}
+    
+    # Process every sheet and store in dictionary
+    for sheet_name in excel.sheet_names:
+        raw_df = pd.read_excel(file.name, sheet_name=sheet_name)
+        all_sheets_data[sheet_name] = clean_and_process_sheet(raw_df)
+    
+    sheet_names = list(all_sheets_data.keys())
+    first_sheet_name = sheet_names[0]
+    first_df = all_sheets_data[first_sheet_name]
+    
+    # Generate View for First Sheet
+    fig, months, ranges = build_chart_and_filters(first_df)
     
-    # FIX: Return new component instances instead of .update()
     return (
-        final_df, 
-        fig, 
-        final_df, 
-        gr.Dropdown(choices=unique_months, value=unique_months[0] if unique_months else None),
-        gr.Dropdown(choices=correct_order, value=correct_order[0])
+        all_sheets_data,                                      # State: All Data
+        first_df,                                             # State: Current Sheet Data
+        gr.Dropdown(choices=sheet_names, value=first_sheet_name), # Sheet Selector
+        fig,                                                  # Chart
+        first_df,                                             # Table
+        gr.Dropdown(choices=months, value=months[0] if months else None), # Month Filter
+        gr.Dropdown(choices=ranges, value=ranges[0] if ranges else None)  # Range Filter
     )
 
-def filter_table(full_data, selected_month, selected_range):
-    if full_data is None or full_data.empty: return None
+def change_sheet(selected_sheet, all_data):
+    """
+    Switches the view when a new sheet is selected from dropdown.
+    """
+    if not selected_sheet or not all_data:
+        return None, None, None, gr.Dropdown(choices=[]), gr.Dropdown(choices=[])
+        
+    new_df = all_data[selected_sheet]
+    fig, months, ranges = build_chart_and_filters(new_df)
+    
+    return (
+        new_df, # Update Current Sheet State
+        fig,    # Update Chart
+        new_df, # Update Table
+        gr.Dropdown(choices=months, value=months[0] if months else None), # Update Month Choices
+        gr.Dropdown(choices=ranges, value=ranges[0] if ranges else None)  # Update Range Choices
+    )
+
+def filter_table(current_df, selected_month, selected_range):
+    """
+    Filters the CURRENTLY active sheet's dataframe.
+    """
+    if current_df is None or current_df.empty: return None
     
-    # Filter by specific month AND range
     target_col = f"{selected_month}_Rank Range"
     
-    if target_col in full_data.columns:
-        filtered_df = full_data[full_data[target_col] == selected_range]
+    if target_col in current_df.columns:
+        filtered_df = current_df[current_df[target_col] == selected_range]
         return filtered_df
     
-    return full_data
+    return current_df
+
+def reset_table(current_df):
+    return current_df
 
-def reset_table(full_data):
-    return full_data
 
-# Gradio Interface
-with gr.Blocks() as demo:
-    full_data_state = gr.State()
+# --- 3. UI Layout ---
 
-    gr.Markdown("# 🚀 SEO Market Intelligence Dashboard")
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    # STATES
+    all_sheets_state = gr.State({}) # Stores DICTIONARY of all sheets {name: df}
+    current_sheet_state = gr.State(pd.DataFrame()) # Stores currently visible sheet DF
+
+    gr.Markdown("# 🚀 SEO Multi-Sheet Dashboard")
     
+    # TOP ROW: Upload & Sheet Selection
     with gr.Row():
-        file_input = gr.File(label="Upload SEO Excel", file_types=[".xlsx"])
-        process_btn = gr.Button("📊 Build Dashboard", variant="primary")
+        file_input = gr.File(label="Upload Excel (Multiple Sheets Supported)", file_types=[".xlsx"])
+        with gr.Column():
+            process_btn = gr.Button("📊 Load File", variant="primary")
+            # NEW: Sheet Selector Dropdown
+            sheet_dropdown = gr.Dropdown(label="📑 Select Sheet to Analyze", choices=[], interactive=True)
 
-    # The Chart
+    # VISUALIZATION
     plot_output = gr.Plot(label="Keyword Distribution")
     
-    gr.Markdown("### 🔍 Filter Keywords")
+    # FILTER ROW
+    gr.Markdown("### 🔍 Filter Data")
     with gr.Row():
         month_dropdown = gr.Dropdown(label="Select Month", choices=[])
         range_dropdown = gr.Dropdown(label="Select Rank Range", choices=[])
         filter_btn = gr.Button("Apply Filter")
-        reset_btn = gr.Button("Show All")
+        reset_btn = gr.Button("Show All Rows")
 
     table_output = gr.DataFrame(interactive=False)
 
-    # 1. Process File
+    # --- EVENTS ---
+    
+    # 1. Upload & Process (Loads all sheets, defaults to first)
     process_btn.click(
-        fn=process_data,
+        fn=process_upload_initial,
         inputs=file_input,
-        outputs=[table_output, plot_output, full_data_state, month_dropdown, range_dropdown]
+        outputs=[
+            all_sheets_state,    # Save all sheets
+            current_sheet_state, # Save current sheet
+            sheet_dropdown,      # Update dropdown options
+            plot_output,         # Show chart
+            table_output,        # Show table
+            month_dropdown,      # Update filters
+            range_dropdown
+        ]
+    )
+    
+    # 2. Change Sheet (User selects "RD Monthly" etc.)
+    sheet_dropdown.change(
+        fn=change_sheet,
+        inputs=[sheet_dropdown, all_sheets_state],
+        outputs=[
+            current_sheet_state,
+            plot_output,
+            table_output,
+            month_dropdown,
+            range_dropdown
+        ]
     )
 
-    # 2. Filter Table
+    # 3. Filter Data (Applied to current sheet)
     filter_btn.click(
         fn=filter_table,
-        inputs=[full_data_state, month_dropdown, range_dropdown],
+        inputs=[current_sheet_state, month_dropdown, range_dropdown],
         outputs=table_output
     )
 
-    # 3. Reset Table
+    # 4. Reset Data
     reset_btn.click(
         fn=reset_table,
-        inputs=full_data_state,
+        inputs=current_sheet_state,
         outputs=table_output
     )