Fix revenue distribution chart: prioritize positive values and use Booking Amount

streamlit_app/app.py

@@ -474,12 +474,23 @@ if page == "📊 Overview":
     
     with col2:
         st.subheader("Revenue Distribution")
+        
+        # Prioritize certain columns for revenue
+        priority_cols = ['Total Amount', 'Booking Amount', 'Received Amount']
         revenue_candidates = [
             col for col in df_filtered.columns
             if any(k in col.lower() for k in ['amount', 'revenue', 'total', 'rate', 'price', 'cost'])
         ]
+        
+        # Sort by priority
+        revenue_candidates = sorted(
+            revenue_candidates,
+            key=lambda x: (priority_cols.index(x) if x in priority_cols else len(priority_cols))
+        )
+        
         best_col = None
         best_score = 0
+        
         for col in revenue_candidates:
             cleaned = (
                 df_filtered[col]
@@ -487,8 +498,11 @@ if page == "📊 Overview":
                 .str.replace(r'[^0-9.-]', '', regex=True)
             )
             vals = pd.to_numeric(cleaned, errors='coerce')
-            score = vals.notna().sum()
-            if score > best_score and vals.sum(skipna=True) > 0:
+            # Filter to positive values only (revenue should be > 0)
+            positive_vals = vals[vals > 0]
+            score = len(positive_vals)
+            
+            if score > best_score and positive_vals.sum() > 0:
                 best_score = score
                 best_col = col
 
@@ -498,16 +512,38 @@ if page == "📊 Overview":
                 .astype(str)
                 .str.replace(r'[^0-9.-]', '', regex=True)
             )
-            revenue_values = pd.to_numeric(cleaned, errors='coerce').dropna()
-            if not revenue_values.empty:
-                fig = px.histogram(
-                    revenue_values,
-                    nbins=30,
-                    title="Revenue Distribution",
-                    color_discrete_sequence=['#636EFA']
-                )
-                fig.update_layout(height=400, xaxis_title=best_col, yaxis_title="Count")
-                st.plotly_chart(fig, use_container_width=True)
+            revenue_values = pd.to_numeric(cleaned, errors='coerce')
+            # Filter to positive values only
+            revenue_values = revenue_values[revenue_values > 0].dropna()
+            
+            if not revenue_values.empty and len(revenue_values) > 5:
+                try:
+                    # Create histogram
+                    fig = px.histogram(
+                        revenue_values,
+                        nbins=min(30, len(revenue_values)//5),
+                        title=f"Revenue Distribution - {best_col}",
+                        color_discrete_sequence=['#636EFA'],
+                        labels={'value': best_col, 'count': 'Count'}
+                    )
+                    fig.update_layout(
+                        height=400,
+                        xaxis_title=best_col,
+                        yaxis_title="Count",
+                        hovermode='x unified'
+                    )
+                    fig.update_traces(marker=dict(line=dict(width=0)))
+                    st.plotly_chart(fig, use_container_width=True)
+                except Exception as e:
+                    st.warning(f"Could not create histogram: {e}")
+                    # Fallback: show summary stats
+                    st.metric("Average Revenue", f"₹{revenue_values.mean():,.0f}")
+                    st.metric("Median Revenue", f"₹{revenue_values.median():,.0f}")
+                    st.metric("Max Revenue", f"₹{revenue_values.max():,.0f}")
+            elif revenue_values.empty:
+                st.info(f"No positive revenue values found in {best_col}")
+            else:
+                st.info("Insufficient data for distribution chart")
             else:
                 st.info("No revenue values available for the current filters.")
         else: