Update app.py

app.py

@@ -75,26 +75,38 @@ def build_report():
 
     scanned = 0
     mappable = 0
+    skipped_session_start = 0
+    missing_country = 0
+    invalid_country_code = 0
 
     for row in load_rows_streaming():
         scanned += 1
+    
+        # 1) Skip session starts
         event_type = str(row.get("event", "") or "").strip().lower()
         if event_type == "session_start":
+            skipped_session_start += 1
             continue
+    
+        # 2) Missing country
         country = normalize_country_name(row.get("final_country"))
         if not country:
+            missing_country += 1
             continue
-
-        country_counts[country] += 1
-
+    
+        # 3) Invalid / missing country code
         iso3 = iso2_to_iso3(row.get("final_country_code"))
         if not iso3:
+            invalid_country_code += 1
             continue
 
+        country_counts[country] += 1
+
         iso3_counts[iso3] += 1
         iso3_to_name.setdefault(iso3, country)
         mappable += 1
 
+
     # Table dataframe (country name + usage events)
     table_df = (
         pd.DataFrame([{"country": k, "usage events": v} for k, v in country_counts.items()])
@@ -205,9 +217,22 @@ def build_report():
         font=dict(size=20),
     )
 
+    accounted = (
+        skipped_session_start
+        + missing_country
+        + invalid_country_code
+        + mappable
+    )
+
     summary = (
-        f"Rows scanned: {scanned:,} • Rows mappable: {mappable:,} • "
-        f"Countries (table): {len(table_df):,} • Countries (map): {len(map_df):,} • "
+        f"Rows scanned: {scanned:,}\n"
+        f"- Session starts skipped: {skipped_session_start:,}\n"
+        f"- Missing country: {missing_country:,}\n"
+        f"- Invalid country code: {invalid_country_code:,}\n"
+        f"- Rows mappable: {mappable:,}\n\n"
+        f"Accounted rows: {accounted:,} / {scanned:,}\n"
+        f"Countries (table): {len(table_df):,}\n"
+        f"Countries (map): {len(map_df):,}\n"
         f"Total usage events: {int(table_df['usage events'].sum()) if len(table_df) else 0:,}"
     )