initial commit

generate_sample_data.py

@@ -0,0 +1,43 @@
+import duckdb
+import pandas as pd
+from datetime import datetime, timedelta
+import random
+
+def generate_data():
+    riders = ["Rider Alpha", "Rider Beta", "Rider Gamma", "Rider Delta"]
+    tenants = ["Tenant A", "Tenant B", "Tenant C"]
+    statuses = ["completed", "pending", "cancelled"]
+    
+    data = []
+    for i in range(50):
+        rider = random.choice(riders)
+        tenant = random.choice(tenants)
+        order_amt = round(random.uniform(20.0, 100.0), 2)
+        delivery_amt = round(order_amt * 0.1, 2)
+        delivery_charges = round(random.uniform(2.0, 5.0), 2)
+        status = random.choice(statuses)
+        
+        pickup = datetime.now() - timedelta(minutes=random.randint(10, 1000))
+        # ridertime in minutes
+        rider_time = random.randint(15, 45)
+        delivery = pickup + timedelta(minutes=rider_time)
+        
+        data.append({
+            "ridername": rider,
+            "tenantname": tenant,
+            "orderamount": order_amt,
+            "deliveryamt": delivery_amt,
+            "deliverycharges": delivery_charges,
+            "orderstatus": status,
+            "pickuptime": pickup.isoformat(),
+            "deliverytime": delivery.isoformat(),
+            "ridertime": rider_time  # explicitly adding this column for the user's example
+        })
+    
+    df = pd.DataFrame(data)
+    conn = duckdb.connect()
+    conn.execute("COPY df TO 'c:/bitbucket/backup/analytics/ai-service/data/deliveries.parquet' (FORMAT PARQUET)")
+    print("Generated c:/bitbucket/backup/analytics/ai-service/data/deliveries.parquet")
+
+if __name__ == "__main__":
+    generate_data()