Automate Oracle to Microsoft Fabric Migration in Minutes

/* Oracle PL/SQL Iterative Cursor */
CREATE OR REPLACE PROCEDURE calculate_discounts AS
  CURSOR c_orders IS 
    SELECT order_id, customer_id, total_amount, NVL(discount_pct, 0) as pct
    FROM orders 
    WHERE order_status = 'COMPLETED';
BEGIN
  FOR r IN c_orders LOOP
    UPDATE orders_summary
    SET net_amount = r.total_amount * (1 - (r.pct / 100))
    WHERE order_id = r.order_id;
  END LOOP;
  COMMIT;
END calculate_discounts;

# Microsoft Fabric Vectorized PySpark Transformation
from pyspark.sql.functions import col, coalesce, when

df_orders = spark.read.table("orders").filter(col("order_status") == "COMPLETED")

df_summary = df_orders.withColumn(
    "net_amount",
    col("total_amount") * (1.0 - (coalesce(col("discount_pct"), col("lit(0)")) / 100.0))
)
df_summary.write.format("delta").mode("overwrite").saveAsTable("orders_summary")

/* Oracle DECODE & NVL Aggregation */
SELECT 
  department_id,
  NVL(manager_id, 0) AS mgr_id,
  SUM(DECODE(employee_type, 'FULL_TIME', salary * 1.1, salary)) AS adjusted_payroll
FROM hr.employees
GROUP BY department_id, NVL(manager_id, 0)
HAVING SUM(salary) > 50000;

-- Microsoft Fabric Spark SQL / Synapse T-SQL
SELECT 
  department_id,
  COALESCE(manager_id, 0) AS mgr_id,
  SUM(CASE WHEN employee_type = 'FULL_TIME' THEN salary * 1.1 ELSE salary END) AS adjusted_payroll
FROM hr.employees
GROUP BY department_id, COALESCE(manager_id, 0)
HAVING SUM(salary) > 50000;

/* Oracle Package Body */
CREATE OR REPLACE PACKAGE BODY pkg_financial_metrics AS
  FUNCTION get_annual_run_rate(p_cust_id IN NUMBER, p_year IN NUMBER) 
  RETURN NUMBER IS
    v_total NUMBER;
  BEGIN
    SELECT SUM(amount) INTO v_total 
    FROM fact_revenue 
    WHERE customer_id = p_cust_id AND fiscal_year = p_year;
    RETURN NVL(v_total, 0) * 1.05;
  END get_annual_run_rate;
END pkg_financial_metrics;

# Microsoft Fabric Modular PySpark Routine
from pyspark.sql.functions import col, sum as _sum, coalesce, lit

def get_annual_run_rate(df_revenue, customer_id: int, fiscal_year: int) -> float:
    """Calculates customer annual run rate directly in OneLake."""
    filtered_df = df_revenue.filter(
        (col("customer_id") == customer_id) & (col("fiscal_year") == fiscal_year)
    )
    result = filtered_df.select(_sum("amount")).collect()[0][0]
    return (result or 0.0) * 1.05