Added window functions

Co-Authored-By: Sanjay Viswanathan <[email protected]>

Author: Adez017

docs/sql/SQL-Advance/window-functions.md

@@ -414,247 +414,6 @@ N FOLLOWING            -- N rows after current
        </TabItem>
     </Tabs>
 
-## Real-World Use Cases
-
-:::success
-**Business Analytics & Reporting**
-
-1. **Sales Performance Dashboard**
-```sql
-WITH daily_metrics AS (
-    SELECT 
-        sale_date,
-        salesperson_id,
-        salesperson_name,
-        region,
-        SUM(sale_amount) AS daily_sales
-    FROM sales
-    WHERE sale_date >= '2024-01-01'
-    GROUP BY sale_date, salesperson_id, salesperson_name, region
-)
-SELECT 
-    sale_date,
-    salesperson_name,
-    region,
-    daily_sales,
-    -- Regional context
-    AVG(daily_sales) OVER (PARTITION BY region, sale_date) AS region_avg_today,
-    RANK() OVER (PARTITION BY region, sale_date ORDER BY daily_sales DESC) AS daily_region_rank,
-    -- Personal trends
-    LAG(daily_sales) OVER (PARTITION BY salesperson_id ORDER BY sale_date) AS yesterday_sales,
-    daily_sales - LAG(daily_sales) OVER (PARTITION BY salesperson_id ORDER BY sale_date) AS day_over_day_change,
-    -- Running metrics
-    SUM(daily_sales) OVER (
-        PARTITION BY salesperson_id 
-        ORDER BY sale_date
-        ROWS UNBOUNDED PRECEDING
-    ) AS ytd_sales,
-    AVG(daily_sales) OVER (
-        PARTITION BY salesperson_id 
-        ORDER BY sale_date
-        ROWS BETWEEN 6 PRECEDING AND CURRENT ROW
-    ) AS rolling_7day_avg
-FROM daily_metrics
-ORDER BY sale_date DESC, region, daily_sales DESC;
-```
-
-2. **Customer Segmentation**
-```sql
-WITH customer_stats AS (
-    SELECT 
-        customer_id,
-        customer_name,
-        COUNT(*) AS total_orders,
-        SUM(order_amount) AS lifetime_value,
-        MAX(order_date) AS last_order_date,
-        MIN(order_date) AS first_order_date,
-        DATEDIFF(CURRENT_DATE, MAX(order_date)) AS days_since_last_order
-    FROM orders
-    GROUP BY customer_id, customer_name
-)
-SELECT 
-    customer_id,
-    customer_name,
-    lifetime_value,
-    total_orders,
-    days_since_last_order,
-    -- Value segmentation
-    NTILE(5) OVER (ORDER BY lifetime_value DESC) AS value_quintile,
-    -- Recency scoring
-    NTILE(5) OVER (ORDER BY days_since_last_order) AS recency_score,
-    -- Frequency ranking
-    PERCENT_RANK() OVER (ORDER BY total_orders) AS frequency_percentile,
-    -- Overall ranking
-    RANK() OVER (ORDER BY lifetime_value DESC) AS customer_rank,
-    -- Compare to average
-    lifetime_value - AVG(lifetime_value) OVER () AS value_vs_avg,
-    CASE 
-        WHEN NTILE(4) OVER (ORDER BY lifetime_value DESC) = 1 
-             AND days_since_last_order < 90 THEN 'Champion'
-        WHEN NTILE(4) OVER (ORDER BY lifetime_value DESC) = 1 THEN 'At Risk VIP'
-        WHEN days_since_last_order < 30 THEN 'Active'
-        WHEN days_since_last_order > 180 THEN 'Churned'
-        ELSE 'Regular'
-    END AS segment
-FROM customer_stats;
-```
-
-3. **Inventory Management**
-```sql
-SELECT 
-    product_name,
-    stock_date,
-    quantity_sold,
-    current_stock,
-    reorder_point,
-    -- Moving average demand
-    AVG(quantity_sold) OVER (
-        PARTITION BY product_name 
-        ORDER BY stock_date
-        ROWS BETWEEN 29 PRECEDING AND CURRENT ROW
-    ) AS avg_daily_demand_30d,
-    -- Days until reorder needed
-    CASE 
-        WHEN AVG(quantity_sold) OVER (
-            PARTITION BY product_name 
-            ORDER BY stock_date
-            ROWS BETWEEN 29 PRECEDING AND CURRENT ROW
-        ) > 0 THEN
-            FLOOR(current_stock / AVG(quantity_sold) OVER (
-                PARTITION BY product_name 
-                ORDER BY stock_date
-                ROWS BETWEEN 29 PRECEDING AND CURRENT ROW
-            ))
-        ELSE NULL
-    END AS days_of_stock_remaining,
-    -- Stock status
-    CASE 
-        WHEN current_stock < reorder_point THEN 'REORDER NOW'
-        WHEN current_stock < reorder_point * 1.5 THEN 'Low Stock'
-        ELSE 'Adequate'
-    END AS stock_status
-FROM inventory_daily
-ORDER BY product_name, stock_date DESC;
-```
-:::
-
-## Performance Tips
-
-:::warning
-**Window Function Performance Considerations:**
-
-1. **Index Your Ordered Columns**
-```sql
--- If you frequently order by sale_date:
-CREATE INDEX idx_sales_date ON sales(sale_date);
-
--- For partitioned queries:
-CREATE INDEX idx_sales_dept_date ON sales(department, sale_date);
-```
-
-2. **Use OVER () Wisely**
-```sql
--- Bad: Recalculating same window multiple times
-SELECT 
-    SUM(amount) OVER (PARTITION BY dept ORDER BY date),
-    AVG(amount) OVER (PARTITION BY dept ORDER BY date),
-    COUNT(*) OVER (PARTITION BY dept ORDER BY date)
-FROM sales;
-
--- Good: Use WINDOW clause (PostgreSQL)
-SELECT 
-    SUM(amount) OVER w,
-    AVG(amount) OVER w,
-    COUNT(*) OVER w
-FROM sales
-WINDOW w AS (PARTITION BY dept ORDER BY date);
-
--- Or use CTE to calculate once
-WITH dept_totals AS (
-    SELECT dept, SUM(amount) as total
-    FROM sales
-    GROUP BY dept
-)
-SELECT s.*, dt.total
-FROM sales s
-JOIN dept_totals dt ON s.dept = dt.dept;
-```
-
-3. **Limit Your Windows**
-```sql
--- Avoid if possible: UNBOUNDED FOLLOWING forces scanning entire partition
-LAST_VALUE(x) OVER (ORDER BY date ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)
-
--- Better: Use subquery if you need the last value
-```
-
-4. **Filter Before Windowing**
-```sql
--- Good: Filter reduces rows before window function
-SELECT 
-    ROW_NUMBER() OVER (PARTITION BY dept ORDER BY salary DESC) as rn,
-    *
-FROM employees
-WHERE hire_date >= '2020-01-01'  -- Filter first
-  AND status = 'Active';
-```
-:::
-
-## Common Mistakes to Avoid
-
-:::danger
-**Pitfall #1: Wrong Frame with LAST_VALUE**
-```sql
--- Wrong: This gives current row, not last row!
-LAST_VALUE(salary) OVER (PARTITION BY dept ORDER BY hire_date)
-
--- Correct: Need to specify the full frame
-LAST_VALUE(salary) OVER (
-    PARTITION BY dept 
-    ORDER BY hire_date
-    ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING
-)
-```
-
-**Pitfall #2: Using WHERE with Window Functions**
-```sql
--- Wrong: Window functions can't be in WHERE clause
-SELECT * FROM sales
-WHERE ROW_NUMBER() OVER (ORDER BY amount) <= 10;
-
--- Correct: Use CTE or subquery
-WITH ranked AS (
-    SELECT *, ROW_NUMBER() OVER (ORDER BY amount DESC) as rn
-    FROM sales
-)
-SELECT * FROM ranked WHERE rn <= 10;
-```
-
-**Pitfall #3: Forgetting ORDER BY for Sequential Functions**
-```sql
--- Wrong: LAG without ORDER BY is meaningless
-LAG(amount) OVER (PARTITION BY customer_id)
-
--- Correct: Must specify order
-LAG(amount) OVER (PARTITION BY customer_id ORDER BY order_date)
-```
-
-**Pitfall #4: PARTITION BY vs GROUP BY Confusion**
-```sql
--- GROUP BY: Collapses rows
-SELECT department, COUNT(*), AVG(salary)
-FROM employees
-GROUP BY department;  -- Returns one row per department
-
--- PARTITION BY: Keeps all rows
-SELECT 
-    employee_name,
-    department,
-    salary,
-    AVG(salary) OVER (PARTITION BY department) as dept_avg
-FROM employees;  -- Returns every employee with dept average
-```
-:::
 
 ## Window Functions Quick Reference
 

sidebars.ts

@@ -151,6 +151,7 @@ const sidebars: SidebarsConfig = {
           items: [
             'sql/SQL-Advance/sql-subqueries',
             'sql/SQL-Advance/common-table-expressions',
+            'sql/SQL-Advance/window-functions',
           ],
         },
       ],