Merge pull request #355 from kapilbk1996/SQL-187

SQL-187:- Deleting One of Two Identical Rows in SQL

Author: maibin

sql-queries-11/delete-identical-rows/create-table.sql

@@ -0,0 +1,16 @@
+-- Create sample table
+CREATE TABLE DuplicateRecords (
+RecordID SERIAL PRIMARY KEY, -- Use INT AUTO_INCREMENT for MySQL
+Value1 VARCHAR(50),
+Value2 INT,
+Value3 DATE
+);
+
+-- Insert sample data with duplicates
+INSERT INTO DuplicateRecords (Value1, Value2, Value3) VALUES
+('Apple', 10, '2023-01-01'),
+('Banana', 20, '2023-02-01'),
+('Apple', 10, '2023-01-01'), -- Identical to the first row (logical duplicate)
+('Cherry', 30, '2023-03-01'),
+('Banana', 20, '2023-02-01'), -- Identical to the second row (logical duplicate)
+('Apple', 10, '2023-01-01'); -- Another identical row

sql-queries-11/delete-identical-rows/using-delete-with-self-join.sql

@@ -0,0 +1,26 @@
+-- PostgreSQL
+DELETE FROM DuplicateRecords DR1
+USING DuplicateRecords DR2
+WHERE
+    DR1.Value1 = DR2.Value1 AND
+    DR1.Value2 = DR2.Value2 AND
+    DR1.Value3 = DR2.Value3 AND
+    DR1.RecordID > DR2.RecordID;
+
+-- MySQL
+DELETE FROM DR1
+USING DuplicateRecords DR1
+JOIN DuplicateRecords DR2 ON
+    DR1.Value1 = DR2.Value1 AND
+    DR1.Value2 = DR2.Value2 AND
+    DR1.Value3 = DR2.Value3 AND
+    DR1.RecordID > DR2.RecordID;
+
+-- MS SQL Server
+DELETE DR1
+FROM DuplicateRecords DR1
+JOIN DuplicateRecords DR2 
+    ON DR1.Value1 = DR2.Value1 
+    AND DR1.Value2 = DR2.Value2 
+    AND DR1.Value3 = DR2.Value3 
+    AND DR1.RecordID > DR2.RecordID;

sql-queries-11/delete-identical-rows/using-min-max-subquery.sql

@@ -0,0 +1,17 @@
+-- PostgreSQL / MS SQL Server
+DELETE FROM DuplicateRecords
+WHERE RecordID NOT IN (
+    SELECT MIN(RecordID)
+    FROM DuplicateRecords
+    GROUP BY Value1, Value2, Value3
+);
+
+-- MySQL
+DELETE FROM DuplicateRecords
+WHERE RecordID NOT IN (
+    SELECT T2.MinRecordID FROM (
+        SELECT MIN(RecordID) AS MinRecordID
+        FROM DuplicateRecords
+        GROUP BY Value1, Value2, Value3
+    ) AS T2
+);

sql-queries-11/delete-identical-rows/using-row-number.sql

@@ -0,0 +1,13 @@
+-- PostgreSQL / MS SQL Server / MySQL Server
+WITH CTE_DuplicateRecords AS (
+    SELECT
+        RecordID,
+        Value1,
+        Value2,
+        Value3,
+        ROW_NUMBER() OVER (PARTITION BY Value1, Value2, Value3 ORDER BY RecordID) as rn
+    FROM
+        DuplicateRecords
+)
+DELETE FROM DuplicateRecords
+WHERE RecordID IN (SELECT RecordID FROM CTE_DuplicateRecords WHERE rn > 1);