Choose which rows MERGE updates

TutorialApplies to: Unified engine deployments

Update selected fields and apply conditions without changing the meaning of a match. Keep the business key in ON. Put action rules such as a newer source version in WHEN MATCHED AND. A row that fails ON is an unmatched source row and may qualify for INSERT. Updating one column leaves other target columns unchanged. Expressions can read both source and target values.

Before you start

You need a writable Iceberg catalog with namespace creation, table creation, and MERGE support. These examples create format-v2 tables in a private namespace. They use inline data and do not need an external sample dataset. Choose an unused namespace. The examples use learn_merge_02. Replace it everywhere if it already exists. In Workbench, select your writable catalog before running any SQL. Keep that catalog selected throughout the lesson. Opening an example does not execute it.

Set up the namespace

Run this setup once:

SET spark.sql.caseSensitive = true;
CREATE NAMESPACE learn_merge_02;

Each example creates a separate table. Run an example’s setup once. To start it again, remove only that example’s table first, using the cleanup section. Scripts execute as separate operations; they are not a multi-statement transaction.

1. Change balance while preserving the customer name

Start with these target rows:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

CREATE TABLE learn_merge_02.selected_columns (id BIGINT, name STRING, balance BIGINT, version BIGINT, is_deleted BOOLEAN)
USING iceberg TBLPROPERTIES ('format-version' = '2');
INSERT INTO learn_merge_02.selected_columns VALUES (1, 'Ada', 100, 1, false), (2, 'Ben', 200, 2, false), (3, 'Cyd', 300, 1, false);

Apply the incoming source below:

MERGE INTO learn_merge_02.selected_columns AS t
USING (SELECT 2 AS id, 'Incoming' AS name, 250 AS balance, 3 AS version, false AS is_deleted) AS s
ON t.id = s.id
WHEN MATCHED THEN UPDATE SET balance = s.balance;

Check all rows:

SELECT id, name, balance, version, is_deleted FROM learn_merge_02.selected_columns;

Expect these exact rows, in any display order:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	250	2	false
3	Cyd	300	1	false

2. Ignore an older customer version

Start with these target rows:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

CREATE TABLE learn_merge_02.conditional_update (id BIGINT, name STRING, balance BIGINT, version BIGINT, is_deleted BOOLEAN)
USING iceberg TBLPROPERTIES ('format-version' = '2');
INSERT INTO learn_merge_02.conditional_update VALUES (1, 'Ada', 100, 1, false), (2, 'Ben', 200, 2, false), (3, 'Cyd', 300, 1, false);

Apply the incoming source below:

MERGE INTO learn_merge_02.conditional_update AS t
USING (SELECT 1 AS id, 'Old Ada' AS name, 50 AS balance, 0 AS version, false AS is_deleted UNION ALL SELECT 2, 'Ben', 250, 3, false) AS s
ON t.id = s.id
WHEN MATCHED AND s.version > t.version THEN UPDATE SET name = s.name, balance = s.balance, version = s.version, is_deleted = s.is_deleted
WHEN NOT MATCHED THEN INSERT (id,name,balance,version,is_deleted) VALUES (s.id,s.name,s.balance,s.version,s.is_deleted);

Check all rows:

SELECT id, name, balance, version, is_deleted FROM learn_merge_02.conditional_update;

Expect these exact rows, in any display order:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	250	3	false
3	Cyd	300	1	false

3. Keep the version condition in the matched action

Start with these target rows:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

CREATE TABLE learn_merge_02.on_versus_when (id BIGINT, name STRING, balance BIGINT, version BIGINT, is_deleted BOOLEAN)
USING iceberg TBLPROPERTIES ('format-version' = '2');
INSERT INTO learn_merge_02.on_versus_when VALUES (1, 'Ada', 100, 1, false), (2, 'Ben', 200, 2, false), (3, 'Cyd', 300, 1, false);

Apply the incoming source below:

MERGE INTO learn_merge_02.on_versus_when AS t
USING (SELECT 2 AS id, 'Older Ben' AS name, 50 AS balance, 1 AS version, false AS is_deleted) AS s
ON t.id = s.id
WHEN MATCHED AND s.version > t.version THEN UPDATE SET name = s.name, balance = s.balance, version = s.version, is_deleted = s.is_deleted
WHEN NOT MATCHED THEN INSERT (id,name,balance,version,is_deleted) VALUES (s.id,s.name,s.balance,s.version,s.is_deleted);

Check all rows:

SELECT id, name, balance, version, is_deleted FROM learn_merge_02.on_versus_when;

Expect these exact rows, in any display order:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

4. See why an ON filter can turn an old row into an insert

Start with these target rows:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

CREATE TABLE learn_merge_02.on_condition (id BIGINT, name STRING, balance BIGINT, version BIGINT, is_deleted BOOLEAN)
USING iceberg TBLPROPERTIES ('format-version' = '2');
INSERT INTO learn_merge_02.on_condition VALUES (1, 'Ada', 100, 1, false), (2, 'Ben', 200, 2, false), (3, 'Cyd', 300, 1, false);

Apply the incoming source below:

MERGE INTO learn_merge_02.on_condition AS t
USING (SELECT 2 AS id, 'Older Ben' AS name, 50 AS balance, 1 AS version, false AS is_deleted) AS s
ON t.id = s.id AND s.version > t.version
WHEN MATCHED THEN UPDATE SET name = s.name, balance = s.balance, version = s.version
WHEN NOT MATCHED THEN INSERT (id,name,balance,version,is_deleted) VALUES (s.id,s.name,s.balance,s.version,s.is_deleted);

Check all rows:

SELECT id, name, balance, version, is_deleted FROM learn_merge_02.on_condition;

Expect these exact rows, in any display order:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false
2	Older Ben	50	1	false

Customer 2 now appears twice. The older version did not match the ON condition, so the INSERT clause added it. This is the reason to keep the version guard in the action.

5. Calculate a changed value from target and source

Start with these target rows:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	200	2	false
3	Cyd	300	1	false

CREATE TABLE learn_merge_02.expression (id BIGINT, name STRING, balance BIGINT, version BIGINT, is_deleted BOOLEAN)
USING iceberg TBLPROPERTIES ('format-version' = '2');
INSERT INTO learn_merge_02.expression VALUES (1, 'Ada', 100, 1, false), (2, 'Ben', 200, 2, false), (3, 'Cyd', 300, 1, false);

Apply the incoming source below:

MERGE INTO learn_merge_02.expression AS t
USING (SELECT 2 AS id, 'Ben' AS name, 25 AS balance, 3 AS version, false AS is_deleted) AS s
ON t.id = s.id
WHEN MATCHED THEN UPDATE SET balance = t.balance + s.balance;

Check all rows:

SELECT id, name, balance, version, is_deleted FROM learn_merge_02.expression;

Expect these exact rows, in any display order:

id	name	balance	version	is_deleted
1	Ada	100	1	false
2	Ben	225	2	false
3	Cyd	300	1	false

Change one thing

Start the relevant example again from its original target rows. Drop only that example table, rerun its CREATE TABLE and INSERT statements, then apply the changed MERGE. Keep the lesson namespace. In the selected-columns example, also assign name = s.name. Customer 2 should now be named Incoming. All other rows should stay unchanged.

Check the result

Compare every row and field with the expected tables, including unchanged customers and the number of copies of each key. A Succeeded status alone is not the result check. Wait for confirmed publication before reading the target.

Clean up

Remove only the private tables created by this lesson:

DROP TABLE IF EXISTS learn_merge_02.selected_columns;
DROP TABLE IF EXISTS learn_merge_02.conditional_update;
DROP TABLE IF EXISTS learn_merge_02.on_versus_when;
DROP TABLE IF EXISTS learn_merge_02.on_condition;
DROP TABLE IF EXISTS learn_merge_02.expression;
DROP NAMESPACE learn_merge_02;

Continue

Combine MERGE actions or review the MERGE SQL reference.

Transformation Engine

Learn Laminar

Workloads

SQL reference

HTTP API

Recipes

Operations

Choose which rows MERGE updates

Before you start

Set up the namespace

1. Change balance while preserving the customer name

2. Ignore an older customer version

3. Keep the version condition in the matched action

4. See why an ON filter can turn an old row into an insert

5. Calculate a changed value from target and source

Change one thing

Check the result

Clean up

Continue

​Before you start

​Set up the namespace

​1. Change balance while preserving the customer name

​2. Ignore an older customer version

​3. Keep the version condition in the matched action

​4. See why an ON filter can turn an old row into an insert

​5. Calculate a changed value from target and source

​Change one thing

​Check the result

​Clean up

​Continue

Before you start

Set up the namespace

1. Change balance while preserving the customer name

2. Ignore an older customer version

3. Keep the version condition in the matched action

4. See why an ON filter can turn an old row into an insert

5. Calculate a changed value from target and source

Change one thing

Check the result

Clean up

Continue