MERGE

Important Notice: Migration of Documentation Website

Please be advised that our documentation website is currently undergoing a migration to a new platform. To ensure you have access to the most up-to-date information, we kindly request that you visit our new documentation website located at:

Official Snowflake Snowconvert Documentation

For any immediate assistance or if you encounter any issues, please contact our support team at [email protected].

Thank you for your understanding.

Grammar Syntax

MERGE INTO target_table 
USING source_table [ [ AS ] alias ] 
ON match_condition 
[ WHEN MATCHED THEN { UPDATE SET col_name = { expr } [,...] | DELETE }
WHEN NOT MATCHED THEN INSERT [ ( col_name [,...] ) ] VALUES ( { expr } [, ...] ) |
REMOVE DUPLICATES ]

For more information please refer to Redshift MERGE documentation.

Sample Source Patterns

UPDATE - INSERT

There are no differences between both languages. The code is kept in its original form.

Input Code:

Redshift

IN -> Redshift_01.sql

MERGE INTO target USING source ON target.id = source.id
WHEN MATCHED THEN UPDATE SET id = source.id, name = source.name
WHEN NOT MATCHED THEN INSERT VALUES (source.id, source.name);

Output Code:

Snowflake

OUT -> Redshift_01.sql

--** SSC-FDM-RS0005 - REDSHIFT MERGE STATEMENT DOESN'T ALLOW DUPLICATES IN THE SOURCE TABLE. SNOWFLAKE BEHAVIOR MAY DIFFER IF THERE ARE DUPLICATE VALUES. **
MERGE INTO target USING source ON target.id = source.id
WHEN MATCHED THEN UPDATE SET id = source.id, name = source.name
WHEN NOT MATCHED THEN INSERT VALUES (source.id, source.name);

DELETE - INSERT

There are no differences between both languages. The code is kept in its original form.

Input Code:

Redshift

IN -> Redshift_02.sql

MERGE INTO target USING source ON target.id = source.id
WHEN MATCHED THEN DELETE
WHEN NOT MATCHED THEN INSERT VALUES (source.id, source.name);

Output Code:

Snowflake

OUT -> Redshift_02.sql

--** SSC-FDM-RS0005 - REDSHIFT MERGE STATEMENT DOESN'T ALLOW DUPLICATES IN THE SOURCE TABLE. SNOWFLAKE BEHAVIOR MAY DIFFER IF THERE ARE DUPLICATE VALUES. **
MERGE INTO target USING source ON target.id = source.id
WHEN MATCHED THEN DELETE
WHEN NOT MATCHED THEN INSERT VALUES (source.id, source.name);

REMOVE DUPLICATES

The REMOVE DUPLICATES clause is not supported in Snowflake, however, there is a workaround that could emulate the original behavior.

The output code will have three new statements:

• A TEMPORARY TABLE with the duplicate values from the source and target table that matches the condition

• An INSERT statement that adds the pending values to the target table after the merge

• A DROP statement that drops the generated temporary table.

These are necessary since the DROP DUPLICATES behavior removes the duplicate values from the target table and then inserts the values that match the condition from the source table.

Input Code:

Redshift

IN -> Redshift_03.sql

CREATE TABLE target (id INT, name CHAR(10));
CREATE TABLE source (id INT, name CHAR(10));

INSERT INTO target VALUES (30, 'Tony'), (30, 'Daisy'), (11, 'Alice'), (23, 'Bill'), (23, 'Nikki');
INSERT INTO source VALUES (23, 'David'), (22, 'Clarence');

MERGE INTO target USING source ON target.id = source.id REMOVE DUPLICATES;

Results

Output Code:

Snowflake

OUT -> Redshift_03.sql

CREATE TABLE target (id INT, name CHAR(10))
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "03/03/2025",  "domain": "test" }}';

CREATE TABLE source (id INT, name CHAR(10))
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "03/03/2025",  "domain": "test" }}';

INSERT INTO target
VALUES (30, 'Tony'), (30, 'Daisy'), (11, 'Alice'), (23, 'Bill'), (23, 'Nikki');

INSERT INTO source
VALUES (23, 'David'), (22, 'Clarence');

CREATE TEMPORARY TABLE source_duplicates AS
SELECT DISTINCT
source.*
FROM
source
INNER JOIN
target
ON target.id = source.id;
--** SSC-FDM-RS0005 - REDSHIFT MERGE STATEMENT DOESN'T ALLOW DUPLICATES IN THE SOURCE TABLE. SNOWFLAKE BEHAVIOR MAY DIFFER IF THERE ARE DUPLICATE VALUES. **
MERGE INTO target
USING source ON target.id = source.id
WHEN MATCHED THEN
DELETE
WHEN NOT MATCHED THEN
INSERT
VALUES (source.id, source.name);
INSERT INTO target

SELECT
*
FROM
source_duplicates;

DROP TABLE IF EXISTS source_duplicates CASCADE;

Results

Known Issues

There are no known issues.

Related EWIs

1. SSC-EWI-RS0009: Semantic information not found for the source table.

2. SSC-FDM-RS0005: Duplicates not allowed in source table.