SnowConvert AI – Mensagens gerais de análise de desempenho¶

SSC-PRF-0001¶

Esta instrução faz uso de operações de busca em massa de cursor.

Descrição¶

Esse aviso indica que a instrução usa operações em massa de busca de cursor. Essas operações permitem que você recupere várias linhas de dados de um cursor de uma só vez, em vez de uma linha de cada vez. O uso de operações em massa melhora o desempenho ao reduzir o número de comunicações necessárias entre o cliente e o servidor.

Esse padrão pode se tornar complexo se não for implementado corretamente. Por exemplo, a recuperação de muitas linhas em uma única operação de busca pode consumir muita memória. É fundamental manter um equilíbrio entre o número de linhas buscadas e os recursos de memória disponíveis.

Exemplo de código¶

Oracle¶
Entrada¶
 CREATE OR REPLACE PROCEDURE oracle_cursor_fetch_bulk AS
--cursor and variable declarations
BEGIN
    OPEN c1;
    FETCH c1 BULK COLLECT INTO col1;
    CLOSE c1;
END;

Saída¶
 CREATE OR REPLACE PROCEDURE oracle_cursor_fetch_bulk ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"oracle"}}'
EXECUTE AS CALLER
AS
--cursor and variable declarations
$$
    BEGIN
        OPEN c1;
        --** SSC-PRF-0001 - THIS STATEMENT HAS USAGES OF CURSOR FETCH BULK OPERATIONS **
        c1 := (
            CALL FETCH_BULK_COLLECTION_RECORDS_UDF(:c1)
        );
        col1 := :c1:RESULT;
        CLOSE c1;
    END;
$$;

Práticas recomendadas¶

Para obter suporte adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0002¶

Colunas que não diferenciam maiúsculas de minúsculas podem diminuir o desempenho das consultas.

Descrição¶

O uso de agrupamento no Snowflake pode afetar o desempenho da consulta, principalmente nas cláusulas WHERE. Para saber mais sobre como o agrupamento afeta o desempenho, consulte Implicações do uso do agrupamento para o desempenho.

Um aviso foi gerado para indicar que uma coluna foi criada com agrupamento sem distinção entre maiúsculas e minúsculas. O uso dessa coluna em consultas pode causar um desempenho mais lento.

Exemplos de código¶

Saída¶

 CREATE TABLE exampleTable
(
    col1 CHAR(10),
    col2 CHAR(20) COLLATE 'en-ci' /*** SSC-PRF-0002 - CASE INSENSITIVE COLUMNS CAN DECREASE THE PERFORMANCE OF QUERIES ***/
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"teradata"}}'
;

Oracle¶

Entrada¶

 CREATE TABLE exampleTable (
    col1 VARCHAR(50) COLLATE BINARY_CI,
    col2 VARCHAR(50) COLLATE BINARY_CS
);

Saída¶

 CREATE OR REPLACE TABLE exampleTable (
       col1 VARCHAR(50) COLLATE BINARY_CI /*** SSC-PRF-0002 - CASE INSENSITIVE COLUMNS CAN DECREASE THE PERFORMANCE OF QUERIES ***/,
       col2 VARCHAR(50) COLLATE BINARY_CS
   )
   COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"oracle"}}'
   ;

Microsoft SQL Server¶

Entrada¶

 CREATE TABLE exampleTable (
    col1 VARCHAR(50) COLLATE Latin1_General_CI_AS,
    col2 VARCHAR(50) COLLATE Latin1_General_CS_AS
);

Saída¶

 CREATE OR REPLACE TABLE exampleTable (
    col1 VARCHAR(50) COLLATE 'EN-CI-AS' /*** SSC-PRF-0002 - CASE INSENSITIVE COLUMNS CAN DECREASE THE PERFORMANCE OF QUERIES ***/,
    col2 VARCHAR(50) COLLATE 'EN-CS-AS'
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},"attributes":{"component":"transact"}}'
;

Práticas recomendadas¶

Se o desempenho do seu aplicativo for significativamente afetado pelo agrupamento sem distinção entre maiúsculas e minúsculas, considere a possibilidade de reescrever o código para evitar seu uso. Entretanto, se o impacto no desempenho for aceitável, você pode ignorar esse aviso.
Para obter assistência adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0003¶

A busca dentro de um loop é considerada um padrão complexo, o que pode degradar o desempenho do Snowflake.

Gravidade¶

Low

Descrição¶

Esse aviso aparece quando uma instrução FETCH é detectada em um loop. A instrução FETCH recupera e processa linhas individuais de um conjunto de resultados, uma de cada vez.

O processamento de grandes conjuntos de dados usando cursores dentro de loops pode se tornar complexo, especialmente quando:

Estão envolvidas várias junções de tabelas
São necessários cálculos complexos
Um grande número de linhas precisa ser processado

Esse padrão pode levar a problemas de desempenho e pode ser difícil de manter à medida que o volume de dados aumenta.

Exemplo de código¶

Teradata¶

Entrada¶

 REPLACE PROCEDURE teradata_fetch_inside_loop()
DYNAMIC RESULT SETS 1
BEGIN
    DECLARE col_name VARCHAR(200);
    DECLARE col_int INTEGER DEFAULT 0;
    DECLARE cursor_var CURSOR FOR SELECT some_column FROM tabla1;
    WHILE (col_int <> 0) DO		
        FETCH cursor_var INTO col_name;
        SET col_int = col_int + 1;
    END WHILE;
END;

Saída¶

 --** SSC-FDM-0007 - MISSING DEPENDENT OBJECT "tabla1" **
CREATE OR REPLACE PROCEDURE teradata_fetch_inside_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "teradata",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
    DECLARE
        col_name VARCHAR(200);
        col_int INTEGER DEFAULT 0;
    BEGIN
         
         
        LET cursor_var CURSOR
        FOR
            SELECT
                some_column FROM
                tabla1;
                WHILE (:col_int <> 0) LOOP
            --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
                    FETCH cursor_var INTO col_name;
            col_int := col_int + 1;
                END LOOP;
    END;
$$;

Oracle¶

Entrada¶

 CREATE PROCEDURE oracle_fetch_inside_loop
IS
  var1 table1.column1%TYPE;
  CURSOR cursor1 IS SELECT COLUMN_NAME FROM table1; 
BEGIN
  WHILE true LOOP
    FETCH cursor1 INTO var1;
    EXIT WHEN cursor1%NOTFOUND;
  END LOOP;  
END;

Saída¶

 CREATE OR REPLACE PROCEDURE oracle_fetch_inside_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "oracle",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
  DECLARE
    var1 VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'table1.column1%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
    --** SSC-PRF-0009 - PERFORMANCE REVIEW - CURSOR USAGE **
    cursor1 CURSOR
    FOR
      SELECT COLUMN_NAME FROM
        table1;
  BEGIN
    WHILE (true)
                 --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
                 LOOP
      --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
         FETCH cursor1 INTO
        :var1;
         IF (var1 IS NULL) THEN
        EXIT;
         END IF;
       END LOOP;
  END;
$$;

SQL Server¶

Entrada¶

 CREATE OR ALTER PROCEDURE transact_fetch_inside_loop
AS
BEGIN
    DECLARE cursor1 CURSOR
        FOR SELECT col1 FROM my_table;
    WHILE 1=0
    BEGIN
       FETCH NEXT FROM @cursor1 INTO @variable1;
    END
END;

Saída¶

 CREATE OR REPLACE PROCEDURE transact_fetch_inside_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"transact"}}'
EXECUTE AS CALLER
AS
$$
    DECLARE
        --** SSC-FDM-TS0013 - SNOWFLAKE SCRIPTING CURSOR ROWS ARE NOT MODIFIABLE **
        cursor1 CURSOR
        FOR
            SELECT
                col1
            FROM
                my_table;
    BEGIN
         
        WHILE (1=0) LOOP
            --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
            FETCH
                CURSOR1
                INTO
                :VARIABLE1;
        END LOOP;
    END;
$$;

Práticas recomendadas¶

Para melhorar o desempenho e evitar padrões complexos, use operações baseadas em conjuntos em vez de loops. Substitua o processamento linha por linha por instruções SQL (SELECT, UPDATE, DELETE) que operam em várias linhas simultaneamente usando cláusulas WHERE. Essa abordagem é mais eficiente e mais fácil de manter.

Oracle¶

 CREATE OR REPLACE PROCEDURE cursor_fetch_inside_loop 
AS
  record_employee employees%rowtype;
  CURSOR emp_cursor IS SELECT * FROM employees;
BEGIN
  OPEN emp_cursor;
  LOOP
    FETCH emp_cursor INTO record_employee;
    EXIT WHEN emp_cursor%notfound;
    INSERT INTO new_employees VALUES (record_employee.first_name, record_employee.last_name);
  END LOOP;
  CLOSE emp_cursor;
END;

Snowflake¶

 CREATE OR REPLACE PROCEDURE cursor_fetch_inside_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "oracle",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
  DECLARE
    record_employee OBJECT !!!RESOLVE EWI!!! /*** SSC-EWI-0036 - ROWTYPE DATA TYPE CONVERTED TO OBJECT ***/!!! := OBJECT_CONSTRUCT();
    --** SSC-PRF-0009 - PERFORMANCE REVIEW - CURSOR USAGE **
    emp_cursor CURSOR
    FOR
      SELECT
        OBJECT_CONSTRUCT( *) sc_cursor_record FROM
        employees;
  BEGIN
    OPEN emp_cursor;
    --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
    LOOP
      --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
      FETCH emp_cursor INTO
        :record_employee;
      IF (record_employee IS NULL) THEN
        EXIT;
      END IF;
      INSERT INTO new_employees
      SELECT
        :record_employee:FIRST_NAME,
        :record_employee:LAST_NAME;
    END LOOP;
  CLOSE emp_cursor;
  END;
$$;

As operações baseadas em conjuntos podem ser usadas para processar dados com mais eficiência.

 CREATE OR REPLACE PROCEDURE cursor_fetch_inside_loop AS
BEGIN
  INSERT INTO new_employees (first_name, last_name)
  SELECT first_name, last_name FROM employees;
END;

As operações baseadas em conjuntos podem ser usadas para processar dados com mais eficiência.

 CREATE OR REPLACE PROCEDURE cursor_fetch_inside_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"oracle"}}'
EXECUTE AS CALLER
AS
$$
  BEGIN
    INSERT INTO new_employees(first_name, last_name)
    SELECT first_name, last_name FROM
      employees;
  END;
$$;

Práticas recomendadas¶

Para obter suporte adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0004¶

Esta instrução faz uso de cursor for loop.

Gravidade¶

Nenhum

Descrição¶

Esse aviso indica que a declaração contém loops do cursor for. Um loop de cursor for é uma estrutura de programação que processa os resultados da consulta uma linha de cada vez, permitindo que o usuário trabalhe com registros individuais de um conjunto de resultados.

Esse aviso ajuda a identificar possíveis problemas de desempenho nos loops do cursor FOR. Podem ocorrer problemas de desempenho quando:

A instrução SELECT dentro do cursor retorna um grande conjunto de dados
O loop contém operações complexas
O loop contém loops aninhados

Embora o SnowConvert AI possa detectar esses padrões, você deve revisar e otimizar o código para garantir uma execução eficiente.

Exemplo de código¶

Teradata¶

Entrada¶

 REPLACE PROCEDURE teradata_cursor_for_loop()    
BEGIN
    FOR fUsgClass AS cUsgClass CURSOR FOR
        (SELECT col1
        FROM sample_table)
    DO
        SET var1 = fUsgClass.col1;
    END FOR;
END;

Saída¶

 --** SSC-FDM-0007 - MISSING DEPENDENT OBJECT "sample_table" **
CREATE OR REPLACE PROCEDURE teradata_cursor_for_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "teradata",  "convertedOn": "07/16/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
    DECLARE
        !!!RESOLVE EWI!!! /*** SSC-EWI-0110 - TRANSFORMATION NOT PERFORMED DUE TO MISSING DEPENDENCIES ***/!!!
        temp_fUsgClass_col1;
    BEGIN
        LET cUsgClass CURSOR
        FOR
            SELECT
                col1
                   FROM
                sample_table;
        --** SSC-PRF-0004 - THIS STATEMENT HAS USAGES OF CURSOR FOR LOOP **
        FOR fUsgClass IN cUsgClass DO
            temp_fUsgClass_col1 := fUsgClass.col1;
            var1 := :temp_fUsgClass_col1;
        END FOR;
    END;
$$;

Oracle¶

Entrada¶

 CREATE OR REPLACE PROCEDURE oracle_cursor_for_loop AS
BEGIN
    FOR r1 IN (SELECT col1 FROM sample_table) LOOP
        NULL;
    END LOOP;
END;

Saída¶

 CREATE OR REPLACE PROCEDURE oracle_cursor_for_loop ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"oracle"}}'
EXECUTE AS CALLER
AS
$$
    BEGIN
        LET temporary_for_cursor_0 CURSOR
        FOR
            (SELECT col1 FROM
                    sample_table
            );
        --** SSC-PRF-0004 - THIS STATEMENT HAS USAGES OF CURSOR FOR LOOP **
        FOR r1 IN temporary_for_cursor_0 DO
            NULL;
        END FOR;
    END;
$$;

Práticas recomendadas¶

Para obter suporte adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0005¶

A instrução abaixo apresenta usos de cursores aninhados.

Nota

Para facilitar a leitura, simplificamos algumas seções do código neste exemplo.

Gravidade¶

Nenhum

Descrição¶

Esse aviso indica que a instrução contém cursores aninhados. Um cursor é um recurso do banco de dados que permite processar linhas de um resultado de consulta, uma de cada vez. Os cursores aninhados ocorrem quando o usuário usa um cursor dentro do loop de outro cursor, o que pode afetar o desempenho e deve ser usado com cautela.

Os cursores aninhados podem reduzir significativamente o desempenho do seu código, principalmente ao trabalhar com grandes quantidades de dados. Isso ocorre porque toda vez que um cursor opera, ele precisa se comunicar com o servidor de banco de dados, criando atrasos e sobrecarga de processamento adicionais.

Exemplos de código¶

SQL Server¶

Entrada¶

 CREATE OR ALTER PROCEDURE procedureSample
AS
BEGIN
  DECLARE
    @outer_category_id INT,
    @outer_category_name NVARCHAR(50),
    @inner_product_name NVARCHAR(50);

  -- Define the outer cursor
  DECLARE outer_cursor CURSOR FOR 
    SELECT category_id, category_name FROM categories;

  -- Open the outer cursor
  OPEN @outer_cursor;

  -- Fetch the first row from the outer cursor
  FETCH NEXT FROM outer_cursor INTO @outer_category_id, @outer_category_name;

  -- Start the outer loop
  WHILE @@FETCH_STATUS = 0
  BEGIN

    PRINT 'Category: ' + @outer_category_name;
	
    -- Define the inner cursor
    DECLARE inner_cursor CURSOR FOR
      SELECT product_name FROM products WHERE category_id = @outer_category_id;
    
    -- Open the inner cursor
    OPEN inner_cursor;
	FETCH NEXT FROM inner_cursor INTO @inner_product_name;

    WHILE @@FETCH_STATUS = 0
    BEGIN
      PRINT 'Product: ' + @inner_product_name + ' Category: ' + CAST(@outer_category_id AS NVARCHAR(10));

      -- Fetch the next row from the inner cursor
      FETCH NEXT FROM inner_cursor INTO @inner_product_name;
    END;

    -- Close the inner cursor
    CLOSE inner_cursor;
    DEALLOCATE inner_cursor;

    -- Fetch the next row from the outer cursor
    FETCH NEXT FROM outer_cursor INTO @outer_category_id, @outer_category_name;
  END;

  -- Close the outer cursor
  CLOSE outer_cursor;
  DEALLOCATE outer_cursor;
  
END;

Saída¶

 CREATE OR REPLACE PROCEDURE procedureSample ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},"attributes":{"component":"transact"}}'
EXECUTE AS CALLER
AS
$$
	DECLARE
		OUTER_CATEGORY_ID INT;
		OUTER_CATEGORY_NAME VARCHAR(50);
		INNER_PRODUCT_NAME VARCHAR(50);

		-- Define the outer cursor
		--** SSC-FDM-TS0013 - SNOWFLAKE SCRIPTING CURSOR ROWS ARE NOT MODIFIABLE **
		outer_cursor CURSOR
		FOR
			SELECT
				category_id,
				category_name
			FROM
				categories;

		-- Define the inner cursor
		--** SSC-FDM-TS0013 - SNOWFLAKE SCRIPTING CURSOR ROWS ARE NOT MODIFIABLE **
		inner_cursor CURSOR
		FOR
			SELECT
				product_name
			FROM
				products
			WHERE
				category_id = :OUTER_CATEGORY_ID;
	BEGIN
		 
		 

		-- Open the outer cursor
		--** SSC-PRF-0005 - THE STATEMENT BELOW HAS USAGES OF NESTED CURSORS. **
		OPEN OUTER_CURSOR;
  -- Fetch the first row from the outer cursor
		FETCH
			outer_cursor
			INTO
			:OUTER_CATEGORY_ID,
			:OUTER_CATEGORY_NAME;

			-- Start the outer loop

			  -- Define the inner cursor
			WHILE (:FETCH_STATUS = 0) LOOP
			!!!RESOLVE EWI!!! /*** SSC-EWI-0073 - PENDING FUNCTIONAL EQUIVALENCE REVIEW FOR 'PRINT' NODE ***/!!!

			  PRINT 'Category: ' + @outer_category_name;
			 

			-- Open the inner cursor
			OPEN inner_cursor;
			--** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
			FETCH
				inner_cursor
			INTO
				:INNER_PRODUCT_NAME;
			WHILE (:FETCH_STATUS = 0) LOOP
				!!!RESOLVE EWI!!! /*** SSC-EWI-0073 - PENDING FUNCTIONAL EQUIVALENCE REVIEW FOR 'PRINT' NODE ***/!!!
				PRINT 'Product: ' + @inner_product_name + ' Category: ' + CAST(@outer_category_id AS NVARCHAR(10));
				-- Fetch the next row from the inner cursor
				--** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
				FETCH
					inner_cursor
				INTO
					:INNER_PRODUCT_NAME;
			END LOOP;
			-- Close the inner cursor
			CLOSE inner_cursor;
			!!!RESOLVE EWI!!! /*** SSC-EWI-0058 - FUNCTIONALITY FOR 'DEALLOCATE' IS NOT CURRENTLY SUPPORTED BY SNOWFLAKE SCRIPTING ***/!!!
			  DEALLOCATE inner_cursor;
			-- Fetch the next row from the outer cursor
			--** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
			FETCH
				outer_cursor
			INTO
				:OUTER_CATEGORY_ID,
				:OUTER_CATEGORY_NAME;
			END LOOP;
  -- Close the outer cursor
			CLOSE outer_cursor;
			!!!RESOLVE EWI!!! /*** SSC-EWI-0058 - FUNCTIONALITY FOR 'DEALLOCATE' IS NOT CURRENTLY SUPPORTED BY SNOWFLAKE SCRIPTING ***/!!!
			DEALLOCATE outer_cursor;
	END;
$$;

Oracle¶

Cursor explícito¶

Entrada¶

 CREATE OR REPLACE PROCEDURE procedureSample AS
BEGIN
DECLARE
  CURSOR outer_cursor IS
    SELECT category_id, category_name FROM categories;

  CURSOR inner_cursor (p_category_id NUMBER) IS
    SELECT product_name FROM products WHERE category_id = p_category_id;

  outer_category_id categories.category_id%TYPE;
  outer_category_name categories.category_name%TYPE;
  inner_product_name products.product_name%TYPE;
BEGIN

  OPEN outer_cursor;
  FETCH outer_cursor INTO outer_category_id, outer_category_name;

  LOOP
    EXIT WHEN outer_cursor%NOTFOUND;
    DBMS_OUTPUT.PUT_LINE('Category: ' || outer_category_name);

    OPEN inner_cursor(outer_category_id);
    LOOP
        FETCH inner_cursor INTO inner_product_name;
        EXIT WHEN inner_cursor%NOTFOUND;
        DBMS_OUTPUT.PUT_LINE('Product: ' || inner_product_name || ' Category: ' || outer_category_id);
    END LOOP;
    CLOSE inner_cursor;

    FETCH outer_cursor INTO outer_category_id, outer_category_name;
  END LOOP;

  CLOSE outer_cursor;
END;
END;

Saída¶

 CREATE OR REPLACE PROCEDURE procedureSample ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "oracle",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
  BEGIN
    DECLARE
      --** SSC-PRF-0009 - PERFORMANCE REVIEW - CURSOR USAGE **
      outer_cursor CURSOR
      FOR
        SELECT category_id, category_name FROM
          categories;
      --** SSC-PRF-0009 - PERFORMANCE REVIEW - CURSOR USAGE **
      inner_cursor CURSOR
      FOR
        SELECT product_name FROM
          products
        WHERE category_id = ?;
      outer_category_id VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'categories.category_id%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
      outer_category_name VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'categories.category_name%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
      inner_product_name VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'products.PRODUCT_NAME%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
      call_results VARIANT;
    BEGIN
      --** SSC-PRF-0005 - THE STATEMENT BELOW HAS USAGES OF NESTED CURSORS. **
      OPEN outer_cursor USING ('DEFAULT VALUE NOT FOUND');
      FETCH outer_cursor INTO
        :outer_category_id,
        :outer_category_name;
      --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
      LOOP
        IF (outer_category_id IS NULL) THEN
          EXIT;
        END IF;
        --** SSC-FDM-OR0035 - CHECK UDF IMPLEMENTATION FOR DBMS_OUTPUT.PUT_LINE_UDF. **
        call_results := (
          CALL DBMS_OUTPUT.PUT_LINE_UDF('Category: ' || NVL(:outer_category_name :: STRING, ''))
        );
        OPEN inner_cursor USING (:outer_category_id);
        --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
        LOOP
          --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
            FETCH inner_cursor INTO
            :inner_product_name;
          IF (inner_product_name IS NULL) THEN
            EXIT;
          END IF;
          --** SSC-FDM-OR0035 - CHECK UDF IMPLEMENTATION FOR DBMS_OUTPUT.PUT_LINE_UDF. **
          call_results := (
            CALL DBMS_OUTPUT.PUT_LINE_UDF('Product: ' || NVL(:inner_product_name :: STRING, '') || ' Category: ' || NVL(:outer_category_id :: STRING, ''))
          );
        END LOOP;
        CLOSE inner_cursor;
        --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
        FETCH outer_cursor INTO
          :outer_category_id,
          :outer_category_name;
      END LOOP;
      CLOSE outer_cursor;
      RETURN call_results;
    END;
  END;
$$;

Cursor implícito¶

Entrada¶

 CREATE OR REPLACE PROCEDURE procedureSample AS
BEGIN
DECLARE
   inner_category_id categories.category_name%TYPE;
   inner_product_name products.product_name%TYPE;
   inner_cursor SYS_REFCURSOR;
BEGIN
   FOR outer_cursor IN (SELECT category_id, category_name FROM categories)
   LOOP
      OPEN inner_cursor
       FOR SELECT product_name, category_id FROM products WHERE category_id = outer_cursor.category_id;
      LOOP
         FETCH inner_cursor INTO inner_product_name, inner_category_id;
         EXIT WHEN inner_cursor%NOTFOUND;
         dbms_output.put_line( 'Category id: '|| outer_cursor.category_id);
         dbms_output.put_line('Product name: ' || inner_product_name);
      END LOOP;
      CLOSE inner_cursor;
   END LOOP;
END;
END;

Saída¶

 CREATE OR REPLACE PROCEDURE procedureSample ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "oracle",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
EXECUTE AS CALLER
AS
$$
   BEGIN
      DECLARE
         inner_category_id VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'categories.category_name%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
         inner_product_name VARIANT !!!RESOLVE EWI!!! /*** SSC-EWI-OR0129 - TYPE ATTRIBUTE 'products.product_name%TYPE' COULD NOT BE RESOLVED, SO IT WAS TRANSFORMED TO VARIANT ***/!!!;
         inner_cursor_res RESULTSET;
         call_results VARIANT;
      BEGIN
         LET temporary_for_cursor_0 CURSOR
         FOR
            (SELECT category_id, category_name FROM
                  categories
            );
         --** SSC-PRF-0004 - THIS STATEMENT HAS USAGES OF CURSOR FOR LOOP **
         --** SSC-PRF-0005 - THE STATEMENT BELOW HAS USAGES OF NESTED CURSORS. **
         FOR outer_cursor IN temporary_for_cursor_0 DO
            LET inner_cursor CURSOR
            FOR
               SELECT product_name, category_id FROM
                  products
               WHERE category_id = outer_cursor.category_id;
            OPEN inner_cursor;
            --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
                 LOOP
               --** SSC-PRF-0003 - FETCH INSIDE A LOOP IS CONSIDERED A COMPLEX PATTERN, THIS COULD DEGRADE SNOWFLAKE PERFORMANCE. **
                    FETCH inner_cursor INTO
                  :inner_product_name,
                  :inner_category_id;
               IF (inner_product_name IS NULL) THEN
                  EXIT;
               END IF;
               --** SSC-FDM-OR0035 - CHECK UDF IMPLEMENTATION FOR DBMS_OUTPUT.PUT_LINE_UDF. **
               call_results := (
                  CALL dbms_output.put_line( 'Category id: ' || NVL(outer_cursor.category_id :: STRING, ''))
               );
               --** SSC-FDM-OR0035 - CHECK UDF IMPLEMENTATION FOR DBMS_OUTPUT.PUT_LINE_UDF. **
               call_results := (
                  CALL dbms_output.put_line('Product name: ' || NVL(:inner_product_name :: STRING, ''))
               );
                 END LOOP;
                 CLOSE inner_cursor;
         END FOR;
         RETURN call_results;
      END;
   END;
$$;

Práticas recomendadas¶

Os cursores aninhados devem ser evitados, pois podem afetar negativamente o desempenho e tornar o código mais complexo.
Em vez de cursores aninhados, use recursos SQL, como:
- Funções SQL
- Junções
- Subconsultas
- Funções de janela
- Expressões de tabela comuns (CTEs)
- Consultas recursivas Essas alternativas processam dados em massa e são mais eficientes.
Para obter assistência adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0006¶

O cursor aninhado dentro de consulta não é compatível com o Snowflake.

Gravidade¶

Nenhum

Descrição¶

Essa mensagem aparece quando uma consulta contém uma definição de cursor. Quando uma expressão de cursor é avaliada, ela retorna e abre automaticamente um cursor aninhado. Para obter mais detalhes, consulte Expressão de cursor Oracle.

Exemplos de código¶

Entrada¶

 SELECT
  category_id,
  category_name,
  CURSOR (
    SELECT
      product_id,
      product_name || ', ' || category_id
    FROM
      products e
    WHERE
      e.category_id = d.category_id
  ) EMP_CUR
FROM
  categories d;

Saída¶

 SELECT
  category_id,
  category_name,
  --** SSC-PRF-0006 - NESTED CURSOR INSIDE QUERY IS NOT SUPPORTED IN SNOWFLAKE. **
  CURSOR
    !!!RESOLVE EWI!!! /*** SSC-EWI-0108 - THE FOLLOWING SUBQUERY MATCHES AT LEAST ONE OF THE PATTERNS CONSIDERED INVALID AND MAY PRODUCE COMPILATION ERRORS ***/!!! (
    SELECT
      product_id,
      NVL(
      product_name :: STRING, '') || ', ' || NVL(category_id :: STRING, '')
    FROM
      products e
    WHERE
      e.category_id = d.category_id
  ) EMP_CUR
FROM
  categories d;

Práticas recomendadas¶

Recomendamos evitar cursores, pois eles podem afetar negativamente o desempenho e tornar o código mais complexo.
Em vez de usar cursores aninhados, considere estas alternativas:
- Funções SQL
- Junções
- Subconsultas
- Funções de janela
- Expressões de tabela comuns (CTEs)
- Consultas recursivas Essas opções são melhores para processar grandes quantidades de dados com eficiência.
Para obter assistência adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0007¶

PERFORMANCE REVIEW - CLUSTER BY

Descrição¶

Indica onde o uso de CLUSTER BY pode causar problemas de desempenho.

Exemplo de código¶

Teradata:¶

 CREATE MULTISET TABLE T_2008,
NO FALLBACK,
NO BEFORE JOURNAL,
NO AFTER JOURNAL,
CHECKSUM = DEFAULT,
DEFAULT MERGEBLOCKRATIO
(
      COL1 NUMBER(20,0) NOT NULL,
      COL2 INTEGER,
      COL3 VARCHAR(4) CHARACTER SET LATIN NOT CASESPECIFIC,
      COL4 DATE FORMAT 'YYYY-MM-DD'
)
PRIMARY INDEX 
( 
      COL1, COL2
)
PARTITION BY ( RANGE_N(COL4 BETWEEN DATE '2010-01-01' AND DATE '2025-12-31' EACH INTERVAL '1' YEAR ),
CASE_N(
COL3  = 'T',
COL3 = 'M',
COL3 = 'L') ); -- PARTITION BY transformed to CLUSTER BY

Snowflake:¶

CREATE OR REPLACE TABLE T_2008
(
      COL1 NUMBER(20,0) NOT NULL,
      COL2 INTEGER,
      COL3 VARCHAR(4),
      COL4 DATE
)
--** SSC-PRF-0007 - PERFORMANCE REVIEW - CLUSTER BY **
CLUSTER BY (
             !!!RESOLVE EWI!!! /*** SSC-EWI-0031 - RANGE_N FUNCTION NOT SUPPORTED ***/!!!
             RANGE_N(COL4 BETWEEN DATE '2010-01-01' AND DATE '2025-12-31' EACH INTERVAL '1' YEAR ),
!!!RESOLVE EWI!!! /*** SSC-EWI-0031 - CASE_N FUNCTION NOT SUPPORTED ***/!!!
CASE_N(
COL3  = 'T',
COL3 = 'M',
COL3 = 'L'))
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "teradata",  "convertedOn": "07/02/2025",  "domain": "no-domain-provided" }}'
; -- PARTITION BY transformed to CLUSTER BY

Transact:¶

 CREATE TABLE my_table (
    enterprise_cif INT,
    name NVARCHAR(100),
    address NVARCHAR(255),
    created_at DATETIME
) 
WITH (
    DISTRIBUTION = HASH(enterprise_cif),
    CLUSTERED INDEX (enterprise_cif)
);

Snowflake:¶

 CREATE OR REPLACE TABLE my_table (
  enterprise_cif INT,
  name VARCHAR(100),
  address VARCHAR(255),
  created_at TIMESTAMP_NTZ(3)
)
--** SSC-PRF-0007 - PERFORMANCE REVIEW - CLUSTER BY **
CLUSTER BY (enterprise_cif)
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "transact",  "convertedOn": "10/09/2024" }}'
;

Práticas recomendadas¶

Revise o código para identificar possíveis problemas de desempenho. Mais informações sobre esse tópico podem ser encontradas aqui.
Se precisar de mais suporte, envie um e-mail para snowconvert-support@snowflake.com

SSC-PRF-0008¶

Descrição¶

This message appears when SnowConvert AI detects loop usage in procedural code. Loops such as LOOP, WHILE, and FOR can lead to row-by-row processing and may degrade performance in Snowflake, especially when the loop iterates over large datasets or contains complex logic. The message is informational and prompts a review of the pattern.

Exemplo de código¶

PostgreSQL:¶

CREATE OR REPLACE FUNCTION loop_example() RETURNS void AS $$
BEGIN
  FOR i IN 1..10 LOOP
    NULL;
  END LOOP;
END;
$$ LANGUAGE plpgsql;

Snowflake:¶

CREATE OR REPLACE PROCEDURE loop_example ()
RETURNS VARCHAR
LANGUAGE SQL
EXECUTE AS CALLER
AS
$$
  BEGIN
    --** SSC-PRF-0008 - PERFORMANCE REVIEW - LOOP USAGE **
    FOR i IN 1 TO 10
                     LOOP
      NULL;
    END LOOP;
  END;
$$;
;

Best practices¶

Prefer set-based SQL operations (SELECT, INSERT, UPDATE, DELETE) over row-by-row loops.
Avoid nested loops when possible; use joins, CTEs, or window functions instead.
If loops are required, keep iterations small and limit expensive operations inside the loop.
Consider refactoring procedural logic into single statements or bulk operations.

SSC-PRF-0009¶

CURSOR usage review

Gravidade¶

Nenhum

Descrição¶

This message appears when SnowConvert AI detects a cursor declaration in procedural code. Cursors allow row-by-row processing of query results, which can lead to performance issues in Snowflake, especially when processing large datasets.

While cursors are valid in Snowflake Scripting, they introduce overhead because:

Each row is processed individually rather than as a set
Multiple round trips to the database may be required
Memory usage can be higher compared to set-based operations

This warning is informational and prompts a review of whether the cursor usage is necessary or can be replaced with more efficient set-based operations.

Exemplo de código¶

Oracle¶

Entrada¶

CREATE OR REPLACE PROCEDURE get_first_employee AS
  CURSOR emp_cursor IS SELECT employee_id, first_name FROM employees;
  v_emp_id NUMBER;
  v_name VARCHAR2(100);
BEGIN
  OPEN emp_cursor;
  FETCH emp_cursor INTO v_emp_id, v_name;
  CLOSE emp_cursor;
END;

Saída¶

CREATE OR REPLACE PROCEDURE get_first_employee ()
RETURNS VARCHAR
LANGUAGE SQL
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"oracle"}}'
EXECUTE AS CALLER
AS
$$
  DECLARE
    --** SSC-PRF-0009 - PERFORMANCE REVIEW - CURSOR USAGE **
    emp_cursor CURSOR
    FOR
      SELECT employee_id, first_name FROM
        employees;
    v_emp_id NUMBER(38, 18);
    v_name VARCHAR(100);
  BEGIN
    OPEN emp_cursor;
    FETCH emp_cursor INTO
      :v_emp_id,
      :v_name;
    CLOSE emp_cursor;
  END;
$$;

Práticas recomendadas¶

Replace cursor-based row-by-row processing with set-based SQL operations (SELECT, INSERT, UPDATE, DELETE) whenever possible.
Use JOINs, subqueries, CTEs (Common Table Expressions), or window functions instead of cursors to process multiple rows efficiently.
If cursors are unavoidable, minimize the work done inside the cursor loop and avoid nested cursors.
Consider using MERGE statements for upsert operations instead of cursor-based conditional INSERT/UPDATE logic.
Para obter assistência adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com

SSC-PRF-0010¶

Partition by removed, at least one of the specified expressions have no iceberg partition transform equivalent

Gravidade¶

Nenhum

Descrição¶

Snowflake supports the PARTITION BY clause in Iceberg tables, however, only Iceberg partition transforms are supported. When transforming paritioning into Iceberg tables, SnowConvert AI will generate the equivalent partition transforms for supported cases. When no partition transform equivalent can be generated for the partition expressions, the PARTITION BY will be removed from the table by commenting it out with this PRF.

This PRF is only generated when SnowConvert AI migrates tables into Iceberg tables using the Tables translation conversion setting.

Exemplos de código¶

Entrada¶

 -- Additional Params: --TablesTransformationTarget SnowflakeIceberg
CREATE TABLE FINANCE.FINANCE_TABLE
(
  customerName VARCHAR(30),
  accountBalance VARCHAR(20)
)
PARTITION BY CASE_N(
accountBalance <  0 ,
accountBalance >=  0);

Saída¶

 -- Additional Params: --TablesTransformationTarget SnowflakeIceberg
CREATE OR REPLACE ICEBERG TABLE FINANCE.FINANCE_TABLE
  (
 customerName VARCHAR,
 accountBalance VARCHAR
  )
  CATALOG = 'SNOWFLAKE'
--  --** SSC-PRF-0010 - PARTITION BY REMOVED, AT LEAST ONE OF THE SPECIFIED EXPRESSIONS HAVE NO ICEBERG PARTITION TRANSFORM EQUIVALENT **
--  PARTITION BY CASE_N(
--  accountBalance <  0 ,
--  accountBalance >=  0)
  COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 1,  "minor": 0,  "patch": "0.0" }, "attributes": {  "component": "teradata",  "convertedOn": "12/16/2025",  "domain": "no-domain-provided",  "migrationid": "9CebAVkM33qsfTnTrMh3Dw==" }}'
;

Práticas recomendadas¶

Analyze the impact of partitioning in the performance of queries over the generated Iceberg tables, if the difference is neglible then this PRF can be safely ignored.
Para obter assistência adicional, entre em contato conosco pelo e-mail snowconvert-support@snowflake.com