SnowConvert : expressions de Redshift¶

Listes d’expressions¶

Description¶

Une liste d’expressions est une combinaison d’expressions et peut apparaître dans des conditions d’appartenance et de comparaison (clauses WHERE) et dans des clauses GROUP BY. (Référence linguistique Redshift SQL Listes d’expressions).

Cette syntaxe est entièrement prise en charge par Snowflake.

Grammar Syntax

 expression , expression , ... | (expression, expression, ...)

Copy

Modèles d’échantillons de sources

Données de configuration

 CREATE TABLE table1 (
    quantity VARCHAR(50),
    fruit VARCHAR(50)
);

CREATE TABLE table2 (
    quantity VARCHAR(50),
    fruit VARCHAR(50)
);

CREATE TABLE table3 (
    id INT,
    name VARCHAR(50),
    quantity INT,
    fruit VARCHAR(50),
    price INT
);

INSERT INTO table1 (quantity, fruit)
VALUES
    ('one', 'apple'),
    ('two', 'banana'),
    ('three', 'cherry');

INSERT INTO table2 (quantity, fruit)
VALUES
    ('one', 'apple'),
    ('two', 'banana'),
    ('four', 'orange');

INSERT INTO table3 (id, name, quantity, fruit, price)
VALUES
    (1, 'Alice', 1, 'apple', 100),
    (2, 'Bob', 5, 'banana', 200),
    (3, 'Charlie', 10, 'cherry', 300),
    (4, 'David', 15, 'orange', 400);

Copy

Clause IN

Code d’entrée :

SELECT *
FROM table3
WHERE quantity IN (1, 5, 10);

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Code de sortie :

 SELECT *
FROM
    table3
WHERE quantity IN (1, 5, 10);

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Comparaisons

Input Code:

 SELECT *
FROM table3
WHERE (quantity, fruit) = (1, 'apple');

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100

Output Code:

 SELECT *
FROM
    table3
WHERE (quantity, fruit) = (1, 'apple');

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100

Note

Les comparaisons de listes d’expressions avec les opérateurs suivants peuvent avoir un comportement différent dans Snowflake (<, <= , >, >=). Ces opérateurs sont transformés en opérations logiques AND pour obtenir une équivalence totale dans Snowflake.

Code d’entrée :¶

 SELECT (1,8,20) < (2,2,0) as r1,
       (1,null,2) > (1,0,8) as r2,
       (null,null,2) < (1,0,8) as r3,
       (1,0,null) <= (1,1,0) as r4,
       (1,1,0) >= (1,1,20) as r5;

Copy

R1	R2	R3	R4	R5
FALSE	FALSE	NULL	NULL	FALSE

Code de sortie :¶

 SELECT
    (1 < 2
    AND 8 < 2
    AND 20 < 0) as r1,
    (1 > 1
    AND null > 0
    AND 2 > 8) as r2,
    (null < 1
    AND null < 0
    AND 2 < 8) as r3,
    (1 <= 1
    AND 0 <= 1
    AND null <= 0) as r4,
    (1 >= 1
    AND 1 >= 1
    AND 0 >= 20) as r5;

Copy

R1	R2	R3	R4	R5
FALSE	FALSE	NULL	NULL	FALSE

Tuples imbriqués¶

Code d’entrée :¶

 SELECT *
FROM table3
WHERE (quantity, fruit) IN ((1, 'apple'), (5, 'banana'), (10, 'cherry'));

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Code de sortie¶

 SELECT *
FROM
    table3
WHERE (quantity, fruit) IN ((1, 'apple'), (5, 'banana'), (10, 'cherry'));

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Instruction de casse¶

Code d’entrée :¶

 SELECT
    CASE
        WHEN quantity IN (1, 5, 10) THEN 'Found'
        ELSE 'Not Found'
    END AS result
FROM table3;

Copy

RESULT
Trouvé.
Trouvé.
Trouvé.
Introuvable
Introuvable
Introuvable

Code de sortie¶

 SELECT
    CASE
        WHEN quantity IN (1, 5, 10) THEN 'Found'
        ELSE 'Not Found'
    END AS result
FROM
    table3;

Copy

RESULT
Trouvé.
Trouvé.
Trouvé.
Introuvable
Introuvable
Introuvable

Expressions multiples¶

Code d’entrée :¶

 SELECT *
FROM table3
WHERE (quantity, fruit) IN ((1, 'apple'), (5, 'banana'), (10, 'cherry'))
  AND price IN (100, 200, 300);

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Code de sortie¶

 SELECT *
FROM
    table3
WHERE (quantity, fruit) IN ((1, 'apple'), (5, 'banana'), (10, 'cherry'))
  AND price IN (100, 200, 300);

Copy

ID	NAME	QUANTITY	FRUIT	PRICE
1	Alice	1	apple	100
2	Bob	5	banana	200
3	Charlie	10	cherry	300

Jointures¶

Code d’entrée :¶

 SELECT *
FROM table1 t1
JOIN table2 t2
    ON (t1.quantity, t1.fruit) = (t2.quantity, t2.fruit)
WHERE t1.quantity = 'one' AND t1.fruit = 'apple';

Copy

QUANTITY	FRUIT	QUANTITY	FRUIT
un	apple	un	apple

Code de sortie¶

 SELECT *
FROM
table1 t1
JOIN
        table2 t2
    ON (t1.quantity, t1.fruit) = (t2.quantity, t2.fruit)
WHERE t1.quantity = 'one' AND t1.fruit = 'apple';

Copy

QUANTITY	FRUIT	QUANTITY	FRUIT
un	apple	un	apple

Problèmes connus ¶

Aucun problème n’a été constaté.

Expressions composées¶

Description ¶

Une expression composée est une série d’expressions simples reliées par des opérateurs arithmétiques. Une expression simple utilisée dans une expression composée doit renvoyer une valeur numérique. (Référence linguistique RedShift SQL Expressions composées)

Grammar Syntax ¶

 expression operator {expression | (compound_expression)}

Copy

Table de conversion¶

Redshift	Snowflake	Comments
`\|\|` (Concatenation)	`\|\|`	Fully supported by Snowflake

Modèles d’échantillons de sources¶

Code d’entrée :¶

 CREATE TABLE concatenation_demo (
    col1 VARCHAR(20),
    col2 INTEGER,
    col3 DATE
);

INSERT INTO concatenation_demo (col1, col2, col3) VALUES
('Hello', 42, '2023-12-01'),
(NULL, 0, '2024-01-01'),
('Redshift', -7, NULL);

SELECT 
    col1 || ' has number ' || col2 AS concat_string_number
FROM concatenation_demo;

SELECT 
    col1 || ' on ' || col3 AS concat_string_date
FROM concatenation_demo;

SELECT
    COALESCE(col1, 'Unknown') || ' with number ' || COALESCE(CAST(col2 AS VARCHAR), 'N/A') AS concat_with_null_handling
FROM concatenation_demo;

Copy

concat_string_number
Hello a le numéro 42
<NULL>
Redshift a le numéro -7

concat_string_date
Bonjour le 2023-12-01
<NULL>
<NULL>

concat_with_null_handling
Bonjour avec le numéro 42
Inconnu avec le numéro 0
Redshift avec le numéro -7

Code de sortie :

 CREATE TABLE concatenation_demo (
    col1 VARCHAR(20),
    col2 INTEGER,
    col3 DATE
)
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "12/16/2024",  "domain": "test" }}';

INSERT INTO concatenation_demo (col1, col2, col3) VALUES
('Hello', 42, '2023-12-01'),
(NULL, 0, '2024-01-01'),
('Redshift', -7, NULL);

SELECT
    col1 || ' has number ' || col2 AS concat_string_number
FROM
    concatenation_demo;

SELECT
    col1 || ' on ' || col3 AS concat_string_date
FROM
    concatenation_demo;

SELECT
    COALESCE(col1, 'Unknown') || ' with number ' || COALESCE(CAST(col2 AS VARCHAR), 'N/A') AS concat_with_null_handling
FROM
    concatenation_demo;

Copy

concat_string_number
Hello a le numéro 42
<NULL>
Redshift a le numéro -7

concat_string_date
Bonjour le 2023-12-01
<NULL>
<NULL>

concat_with_null_handling
Bonjour avec le numéro 42
Inconnu avec le numéro 0
Redshift avec le numéro -7

Problèmes connus¶

Aucun problème n’a été constaté.

EWIs connexes¶

Il n’y a pas de problème connu.

Opérateurs arithmétiques¶

Traduction des opérateurs arithmétiques

Table de conversion¶

Redshift	Snowflake	Comments
+/- (positive and negative sign/operator)	+/-	Fully supported by Snowflake
^ (exponentiation)	POWER	Fully supported by Snowflake
* (multiplication)	*	Fully supported by Snowflake
/ (division)	/	Redshift division between integers always returns integer value, FLOOR function is added to emulate this behavior.
% (modulo)	%	Fully supported by Snowflake
+ (addition)	+ and \|\|	Fully supported by Snowflake. When string are added, it is transformed to a concat.
- (subtraction)	-	Fully supported by Snowflake
@ (absolute value)	ABS	Fully supported by Snowflake
\|/ (square root)	SQRT	Fully supported by Snowflake
\|\|/ (cube root)	CBRT	Fully supported by Snowflake

Modèles d’échantillons de sources¶

Addition, soustraction, positif et négatif¶

Code d’entrée :

 CREATE TABLE test_math_operations (
    base_value DECIMAL(10, 2),
    multiplier INT,
    divisor INT,
    description VARCHAR(100),
    created_at TIMESTAMP,
    category VARCHAR(50)
);


INSERT INTO test_math_operations (base_value, multiplier, divisor, description, created_at, category)
VALUES
(100.50, 2, 5, 'Basic test', '2024-12-01 10:30:00', 'Type A'),
(250.75, 3, 10, 'Complex operations', '2024-12-02 15:45:00', 'Type B'),
(-50.25, 5, 8, 'Negative base value', '2024-12-03 20:00:00', 'Type C'),
(0, 10, 2, 'Zero base value', '2024-12-04 09:15:00', 'Type D');


SELECT +base_value AS positive_value,
       -base_value AS negative_value,
       (base_value + multiplier - divisor) AS add_sub_result,
       created_at + INTERVAL '1 day' AS next_day,
       created_at - INTERVAL '1 hour' AS one_hour_before,
       description + category as string_sum,
       base_value + '5' as int_string_sum,
       '5' + base_value as string_int_sum
FROM test_math_operations;

Copy

positive_value	negative_value	add_sub_result	next_day	one_hour_before	string_sum	int_string_sum	string_int_sum
100.50	-100.50	97.50	2024-12-02 10:30:00.000000	2024-12-01 09:30:00.000000	Basic testType A	105.5	105.5
250.75	-250.75	243.75	2024-12-03 15:45:00.000000	2024-12-02 14:45:00.000000	Complex operationsType B	255.75	255.75
-50.25	50.25	-53.25	2024-12-04 20:00:00.000000	2024-12-03 19:00:00.000000	Negative base valueType C	-45.25	-45.25
0.00	0.00	8.00	2024-12-05 09:15:00.000000	2024-12-04 08:15:00.000000	Zero base valueType D	5	5

Code de sortie :

 CREATE TABLE test_math_operations (
    base_value DECIMAL(10, 2),
    multiplier INT,
    divisor INT,
    description VARCHAR(100),
    created_at TIMESTAMP,
    category VARCHAR(50)
)
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "12/16/2024",  "domain": "test" }}';


INSERT INTO test_math_operations (base_value, multiplier, divisor, description, created_at, category)
VALUES
(100.50, 2, 5, 'Basic test', '2024-12-01 10:30:00', 'Type A'),
(250.75, 3, 10, 'Complex operations', '2024-12-02 15:45:00', 'Type B'),
(-50.25, 5, 8, 'Negative base value', '2024-12-03 20:00:00', 'Type C'),
(0, 10, 2, 'Zero base value', '2024-12-04 09:15:00', 'Type D');


SELECT +base_value AS positive_value,
       -base_value AS negative_value,
       (base_value + multiplier - divisor) AS add_sub_result,
       created_at + INTERVAL '1 day' AS next_day,
       created_at - INTERVAL '1 hour' AS one_hour_before,
       description || category as string_sum,
       base_value + '5' as int_string_sum,
       '5' + base_value as string_int_sum
FROM
       test_math_operations;

Copy

positive_value	negative_value	add_sub_result	next_day	one_hour_before	string_sum	int_string_sum	string_int_sum
100.5	-100.5	97.5	2024-12-02 10:30:00	2024-12-01 09:30:00	Basic testType A	105.5	105.5
250.75	-250.75	243.75	2024-12-03 15:45:00	2024-12-02 14:45:00	Complex operationsType B	255.75	255.75
-50.25	50.25	-53.25	2024-12-04 20:00:00	2024-12-03 19:00:00	Negative base valueType C	-45.25	-45.25
0	0	8	2024-12-05 09:15:00	2024-12-04 08:15:00	Zero base valueType D	5	5

Exponentiation, multiplication, division et modulo¶

Code d’entrée :

 CREATE TABLE test_math_operations (
    base_value DECIMAL(10, 2),
    multiplier INT,
    divisor INT,
    mod_value INT,
    exponent INT
);

INSERT INTO test_math_operations (base_value, multiplier, divisor, mod_value, exponent)
VALUES
(100.50, 2, 5, 3, 2),
(250.75, 3, 10, 7, 3),
(-50.25, 5, 8, 4, 4),
(0, 10, 2, 1, 5);

SELECT
    base_value ^ exponent AS raised_to_exponent,
    (base_value * multiplier) AS multiplied_value,
    (base_value / divisor) AS divided_value,
    base_value::int / divisor as int_division,
    (mod_value % 2) AS modulo_result,
    (base_value + multiplier - divisor) AS add_sub_result,
    (base_value + (multiplier * (divisor - mod_value))) AS controlled_eval
FROM
    test_math_operations;

Copy

raised_to_exponent	multiplied_value	divided_value	int_division	modulo_result	add_sub_result	controlled_eval
10100.25	201	20.1	20	1	97.5	104.5
15766047.296875	752.25	25.075	25	1	243.75	259.75
6375940.62890625	-251.25	-6.28125	-6	0	-53.25	-30.25
0	0	0	0	1	8	10

Code de sortie :

 CREATE TABLE test_math_operations (
    base_value DECIMAL(10, 2),
    multiplier INT,
    divisor INT,
    mod_value INT,
    exponent INT
)
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "12/10/2024",  "domain": "test" }}';

INSERT INTO test_math_operations (base_value, multiplier, divisor, mod_value, exponent)
VALUES
(100.50, 2, 5, 3, 2),
(250.75, 3, 10, 7, 3),
(-50.25, 5, 8, 4, 4),
(0, 10, 2, 1, 5);

SELECT
    POWER(
    base_value, exponent) AS raised_to_exponent,
    (base_value * multiplier) AS multiplied_value,
    (base_value / divisor) AS divided_value,
    FLOOR(
    base_value::int / divisor) as int_division,
    (mod_value % 2) AS modulo_result,
    (base_value + multiplier - divisor) AS add_sub_result,
    (base_value + (multiplier * (divisor - mod_value))) AS controlled_eval
FROM
    test_math_operations;

Copy

raised_to_exponent	multiplied_value	divided_value	int_division	modulo_result	add_sub_result	controlled_eval
10100.25	201	20.1	20	1	97.5	104.5
15766047.2969	752.25	25.075	25	1	243.75	259.75
6375940.6289	-251.25	-6.2812	-7	0	-53.25	-30.25
0	0	0	0	1	8	10

Valeur absolue, racine carrée et racine cubique¶

Code d’entrée :

 CREATE TABLE unary_operators
(
    col1 INTEGER,
    col2 INTEGER
);

INSERT INTO unary_operators VALUES
(14, 10),
(-8, 8),
(975, 173),
(-1273, 187);

SELECT
|/ col2 AS square_root,
||/ col1 AS cube_root,
@ col1 AS absolute_value
FROM unary_operators;

Copy

+-------------------+--------------------+--------------+
|square_root        |cube_root           |absolute_value|
+-------------------+--------------------+--------------+
|3.1622776601683795 |2.4101422641752306  |14            |
|2.8284271247461903 |-2                  |8             |
|13.152946437965905 |9.915962413403873   |975           |
|13.674794331177344 |-10.837841647592736 |1273          |
+-------------------+--------------------+--------------+

Copy

Code de sortie :

 CREATE TABLE unary_operators
(
    col1 INTEGER,
    col2 INTEGER
)
COMMENT = '{ "origin": "sf_sc", "name": "snowconvert", "version": {  "major": 0,  "minor": 0,  "patch": "0" }, "attributes": {  "component": "redshift",  "convertedOn": "12/17/2024",  "domain": "test" }}';

INSERT INTO unary_operators
VALUES
(14, 10),
(-8, 8),
(975, 173),
(-1273, 187);

SELECT
    SQRT(col2) AS square_root,
    CBRT(col1) AS cube_root,
    ABS(col1) AS absolute_value
FROM
    unary_operators;

Copy

+-------------+--------------+--------------+
|square_root  |cube_root     |absolute_value|
+-------------+--------------+--------------+
|3.16227766   |2.410142264   |14            |
|2.828427125  |-2            |8             |
|13.152946438 |9.915962413   |975           |
|13.674794331 |-10.837841648 |1273          |
+-------------+--------------+--------------+

Copy

Problèmes connus¶

Dans Snowflake, il est possible d’utiliser les opérateurs unaires +et - avec des valeurs de chaîne, mais dans Redshift, ce n’est pas valable.

EWIs connexes¶

Pas d’EWIs connexes.

Opérateurs binaires¶

Traduction pour les opérateurs Bitwise

Table de conversion¶

Redshift	Snowflake	Comments
`&` (AND)	`BITAND`	Fully supported by Snowflake
`\|` (OR)	`BITOR`	Fully supported by Snowflake
`<<` (Shift Left)	`BITSHIFTLEFT`
`>>` (Shift Right)	`BITSHIFTRIGHT`
`#` (XOR)	`BITXOR`	Fully supported by Snowflake
`~` (NOT)	`BITNOT`	Fully supported by Snowflake

Modèles d’échantillons de sources¶

Données de configuration¶

Redshift

 CREATE TABLE bitwise_demo (
    col1 INTEGER,
    col2 INTEGER,
    col3 INTEGER,
    col4 VARBYTE(5),
    col5 VARBYTE(7)
);

INSERT INTO bitwise_demo (col1, col2, col3, col4, col5) VALUES
-- Binary: 110, 011, 1111, 0100100001100101011011000110110001101111, 0100100001101001
(6, 3, 15, 'Hello'::VARBYTE, 'Hi'::VARBYTE),
-- Binary: 1010, 0101, 0111, 0100000101000010, 01000011
(10, 5, 7, 'AB'::VARBYTE, 'C'::VARBYTE),   
-- Binary: 11111111, 10000000, 01000000, 010000100111100101100101, 01000111011011110110111101100100010000100111100101100101
(255, 128, 64, 'Bye'::VARBYTE, 'GoodBye'::VARBYTE),
-- Edge case with small numbers and a negative number
(1, 0, -1, 'Hey'::VARBYTE, 'Ya'::VARBYTE);

Copy

Snowflake

 CREATE TABLE bitwise_demo (
    col1 INTEGER,
    col2 INTEGER,
    col3 INTEGER,
    col4 BINARY(5),
    col5 BINARY(7)
);

-- Binary: 110, 011, 1111, 0100100001100101011011000110110001101111, 0100100001101001
INSERT INTO bitwise_demo (col1, col2, col3, col4, col5) SELECT 6, 3, 15, TO_BINARY(HEX_ENCODE('Hello')), TO_BINARY(HEX_ENCODE('Hi'));  
-- Binary: 1010, 0101, 0111, 0100000101000010, 01000011 
INSERT INTO bitwise_demo (col1, col2, col3, col4, col5) SELECT 10, 5, 7, TO_BINARY(HEX_ENCODE('AB')), TO_BINARY(HEX_ENCODE('C'));   
-- Binary: 11111111, 10000000, 01000000, 010000100111100101100101, 01000111011011110110111101100100010000100111100101100101
INSERT INTO bitwise_demo (col1, col2, col3, col4, col5) SELECT 255, 128, 64, TO_BINARY(HEX_ENCODE('Bye')), TO_BINARY(HEX_ENCODE('GoodBye')); 
-- Edge case with small numbers and a negative number
INSERT INTO bitwise_demo (col1, col2, col3, col4, col5) SELECT 1, 0, -1, TO_BINARY(HEX_ENCODE('Hey')), TO_BINARY(HEX_ENCODE('Ya'));

Copy

Opérateurs bitwise sur les valeurs entières¶

Code d’entrée :

 SELECT
    -- Bitwise AND
    col1 & col2 AS bitwise_and,  -- col1 AND col2

    -- Bitwise OR
    col1 | col2 AS bitwise_or,   -- col1 OR col2

    -- Left Shift
    col3 << 1 AS left_shift_col3, -- col3 shifted left by 1

    -- Right Shift
    col3 >> 1 AS right_shift_col3, -- col3 shifted right by 1

    -- XOR
    col1 # col2 AS bitwise_xor, -- col1 XOR col2

    -- NOT
    ~ col3 AS bitwise_not -- NOT col3

FROM bitwise_demo;

Copy

+-------------+------------+-----------------+------------------+-------------+-------------+
| bitwise_and | bitwise_or | left_shift_col3 | right_shift_col3 | bitwise_xor | bitwise_not |
+-------------+------------+-----------------+------------------+-------------+-------------+
|2            |7           |30               |7                 |5            |-16          |
|0            |15          |14               |3                 |15           |-8           |
|128          |255         |128              |32                |127          |-65          |
|0            |1           |-2               |-1                |1            |0            |
+-------------+------------+-----------------+------------------+-------------+-------------+

Copy

Code de sortie :

 SELECT
        BITAND(
        -- Bitwise AND
        col1, col2) AS bitwise_and,  -- col1 AND col2
        BITOR(

        -- Bitwise OR
        col1, col2) AS bitwise_or,   -- col1 OR col2
        -- Left Shift
        --** SSC-FDM-PG0010 - RESULTS MAY VARY DUE TO THE BEHAVIOR OF SNOWFLAKE'S BITSHIFTLEFT BITWISE FUNCTION **
        BITSHIFTLEFT(
        col3, 1) AS left_shift_col3, -- col3 shifted left by 1
        -- Right Shift
        --** SSC-FDM-PG0010 - RESULTS MAY VARY DUE TO THE BEHAVIOR OF SNOWFLAKE'S BITSHIFTRIGHT BITWISE FUNCTION **
        BITSHIFTRIGHT(
        col3, 1) AS right_shift_col3, -- col3 shifted right by 1
        BITXOR(

        -- XOR
        col1, col2) AS bitwise_xor, -- col1 XOR col2
        -- NOT
        BITNOT(col3) AS bitwise_not -- NOT col3
FROM
        bitwise_demo;

Copy

+-------------+------------+-----------------+------------------+-------------+-------------+
| bitwise_and | bitwise_or | left_shift_col3 | right_shift_col3 | bitwise_xor | bitwise_not |
+-------------+------------+-----------------+------------------+-------------+-------------+
|2            |7           |30               |7                 |5            |-16          |
|0            |15          |14               |3                 |15           |-8           |
|128          |255         |128              |32                |127          |-65          |
|0            |1           |-2               |-1                |1            |0            |
+-------------+------------+-----------------+------------------+-------------+-------------+

Copy

Opérateurs binaires sur des données binaires¶

Pour les fonctions BITAND, BITOR et BITXOR, le paramètre'LEFT' est ajouté pour insérer un espacement dans le cas où les deux valeurs binaires ont une longueur différente, ceci afin d’éviter des erreurs lors de la comparaison des valeurs dans Snowflake.

Redshift

 SELECT
    -- Bitwise AND
    col4 & col5 AS bitwise_and,  -- col4 AND col5

    -- Bitwise OR
    col4 | col5 AS bitwise_or,   -- col4 OR col5

    -- XOR
    col4 # col5 AS bitwise_xor, -- col4 XOR col5

    -- NOT
    ~ col4 AS bitwise_not -- NOT col4

FROM bitwise_demo;

Copy

+-----------------+-----------------+-----------------+-------------+
| bitwise_and     | bitwise_or      | bitwise_xor     | bitwise_not |
+-----------------+-----------------+-----------------+-------------+
|0x0000004869     |0x48656C6C6F     |0x48656C2406     |0xB79A939390 |
|0x0042           |0x4143           |0x4101           |0xBEBD       |
|0x00000000427965 |0x476F6F64427965 |0x476F6F64000000 |0xBD869A     |
|0x004161         |0x487D79         |0x483C18         |0xB79A86     |
+-----------------+-----------------+-----------------+-------------+

Copy

Snowflake

 SELECT
    BITAND(
    -- Bitwise AND
    col4, col5, 'LEFT') AS bitwise_and,  -- col4 AND col5
    BITOR(

    -- Bitwise OR
    col4, col5, 'LEFT') AS bitwise_or,   -- col4 OR col5

    -- XOR
    BITXOR(col4, col5, 'LEFT') AS bitwise_xor, -- col4 XOR col5

    -- NOT
    BITNOT(col4) AS bitwise_not -- NOT col4
    
    FROM bitwise_demo;

Copy

+---------------+---------------+---------------+-------------+
| bitwise_and   | bitwise_or    | bitwise_xor   | bitwise_not |
+---------------+---------------+---------------+-------------+
|0000004869     |48656C6C6F     |48656C2406     |B79A939390   |
|0042           |4143           |4101           |BEBD         |
|00000000427965 |476F6F64427965 |476F6F64000000 |BD869A       |
|004161         |487D79         |483C18         |B79A86       |
+---------------+---------------+---------------+-------------+

Copy

Problèmes connus¶

Aucun problème n’a été constaté.

EWIs connexes¶

SSC-FDM-PG0010 : les résultats peuvent varier en raison du comportement de la fonction bitwise de Snowflake.