Want to use multiple aggregate function with snowflake pivot columns function - sql

Question

CREATE TABLE person (id INT, name STRING, date date, class INT, address STRING);
INSERT INTO person VALUES
(100, 'John', 30-1-2021, 1, 'Street 1'),
(200, 'Mary', 20-1-2021, 1, 'Street 2'),
(300, 'Mike', 21-1-2021, 3, 'Street 3'),
(100, 'John', 15-5-2021, 4, 'Street 4');
SELECT * FROM person
PIVOT (
**SUM(age) AS a, MAX(date) AS c**
FOR name IN ('John' AS john, 'Mike' AS mike)
);
This is databricks sql code above, how do I implement the same logic in snowflake

Answer 1

Below is the syntax for PIVOT in Snowflake:
SELECT ...
FROM ...
PIVOT ( <aggregate_function> ( <pivot_column> )
FOR <value_column> IN ( <pivot_value_1> [ , <pivot_value_2> ... ] ) )
[ ... ]
In case of Snowflake, your AS keyword will be outside the PIVOT function.
Check this example for your reference:
select *
from monthly_sales
pivot(sum(amount) for month in ('JAN', 'FEB', 'MAR', 'APR'))
as p
order by empid;
Visit this official document and check the given examples for better understanding.

Answer 2

Firstly, there is no "AGE" column as I can see from your table DDL.
Secondly, I do not think you can pivot on multiple aggregation functions, as the value will be put under the mentioned columns "JOHN" and "MIKE" for their corresponding aggregated values, it can't fit into two separate values. I don't know how your DataBricks example would work.
Your example will look something like below in Snowflake, after removing one aggregation function:
SELECT *
FROM
person
PIVOT (
MAX(date) FOR name IN ('John', 'Mike')
)
as p (id, class, address, john, mike)
;

Answer 3

Snowflake does not support multiple aggregate expressions in the PIVOT
And as noted by others, your AGE is missing, and you also do not have a ORDER BY clause, which makes rolling your own SQL harder.
SELECT
SUM(IFF(name='John',age,null)) AS john_sum_age,
MAX(IFF(name='John',date,null)) AS john_max_date,
SUM(IFF(name='Mike',age,null)) AS mike_age,
MAX(IFF(name='Mike',date,null)) AS mike_max_date
FROM person
if you had the ORDER BY in your example it would become the GROUP BY clause in this form
SELECT
<gouping_columns>,
SUM(IFF(name='John',age,null)) AS john_sum_age,
MAX(IFF(name='John',date,null)) AS john_max_date,
SUM(IFF(name='Mike',age,null)) AS mike_age,
MAX(IFF(name='Mike',date,null)) AS mike_max_date
FROM person
GROUP BY <gouping_columns>