模拟 COUNTIF 的雪花子查询

Question

我有以下查询，为什么我试图跨两个表模拟 COUNTIF：

WITH Range__A1_A4 (val) AS (
    SELECT 1 UNION ALL 
    SELECT 2 UNION ALL 
    SELECT 8 UNION ALL 
    SELECT 10
), NFL (SEASON, TEAM, WINS) AS (
    SELECT 2010, 'KC', 10 UNION ALL
    SELECT 2011, 'GB', 11
)
SELECT 
    (SELECT SUM(CASE WHEN NFL.WINS>R.val THEN 1 ELSE 0 END) FROM Range__A1_A4 R) ct,
    *
FROM NFL;

由此我得到以下错误：

Unsupported subquery type cannot be evaluated

这个查询有什么问题，修复它的正确方法是什么？

Answer 1

我会这样写：

WITH Range__A1_A4 (val) AS (
    SELECT * FROM VALUES 
        (1),
        (2),
        (8),
        (10)
), NFL (SEASON, TEAM, WINS) AS (
    SELECT * FROM VALUES 
        (2010, 'KC', 10),
        (2011, 'GB', 11)
)
SELECT 
    n.*,
    SUM(iff( n.wins > r.val, 1, 0)) as ct_a,
    count_if(n.wins > r.val) as ct_b
FROM NFL AS n
JOIN Range__A1_A4 AS r
group by 1,2,3;

SEASON	TEAM	WINS	CT	CT_B
2,010	KC	10	3	3
2,011	GB	11	4	4

模拟 COUNTIF 的雪花子查询

Snowflake subquery to simulate COUNTIF

sql

snowflake-cloud-data-platform