SQL 使用 group_concat 每个串联项目的计数而不是一行中的总计数

SQL using group_concat with count of each concatenated item rather total count in a row

我改写成this fiddle,但是报错

有人可以帮忙吗?

就此而言,要解决错误并获得正确的结果,您只需要 return 来自聚合子查询的 CUSTOMERID (该列已经是 [=13 的一部分=] 子句).

更新查询:

 SELECT SELLER, 
       COUNT(*) as COUNT_OF_CUSTIDS,
       GROUP_CONCAT(CUSTOMERID, ' (', cnt, ')' SEPARATOR '|') AS CUSTIDS
FROM (
    SELECT SELLER, CUSTOMERID, COUNT(*) as cnt
    FROM docs
    GROUP BY SELLER, CUSTOMERID
) t
GROUP BY SELLER
HAVING COUNT(*) > 1

Returns:

| SELLER | COUNT_OF_CUSTIDS | CUSTIDS              |
| ------ | ---------------- | -------------------- |
| S1     | 3                | C2 (2)|C1 (2)|C3 (3) |
| S2     | 3                | C3 (1)|C2 (1)|C1 (2) |

Demo on DB Fiddle