如何按文件状态显示最后一行的总数？

Question

我必须按文件状态显示最后一行的总数。

tbl_bankdata

bank_id | b_orderno| b_bankname| lead_id
     1  | 01-01    | 1         |  1
     2  | 01-02    | 2         |  1
     3  | 02-01    | 3         |  2
     4  | 03-01    | 1         |  3

tbl_fileStatus

f_id | f_bankid| f_filestatus
  1  |      1  | 1
  2  |      2  | 1
  3  |      2  | 2
  4  |      1  | 2
  5  |      1  | 3
  6  |      3  | 2 
  7  |      3  | 3  

我有两张桌子tbl_bankdata 和tbl_fileStatus。我在tbl_fileStatus 中发送bank_id 作为f_bank_id。

现在我必须显示最后一个 f_bankid 计数。

例如，我必须获取 f_filestatus=1 的计数。所以我的输出将是0。为什么0 因为f_bankid 1 and 2 有一个f_filestatus=1 但f_bankid 1 and 2 的最后一行带有f_filestatus 3 and 2。

如果我必须计算f_filestatus=2，那么我将得到输出1，如果计算f_filestatus=3，那么输出将是2。为什么是 2，因为 f_bank_id 1 有 f_filestatus 3 和 f_bank_id 3 有 f_filestatus 3

这是我的查询

select (
    SELECT COUNT(f_id) 
    FROM tbl_fileStatus 
    WHERE f_filestatus=1 and f_id IN (
        SELECT MAX(f_id) FROM tbl_fileStatus GROUP BY f_bankid
    )
) as tcount

你能帮我解决这个问题吗？

@forpas 建议后

SELECT (SELECT Count(DISTINCT f_bankid)
    FROM   tbl_filestatus t
    WHERE  1 = (SELECT f_filestatus
                FROM   tbl_filestatus
                WHERE  f_bankid = t.f_bankid
                ORDER  BY f_id DESC
                LIMIT  1)) AS tcount1,
   (SELECT Count(DISTINCT f_bankid)
    FROM   tbl_filestatus t
    WHERE  2 = (SELECT f_filestatus
                FROM   tbl_filestatus
                WHERE  f_bankid = t.f_bankid
                ORDER  BY f_id DESC
                LIMIT  1)) AS tcount2,
   (SELECT Count(DISTINCT f_bankid)
    FROM   tbl_filestatus t
    WHERE  3 = (SELECT f_filestatus
                FROM   tbl_filestatus
                WHERE  f_bankid = t.f_bankid
                ORDER  BY f_id DESC
                LIMIT  1)) AS tcount3  

Answer 1

您可以使用下一个查询，但有一个例外 - 查询不返回 0 结果，但可以在可视化级别解决：

SELECT COUNT(*) FROM tbl_fileStatus 
JOIN (
    SELECT f_bankid, MAX(f_filestatus) AS max_f_filestatus
    FROM tbl_fileStatus GROUP BY f_bankid
) AS m ON m.max_f_filestatus = tbl_fileStatus.f_filestatus and
  m.f_bankid = tbl_fileStatus.f_bankid
  WHERE tbl_fileStatus.f_filestatus = 3 -- or 2 or any other value :)

Test SQL here

Answer 2

使用相关子查询：

SELECT COUNT(DISTINCT f_bankid) AS tcount
FROM tbl_fileStatus t
WHERE ? = (SELECT f_filestatus FROM tbl_fileStatus WHERE f_bankid = t.f_bankid ORDER BY f_id DESC LIMIT 1)

将? 替换为您搜索的f_bankid。
请参阅demo。

在MySql 8.0+中可以使用FIRST_VALUE()窗口函数：

SELECT COUNT(*) AS tcount
FROM (
  SELECT DISTINCT f_bankid, 
         FIRST_VALUE(f_filestatus) OVER (PARTITION BY f_bankid ORDER BY f_id DESC) f_filestatus
  FROM tbl_fileStatus
) t
WHERE f_filestatus = ?

请参阅demo。

如果您想要 1 行中所有 f_filestatus 的结果：

SELECT
  SUM(f_filestatus = 1) AS tcount1,
  SUM(f_filestatus = 2) AS tcount2,
  SUM(f_filestatus = 3) AS tcount3
FROM (
  SELECT t.f_bankid, t.f_filestatus
  FROM tbl_fileStatus t
  WHERE t.f_id = (SELECT f_id FROM tbl_fileStatus WHERE f_bankid = t.f_bankid ORDER BY f_id DESC LIMIT 1)
) t

请参阅demo。
结果：

> tcount1 | tcount2 | tcount3
> ------: | ------: | ------:
>       0 |       1 |       2

Answer 3

在 MySQL 8.0 中，我会推荐row_number()：

select count(*) as cnt
from (
    select fs.*,
        row_number() over(partition by f_bank_id order by f_id desc) rn
    from tbl_filestatus fs
) fs
where rn = 1 and filestatus = ?

适用于所有版本的替代方法是相关子查询：

select count(*) as cnt
from tbl_filestatus fs
where fs.f_id = (
    select max(fs1.f_id)
    from tbl_filestatus fs1
    where fs1.f_bank_id = fs.bank_id and fs1.filestatus = ?
)

为了提高第二个查询的性能，请考虑在(bank_id, filestatus, fs_id desc) 上创建一个索引。使用此索引，该查询可能比使用row_number() 的第一种方法运行得更快，即使在 MySQL 8.0 中也是如此。

如果您想一次计算所有filestatuses：

select filestatus, count(*) as cnt
from tbl_filestatus fs
where fs.f_id = (
    select max(fs1.f_id)
    from tbl_filestatus fs1
    where fs1.f_bank_id = fs.bank_id
)
group by filestatus