SELECT * FROM p_record WHERE create_time BETWEEN ('2020-06-10 10:40:58','2023-06-10 10:40:58'); 该sql存在什么问题

该 SQL 语句存在语法错误，应该使用 AND 连接两个时间点，而不是将它们放在括号内作为一个参数。正确的 SQL 语句应该是：

SELECT * FROM p_record WHERE create_time BETWEEN '2020-06-10 10:40:58' AND '2023-06-10 10:40:58';

注意，这里的时间点应该是数据库中 `create_time` 字段的数据类型，如果 `create_time` 的数据类型不是时间类型，那么这个 SQL 语句也会出现错误。同时，该 SQL 语句没有进行排序或限制查询结果的数量，这可能会导致查询结果很大，占用过多的系统资源。

相关问题

SELECT CUST_ID agentId, AGE_NAME agentName, countReleased, dayAmountReleased, CASE o.agelimitType WHEN '1' THEN TO_CHAR(o.agelimitNum) ELSE TO_CHAR(o.issueamtlimit) END countLimit, CASE o.agelimitType WHEN '1' THEN '不限' ELSE TO_CHAR(o.daylimitamount) END dayMaxCount FROM (SELECT m.CUST_ID, m.AGE_NAME, ad.countReleased, t.dayAmountReleased, d.daylimitamount, d.issueamtlimit, (SELECT AGE_LIMIT_NUM FROM P_ACT_ALLOW_MANAGER WHERE ACTIVITY_ID = #{activityId} AND rownum = 1 ) AS agelimitNum, (SELECT AGE_LIMIT_NUM_TYPE FROM P_ACT_MANAGER_ACTIVITY WHERE ACTIVITY_ID = #{activityId} ) AS agelimitType FROM (SELECT c1.CUST_ID, c1.AGE_NAME FROM (SELECT CUST_ID, AGE_NAME FROM PM_AGET_INFO START WITH CUST_ID = #{agetId} CONNECT BY PRIOR CUST_ID = FATH_CUST_ID ) c1 INNER JOIN (SELECT DISTINCT CUST_ID, AGE_NAME FROM PM_AGET_INFO START WITH CUST_ID in (SELECT CUST_ID FROM P_ACT_ALLOW_MANAGER WHERE ACTIVITY_ID = #{activityId} ) CONNECT BY PRIOR CUST_ID = FATH_CUST_ID ) c2 ON c1.CUST_ID = c2.CUST_ID ) m LEFT JOIN (SELECT CUST_ID, count(RECORD_ID) countReleased FROM P_ACT_MANAGER_RECORD WHERE ACTIVITY_ID = #{activityId} AND RECORD_STATUS NOT IN (2, 5) AND to_date(CREATE_TIME, 'yyyy-MM-dd hh24:mi:ss') BETWEEN to_date(#{beginTime}, 'yyyyMMddhh24miss') AND to_date(#{endTime}, 'yyyyMMddhh24miss') GROUP BY CUST_ID ) ad ON m.CUST_ID = ad.CUST_ID LEFT JOIN (SELECT AGE_ID cust_id, DAY_LIMIT_AMOUNT daylimitAmount, ISSUE_AMT_LIMIT issueAmtLimit FROM P_ACT_MANAGER_LIMIT_ALLOT WHERE ACT_ID =#{activityId} ) d ON m.CUST_ID = d.CUST_ID LEFT JOIN (SELECT CUST_ID, count(RECORD_ID) dayAmountReleased FROM P_ACT_MANAGER_RECORD WHERE ACTIVITY_ID =#{activityId} AND RECORD_STATUS NOT IN (2, 5) AND to_date(CREATE_TIME, 'yyyy-MM-dd hh24:mi:ss') BETWEEN to_date(#{todayBegin}, 'yyyyMMddhh24miss') AND to_date(#{todayend}, 'yyyyMMddhh24miss') GROUP BY CUST_ID ) t ON m.CUST_ID = t.CUST_ID) o WHERE 1 = 1 AND o.CUST_ID = #{agentId} AND o.AGE_NAME = #{agentName} ORDER BY o.countReleased DESC NULLS LAST 如何优化此段sql代码,使得更加简洁？

这段 SQL 代码可以进行以下优化：

1. 使用 WITH 子句来代替嵌套查询，增加可读性。

2. 使用 LEFT JOIN 和 COALESCE 函数来代替子查询，避免多次查询。

3. 使用 CASE WHEN THEN ELSE END 语句来代替 IF 函数。

优化后的 SQL 代码如下：

WITH
cte1 AS (
SELECT
CUST_ID,
AGE_NAME
FROM
PM_AGET_INFO
START WITH
CUST_ID = #{agetId}
CONNECT BY
PRIOR CUST_ID = FATH_CUST_ID
),
cte2 AS (
SELECT
DISTINCT CUST_ID,
AGE_NAME
FROM
PM_AGET_INFO
START WITH
CUST_ID IN (
SELECT
CUST_ID
FROM
P_ACT_ALLOW_MANAGER
WHERE
ACTIVITY_ID = #{activityId}
)
CONNECT BY
PRIOR CUST_ID = FATH_CUST_ID
),
cte3 AS (
SELECT
CUST_ID,
count(RECORD_ID) countReleased
FROM
P_ACT_MANAGER_RECORD
WHERE
ACTIVITY_ID = #{activityId}
AND RECORD_STATUS NOT IN (2, 5)
AND to_date(CREATE_TIME, 'yyyy-MM-dd hh24:mi:ss') BETWEEN to_date(#{beginTime}, 'yyyyMMddhh24miss') AND to_date(#{endTime}, 'yyyyMMddhh24miss')
GROUP BY
CUST_ID
),
cte4 AS (
SELECT
CUST_ID,
count(RECORD_ID) dayAmountReleased
FROM
P_ACT_MANAGER_RECORD
WHERE
ACTIVITY_ID =#{activityId}
AND RECORD_STATUS NOT IN (2, 5)
AND to_date(CREATE_TIME, 'yyyy-MM-dd hh24:mi:ss') BETWEEN to_date(#{todayBegin}, 'yyyyMMddhh24miss') AND to_date(#{todayend}, 'yyyyMMddhh24miss')
GROUP BY
CUST_ID
),
cte5 AS (
SELECT
AGE_LIMIT_NUM
FROM
P_ACT_ALLOW_MANAGER
WHERE
ACTIVITY_ID = #{activityId}
AND rownum = 1
),
cte6 AS (
SELECT
AGE_LIMIT_NUM_TYPE
FROM
P_ACT_MANAGER_ACTIVITY
WHERE
ACTIVITY_ID = #{activityId}
),
cte7 AS (
SELECT
AGE_ID cust_id,
DAY_LIMIT_AMOUNT daylimitAmount,
ISSUE_AMT_LIMIT issueAmtLimit
FROM
P_ACT_MANAGER_LIMIT_ALLOT
WHERE
ACT_ID =#{activityId}
)
SELECT
m.CUST_ID agentId,
m.AGE_NAME agentName,
COALESCE(ad.countReleased, 0) countReleased,
COALESCE(t.dayAmountReleased, 0) dayAmountReleased,
CASE
WHEN o.agelimitType = '1' THEN TO_CHAR(o.agelimitNum)
ELSE TO_CHAR(o.issueamtlimit)
END countLimit,
CASE
WHEN o.agelimitType = '1' THEN '不限'
ELSE TO_CHAR(o.daylimitamount)
END dayMaxCount
FROM
cte1 m
INNER JOIN cte2 c ON m.CUST_ID = c.CUST_ID AND m.AGE_NAME = c.AGE_NAME
LEFT JOIN cte3 ad ON m.CUST_ID = ad.CUST_ID
LEFT JOIN cte4 t ON m.CUST_ID = t.CUST_ID
LEFT JOIN cte5 ON 1 = 1
LEFT JOIN cte6 ON 1 = 1
LEFT JOIN cte7 d ON m.CUST_ID = d.CUST_ID
LEFT JOIN (
SELECT
COALESCE(AGE_LIMIT_NUM, 0) agelimitNum,
COALESCE(AGE_LIMIT_NUM_TYPE, '1') agelimitType
FROM
dual
) o ON 1 = 1
WHERE
m.CUST_ID = #{agentId}
AND m.AGE_NAME = #{agentName}
ORDER BY
COALESCE(ad.countReleased, 0) DESC NULLS LAST

这样可以使得代码更加简洁易读，并且避免了重复查询和嵌套查询的问题。

<select id="getToallocatedBusinessNum" resultType="Integer"> SELECT COUNT(DISTINCT(id)) AS toallocatedBusinessNum FROM `tb_business` WHERE id NOT IN ( SELECT assign_id FROM `tb_assign_record` WHERE create_time BETWEEN #{startTime} AND #{endTime} AND latest = 1 AND `type` = 1 ) AND create_time BETWEEN #{startTime} AND #{endTime} </select>写成外连接

可以将原先的子查询改写成一个左外连接，如下所示：

<select id="getToallocatedBusinessNum" resultType="Integer">
  SELECT COUNT(DISTINCT b.id) AS toallocatedBusinessNum
  FROM `tb_business` b
  LEFT JOIN (
    SELECT assign_id
    FROM `tb_assign_record`
    WHERE create_time BETWEEN #{startTime} AND #{endTime}
      AND latest = 1
      AND `type` = 1
  ) ar ON b.id = ar.assign_id
  WHERE ar.assign_id IS NULL
    AND b.create_time BETWEEN #{startTime} AND #{endTime}
</select>

这里使用了一个左外连接，将 `tb_business` 表和子查询的结果集进行连接，连接条件是 `tb_business.id = tb_assign_record.assign_id`，然后在 `WHERE` 子句中过滤掉 `tb_assign_record.assign_id IS NULL` 的记录，即未分配的业务。最后统计 `DISTINCT` 的业务数即可。