如何获取每个员工每天的第一个和最后一个数据

Question

嗨，我正在尝试编写一个查询，它会给我输出

每天先入后出的用户 reader 姓名不是 Locker 并确定他是迟到还是早出

示例数据：

EVENTID USERID  DATE_TIME   READERID    READERNAME  SNAME   LOGTYPE USERNAME    RESULT  REMARKS DURATION

177792329   50078   2020-01-02 05:45:34 544381431   LOCKER  RF-T-33 [IN]    IN   SUBIA, MAY             
177792497   50078   2020-01-02 05:45:50 544343461   SENSORS RF-T-19 [IN]    IN   SUBIA, MAY             
177813432   50078   2020-01-02 06:35:51 544352385   SENSORS RF-T-20 [OUT]   OUT  SUBIA, MAY             
177813495   50078   2020-01-02 06:36:13 544381436   LOCKER  RF-T-36 [OUT]   OUT  SUBIA, MAY             
177950118   50078   2020-01-03 05:50:48 544352385   SENSORS RF-T-20 [OUT]   OUT  SUBIA, MAY             
177954602   50078   2020-01-03 05:54:11 544381436   LOCKER  RF-T-36 [OUT]   OUT  SUBIA, MAY             
177967366   50078   2020-01-03 06:07:25 544381431   LOCKER  RF-T-33 [IN]    IN   SUBIA, MAY     00:13       13
177967448   50078   2020-01-03 06:07:38 544343461   SENSORS RF-T-19 [IN]    IN   SUBIA, MAY             
177977184   50078   2020-01-03 08:02:59 544352385   SENSORS RF-T-20 [OUT]   OUT  SUBIA, MAY             
177977214   50078   2020-01-03 08:03:21 544381436   LOCKER  RF-T-36 [OUT]   OUT  SUBIA, MAY             
177979653   50078   2020-01-03 08:23:54 544381431   LOCKER  RF-T-33 [IN]    IN   SUBIA, MAY     00:20       20

预期输出

先入

USERNAME    DATE_TIME   First_IN_Time
 SUBIA, MAY     2020-01-02 05:45:50   05:45:34 
SUBIA, MAY      2020-01-03 05:45:50   06:07:25

最后一个

USERNAME    DATE_TIME   last_out_Time
 SUBIA, MAY     2020-01-02 06:35:51  06:35:51
SUBIA, MAY      2020-01-03 05:45:50   08:02:59

tblshift

  SHIFTCODE   DATESHIFT       STARTSHIFT                      ENDSHIFT                                ISRESTDAY

    50078           1/24/2020        1/1/1970 2:00:00.000000 PM  1/1/1970 10:00:00.000000 PM              0
    50078        1/25/2020       1/1/1970 2:00:00.000000 PM  1/1/1970 10:00:00.000000 PM              0
    50078        1/26/2020       1/1/1970 2:00:00.000000 PM  1/1/1970 10:00:00.000000 PM             0
    50078        1/27/2020       1/1/1970 2:00:00.000000 PM  1/1/1970 10:00:00.000000 PM             0
    50078        2/10/2020       1/1/1970 10:00:00.000000 PM 1/1/1970 6:00:00.000000 AM              0
    50078        2/11/2020       1/1/1970 10:00:00.000000 PM 1/1/1970 6:00:00.000000 AM              0
    50078        2/12/2020       1/1/1970 10:00:00.000000 PM 1/1/1970 6:00:00.000000 AM

我有什么尝试

Select
  max(A.DATE_TIME) keep (dense_rank first order by A.DATE_TIME) first_IN,
 --Max(A.DATE_TIME)  first_IN,
 A.USERNAME 
--min (A.DATE_TIME ) first_IN, max (A.DATE_TIME ) last_out,A.USERNAME 
--case
--end
from tblaccesslogs a 

where A.LOGTYPE = 'IN' and A.READERNAME <> 'LOCKER'
 group by  A.USERNAME

更新我需要确定员工是否早退基于 table tblshift

希望有人能帮我解决这个问题。

Answer 1

您可以使用 aggregate 函数在单个查询中获取上下班时间，如下所示，并将其与理想的班次开始时间进行比较，如下所示：

SELECT
    T.USERNAME,
    CASE
        WHEN FIRST_IN_TIME > 
            TRUNC(FIRST_IN_TIME) + ( STARTSHIFT - DATE '1970-01-01' ) THEN 'LATE'
        ELSE 'ONTIME'
    END AS IN_STATUS
FROM
    (
        SELECT
            A.USERNAME,
            A.USERID,
            TRUNC(A.DATE_TIME) AS DATE_TIME,
            MIN(CASE WHEN A.LOGTYPE = 'IN' THEN A.DATE_TIME END) AS FIRST_IN_TIME,
            MAX(CASE WHEN A.LOGTYPE = 'OUT' THEN A.DATE_TIME END) AS LAST_OUT_TIME
        FROM TBLACCESSLOGS A
        WHERE A.READERNAME <> 'LOCKER'
        GROUP BY A.USERNAME, A.USERID, TRUNC(A.DATE_TIME)
    ) T
    JOIN TBLSHIFTS S ON T.USERID = S.SHIFTCODE AND TRUNC(T.DATE_TIME) = TRUNC(S.DATESHIFT);

干杯！！

Answer 2

使用ROW_NUMBER解析函数：

create table tblaccesslogs (
  EVENTID,
  USERID,
  DATE_TIME,
  READERID,
  READERNAME,
  SNAME,
  LOGTYPE,
  USERNAME
) AS
SELECT 177792329,   50078,  DATE '2020-01-02' + INTERVAL '05:45:34' HOUR TO SECOND, 544381431,   'LOCKER',  'RF-T-33', 'IN',   'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177792497,   50078,  DATE '2020-01-02' + INTERVAL '05:45:50' HOUR TO SECOND, 544343461,   'SENSORS', 'RF-T-19', 'IN',   'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177813432,   50078,  DATE '2020-01-02' + INTERVAL '06:35:51' HOUR TO SECOND, 544352385,   'SENSORS', 'RF-T-20', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177813495,   50078,  DATE '2020-01-02' + INTERVAL '06:36:13' HOUR TO SECOND, 544381436,   'LOCKER',  'RF-T-36', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177950118,   50078,  DATE '2020-01-03' + INTERVAL '05:50:48' HOUR TO SECOND, 544352385,   'SENSORS', 'RF-T-20', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177954602,   50078,  DATE '2020-01-03' + INTERVAL '05:54:11' HOUR TO SECOND, 544381436,   'LOCKER',  'RF-T-36', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177967366,   50078,  DATE '2020-01-03' + INTERVAL '06:07:25' HOUR TO SECOND, 544381431,   'LOCKER',  'RF-T-33', 'IN',   'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177967448,   50078,  DATE '2020-01-03' + INTERVAL '06:07:38' HOUR TO SECOND, 544343461,   'SENSORS', 'RF-T-19', 'IN',   'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177977184,   50078,  DATE '2020-01-03' + INTERVAL '08:02:59' HOUR TO SECOND, 544352385,   'SENSORS', 'RF-T-20', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177977214,   50078,  DATE '2020-01-03' + INTERVAL '08:03:21' HOUR TO SECOND, 544381436,   'LOCKER',  'RF-T-36', 'OUT',  'SUBIA, MAY' FROM DUAL UNION ALL
SELECT 177979653,   50078,  DATE '2020-01-03' + INTERVAL '08:23:54' HOUR TO SECOND, 544381431,   'LOCKER',  'RF-T-33', 'IN',   'SUBIA, MAY' FROM DUAL;

然后：

SELECT *
FROM   (
  SELECT t.*,
         ROW_NUMBER() OVER (
           PARTITION BY USERNAME, LOGTYPE, TRUNC( DATE_TIME )
           ORDER BY DATE_TIME ASC
         ) AS first_date_time_rn
  FROM   tblaccesslogs t
  WHERE  READERNAME <> 'LOCKER'
)
WHERE  logtype = 'IN'
AND    first_date_time_rn = 1

输出：

  EVENTID | USERID | DATE_TIME           |  READERID | READERNAME | SNAME   | LOGTYPE | USERNAME   | FIRST_DATE_TIME_RN
--------: | -----: | :------------------ | --------: | :--------- | :------ | :------ | :--------- | -----------------:
177792497 |  50078 | 2020-01-02 05:45:50 | 544343461 | SENSORS    | RF-T-19 | IN      | SUBIA, MAY |                  1
177967448 |  50078 | 2020-01-03 06:07:38 | 544343461 | SENSORS    | RF-T-19 | IN      | SUBIA, MAY |                  1

和：

SELECT *
FROM   (
  SELECT t.*,
         ROW_NUMBER() OVER (
           PARTITION BY USERNAME, LOGTYPE, TRUNC( DATE_TIME )
           ORDER BY DATE_TIME DESC
         ) AS last_date_time_rn
  FROM   tblaccesslogs t
  WHERE  READERNAME <> 'LOCKER'
)
WHERE  logtype = 'OUT'
AND    last_date_time_rn = 1

输出：

  EVENTID | USERID | DATE_TIME           |  READERID | READERNAME | SNAME   | LOGTYPE | USERNAME   | LAST_DATE_TIME_RN
--------: | -----: | :------------------ | --------: | :--------- | :------ | :------ | :--------- | ----------------:
177813432 |  50078 | 2020-01-02 06:35:51 | 544352385 | SENSORS    | RF-T-20 | OUT     | SUBIA, MAY |                 1
177977184 |  50078 | 2020-01-03 08:02:59 | 544352385 | SENSORS    | RF-T-20 | OUT     | SUBIA, MAY |                 1

db<>fiddle here

如何获取每个员工每天的第一个和最后一个数据

How to get first and last data on each day per employee

sql

oracle

oracle11g

greatest-n-per-group