Question

这就是我的表/输出：

ID       ID_2        Status              Timestamp
-------  ----------- ------------------- --------------------
4613840  19668170    Submitted           05-06-2015 16:37:00
4613840  19668330    Submitted           05-06-2015 16:44:00
4613840  19668409    In Progress         05-06-2015 16:48:00
4613840  19669674    SupplierPend        05-06-2015 17:43:00
4613840  19705863    SupplierPend        09-06-2015 15:01:00
4613840  19735270    In Progress         12-06-2015 11:38:00
4613840  19735282    Information Pend    12-06-2015 11:38:00
4613840  19735383    Closed              12-06-2015 11:42:00

我需要在末尾添加另一列，其中包含日期之间的差异（以分钟为单位）。

预期产出：

ID       ID_2        Status              Timestamp            Result
-------  ----------- ------------------- -------------------  ---------
4613840  19668170    Submitted          05-06-2015 16:37:00   0:07:00
4613840  19668330    Submitted          05-06-2015 16:44:00   0:04:00
4613840  19668409    In Progress        05-06-2015 16:48:00   0:55:00
4613840  19669674    SupplierPend       05-06-2015 17:43:00   93:18:00
4613840  19705863    SupplierPend       09-06-2015 15:01:00   68:37:00
4613840  19735270    In Progress        12-06-2015 11:38:00   0:00:00
4613840  19735282    Information Pend   12-06-2015 11:38:00   0:04:00
4613840  19735383    Closed             12-06-2015 11:42:00

第一行的结果是时间戳与ID 19668330和时间戳与ID 19668170之间的差异。

Answer 1

如果您希望以分钟为单位的时间到下一行（按时间顺序排列），请使用lead()。以下是一天中的分数差异：

select t.*,
       (lead(timestamp) over (partition by id order by timestamp) - timestamp
       ) as DayFrac
from table t;

这会将值转换为分钟：

select t.*,
       (lead(timestamp) over (partition by id order by timestamp) - timestamp
       ) * 60 * 24 as Minutes
from table t;

Answer 2

尝试类似：

with x as (
select 1 as id, 'Start' as status, to_timestamp('20150721 07:15:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
union all
select 1 as id, 'Processing' as status, to_timestamp('20150721 08:00:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
union all
select 1 as id, 'Completed' as status, to_timestamp('20150721 08:05:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
union all
select 2 as id, 'Start' as status, to_timestamp('20150721 08:15:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
union all
select 2 as id, 'Processing' as status, to_timestamp('20150721 08:30:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
union all
select 2 as id, 'Completed' as status, to_timestamp('20150721 08:32:00', 'YYYYMMDD HH24:MI:SS') as ts from dual
)
select id, status, ts1, ts2 - ts1 as timespan
from (
  select id, status, ts as ts1, lead(ts, 1) over (partition by id order by ts asc) ts2
  from x
)

输出：

ID  STATUS  TS1 TIMESPAN
1   Start   7/21/2015 7:15:00.000000000 AM  +00 00:45:00.000000
1   Processing  7/21/2015 8:00:00.000000000 AM  +00 00:05:00.000000
1   Completed   7/21/2015 8:05:00.000000000 AM  
2   Start   7/21/2015 8:15:00.000000000 AM  +00 00:15:00.000000
2   Processing  7/21/2015 8:30:00.000000000 AM  +00 00:02:00.000000
2   Completed   7/21/2015 8:32:00.000000000 AM

Oracle SQL - 多个ID的两个日期之间的差异

2 个答案: