滞后功能正常

我得到这个查询滞后功能正常

CREATE OR REPLACE View QTY_VIEW As 
SELECT  
    DATA_VIEW.RAW_ID as RAW_ID, 
    DATA_VIEW.CREATE_DATE as CREATE_DATE, 
    DATA_VIEW.UPDATE_DATE as UPDATE_DATE, 
    DATA_VIEW.INITIAL_QTY as INITIAL_QTY, 
    DATA_VIEW.ACTUAL_QTY as ACTUAL_QTY, 
    DATA_VIEW.USED_QTY as USED_QTY, 
    DATA_VIEW.LAST_USED_BY as LAST_USED_BY, 
    CASE 
    WHEN DATA_VIEW.ACTUAL_QTY = DATA_VIEW.INITIAL_QTY 
     THEN 0    
    WHEN LAG(DATA_VIEW.USED_QTY, 1, 0) 
      OVER (ORDER BY DATA_VIEW.RAW_ID ASC,DATA_VIEW.ACTUAL_QTY,DATA_VIEW.UPDATE_DATE) 
      = DATA_VIEW.USED_QTY - DATA_VIEW.USED_QTY 
     THEN 0    
    ELSE 
     LAG(DATA_VIEW.USED_QTY, 1, 0) 
     OVER (ORDER BY DATA_VIEW.RAW_ID ASC,DATA_VIEW.ACTUAL_QTY,DATA_VIEW.UPDATE_DATE) 
     - DATA_VIEW.USED_QTY 
    END as SINGLE_USE_QTY 
FROM 
    DATA_VIEW 
order by RAW_ID asc, UPDATE_DATE,ACTUAL_QTY ;

我觉得我没有使用LAG正常运行作为最后结果为负值，而不是积极的。你能帮我请一些建议吗？的DATA_VIEW是而QTY_VIEW就是这样我做错了什么？

来源

2017-06-26 Andrei Paul

你的主要'ORDER BY'不匹配'LAG'中使用'ORDER BY'，所以显示的最后一条记录可能不是LAG中使用的最后一条记录？ –

ORDER BY似乎是问题，因为我有不同的顺序。谢谢 –

夫妇事情似乎“腥”给我..

1）你排序上RAW_ID ..这是一个字符串...它包含数字，... 您的排序顺序将整理raw_1，然后raw_10 ..然后raw_11（etc）...然后raw_19，然后raw_2 ..然后raw_20 ..然后raw_21 ... etc ...然后raw_3 ..等等您应该考虑使用一个PROPER序列...纯数值..所以你可以适当地分类。

2）你的第二个CASE子句中..你..有本质：当LAG（东西）=数量 - 数量... 这基本上是：当LAG（东西）= 0 ... 是这个意？如果你是第一次出现后，它可能是更明确地说：当LAG（东西）IS NULL你显示为您的结果

3）..绝对不符合他们的查询/数据您提供的。 ..我重新键入所有的数据后（感谢图像btw ..下一次，请将实际的文本，以便我们可以复制/粘贴..使我们的工作更容易帮助你：P）我不得到和你一样的结果... 所以请重新发布一个完整的测试案例...与适当的数据，查询..和这样的文本格式...所以我们可以验证，我们不会在示例中引入错别字;）

这就是我从你发布的内容中得到的：

RAW_ID UPDATE_DA INITIAL_QTY ACTUAL_QTY USED_QTY SINGLE_USE_QTY 
    ------ --------- ----------- ---------- ---------- -------------- 
    raw_1 06-JUN-17  20000 48.52559 19951.47    0 
    raw_10 06-JUN-17  20000 20.76559 19979.23   -27.76 
    raw_11 06-JUN-17  20000 17.29559 19982.7   -3.47 
    raw_12 06-JUN-17  20000 13.82559 19986.17   -3.47 
    raw_13 06-JUN-17  20000 10.35559 19989.64   -3.47 
    raw_14 06-JUN-17  20000 6.885593 19996.11   -6.47 
    raw_15 06-JUN-17  20000 3.415594 19996.58   -.47 
    raw_2 06-JUN-17  20000 45.0556 19954.94   41.64 
    raw_3 06-JUN-17  20000 41.58559 19958.41   -3.47 
    raw_4 06-JUN-17  20000 38.11559 19961.88   -3.47 
    raw_5 06-JUN-17  20000 38.11559 19961.88    0 
    raw_6 06-JUN-17  20000 34.6456 19965.35   -3.47 
    raw_7 06-JUN-17  20000 31.17559 19968.82   -3.47 
    raw_8 06-JUN-17  20000 27.70559 19972.29   -3.47 
    raw_9 06-JUN-17  20000 24.23559 19975.76   -3.47 

    15 rows selected.

这里是我重新工作的查询..但不知道你真正想要的..

with data_view as (
    select 'raw_1' raw_id, 1 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 48.52559 actual_qty, 19951.47 used_qty from dual union all 
    select 'raw_2' raw_id, 2 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 45.0556 actual_qty, 19954.94 used_qty from dual union all 
    select 'raw_3' raw_id, 3 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 41.58559 actual_qty, 19958.41 used_qty from dual union all 
    select 'raw_4' raw_id, 4 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 38.11559 actual_qty, 19961.88 used_qty from dual union all 
    select 'raw_5' raw_id, 5 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 38.11559 actual_qty, 19961.88 used_qty from dual union all 
    select 'raw_6' raw_id, 6 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 34.6456 actual_qty, 19965.35 used_qty from dual union all 
    select 'raw_7' raw_id, 7 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 31.17559 actual_qty, 19968.82 used_qty from dual union all 
    select 'raw_8' raw_id, 8 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 27.70559 actual_qty, 19972.29 used_qty from dual union all 
    select 'raw_9' raw_id, 9 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 24.23559 actual_qty, 19975.76 used_qty from dual union all 
    select 'raw_10' raw_id, 10 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 20.76559 actual_qty, 19979.23 used_qty from dual union all 
    select 'raw_11' raw_id, 11 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 17.29559 actual_qty, 19982.7 used_qty from dual union all 
    select 'raw_12' raw_id, 12 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 13.82559 actual_qty, 19986.17 used_qty from dual union all 
    select 'raw_13' raw_id, 13 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 10.35559 actual_qty, 19989.64 used_qty from dual union all 
    select 'raw_14' raw_id, 14 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 6.885593 actual_qty, 19996.11 used_qty from dual union all 
    select 'raw_15' raw_id, 15 raw_id2, to_date('6/6/2017 16:01','dd/mm/yyyy hh24:mi') update_date, 20000 initial_qty, 3.415594 actual_qty, 19996.58 used_qty from dual 
) 
    SELECT  
     DATA_VIEW.RAW_ID as RAW_ID, 
     DATA_VIEW.RAW_ID2 as RAW_ID2, 
     DATA_VIEW.UPDATE_DATE as UPDATE_DATE, 
     DATA_VIEW.INITIAL_QTY as INITIAL_QTY, 
     DATA_VIEW.ACTUAL_QTY as ACTUAL_QTY, 
     DATA_VIEW.USED_QTY as USED_QTY, 
     LAG(DATA_VIEW.USED_QTY, 1, 0) OVER  
       (ORDER BY DATA_VIEW.RAW_ID2 ASC,DATA_VIEW.ACTUAL_QTY,DATA_VIEW.UPDATE_DATE) 
      as lag_used_qty, 
     CASE 
      WHEN DATA_VIEW.ACTUAL_QTY = DATA_VIEW.INITIAL_QTY 
       THEN 0    
      WHEN LAG(DATA_VIEW.USED_QTY, 1, 0) OVER 
       (ORDER BY DATA_VIEW.RAW_ID2 ASC,DATA_VIEW.ACTUAL_QTY,DATA_VIEW.UPDATE_DATE) IS NULL 
       THEN 0    
      ELSE 
       LAG(DATA_VIEW.USED_QTY, 1, 0) OVER 
       (ORDER BY DATA_VIEW.RAW_ID2 ASC,DATA_VIEW.ACTUAL_QTY,DATA_VIEW.UPDATE_DATE) 
       - DATA_VIEW.USED_QTY 
     END as SINGLE_USE_QTY 
    FROM 
     DATA_VIEW 
    order by RAW_ID2 asc, ACTUAL_QTY, UPDATE_DATE ;

和这里的结果：

RAW_ID RAW_ID2 UPDATE_DA INITIAL_QTY ACTUAL_QTY USED_QTY SINGLE_USE_QTY 
    ------ ---------- --------- ----------- ---------- ---------- -------------- 
    raw_1   1 06-JUN-17  20000 48.52559 19951.47    0 
    raw_2   2 06-JUN-17  20000 45.0556 19954.94   -3.47 
    raw_3   3 06-JUN-17  20000 41.58559 19958.41   -3.47 
    raw_4   4 06-JUN-17  20000 38.11559 19961.88   -3.47 
    raw_5   5 06-JUN-17  20000 38.11559 19961.88    0 
    raw_6   6 06-JUN-17  20000 34.6456 19965.35   -3.47 
    raw_7   7 06-JUN-17  20000 31.17559 19968.82   -3.47 
    raw_8   8 06-JUN-17  20000 27.70559 19972.29   -3.47 
    raw_9   9 06-JUN-17  20000 24.23559 19975.76   -3.47 
    raw_10   10 06-JUN-17  20000 20.76559 19979.23   -3.47 
    raw_11   11 06-JUN-17  20000 17.29559 19982.7   -3.47 
    raw_12   12 06-JUN-17  20000 13.82559 19986.17   -3.47 
    raw_13   13 06-JUN-17  20000 10.35559 19989.64   -3.47 
    raw_14   14 06-JUN-17  20000 6.885593 19996.11   -6.47 
    raw_15   15 06-JUN-17  20000 3.415594 19996.58   -.47 

    15 rows selected.

来源

2017-06-26 13:26:39 Ditto

表格由其他人制作，我无权修改其中的任何内容。我想要做的是从数据库中获取信息，通过3个视图解析它，然后将其保存在其他数据库中，该服务将每天运行。我之前举例说明的是第二个视图，它应该获取数量和结果用于每个步骤的数量。我使用LAG函数来得到结果（RowN - RowN + 1）。至少这是我了解LAG正在做的事情。我跑了一些测试，发现目前的问题在ORDER BY。 –

这种查询需要7-9小时运行。那些日子我会运行更多的测试，我希望那是问题所在。但我知道我没有正确使用LAG函数，并且整个语法都可能不正确。感谢您的信息和帮助。 –

滞后功能正常

回答

相关问题