有没有办法在 BIGQUERY 的 window 帧中识别最新的非空值？
Is there a way to to identify the latest non null value in a window frame in BIGQUERY?
我希望我能得到一些帮助，以识别 BigQuery 中按日期排序的 window 帧中的最新非空值。到目前为止，我尝试了一些 first_value 和分区，但没有成功。
输入：
Date
Key
Status
2021-01-13
1
In Progress
2021-01-13
2
Closed
2021-01-14
1
Waiting
2021-01-14
2
NULL
2021-01-15
1
NULL
2021-01-15
2
NULL
2021-01-16
1
NULL
2021-01-16
2
NULL
2021-01-17
1
In Progress
2021-01-17
2
NULL
2021-01-18
1
NULL
2021-01-18
2
NULL
预期输出为：
Date
Key
Status
2021-01-13
1
In Progress
2021-01-13
2
Closed
2021-01-14
1
Waiting
2021-01-14
2
Closed
2021-01-15
1
Waiting
2021-01-15
2
Closed
2021-01-16
1
Waiting
2021-01-16
2
Closed
2021-01-17
1
In Progress
2021-01-17
2
Closed
2021-01-18
1
In Progress
2021-01-18
2
Closed
希望我已经包含了相关信息。
预先感谢您的帮助
考虑以下方法
select * except(status),
last_value(status ignore nulls) over(partition by key order by date) as status
from your_table
如果应用于您问题中的示例数据 - 输出为
我希望我能得到一些帮助，以识别 BigQuery 中按日期排序的 window 帧中的最新非空值。到目前为止，我尝试了一些 first_value 和分区，但没有成功。
输入：
| Date | Key | Status |
|---|---|---|
| 2021-01-13 | 1 | In Progress |
| 2021-01-13 | 2 | Closed |
| 2021-01-14 | 1 | Waiting |
| 2021-01-14 | 2 | NULL |
| 2021-01-15 | 1 | NULL |
| 2021-01-15 | 2 | NULL |
| 2021-01-16 | 1 | NULL |
| 2021-01-16 | 2 | NULL |
| 2021-01-17 | 1 | In Progress |
| 2021-01-17 | 2 | NULL |
| 2021-01-18 | 1 | NULL |
| 2021-01-18 | 2 | NULL |
预期输出为：
| Date | Key | Status |
|---|---|---|
| 2021-01-13 | 1 | In Progress |
| 2021-01-13 | 2 | Closed |
| 2021-01-14 | 1 | Waiting |
| 2021-01-14 | 2 | Closed |
| 2021-01-15 | 1 | Waiting |
| 2021-01-15 | 2 | Closed |
| 2021-01-16 | 1 | Waiting |
| 2021-01-16 | 2 | Closed |
| 2021-01-17 | 1 | In Progress |
| 2021-01-17 | 2 | Closed |
| 2021-01-18 | 1 | In Progress |
| 2021-01-18 | 2 | Closed |
希望我已经包含了相关信息。 预先感谢您的帮助
考虑以下方法
select * except(status),
last_value(status ignore nulls) over(partition by key order by date) as status
from your_table
如果应用于您问题中的示例数据 - 输出为