如何获取每个 id 的特定行？

Question

要了解具有多种状态的业务流程，

我想根据 created_at 列获取具有以下规则的行：

状态第一行‘created’
最后一行 ‘missing_info’ 在 ‘created’ 之后 (row_no 4)
第一行‘pending’ (row_no 5)
最后一行 ‘missing_info’ 在 ‘pending’ 之后 (row_no 7)
第一行 ‘pending’ after 'missing_info' (row_no 8)
最后一行 ‘successful’ (row_no 10)

下面我突出显示了我要检索的行。

这是 DB-FIDDLE

上的示例数据

这是一般流程：创建 > missing_info > 待定 > 成功。但是也只能这样：创建>成功。

我知道我可以将 QUALIFY 与 window 函数一起使用，并且可以获得 'created' 和 'successful'，如下所示。但是我不知道如何获得临时状态。我怎样才能达到预期的输出？

created AS(

SELECT *

FROM t1

WHERE status = 'created'

QUALIFY ROW_NUMBER() OVER (PARTITION BY STATUS, id ORDER BY created_at) = 1 )

请注意 created 和 successful 是开始和结束状态，因此输出中只有一行。 missing_info 或 pending 等其他状态是临时状态，因此在所需的输出中可以是多个状态。

编辑：

要了解具有多种状态的业务流程，

我想根据 created_at 列获取具有以下两个规则的一些行：

最后一行状态“missing_info”在“待定”之前（第 2 行）
待处理（第 3 行）
最后一行状态“missing_info”在“待定”之前（第 5 行）
待定（第 6 行）

示例数据：

WITH t1 AS (

SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 11:10:00'::timestamp AS created_at UNION ALL

SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 11:20:00'::timestamp AS created_at UNION ALL

SELECT 'A' AS id, 'pending' AS status, '2021-07-15 11:30:00'::timestamp AS created_at UNION ALL

SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 12:10:00'::timestamp AS created_at UNION ALL

SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 12:20:00'::timestamp AS created_at UNION ALL

SELECT 'A' AS id, 'pending' AS status, '2021-07-15 12:30:00'::timestamp AS created_at

    )

SELECT *

FROM t1

期望输出：

Answer 1

Snowflake 实现 MATCH_RECOGNIZE，这是在纯 SQL:

中查找复杂模式的最简单工具

Recognizes matches of a pattern in a set of rows. MATCH_RECOGNIZE accepts a set of rows (from a table, view, subquery, or other source) as input, and returns all matches for a given row pattern within this set. The pattern is defined similarly to a regular expression.

资料准备：

CREATE OR REPLACE TABLE t
AS
WITH t1 AS (
SELECT 'A' AS id, 'created' AS status, '2021-07-15 10:30:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'created' AS status, '2021-07-15 10:38:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 11:10:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 11:12:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'pending' AS status, '2021-07-15 12:05:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 13:36:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'missing_info' AS status, '2021-07-15 14:36:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'pending' AS status, '2021-07-15 12:05:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'successful' AS status, '2021-07-15 16:05:00'::timestamp AS created_at UNION ALL
SELECT 'A' AS id, 'successful' AS status, '2021-07-15 17:00:00'::timestamp AS created_at UNION ALL
SELECT 'B' AS id, 'created' AS status, '2021-07-16 10:30:00'::timestamp AS created_at UNION ALL
SELECT 'B' AS id, 'created' AS status, '2021-07-16 11:30:00'::timestamp AS created_at UNION ALL
SELECT 'B' AS id, 'successful' AS status, '2021-07-16 12:30:00'::timestamp AS created_at
    )     
SELECT * FROM t1;

场景 1 的查询：

SELECT *
FROM t
MATCH_RECOGNIZE (
  PARTITION BY ID
  ORDER BY CREATED_AT
  -- MEASURES MATCH_NUMBER() AS m, --LAST/FIRST/CLASSIFIER/...
  ALL ROWS PER MATCH
  PATTERN (c+m+)
  DEFINE
     c AS status='created'
    ,m AS status='missing_info'
    ,p AS status='pending'
    ,s AS status='succesful'
) mr
ORDER BY ID, CREATED_AT;
-- returns rows 1-4

这里的关键点是作为 Perl 风格的正则表达式提供的模式。在这里，我们正在搜索由一个或多个“missing_info”完成的一个或多个“创建”的模式。

ALL ROWS PER MATCH - return 所有行，但如有必要可以更改为第一行

MEASURES: Specifying Additional Output Columns 可用于提供其他信息，例如 MATCH_NUMBER/MATCH_SEQUENCE_NUMBER/CLASSIFIER 以及更多信息，具体取决于特定需求。

可以通过使用“|”在单个查询中提供更多模式（备选）：(c+m+|pm+|...)

编辑：

"Thanks for the answer! It returns first 4 rows. I was essentially needed 1st and 4th row."

一旦确定了组，就可以过滤第一行和最后一行，例如 QUALIFY。关键是使用我之前提到的 MEASURES：

SELECT *
    FROM t
    MATCH_RECOGNIZE (
      PARTITION BY ID
      ORDER BY CREATED_AT
      MEASURES MATCH_NUMBER() AS mn,
               MATCH_SEQUENCE_NUMBER AS msn
      ALL ROWS PER MATCH
      PATTERN (c+m+)
      DEFINE
         c AS status='created'
        ,m AS status='missing_info'
        ,p AS status='pending'
        ,s AS status='succesful'
    ) mr
    QUALIFY (ROW_NUMBER() OVER(PARTITION BY mn, ID ORDER BY msn) = 1)
          OR(ROW_NUMBER() OVER(PARTITION BY mn, ID ORDER BY msn DESC)=1)
    ORDER BY ID, CREATED_AT;
    -- returns first and last row by group consisted of ID and MATCH_NUMBER

如何获取每个 id 的特定行？

How to get specific rows for each id?

sql

snowflake-cloud-data-platform

window-functions

match-recognize

编辑：