我正在尝试更新 SSRS 中的报告。该查询是由另一位同事使用存储过程构建的，该同事已离开项目并且无法再联系。他为所有批次使用了一个内存表，以便添加一些聚合。
作为输出的列是：
- 批号
- 转换日期
- 源系统
- 公司类型
- MAX(B.FooterLines) AS FooterTotalLines
- MAX(SRC.TotalLines) 作为 SourceTotalLines
- MAX(BalancingLinesTable.BalancingLines) AS 平衡线
- SUM(STG.TotalLines) 作为 StagingTotalLines
- (CAST(STG.Company AS int) * 1000000) + B.BatchID AS CompanyCode
但是我需要添加一个名为平衡线的额外列。我为它开发了查询并且它有效：
SELECT COUNT(0) as "Balancing lines"
FROM [Source].[Staging].[TransactionLine] T
INNER JOIN
[Source].[Staging].HeaderLine H
ON T.HeaderID = H.HeaderID
where H.BatchID = 1234* and H.SourceSystemInstance = 'ABC' and AccountNumber = '98765'
*1234 should be variable -> B.BatchID and also the SourseSystemInstance 'ABC'-> B.SourceSystem
但是，我同事写的存储过程对我来说仍然很陌生。你知道如何将我的工作查询“添加”到他的工作存储过程中吗？
为清楚起见，使用的表格：
- @批次 B
- 标题行 H
- 页脚线 FL
- 'INNER JOIN 表' BalancingLinesTable
- 'INNER JOIN 表' SRC
- 'LEFT JOIN 表' STG
存储过程：
USE [Source]
ALTER PROCEDURE [dbo].[usp_GetSomeCounts] (
@StartAt datetime
)
AS BEGIN
SET NOCOUNT ON;
-- Set a default for start date if none is specified
SELECT @StartAt = ISNULL(@StartAt, DATEADD(dd, -7, GETDATE()))
-- Use an in-memory table for all batches for the specified period
DECLARE @batches TABLE(BatchID int, TransformDate datetime, HeaderID nvarchar(36), CompanyType nvarchar(30), SourceSystem nvarchar(4), FooterLines int)
-- Fill in-memory table batches
INSERT INTO @batches (
BatchID
,TransformDate
,HeaderID
,CompanyType
,SourceSystem
,FooterLines
)
SELECT
H.BatchID
,H.TransformDate
,H.HeaderID
,CompanyTypeID
,H.SourceSystemInstance
,ISNULL(FL.TotalTransactionLines, 0)
FROM
Staging.HeaderLine H WITH (NOLOCK)
INNER JOIN
Staging.FooterLine FL WITH (NOLOCK)
ON
H.HeaderID = FL.HeaderID
WHERE
H.BatchDate >= @StartAt
ORDER BY
H.BatchID
/* Using in-memory table 'batches', count all valid transactions in both Source and Staging
databases. Include the footer totals */
SELECT
B.BatchID
, B.TransformDate
, B.SourceSystem
, B.CompanyType
, MAX(B.FooterLines) AS FooterTotalLines
, MAX(SRC.TotalLines) AS SourceTotalLines
, MAX(BalancingLinesTable.BalancingLines) AS BalancingLines
, SUM(STG.TotalLines) AS StagingTotalLines
, (CAST(STG.Company AS int) * 1000000) + B.BatchID AS CompanyCode
FROM
@batches B
INNER JOIN (
SELECT
B.BatchID
,B.HeaderID
,COUNT(0) as BalancingLines
FROM
@batches B
-- this 'inner join table' BalancingLinesTable is what I added
INNER JOIN
Staging.TransactionLine T WITH (NOLOCK)
ON B.HeaderID = T.HeaderID
INNER JOIN
Staging.HeaderLine H WITH (NOLOCK)
ON T.HeaderID = H.HeaderID
WHERE H.BatchID = B.BatchID AND H.SourceSystemInstance = B.SourceSystem AND AccountNumber = 399990
GROUP BY
B.BatchID
,B.HeaderID
) BalancingLinesTable ON B.BatchID = BalancingLinesTable.BatchID
INNER JOIN (
SELECT
B.BatchID
,B.HeaderID
,COUNT(0) AS TotalLines
FROM
@batches B
INNER JOIN
Staging.TransactionLine T WITH (NOLOCK)
ON
B.HeaderID = T.HeaderID
WHERE
T.LineStatus = 'N'
GROUP BY
B.BatchID
,B.HeaderID ) SRC ON B.BatchID = SRC.BatchID
LEFT JOIN (
SELECT
B.BatchID
,B.HeaderID
,MT.Company
,COUNT(0) AS TotalLines
FROM
@batches B
INNER JOIN
[Production].[dbo].[Mutated_Transactions] MT WITH (NOLOCK)
ON
B.HeaderID = MT.HeaderID
WHERE
MT.RowHasError = 'N'
GROUP BY
B.BatchID
,B.HeaderID
,MT.Company) STG ON SRC.BatchID = STG.BatchID AND SRC.HeaderID = STG.HeaderID
GROUP BY
B.BatchID
,B.TransformDate
,B.SourceSystem
,B.CompanyType
,STG.Company
ORDER BY
B.TransformDate desc
,B.BatchID
结尾