SQL将多个表合并为一个问题

Question

我在 SSMS 中运行以下查询，您可以看到它显示了来自不同数据库的多个表中的结果。

如何将表格合并为一个？

我尝试将 UNION 和 UNION ALL 添加到 SELECT @Sql = COALESCE(@Sql + ' UNION ' + CHAR(13) + CHAR(10), '' ) + 中，但出现 Cannot resolve the collation conflict between "Latin1_General_CI_AS" and "SQL_Latin1_General_CP1_CI_AS" in the UNION operation. 错误

任何想法将不胜感激！

DECLARE @table table(dbname sysname COLLATE Latin1_General_CI_AS)

INSERT INTO @table(dbname)
SELECT NAME COLLATE Latin1_General_CI_AS FROM sys.databases where name like '%AccountsLive'

DECLARE @Sql NVARCHAR(MAX) = NULL;

SELECT  @Sql = COALESCE(@Sql + CHAR(13) + CHAR(10), '' ) +
'SELECT c_broker, det_costheader, cuname, ch_name, CONVERT(VARCHAR(11),ch_date_req,103) AS ''Flight_Date'', IIF(OUT.outstanding>0, ''Yes'', ''No'') AS ''Anything_Outstanding?'', sum(c_grossmargin)/2 AS ''Gross_Margin'' FROM ' 
+ QUOTENAME(dbname) COLLATE Latin1_General_CI_AS + '.dbo.AT_ACS_COMMISSIONS AS COM
LEFT JOIN ' + QUOTENAME(dbname) COLLATE Latin1_General_CI_AS + '.dbo.AT_ACS_PROJECTOUTSTANDING AS OUT
ON COM.det_costheader = OUT.project
WHERE c_broker = ''HKGCGOJT''
GROUP BY det_costheader, c_broker, cuname, ch_name, ch_date_req, outstanding'
FROM    @table

exec( @Sql );

Answer 1

如果任何列在不同数据库中有不同的排序规则，那么您需要向它们添加COLLATE。我添加了几个来向您展示它是如何完成的。

注意事项：

• 这里不需要表变量
• 使用STRING_AGG（或FOR XML）进行聚合，do not use variable coalescing，it is unsafe。
• 如果你打算添加参数，那么使用sp_executesql并将参数传递给动态部分。 不要将参数连接到查询中。

DECLARE @Sql NVARCHAR(MAX);
SET @Sql =
(SELECT
    STRING_AGG(

N'SELECT
    c_broker COLLATE Latin1_General_CI_AS,
    det_costheader,
    cuname COLLATE Latin1_General_CI_AS,
    ch_name,
    CONVERT(VARCHAR(11), ch_date_req, 103) AS [Flight_Date],
    IIF(OUT.outstanding > 0, ''Yes'', ''No'') AS [Anything_Outstanding?],
    sum(c_grossmargin) / 2 AS [Gross_Margin]
FROM ' + QUOTENAME(d.name) + N'.dbo.AT_ACS_COMMISSIONS AS COM
LEFT JOIN ' + QUOTENAME(d.name) + N'.dbo.AT_ACS_PROJECTOUTSTANDING AS OUT
ON COM.det_costheader = OUT.project
WHERE c_broker = ''HKGCGOJT''
GROUP BY det_costheader, c_broker, cuname, ch_name, ch_date_req, outstanding
',
    CAST (N'
UNION ALL
' AS nvarchar(max)) )

FROM sys.databases d
where name like '%AccountsLive'
);

exec( @Sql );

---

如果您使用的是没有STRING_AGG 的旧版SQL Server，则需要使用FOR XML 进行聚合：

DECLARE @Sql NVARCHAR(MAX);
SET @Sql =
STUFF(
(SELECT
    NCHAR(10) + N'UNION ALL' + NCHAR(10) +
N'SELECT
    c_broker COLLATE Latin1_General_CI_AS,
    det_costheader,
    cuname COLLATE Latin1_General_CI_AS,
    ch_name,
    CONVERT(VARCHAR(11), ch_date_req, 103) AS [Flight_Date],
    IIF(OUT.outstanding > 0, ''Yes'', ''No'') AS [Anything_Outstanding?],
    sum(c_grossmargin) / 2 AS [Gross_Margin]
FROM ' + QUOTENAME(d.name) + N'.dbo.AT_ACS_COMMISSIONS AS COM
LEFT JOIN ' + QUOTENAME(d.name) + N'.dbo.AT_ACS_PROJECTOUTSTANDING AS OUT
ON COM.det_costheader = OUT.project
WHERE c_broker = ''HKGCGOJT''
GROUP BY det_costheader, c_broker, cuname, ch_name, ch_date_req, outstanding
'
FROM sys.databases d
where name like '%AccountsLive'
FOR XML PATH(''), TYPE)

.value('text()[1]','nvarchar(max)'), 1, 11, '');

exec( @Sql );

我建议使用PRINT @Sql 进行测试，以确保生成的 SQL 确实有效。