T-SQL中的拆分功能

Question

我有一個包含逗號分隔字符串的字段的表。我想創建的查詢返回的結果如下：

我想創建查詢，返回分割文本。對於exapmle：

數據表：

| ID         | Names       | 
| ------------------------------------ | -------------------------------- | 
| 63F5D993-3AC9-4EEA-8007-B669542BAD9A | John Smith,Kerry King,Tom Arraya | 

要求的結果：

ID         | Names        
------------------------------------ | ----------- 
63F5D993-3AC9-4EEA-8007-B669542BAD9A | John Smith      
------------------------------------- | ----------- 
63F5D993-3AC9-4EEA-8007-B669542BAD9A | Kerry King      
------------------------------------- | ----------- 
63F5D993-3AC9-4EEA-8007-B669542BAD9A | Tom Arraya 

我發現T-SQL，但其作品不太適合我的情況 「分裂」 的功能。我不能這樣執行它：

SELECT dbo.Split(dbo.name, ',') FROM dbo.Mytable 

它只能執行如下：

​​

但它並不適合我。 此外，我試圖寫光標：

DECLARE @bkb varchar(256) 
DECLARE @Bkb_Cursor CURSOR 
SET @Bkb_Cursor = CURSOR SCROLL FOR 
SELECT bkb.best_know_by 
    FROM [sugarcrm_cmsru_dev].[dbo].[contacts] c 
    left JOIN [dbo].[email_addr_bean_rel] eb 
    ON eb.[bean_id] = c.[id] 
    JOIN [dbo].[email_addresses] ea 
    ON ea.[id] = eb.[email_address_id] 
    JOIN [dbo].[contacts_cstm] ccs 
    ON eb.bean_id = ccs.id_c 
    left JOIN [dbo].[BestKnowBy$] bkb 
    ON c.[campaign_id] =bkb.Con_id 
    where c.deleted = 0; 
OPEN @Bkb_Cursor; 
FETCH NEXT FROM @Bkb_Cursor 
INTO @bkb; 
WHILE @@FETCH_STATUS = 0 
BEGIN 
     PRINT dbo.Splitfn(@bkb); 
     FETCH NEXT FROM @Bkb_Cursor INTO @bkb; 
END 
CLOSE @Bkb_Cursor; 
DEALLOCATE @Bkb_Cursor; 
GO 

但它不工作。我收到錯誤「列」dbo「在此上下文中不允許，並且找不到用戶定義的函數或集合」dbo.Splitfn「。」

我該如何解決這個問題？

我的查詢看起來如下：

SELECT c.[id], 
     bkb.best_know_by 
    FROM [sugarcrm_cmsru_dev].[dbo].[contacts] c 
    left JOIN [dbo].[email_addr_bean_rel] eb 
    ON eb.[bean_id] = c.[id] 
    JOIN [dbo].[email_addresses] ea 
    ON ea.[id] = eb.[email_address_id] 
    JOIN [dbo].[contacts_cstm] ccs 
    ON eb.bean_id = ccs.id_c 
    left JOIN [dbo].[BestKnowBy$] bkb 
    ON c.[campaign_id] =bkb.Con_id 
    where c.deleted = 0; 

的bkb.best_know_by字段包含逗號分隔字符串。我如何在這種情況下使用「交叉應用」？

Answer 1

跨應用會做的伎倆

Select A.ID 
     ,Names = B.Item -- << Return Field from your Split Function 
From YourTable A 
Cross Apply (Select * from dbo.Split(A.Names, ',')) B 

有了您的查詢

SELECT c.[id] 
     ,S.*  --<< Removed bkb.best_know_by and Replaced with S.* (don't know your Split() Return Field) 
FROM [sugarcrm_cmsru_dev].[dbo].[contacts] c 
LEFT JOIN [dbo].[email_addr_bean_rel] eb ON eb.[bean_id] = c.[id] 
JOIN [dbo].[email_addresses] ea ON ea.[id] = eb.[email_address_id] 
JOIN [dbo].[contacts_cstm] ccs ON eb.bean_id = ccs.id_c 
LEFT JOIN [dbo].[BestKnowBy$] bkb ON c.[campaign_id] =bkb.Con_id 
Cross Apply dbo.Split(bkb.best_know_by,',') S 
where c.deleted = 0; 

Answer 2

你需要與你的表的結果集組合將執行功能的每一行使用CROSS APPLY：

SELECT st.ID, spl.value 
FROM SplitTest st 
CROSS APPLY string_split(st.Names, ',') spl 

編輯： 關於增加你的查詢對您的問題，您可以執行以下操作：

;WITH CTE_Query AS (
    SELECT c.[id], 
     bkb.best_know_by 
    FROM [sugarcrm_cmsru_dev].[dbo].[contacts] c 
    left JOIN [dbo].[email_addr_bean_rel] eb 
    ON eb.[bean_id] = c.[id] 
    JOIN [dbo].[email_addresses] ea 
    ON ea.[id] = eb.[email_address_id] 
    JOIN [dbo].[contacts_cstm] ccs 
    ON eb.bean_id = ccs.id_c 
    left JOIN [dbo].[BestKnowBy$] bkb 
    ON c.[campaign_id] =bkb.Con_id 
    where c.deleted = 0 
) 
SELECT cte.id, spl.value 
FROM CTE_Query AS cte 
CROSS APPLY string_split(cte.best_know_by, ',') spl