4
我有以下格式的字符串：SQL查詢拆分字符串爲行和列
A:B:C;J:K;P:L:J;
我想冒號後的字符串分割（:)和分號（後啓動一個新行）。 任何人都可以幫我查詢。
輸出示例：
A B C
J K
P L J
A
回答
4
不知道，我理解正確的，但如果你需要數據，三列行集：
declare @str nvarchar(max)
set @str = 'A:B:C;J:K;P:L:J;'
select p.[1] as Column1, p.[2] as Column2, p.[3] as Column3
from (
select T.c.value('.', 'nvarchar(200)') [row], row_number() over (order by @@spid) rn1
from (select cast('<r>' + replace(@str, ';', '</r><r>') + '</r>' as xml) xmlRows) [rows]
cross apply xmlRows.nodes('/r') as T(c)
where T.c.value('.', 'nvarchar(200)') != ''
) t1
cross apply (
select NullIf(T.c.value('.', 'nvarchar(200)'), '') row2,
row_number() over (order by @@spid) rn
from (select cast('<r>' + replace(t1.row, ':', '</r><r>') + '</r>' as xml) xmlRows) [rows]
cross apply xmlRows.nodes('/r') as T(c)
) t2
pivot (max(t2.row2) for t2.rn in ([1], [2], [3])) p
order by p.rn1
輸出
Column1 Column2 Column3
-------- -------- -------
A B C
J K NULL
P L J
7
嘗試這一個 -
溶液＃1：
DECLARE @t VARCHAR(100)
SELECT @t = 'A:B:C;J:K;P:L:J;'
SELECT *
FROM (
SELECT token = t.c.value('.', 'VARCHAR(100)')
FROM
(
SELECT xmls = CAST('<t>' +
REPLACE(
REPLACE(@t, ':', ' '),
';',
'</t><t>') + '</t>' AS XML)
) r
CROSS APPLY xmls.nodes('/t') AS t(c)
) t
WHERE t.token != ''
輸出：
----------
A B C
J K
P L J
溶液＃2：
DECLARE @t VARCHAR(100)
SELECT @t = 'A:B:C;J:K;P:L:J;'
PRINT REPLACE(REPLACE(@t, ':', ' '), ';', CHAR(13) + CHAR(13))
輸出：
A B C
J K
P L J
溶液＃3：
DECLARE @t VARCHAR(100)
SELECT @t = 'A:B:C;J:K;P:L:J;'
SELECT [1], [2], [3]
FROM (
SELECT
t2.id
, t2.name
, rn2 = ROW_NUMBER() OVER (PARTITION BY t2.id ORDER BY 1/0)
FROM (
SELECT
id = t.c.value('@n', 'INT')
, name = t.c.value('@s', 'CHAR(1)')
FROM (
SELECT x = CAST('<t s = "' +
REPLACE(token + ':', ':', '" n = "' + CAST(rn AS VARCHAR(10))
+ '" /><t s = "') + '" />' AS XML)
FROM (
SELECT
token = t.c.value('.', 'VARCHAR(100)')
, rn = ROW_NUMBER() OVER (ORDER BY 1/0)
FROM (
SELECT x = CAST('<t>' + REPLACE(@t, ';', '</t><t>') + '</t>' AS XML)
) r
CROSS APPLY x.nodes('/t') t(c)
) t
) d
CROSS APPLY x.nodes('/t') t(c)
) t2
WHERE t2.name != ''
) t3
PIVOT (
MAX(name) FOR rn2 IN ([1], [2], [3])
) p
輸出：
1 2 3
---- ---- ----
A B C
J K NULL
P L J
1
在SQL中沒有內置的分割函數，但是您可以通過簡單地運行像這樣的語句來創建函數。在您想要執行多部分拆分的其他情況下，此解決方案可重複使用。
CREATE FUNCTION dbo.Split
(
@TextToSplit nvarchar(2000),
@SplitOn nvarchar(5)
)
RETURNS @RtnValue table
(
Id int identity(1,1),
Data nvarchar(100)
)
AS
BEGIN
Declare @Cnt int
Set @Cnt = 1
While (Charindex(@SplitOn,@TextToSplit)>0)
Begin
Insert Into @RtnValue (data)
Select
Data = ltrim(rtrim(Substring(@TextToSplit,1,Charindex(@SplitOn,@TextToSplit)- 1)))
Set @RowData = Substring(@TextToSplit,Charindex(@SplitOn,@TextToSplit)+1,len(@TextToSplit))
Set @Cnt = @Cnt + 1
End
Insert Into @RtnValue (data)
Select Data = ltrim(rtrim(@TextToSplit))
Return
END
然後你就可以像這樣運行：
Declare @SplitThis as varchar(2000) = 'A:B:C;J:K;P:L:J'
DECLARE @Val as varchar(500)
DECLARE MyCursor CURSOR FAST_FORWARD FOR
select data from split(@SplitThis,';')
OPEN MyCursor
FETCH NEXT FROM MyCursor INTO @Val
WHILE @@FETCH_STATUS = 0
BEGIN
select(
select ' ' + data
from split(@Val,':')
for xml Path(''),type).value('.','varchar(100)') as cat
FETCH NEXT FROM MyCursor INTO @Val
END
CLOSE MyCursor
DEALLOCATE MyCursor
+0
祝你的項目好運。 – Brian
1
對於較大的字符串，我發現UDF循環方法比xml方法更有效率。
create Function [Split_Text_Into_Table_Multi_Column] (
@Row_Delimit_Char Varchar(5)
,@Column_Delimit_Char Varchar(5)
,@Text_To_Split Varchar(Max))
Returns
@Output_Table Table (
row_id Int Identity (1, 1)
,column_1 Varchar(Max) Collate Latin1_General_CS_AS
,column_2 Varchar(Max) Collate Latin1_General_CS_AS
,column_3 Varchar(Max) Collate Latin1_General_CS_AS
)
As
Begin
Declare
@Column_Delimit_Index_1 Int
,@Column_Delimit_Index_2 Int
,@String Varchar(Max)
,@String_1 Varchar(Max)
,@String_2 Varchar(Max)
,@String_3 Varchar(Max)
,@String_Len Int
,@Delimit_Len TinyInt
Declare
@Last_Delimit_Index Int = 0
,@Next_Delimit_Index Int = 1
--always treat new line as delimiter
Select
@Text_To_Split = Replace(Replace(@Text_To_Split, Char(10), @Row_Delimit_Char), Char(13), @Row_Delimit_Char)
--begin loop while next delimiter found
While @Next_Delimit_Index > 0
Begin
--find next delimiter
Select
@Next_Delimit_Index = CharIndex(@Row_Delimit_Char, @Text_To_Split, @Last_Delimit_Index + 1)
--get value to insert
Select
@String = Ltrim(Rtrim(Substring(@Text_To_Split, @Last_Delimit_Index + 1, Case
When @Next_Delimit_Index = 0 Then Len(@Text_To_Split) - @Last_Delimit_Index
Else @Next_Delimit_Index - @Last_Delimit_Index - 1
End)))
--Insert Non blank strings only
If @String <> ''
Begin
Select
@Delimit_Len = Len(@Column_Delimit_Char)
Select
@String_Len = DataLength(@String)/*includes trailing spaces*/
,@Column_Delimit_Index_1 = 1
,@Column_Delimit_Index_2 = IsNull(NullIf(CharIndex(@Column_Delimit_Char, @String, @Column_Delimit_Index_1), 0), @String_Len + 1)
,@String_1 =
Case
When @Column_Delimit_Index_2 <= @Column_Delimit_Index_1 Or
@Column_Delimit_Index_2 - @Column_Delimit_Index_1 = 0 Then Null
Else Substring(@String, @Column_Delimit_Index_1, @Column_Delimit_Index_2 - @Column_Delimit_Index_1)
End
,@Column_Delimit_Index_1 = @Column_Delimit_Index_2 + @Delimit_Len
,@Column_Delimit_Index_2 = IsNull(NullIf(CharIndex(@Column_Delimit_Char, @String, @Column_Delimit_Index_1), 0), @String_Len + 1)
,@String_2 =
Case
When @Column_Delimit_Index_2 <= @Column_Delimit_Index_1 Or
@Column_Delimit_Index_2 - @Column_Delimit_Index_1 = 0 Then Null
Else Substring(@String, @Column_Delimit_Index_1, @Column_Delimit_Index_2 - @Column_Delimit_Index_1)
End
,@Column_Delimit_Index_1 = @Column_Delimit_Index_2 + @Delimit_Len
,@Column_Delimit_Index_2 = IsNull(NullIf(CharIndex(@Column_Delimit_Char, @String, @Column_Delimit_Index_1), 0), @String_Len + 1)
,@String_3 =
Case
When @Column_Delimit_Index_2 <= @Column_Delimit_Index_1 Or
@Column_Delimit_Index_2 - @Column_Delimit_Index_1 = 0 Then Null
Else Substring(@String, @Column_Delimit_Index_1, @Column_Delimit_Index_2 - @Column_Delimit_Index_1)
End
Insert Into
@Output_Table (
column_1
,column_2
,column_3)
Values
(
@String_1
,@String_2
,@String_3)
End
--set last last char index
Select
@Last_Delimit_Index = @Next_Delimit_Index
End
Return
End
--sample output
Select *
from [Split_Text_Into_Table_Multi_Column] (',',' ','
917064578 70
917581826 66
917635915 66
917663205 66
918320481 66
918752247 ER
918892315 70
919162748 52
919169913 JS
919169927 70
919187495 52
919207137 52
919226194 70
919252530 52
919252544 KG
919269942 52
919269957 KG
919269961 47
')
輸出
row_id|column_1|column_2|column_3
1|917064578|70|NULL
2|917581826|66|NULL
3|917635915|66|NULL
4|917663205|66|NULL
5|918320481|66|NULL
6|918752247|ER|NULL
7|918892315|70|NULL
8|919162748|52|NULL
9|919169913|JS|NULL
10|919169927|70|NULL
11|919187495|52|NULL
12|919207137|52|NULL
13|919226194|70|NULL
14|919252530|52|NULL
15|919252544|KG|NULL
16|919269942|52|NULL
17|919269957|KG|NULL
18|919269961|47|NULL
相關問題
- 1. 拆分查詢字符串？
- 2. 拆分URL和查詢字符串
- 3. SQL：用查詢拆分逗號分隔的字符串列表？
- 4. java字符串拆分查詢
- 5. 使用MSQL查詢拆分字符串？
- 6. 從查詢字符串拆分請求
- 7. 長查詢字符串被拆分
- 8. 拆分字符串Teradata SQL
- 9. SQL拆分字符串
- 10. 將字符串拆分成行Oracle SQL
- 11. 將字符串列表拆分爲字符串列表列表
- 12. 蟒蛇 - 拆分字符串列表分爲多個字符串
- 13. 將字符串拆分爲兩列（SQL或PL/SQL）
- 14. 將字符串拆分爲字符串
- 15. 將字符串拆分爲字符串
- 16. 將很長的字符串拆分爲行和列
- 17. 如何將字符串列拆分爲4個字符串列？
- 18. C＃拆分字符串 - 將字符串拆分爲數組
- 19. 如何將字符串拆分爲列
- 20. 將字符串拆分爲新列
- 21. 將字符串拆分爲2列Oracle
- 22. 拆分列表轉換爲字符串
- 23. mssql字符串拆分和更新列
- 24. 將SQL字符串拆分爲多個字符串
- 25. 將長字符串拆分爲每行1個字sql
- 26. 奇怪的字符串拆分行爲
- 27. 將字符串拆分爲多行
- 28. 將字符串拆分爲新行
- 29. 字符串拆分方法行爲
- 30. 將字符串拆分爲「。」
該解決方案不是一般的。給定列分隔符和行分隔符，它不能將任何字符串拆分爲表格。它不會處理該字符串A：B：C：X; J：K：X; P：L：J：X; （增加第4欄）。 –