I have a query which is working fine:
BULK INSERT ZIPCodes
FROM 'e:\5-digit Commercial.csv'
WITH
(
FIRSTROW = 2 ,
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n'
)
but now I want to create a stored procedure for it.
I have written below code to make its stored procedure:
create proc dbo.InsertZipCode
@filepath varcha开发者_StackOverflowr(500)='e:\5-digit Commercial.csv'
as
begin
BULK INSERT ZIPCodes
FROM @filepath
WITH
(
FIRSTROW = 2 ,
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n'
)
end
but its showing error:
Msg 102, Level 15, State 1, Procedure InsertZipCode, Line 6 Incorrect syntax near '@filepath'.
Msg 319, Level 15, State 1, Procedure InsertZipCode, Line 7 Incorrect syntax near the keyword 'with'. If this statement is a common table expression, an xmlnamespaces clause or a change tracking context clause, the previous statement must be terminated with a semicolon.
Please tell me what I am doing wrong and what I can do to make it work in stored procedure.
Thanks
There's nothing wrong with your stored procedure code - the point is: the BULK INSERT
command cannot accept a file name as a variable.
This does work:
BULK INSERT ZIPCodes
FROM 'e:\5-digit Commercial.csv'
WITH
but this never works - within a stored proc or not:
DECLARE @filename VARCHAR(255)
SET @filename = 'e:\5-digit Commercial.csv'
BULK INSERT ZIPCodes
FROM @filename
WITH
So you just cannot do it this way, unfortunately. You could consider building up your BULK INSERT
statement as a string (with a fixed file name) and then execute it as dynamic SQL - but I don't really see any other solution.
DECLARE @filepath nvarchar(500)
SET @filepath = N'e:\5-digit Commercial.csv'
DECLARE @bulkinsert NVARCHAR(2000)
SET @bulkinsert =
N'BULK INSERT ZIPCodes FROM ''' +
@filepath +
N''' WITH (FIRSTROW = 2, FIELDTERMINATOR = '','', ROWTERMINATOR = ''\n'')'
EXEC sp_executesql @bulkinsert
There is an alternative to dynamic SQL if you have access to the SQLCmd exe.
The SqlCmd utility allows you to pass string replacement variables using the -v argument.
You can use a template variable named "filepath" which will be replaced when you execute the script via the cmdline.
The SQL script would look like:
BULK INSERT ZIPCodes
FROM '$(filepath)'
WITH
(
FIRSTROW = 2 ,
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n'
)
end
You would then execute the script from a commandline using something like the following:
sqlcmd -b -S SERVER\INSTANCEHERE -E -i "PATH\FILENAMEHERE.Sql" -v FilePath = "e:\5-digit Commercial.csv" -s "|"
The important part of the example is the -v argument:
-v FilePath = "e:\5-digit Commercial.csv"
You just try it out, I think you need to upload this CSV file directly to the 'E' drive. For that you need to have admin rights, I think, or ask someone who is in database administration.
create procedure dbo.InsertZipCode
AS
BEGIN
SET NOCOUNT ON;
BULK
INSERT ZIPCodes from 'e:\5-digit Commercial.csv'
WITH
(
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n'
)
END
create PROC TestInsert
(
@stuName NVARCHAR(50) ,
@XmlData XML
)
AS
BEGIN
BEGIN TRY
INSERT INTO dbo.Test_Student
( stuName
)
VALUES ( @stuName
);
DECLARE @id BIGINT;
SET @id = ( SELECT SCOPE_IDENTITY()
);
INSERT INTO dbo.Test_Qual
( stuid ,
stuclass ,
InstituteId ,
obtmark ,
totalmark ,
per
)
SELECT @id ,
col.value('stuclass[1]', 'nvarchar(50)') AS stuclass ,
col.value('InstituteId[1]', 'int') AS InstituteId ,
col.value('obtmark[1]', 'nvarchar(100)') AS obtmark ,
col.value('totalmark[1]', 'nvarchar(50)') AS totalmark ,
col.value('per[1]', 'nvarchar(50)') AS per
FROM @XmlData.nodes('Parent/child') AS Doc ( col );
SELECT @id AS RegisIdNUH ,
1 AS Flag ,
'Save' AS Msg
FROM dbo.Test_Student R
WHERE R.stuid = @id;
END TRY
BEGIN CATCH
SELECT 0 AS Flag ,
'Some Error occur' AS Msg;
ROLLBACK;
END CATCH;
END;
精彩评论