Use bcp to import csv file to sql 2005 or 2008

问题:

I have a csv file and i need to import it to a table in sql 2005 or 2008. The column names and count in the csv are different from the table column names and count. The csv is splitted by a ';' .

Example

CSV FILEcontents:

FirstName;LastName;Country;Age
Roger;Mouthout;Belgium;55

SQL Person Table

Columns: FName,LName,Country

回答1:

I'd create a temporary table, bulk insert the lot, select into the new table what you need and drop the temporary table.

Something like

CREATE TABLE dbo.TempImport
(
    FirstName varchar(255),
    LastName varchar(255),
    Country varchar(255),
    Age varchar(255)
)
GO
BULK INSERT dbo.TempImport FROM 'PathToMyTextFile' WITH (FIELDTERMINATOR = ';', ROWTERMINATOR = '\n')
GO
INSERT INTO dbo.ExistingTable
(
    FName,
    LName,
    Country
)
SELECT  FirstName,
       LastName,
       Country
FROM       dbo.TempImport
GO
DROP TABLE dbo.TempImport
GO

回答2:

You can use a format file when importing with bcp:

Create a format file for your table:

 bcp [table_name] format nul -f [format_file_name.fmt] -c -T 



 9.0
4
1       SQLCHAR       0       100     ","      1     FName             SQL_Latin1_General_CP1_CI_AS
2       SQLCHAR       0       100     ","      2     LName             SQL_Latin1_General_CP1_CI_AS
3       SQLCHAR       0       100     ","      3     Country           SQL_Latin1_General_CP1_CI_AS
4       SQLCHAR       0       100     "\r\n"   0     Age               SQL_Latin1_General_CP1_CI_AS

Edit the import file. The trick is to add a dummy row for the field you want to skip, and add a '0' as server column order.

Then import the data using this format file, specifying your inputfile, this format file and the seperator:

bcp [table_name] in [data_file_name] -t , -f [format_file_name.fmt] -T

回答3:

I now prefer to use XML format files like this with BULK INSERT or OPENROWSET:

<?xml version="1.0"?>
<BCPFORMAT xmlns="http://schemas.microsoft.com/sqlserver/2004/bulkload/format" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
 <RECORD>
  <FIELD ID="1" xsi:type="CharTerm" TERMINATOR="|" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="2" xsi:type="CharTerm" TERMINATOR="|" MAX_LENGTH="37"/>
  <FIELD ID="3" xsi:type="CharTerm" TERMINATOR="|" MAX_LENGTH="41"/>
  <FIELD ID="4" xsi:type="CharTerm" TERMINATOR="|" MAX_LENGTH="17"/>
  <FIELD ID="5" xsi:type="CharTerm" TERMINATOR="\r\n" MAX_LENGTH="10" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
 </RECORD>
 <ROW>
  <COLUMN SOURCE="1" NAME="i" xsi:type="SQLCHAR"/>
  <COLUMN SOURCE="2" NAME="j" xsi:type="SQLUNIQUEID"/>
  <COLUMN SOURCE="3" NAME="k" xsi:type="SQLNUMERIC" PRECISION="18" SCALE="0"/>
  <COLUMN SOURCE="4" NAME="l" xsi:type="SQLBINARY"/>
  <COLUMN SOURCE="5" NAME="m" xsi:type="SQLVARYCHAR"/>
 </ROW>
</BCPFORMAT>

Then you can use the server-side BULK INSERT command as follows:

BULK INSERT foo FROM '\\mydomain.com\bar\bletch' WITH (FORMATFILE='foo.xml', ERRORFILE='foo.errors',  FIRSTROW = 1, BATCHSIZE=10000)

alternatively, if you want to modify the data 'in-flight', you can use the

INSERT foo(i, j,k)
SELECT foo_delimited.i, foo_delimited.j, foo_delimited.k * 2
 OPENROWSET(BULK 'foo',
                   FORMATFILE= 'foo.xml')
        AS foo_delimited

回答4:

For info, with the same structure, you can use this kind of statement:

bcp schema.Table in "/Samples/AdventureWorksDW/DimCurrency.csv" \
    -S db.url \
    -d databaseName \ 
    -U userName \
    -P pwd  \
    -t ; `# The field separator ` \
    -c  `# Insert as character and doesn't ask the data type`  \ 
    -q `# With quote -- seems to be mandatory` 

See example