bcp_init
Applies to: 
SQL Server
Azure SQL Database
Azure SQL Managed Instance
Azure Synapse Analytics
Analytics Platform System (PDW)
Initializes the bulk copy operation.
Syntax
RETCODE bcp_init (
HDBC hdbc,
LPCTSTR szTable,
LPCTSTR szDataFile,
LPCTSTR szErrorFile,
INT eDirection);
Unicode and ANSI names:
- bcp_initA (ANSI)
- bcp_initW (Unicode)
Arguments
hdbc
Is the bulk copy-enabled ODBC connection handle.
szTable
Is the name of the database table to be copied into or out of. This name can also include the database name or the owner name. For example, pubs.gracie.titles, pubs..titles, gracie.titles, and titles are all legal table names.
If eDirection is DB_OUT, szTable can also be the name of a database view.
If eDirection is DB_OUT and a SELECT statement is specified using bcp_control before bcp_exec is called, bcp_init szTable must be set to NULL.
szDataFile
Is the name of the user file to be copied into or out of. If data is being copied directly from variables by using bcp_sendrow, set szDataFile to NULL.
szErrorFile
Is the name of the error file to be filled with progress messages, error messages, and copies of any rows that, for any reason, could not be copied from a user file to a table. If NULL is passed as szErrorFile, no error file is used.
eDirection
Is the direction of the copy, either DB_IN or DB_OUT. DB_IN indicates a copy from program variables or a user file to a table. DB_OUT indicates a copy from a database table to a user file. You must specify a user file name with DB_OUT.
Returns
SUCCEED or FAIL.
Remarks
Call bcp_init before calling any other bulk-copy function. bcp_init performs the necessary initializations for a bulk copy of data between the workstation and SQL Server.
The bcp_init function must be provided with an ODBC connection handle enabled for use with bulk copy functions. To enable the handle, use SQLSetConnectAttr with SQL_COPT_SS_BCP set to SQL_BCP_ON on an allocated, but not connected, connection handle. Attempting to assign the attribute on a connected handle results in an error.
When a data file is specified, bcp_init examines the structure of the database source or target table, not the data file. bcp_init specifies data format values for the data file based on each column in the database table, view, or SELECT result set. This specification includes the data type of each column, the presence or absence of a length or null indicator and terminator byte strings in the data, and the width of fixed-length data types. bcp_init sets these values as follows:
The data type specified is the data type of the column in the database table, view, or SELECT result set. The data type is enumerated by SQL Server native data types specified in sqlncli.h. Data itself is represented in its computer form. That is, data from a column of integer data type is represented by a four-byte sequence that is big-or little-endian based on the computer that created the data file.
If a database data type is fixed in length, the data file data is also fixed in length. Bulk-copy functions that process data (for example, bcp_exec) parse data rows expecting the length of the data in the data file to be identical to the length of the data specified in the database table, view, or SELECT column list. For example, data for a database column defined as char(13) must be represented by 13 characters for each row of data in the file. Fixed-length data can be prefixed with a null indicator if the database column allows null values.
When terminator-byte sequence is defined, the length of the terminator-byte sequence is set to 0.
When copying to SQL Server, the data file must have data for each column in the database table. When copying from SQL Server, data from all columns in the database table, view, or SELECT result set are copied to the data file.
When copying to SQL Server, the ordinal position of a column in the data file must be identical to the ordinal position of the column in the database table. When copying from SQL Server, bcp_exec places data based on the ordinal position of the column in the database table.
If a database data type is variable in length (for example, varbinary(22)) or if a database column can contain null values, data in the data file is prefixed by a length/null indicator. The width of the indicator varies based on the data type and version of bulk copy.
To change data format values specified for a data file, call bcp_columns and bcp_colfmt.
Bulk copies to SQL Server can be optimized for tables that do not contain indexes by setting the database recovery model to SIMPLE or BULK_LOGGED. For more information, see Prerequisites for Minimal Logging in Bulk Import and ALTER DATABASE.
If no data file is used, you must call bcp_bind to specify the format and location in memory of the data for each column, then copy data rows to the SQL Server using bcp_sendrow.
Example
This sample shows how to use the ODBC bcp_init function with a format file.
Before you compile and run the C++ code, you need to do the following:
Create an ODBC data source called Test. You can associate this data source with any database.
Run the following Transact-SQL on the database:
CREATE TABLE BCPDate (cola int, colb datetime);In the directory where you will run the application, add a file called Bcpfmt.fmt, and add this to the file:
8.0 2 1SQLCHAR04"\t"1colaSQL_Latin1_General_Cp437_Bin 2SQLCHAR08"\r\n"2colbSQL_Latin1_General_Cp437_BinIn the directory where you will run the application, add a file called Bcpodbc.bcp, and add this to the file:
1 2
Now you are ready to compile and run the C++ code.
// compile with: odbc32.lib sqlncli11.lib
#include <stdio.h>
#include <windows.h>
#include <sql.h>
#include <sqlext.h>
#include <odbcss.h>
SQLHENV henv = SQL_NULL_HENV;
HDBC hdbc1 = SQL_NULL_HDBC;
void Cleanup() {
if (hdbc1 != SQL_NULL_HDBC) {
SQLDisconnect(hdbc1);
SQLFreeHandle(SQL_HANDLE_DBC, hdbc1);
}
if (henv != SQL_NULL_HENV)
SQLFreeHandle(SQL_HANDLE_ENV, henv);
}
int main() {
RETCODE retcode;
SDWORD cRows;
// Allocate the ODBC environment and save handle.
retcode = SQLAllocHandle (SQL_HANDLE_ENV, NULL, &henv);
if ( (retcode != SQL_SUCCESS_WITH_INFO) && (retcode != SQL_SUCCESS)) {
printf("SQLAllocHandle(Env) Failed\n\n");
Cleanup();
return(9);
}
// Notify ODBC that this is an ODBC 3.0 app.
retcode = SQLSetEnvAttr(henv, SQL_ATTR_ODBC_VERSION, (SQLPOINTER) SQL_OV_ODBC3, SQL_IS_INTEGER);
if ( (retcode != SQL_SUCCESS_WITH_INFO) && (retcode != SQL_SUCCESS)) {
printf("SQLSetEnvAttr(ODBC version) Failed\n\n");
Cleanup();
return(9);
}
// Allocate ODBC connection handle, set BCP mode, and connect.
retcode = SQLAllocHandle(SQL_HANDLE_DBC, henv, &hdbc1);
if ( (retcode != SQL_SUCCESS_WITH_INFO) && (retcode != SQL_SUCCESS)) {
printf("SQLAllocHandle(hdbc1) Failed\n\n");
Cleanup();
return(9);
}
retcode = SQLSetConnectAttr(hdbc1, SQL_COPT_SS_BCP, (void *)SQL_BCP_ON, SQL_IS_INTEGER);
if ( (retcode != SQL_SUCCESS_WITH_INFO) && (retcode != SQL_SUCCESS)) {
printf("SQLSetConnectAttr(hdbc1) Failed\n\n");
Cleanup();
return(9);
}
// Sample uses Integrated Security. Create SQL Server DSN using Windows NT authentication.
retcode = SQLConnect(hdbc1, (UCHAR*)"Test", SQL_NTS, (UCHAR*)"", SQL_NTS, (UCHAR*)"", SQL_NTS);
if ( (retcode != SQL_SUCCESS) && (retcode != SQL_SUCCESS_WITH_INFO) ) {
printf("SQLConnect() Failed\n\n");
Cleanup();
return(9);
}
// Initialize the bulk copy.
retcode = bcp_init(hdbc1, "BCPDate", "BCPODBC.bcp", NULL, DB_IN);
if ( (retcode != SUCCEED) ) {
printf("bcp_init(hdbc1) Failed\n\n");
Cleanup();
return(9);
}
// Read the format file.
retcode = bcp_readfmt(hdbc1, "BCPFMT.fmt");
if ( (retcode != SUCCEED) ) {
printf("bcp_readfmt(hdbc1) Failed\n\n");
Cleanup();
return(9);
}
// Execute the bulk copy.
retcode = bcp_exec(hdbc1, &cRows);
if ( (retcode != SUCCEED) ) {
printf("bcp_exec(hdbc1) Failed\n\n");
Cleanup();
return(9);
}
printf("Number of rows bulk copied in = %d.\n", cRows);
// Cleanup
SQLDisconnect(hdbc1);
SQLFreeHandle(SQL_HANDLE_DBC, hdbc1);
SQLFreeHandle(SQL_HANDLE_ENV, henv);
}