如何修复TPT脚本错误？ 'MSG='[Microsoft][SQL Server Native Client 11.0]字符串数据，右截断''

Question

我正在使用 Teradata 并行传输 从 SQL Server 中的表中选择数据并将其插入到 Teradata 数据库中的相同表中。我之前已经成功完成此操作，但这是我第一次在通过命令行运行脚本时看到此错误，我得到以下输出：

TD_LOADER: connecting sessions
TD_LOADER: preparing target table(s)
TD_LOADER: entering DML Phase
TD_LOADER: entering Acquisition Phase
ODBC_READER: sending SELECT request
ODBC_READER: TPT17107: Retrieval error for row: 1
ODBC_READER: TPT17101: Fatal error received from ODBC driver:
STATE=01004, CODE=0,
MSG='[Microsoft][SQL Server Native Client 11.0]String data, right truncation'
ODBC_READER: TPT17101: Fatal error received from ODBC driver:
STATE=01004, CODE=0,type here

脚本生成的 .out 文件的一部分显示：

TD_LOADER: connecting sessions
TD_LOADER: preparing target table(s)
TD_LOADER: entering DML Phase
TD_LOADER: entering Acquisition Phase
Job step 'load_the_data' uses the max row size 1024KB.
Job step 'load_the_data' uses the max Using Data Bytes: 1023991.
Job step 'load_the_data' allows data source containing rows that exceed the maximun 
length of 1024KB
Task(APPLY_1[0001]): checkpoint completed, status = Success
Task(SELECT_2[0001]): checkpoint completed, status = Success
ODBC_READER: sending SELECT request
ODBC_READER: TPT17107: Retrieval error for row: 1
ODBC_READER: TPT17101: Fatal error received from ODBC driver:
STATE=01004, CODE=0,
MSG='[Microsoft][SQL Server Native Client 11.0]String data, right truncation' 
ODBC_READER: TPT17101: Fatal error received from ODBC driver:
STATE=01004, CODE=0,

我的所有 TPT 脚本都遵循此模板：

USING CHAR SET ASCII
DEFINE JOB JobName
DESCRIPTION 'Load data from SQL(tablename) to Teradata(tablename)'


(
  DEFINE OPERATOR Database_Commands()
  TYPE DDL
  ATTRIBUTES
  (
    VARCHAR TdpId = @Teradata_Connection,
    VARCHAR UserName = @Teradata_UserId,
    VARCHAR UserPassword = @Teradata_Password,
    VARCHAR Array ErrorList = ['2580'],
    VARCHAR PrivateLogName = 'ddllog.txt',
    VARCHAR Tracelevel='None',
    VARCHAR LogSQL = ''
 );
  
  /*Examples*/
  DEFINE SCHEMA SourceTableSchema
  (
     columnname      INTEGER,
     columnname2     VARCHAR(50),
     columnname3     VARCHAR(85),
     columnname4     VARCHAR(15),
     columnname4     BYTEINT    
  );
  
  
  DEFINE SCHEMA TargetTableSchema
  (
     columnname      INTEGER,
     columnname2     VARCHAR(50),
     columnname3     VARCHAR(85),
     columnname4     VARCHAR(15),
     columnname4     BYTEINT     
  );
  

  DEFINE OPERATOR ODBC_READER
  TYPE ODBC
  SCHEMA SourceTableSchema
  ATTRIBUTES
  (
      VARCHAR PrivateLogName = 'odbc_log.txt',  
      VARCHAR dsnName = 'SQLDatabaseName',                              
      VARCHAR UserName = @ODBC_UserId,        
      VARCHAR UserPassword = @ODBC_Password,
      VARCHAR TraceLevel = 'None',
      VARCHAR SelectStmt = @SQL 
  );
  
   
  DEFINE OPERATOR TD_LOADER()
  TYPE UPDATE
  SCHEMA TargetTableSchema
  ATTRIBUTES
  (
    VARCHAR TargetTable = @TargetTable,
    VARCHAR LogTable    = @ErrDatabase||'.'||@TargetTable||'_LT', 
    VARCHAR WorkTable   = @WorkDatabase||'.'||@TargetTable||'_WT',
    VARCHAR ErrorTable1 = @ErrDatabase||'.'||@TargetTable||'_ET',
    VARCHAR ErrorTable2 = @ErrDatabase||'.'||@TargetTable||'_UV',
    VARCHAR DateForm = 'IntegerDate',
    VARCHAR TdpId = @Teradata_Connection,
    VARCHAR UserName = @Teradata_UserId,
    VARCHAR UserPassword = @Teradata_Password,
    VARCHAR WorkingDatabase = @TargetDatabase,
    VARCHAR PrivateLogName = 'td_loader_log.txt'
  );

STEP RELEASE
  (
     APPLY
           ' release mload '|| @TargetDatabase ||'.'||@TargetTable ||'  ;'
     TO OPERATOR (Database_Commands());
   );


STEP Initialize
  (
     APPLY
           ' delete from '|| @TargetDatabase ||'.'||@TargetTable ||' '||@DeleteScope||' ;'
     TO OPERATOR (Database_Commands());
   );


STEP load_the_data
  (
    APPLY 
         ('INSERT INTO '|| @TargetDatabase ||'.'||@TargetTable||'
                            (   
                           
                             ...columnnames...

                            );'
          )
    TO OPERATOR (TD_LOADER)
       SELECT                      
                                   
              ...columnnames...
    
         FROM OPERATOR (ODBC_READER); 
  );
);

我尝试检查目标表和源表的表定义，但由于目标表是使用源表的定义创建的，因此它们是相同的，因此我不确定它们之间的大小问题是什么。

Answer 1

我在单独的文件中设置 VARCHAR SelectStmt = @SQL 的值。在此文件中，我必须将 select 语句从选择全部的语句更改为显式单独选择每一行并将 NVARCHAR 列转换为 VARCHAR 列的语句（如 Fred 建议的那样）。

CAST(源列名称 AS VARCHAR(目标列长度)) AS 目标列名称