I have a pipe delimited text file. The data is NOT surrounded by double quotes.
Sometime we receive pipes in the data fields themselves. IE |Rob|ert| (instead of Robert)
In SSIS I use a C# component to strip these extraneous pipes using the script below. I just don't know how to convert that into a Java routine to accomplish the same task when reading from a tFileInputDelimited file.
# region Namespaces
using System;
using System.Data;
using Microsoft.SqlServer.Dts.Pipeline.Wrapper;
using Microsoft.SqlServer.Dts.Runtime.Wrapper;
using System.Text.RegularExpressions;
using Microsoft.SqlServer.Dts.Pipeline;
# endregion
[Microsoft.SqlServer.Dts.Pipeline.SSISScriptComponentEntryPointAttribute]
public class ScriptMain : UserComponent
{
public override void ProcessInput(int InputID, PipelineBuffer Buffer)
{
string pattern = "[|]";
Regex reg_exp = new Regex(pattern);
int index = 0;
String DataBackup = "";
while (Buffer.NextRow())
{
try
{
for (int columnIndex = 0;
columnIndex < Buffer.ColumnCount;
columnIndex++)
{
index = columnIndex;
string columnData = null;
if (Buffer.IsNull(columnIndex))
{
columnData = "is NULL";
}
else
{
BufferColumn columnInfo = Buffer.GetColumnInfo(columnIndex);
switch (columnInfo.DataType)
{
case DataType.DT_WSTR:
case DataType.DT_STR:
columnData += Buffer.GetString(columnIndex);
DataBackup = columnData;//Save backup
columnData = reg_exp.Replace(columnData, "");
Buffer.SetString(columnIndex, columnData);
index = 0; //Reset index;
DataBackup = ""; //Reset data
break;
// add code to support more data types here
default:
columnData = "";
index = 0; //Reset index;
DataBackup = ""; //Reset data
break;
}
}
}
}
catch
{
Buffer.SetString(index, DataBackup);
index = 0;
DataBackup = "";
}
}
base.ProcessInput(InputID, Buffer);
}
}
