Shell Programming and Scripting

Hi all,
I am working on a shell script to extract information from a file that has output from Oracle sqlplus. The problem is that the output of a single line is spread across multiple lines and i do not know as how to extract the particular filed at ones,which spans multiple lines. Following is what the file looks like from which i am trying to extract information.

DB1 Jack_ Harris_
Anderson Godwin

DB2 Rich Rick
DB3 Rice Paul

DB4 Paul_ Darren_
Keith Gough


As you can see from the file above, the output is skewed. In the first line the second column has name Jack_Anderson, which is divided in to two lines and there is a blank line between row1 and row 2. My question is how can i extract the individual fields. For eg: the AWK $2 on first line should give me Jack_Anderson (but it only gives Jack_) and $NR==2 is a blank line, how can i eliminate this blank line in the search? I am using AWK to extract the required fields, can anyone shed light on this. I am writing this in KORN shell.

Thanks,
Harris

harris.txt:

Code:

DB1 Jack_ Harris_
Anderson Godwin

DB2 Rich Rick
DB3 Rice Paul

DB4 Paul_ Darren_
Keith Gough

harris.awk:

Code:

BEGIN {
   RS=""
}
{
  for(i=1; i<=NF; i++)
    printf("[Record/Field->%d/%d]:: [%s]\n", FNR, i, $i)
}

nawk -f harris.awk harris.txt
produces:

Code:

[Record/Field->1/1]:: [DB1]
[Record/Field->1/2]:: [Jack_]
[Record/Field->1/3]:: [Harris_]
[Record/Field->1/4]:: [Anderson]
[Record/Field->1/5]:: [Godwin]
[Record/Field->2/1]:: [DB2]
[Record/Field->2/2]:: [Rich]
[Record/Field->2/3]:: [Rick]
[Record/Field->2/4]:: [DB3]
[Record/Field->2/5]:: [Rice]
[Record/Field->2/6]:: [Paul]
[Record/Field->3/1]:: [DB4]
[Record/Field->3/2]:: [Paul_]
[Record/Field->3/3]:: [Darren_]
[Record/Field->3/4]:: [Keith]
[Record/Field->3/5]:: [Gough]

Thank you very much