Shell Programming and Scripting

hi My requirement is this:
I have a file having around 100000 records pipe delimited. Now I want to compare record 1 with record 2 and similarly record3 with record 4, this goes on.. For this purpose i put a script as follows:

#!bin/ksh

ct_line=1
nxt_line=`expr ${ct_line} + 1`

awk -F "|" -v i=${ct_line} -v j=${nxt_line} '{print i j} NR==i {
ct_dt=$1
ct_name=$2
}
NR==j {
nxt_dt=$1
nxt_name=$2
}
{print ct_dt nxt_dt}
i+=1
j=i+1
' <infile.txt >outfile.txt

Expected output:
1 2
20090501 20090502

Actual Output:
1 2
20090501

Its not all processing NR==j.. Why??

If there is any otehr way to do it in awk??
I mean, I need to read two lines at a time and assign their values to respective variables and compare..?/

Its an urgent requirement.. plz help

As a basis for what it is you're trying to do...

Code:

awk -F "|" '
  (NR-1) % 2 == 0 { CT_DT = $1; CT_NAME = $2; next }
  (NR-1) % 2 == 1 { print CT_DT, $1; print CT_NAME, $2 }
' infile.txt > outfile.txt