Shell Programming and Scripting

If I have a flatfile like vote.dat

NAME | SEX | DATETIME | VOTE
Jason|M|2005-12-10 08.01.30|Y
Benson|M|2005-12-10 12.01.00|Y
William|M|2005-12-10 08.01.09|Y
Nick|M|2005-12-11 09.01.07|Y
Pascal|M|2005-12-11 01.01.06|Y
Mickey|F|2005-12-12 12.01.30|Y

How can I write a korn script to have the out put files depend on the date like

20051210.dat
Jason|M|2005-12-10 08.01.30|Y
Benson|M|2005-12-10 12.01.00|Y
William|M|2005-12-10 08.01.09|Y


20051211.dat
Nick|M|2005-12-11 09.01.07|Y
Pascal|M|2005-12-11 01.01.06|Y


20051212.dat
Mickey|F|2005-12-12 12.01.30|Y

Thanks,

Code:

#! /usr/bin/ksh

while IFS=\| read a b c d
do
    print $a\|$b\|$c\|$d >> $(print "$c" | nawk -F[-\ ] '{print $1$2$3".dat"}')
done < vote.dat

Code:

awk 'BEGIN{FS="[ |]"}
NR>1 { f = $3 ".dat" ; gsub(/-/,"",f)
  if (f!=old_f) close(old_f)
  print  > f ; f = old_f
}' vote.dat

Thanks, it works !

OOOp ! one more question ?
If I would like to sort it , how to I sort and remove the duplicate line in file
print $1$2$3".dat" ???, i dont want to sort for vote.dat because yyyyddmm.dat need to be sorted and uniq -d this file , how do you do it ?

Thanks.

==========
#! /usr/bin/ksh

while IFS=\| read a b c d
do
print $a\|$b\|$c\|$d >> $(print "$c" | nawk -F[-\ ] '{print $1$2$3".dat"}')
done < vote.dat

==========