UNIX for Dummies Questions & Answers

I want to add a date/time stamp to the front of each line in an oracle exp log file. I created a shell script:

rm -f expTest.Pipe
mknod expTest.Pipe p
date "+%Y.%m.%d %T"
nawk -f expTest.nawk expTest.Pipe > expTest.Tlog &
exp userid=UID/PW@DB file=expTest.dmp log=expTest.log owner=OWNER 2>> expTest.Pipe
date "+%Y.%m.%d %T"
rm -f expTest.Pipe

and a nawk script (expTest.nawk):

BEGIN {
DT= "date \"+%Y.%m.%d %T\""
DT | getline cur_date
close(DT)
}
{
print cur_date, $0;
}

But when run, the date values are all the same (the start time of the script), even thought this takes about 60 seconds to complete.

Running on Solaris. I've tried awk but it repeatedly fails on the 'DT | ...' line

What am I doing wrong? Thanks in advance.

Just realized BEGIN means something different than I'm use to.

This works:

BEGIN {
start_date=srand()
# print start_date;
}
{
DT="date \"+%Y.%m.%d %T\""
DT | getline cur_date
close(DT)
}
{
print cur_date, $0;
}
END {
end_date=srand()
# print end_date;
print "Run Time (in seconds): ",end_date-start_date;
}