Using the KORN SHELL ONLY, is there any way that I can grep for a string in column/field 1 only? For instance in a data file named inventory, i have: black goodyear tire black michelin tire film window black I would like to grep for the string 'black' in field 1 ONLY. So my expected output should be: black goodyear tire black michelin tire When i tried using grep, it returned all three fields because they all contain the string black. Also, is there a similar awk command that would do the same thing? Thanks in advance for your help!!

WANDr: Here's an awk script matching on just field 1: #!/bin/ksh nawk ' { if(match($1, "black")) print $0 } ' data.file I'm on Solaris and I need to use nawk. You may have to change that to awk. Regards, Nails

Nails, Thank you very much! Do you know of a korn shell script that will do the same thing...possibly using grep?

WANDr: You might try this out. I read each line and break it into 3 fields; echo the first field, and pipe it to grep black. If there's a match, the exit code is set to zero which is true, and the "and" echo's the 3 fields. If you don't like the boolean expression, you can do this within the while loop: echo $f1|grep black > /dev/null if [ $? -eq 0 ] then echo $f1 $f2 $f3 fi I would expect for a large amount of data, the awk script is more efficient. Look out for the less than sign being dropped at the end of the while loop. #!/bin/ksh # break line into 3 fields while read f1 f2 f3 do { echo $f1|grep black > /dev/null ; } && echo $f1 $f2 $f3 done data.file # need a less than sign between "done" and "data.file" Regards, Nails