To all unix expert, I need your help to write a script that will identify and delete duplicate records. Examples: File 1 (comma separated) contains: BBB,23341,inactive AAA,12950,active BBB,3239,active AAA,44098,inactive CCC,40399,inactive The script will look at field 1 and if detect any rows with same field 1 content, delete the row with the string "inactive" in it. Based on the above example, tbe expected result is: AAA,12950,active BBB,3239,active CCC,40399,inactive Thank you for your help.

Based on your sample input, this worked for me: #! /bin/ksh INPUT=file_1 OUTPUT=data.new touch $OUTPUT sort $INPUT | while IFS=, read COL1 COL2 COL3; do grep $COL1 $OUTPUT || print "$COL1,$COL2,$COL3" >> $OUTPUT done -jim

Oops, this one really works: #! /bin/ksh -x INPUT=file1 OUTPUT=data cp /dev/null $OUTPUT cat $INPUT | while IFS=, read COL1 COL2 COL3; do if [[ $COL3 = inactive ]] ; then grep "$COL1.*,active" $INPUT || print "$COL1,$COL2,$COL3" >> $OUTPUT else print "$COL1,$COL2,$COL3" >> $OUTPUT fi done -jim