Question : AWK/GREP combination question

Hello,

Assuming I've got an output similar to the following to parse :

1: 2: 3:

One 12345 Connected Two 67890 Connected

My questions are as folllows :

(1) I would like to display only rows where column 3 status has changed.  "grep -v" is too coarse, it looks at entire rows,  and "awk {print $3} | grep -v" doesn't do the job of outputting whole rows.    Unfortunatley my shell coding expertise is limited to know of other options !

(2) What is an efficient way of continuing the script if there is output from the above ?

1: 2: 3: 4: 5: 6:

count = ` some piped commands | wc -l ` if [$count -gt 0] then do something fi

The above does not seem very efficient, because I would need to call the piped command again in order to be able to use its output.  

Answer : AWK/GREP combination question

command | awk '$3!~"Connected"  {print}'  | while read line
  do
    echo $line | mailx -s 'Lines other than "Connected" found!' [email protected]  
  done

The above will do "nothing" (not send out mail) if no matching lines are found.

wmp