UNIX for Dummies Questions & Answers

Thanks Yoda for the reply but I am getting syntax error with the command provided.Could you please send it again
bash-3.00$ awk '{gsub(/.*\] |".*/,X);NF-=1}1' testing
awk: syntax error near line 1
awk: illegal statement near line 1

Use nawk on Solaris or SunOS.

Apply below modification:

Code:

nawk '{gsub(/.*\] |".*/,X);NF-=1;$1=$1}1' file

Well, more is always better, but i think this will do. You should have told us this at the beginning.

Actually the username is not the fifth but the seventh column if i count correctly. Anyways, i based my own rule on the first "]" character as this seems to be in every line and is easier to match. I suggest you try the following steps out one by one, always feeding the output of the first as input into the next. Examine the resulting intermediate files carefully for each step so that you can find out where something went wrong.

We start with identifying the beginning of the name: as i said, my rule for the beginning of the name is the "]" and i cut away everything up to the first "]" and the following blank:

Code:

sed 's/^[^]]*] //' /path/to/input > /path/to/output

Now for the names end: we search for a double-quote ('"') and delete from the "word" ("word" here means: some string surrounded by space chars) previous to it (the supposed "Device Model") to the end of the line:

Code:

sed 's/ [^ ][^ ]* ".*//' /path/to/input > /path/to/output

If everything works well (but only after you made sure this is the case!) you can combine both steps into one:

Code:

sed 's/^[^]]*] //;s/ [^ ][^ ]* ".*//' /path/to/input > /path/to/output

If this results indeed in identifying the name part show you how to correct the spacing and capitalization, but one step after the other.

I hope this helps.

bakunin

Yoda , Thanks a Ton .

Query you provided worked absolutely fine. It would be great if you can explain the logic too.

Once again Thanks a lot

Code:

nawk '
        {
                # Substitute zero or more occurences of any chars followed by closing square bracket followed by space .*\] with NULL
                # And substitute double quotes followed by zero or more occurrences of any chars ".* with NULL
                gsub ( /.*\] |".*/, X );

                # Remove last field in the record
                NF -= 1;

                # Force record reconstruction to reflect the changes
                $1 = $1
        }

        # 1 == true, so print current record
        1
' file