Perl - Extracting question

[clem9756]

Hello Everyone,

I'm not being familiar with Perl so I wondered if someone could point out how I can translate this to Perl':

# myFile is a 200Mo tab separated value file.
# e.g data in column 5:
Client "JohnDoe" opening a connection from "192.168.0.20"
Administrator connected: "PTMVSB9Z" (10.0.0.1).

---------
perl -i
for each line in "myFile"

add a tab at the end of the line (new empty field)

if column 5 begins with "Client" and contains pattern "opening" or pattern "closing"

then grab the text between the first and second quote in column 5

add the result at the end of the line

fi

add a tab at the end of the line (new empty field)

if column 5 begins with "Admin" or contains pattern "opening a connection"

then grab the IP address in column 5

add the result at the end of the line

fi

...etc
done


Any help would be greatly appreciated!
Thanks

[lar3ry]

Here's some code that should work (untested). Remember that perl's indices are 0-based, for [4] is the fifth column. My regular expressions use /x so they can be better documented, and I use variables to hold "interesting" patterns.

# /usr/bin/perl -w -i.bak

$OPENING_PAT =	'open';
$CLOSING_PAT =	'close';
$ADMIN_PAT =	'(Admin|opening a connection)';
$INET_ADDR =	'\b(\d{1,3}:\d{1,3}:\d{1,3}:\d{1,3})\b';

while (<>) {
    tr /\r\n//d;		# better than "chop/chomp"

    @split = split("\t");

    if ($split[4] =~ /^Client			# match Client at start
			.*			# anything else (don't care)
			${OPENING_PAT}		# opening pattern
			(.*)			# stuff in between
			${CLOSING_PAT}		# closing pattern
		     /x) {
	push @split, $1;			# add new column
    } elsif ($split[4] =~ /^${ADMIN_PAT}	# match Admin/opening... at start
    			    .*			# anything else (don't care)
			    ${INET_ADDR}	# Internet Address
			  /x) {
	push @split, $2;
    }

    print join("\t", @split) . "\n";
}