miller has asked for the wisdom of the Perl Monks concerning the following question:

Hello,
im trying write a little code that checks if a open is open but it do not seem to work for me, i hope you can help 
#!/usr/contrib/bin/perl -w

$portnum=$ARGV[0];

@netstat=`netstat -an`;

@foundlines = grep /^\d+.\d+.\d+.\d+.$portnum/, @netstat;
foreach $port(@foundlines){
   if ( $port =~ /(\d+.\d+.\d+.\d+.$portnum) (\d+.\d+.\d+.\d+.\d+)\s+\
+d+\s+\d\s+\d+\s+\d\s+(\w+)$/){
      if ($3 ne "LISTEN" or $port ne "ESTABLISHED" ){
         print "$1 $3\n";
      }else{
         print "ok";
      }
   }
}

[download]

The problem is if i pick a number like "94" it get also "9494" BUT cuts off the first part of the ip-addr, so : 228.132.245.9494 So how i can define make sure it: 1) only picks the ones that matches precisely the port nummer? 2) and why do it cut of the first part of the ip-addr ?

the std output from a "netstat -an|grep 949" 

131.228.132.245.9494 131.228.132.159.41713 24820      0  8760      0 T
+IME_WAIT
131.228.132.245.94 131.228.132.159.41703 24820      0  8760      0 TIM
+E_WAIT

[download]

The funny part ??? 

root@toybox: perl portstat.pl 94
228.132.245.9494 TIME_WAIT

[download]

this looks right tho 

root@toybox: perl portstat.pl 9494
131.228.132.245.9494 TIME_WAIT

[download]

Here is what i want: 

root@toybox: perl portstat.pl 9494 
131.228.132.245.94 TIME_WAIT

[download]

EDIT::: Pasted the complete code and added some more stuff... This is really giving me a headache. EDIT:::

Thanks alot for the help.
,Miller

janitored by ybiC: Retitle from "Checking status of ports" for better searching, minor format cleanup for legibility

Replies are listed 'Best First'.
Re: Regex to parse netstat output
by Abigail-II (Bishop) on Feb 04, 2004 at 10:26 UTC
    A dot is special in a regex, and will match anything but a newline. So, it will match a digit as well, and that's what is happening here. To match a dot, either put a backslash in front of it, or write is as [.].

    You also might want to use Regexp::Common and $RE{net}{IPv4} to match an ip number.

    Abigail

[reply]
[d/l]
Re: Regex to parse netstat output
by ysth (Canon) on Feb 04, 2004 at 11:23 UTC
    Thanks for posting more detail. Your first regex should have a space at the end, to reject the 9494 when you are looking for 94. Or the second regex should have a ^ anchor at the beginning. And both need to not use . where you want to match a literal period. To match a period only, use \. (backslash period) or [.].

    To see how it's going wrong, look at this: 

    $ perl -we'"131.228.132.245.9494 " =~ /(\d+)(.)(\d+)(.)(\d+)(.)(\d+)(.
+)(94) / &&
> print qq<\n\nfirst digits: "$1", first .: "$2",\n
> second digits: "$3", second dot: "$4",\n
> third digits: "$5", third dot: "$6",\n
> fourth digits: "$7", fourth dot "$8", portnum: "$9">'


first digits: "228", first .: ".",
second digits: "132", second dot: ".",
third digits: "245", third dot: ".",
fourth digits: "9", fourth dot "4", portnum: "94"

    [download]
    Do you see how the missing ^ and unbackslashed . combine to allow it to match?
[reply]
[d/l]
[select]
Re: Regex to parse netstat output
by ysth (Canon) on Feb 04, 2004 at 10:25 UTC
    Can you give us an example of what might be in $ARGV[0] and which pieces you want extracted (and into what variables)? There are several things wrong there to start with. You are setting $portnm but matching against $port, you are using . where you probably mean \. (just . matches any character except a newline).
[reply]
Re: Regex to parse netstat output
by z3d (Scribe) on Feb 04, 2004 at 14:07 UTC
[reply]
Re: Regex to parse netstat output
by sedusedan (Pilgrim) on Oct 04, 2012 at 09:12 UTC
[reply]

Back to Seekers of Perl Wisdom