Month: June 2007

[problem]

You want to listen on a port, maybe to test firewalls! 🙂

Excellent introduction to client server software.

[/problem]

[solution]

Another use bit of socket programming, courtesy of Perl! 🙂

Extremely useful bit of Perl, which can be used with the other post on this site, to confirm network connectivity (port scanner).

This code (see example tab) will listen on a port (although be careful it is over 1024, unless you are running as root/admin).

[/solution]

[example]


perl -MIO::Socket -e '$srv=IO::Socket::INET->
new(LocalPort=>$ARGV[0],Type=>SOCK_STREAM,Reuse=>1,Listen=>5)

or die "Failed trying to listen on $ARGV[0]n";while($cl=$srv->accept()) { while() { print }

} close($socket);' port

Can also use it as a cheap chat service! 🙂 Everything gets echo’d through – so just telnet host port and type away! 🙂

Also see my port scan code at the reference tab.

[/example]

[reference]

[tags]Perl Socket Programming, Network Programming, Perl Port Listen, Perl Client Server demo, Perl Coding School[/tags]

• Perl Docs – Socket Programming
• Perl Port Scanner

[/reference]

[problem]

You want to test a network port, on a remote system – over TCP/IP.

Maybe this is a new setup, or you want confirmation it is working.

Perhaps the firewall rules have just been changed! 🙂

[/problem]

[solution]

A nice small bit of Perl code that I’ve used thousands of times!

In fact I’m running it in most of my production environments,
as a check that the a process is not only running – but also responding.

Yep, you could just use telnet – but some systems have that taken off for security.

Additionally it is n’t as easy to program telnet – I know, I know – you can with expect. 🙂

[/solution]

[example]

Here is just a one liner, but you can easily incorporate this into a script.

perl -MIO::Socket -e '$socket=IO::Socket::INET->
new(Proto=>tcp,PeerAddr=>$ARGV[0],PeerPort=>$ARGV[1]);
if($@) { print "Failed: $@n" } else { print "Succeedn"; }' host port

[/example]

[reference]

[tags]Perl, Network, TCPIP, Port Scanner, Perl Coding School[/tags]

• Perl Docs – socket

[/reference]

[problem]

You want to capture the current epoch. Maybe to use in a log file, or as a filename.

Or maybe you want to calculate the date yesterday.

[/problem]

[solution]

This piece of code is very useful for performing date calculations. You can obtain the current epoch (time in seconds since Jan 1 1970), then add 3600 for 1 hour – or 86400 for 24 hours hence.

[/solution]

[example]

So use in a UNIX variable like this:

epoch=perl -M'English' -e 'print $BASETIME."n";'

To work out 24 hours ago, just subtract 86400.

perl -M'English' -e 'print(($BASETIME-86400)."n");'

Then to see the date yesterday:


$ perl -M'English' -e 'print(($BASETIME-86400)."n");'
1180746252
$ perl -M'English' -e 'print(localtime(1180746252)."n");'
Sat Jun 2 09:04:12 2007


[/example]

[reference]

[tags]Perl, epoch, date manipulation, Perl Coding School[/tags]

• Perl Docs – predefined variables
• Perl Docs – localtime

[/reference]

[recommended]

[/recommended]

[problem]

I wanted to upload a delimited field to mysql db, but hit the problem that the file contained nearly 3000 rows and no schema on the required size of each field.

Therefore I needed to traverse the file and calculate the length of each field. Then at the end, print the largest field found for each column.

[/problem]

[solution]

Perl to the rescue! Pretty easy in Perl, after scratching my head attempting with awk. 🙂

Anyway the code is under the example tab. I’m using the pipe symbol as a delimiter “|” – so just substitute this with your delimiter, cat your file and pipe it through this script.

In the example I show how you can manipulate the file, to produce pipe delimited fields too.

[/solution]

[example]

This is how to run the script. Basically you just need to pipe your output through find largest.

cat yourfile | ./find_largest.pl

Here is the code.

 #!/usr/bin/perl
 
@highest=();
 
while( < STDIN >  ) {
 
    @thisline=split(/|/);
 
    for($i=0;$i<=$#thisline;$i++) {
 
        $thislength=length($thisline[$i]);
 
        if($thislength > $highest[$i]) { $highest[$i]=$thislength; }
 
    }
 
}
 
print(join("|",@highest)."n");
 
exit(0);
 
__END__
 
 
 
So for example, if I want to find the largest fields in one of my web logs - for crunching into a db:
 
 
First off I only want lines starting with a space, then a number.
 
Next I need to replace all multiple spaces between fields, with 1 space.
 
Then I replace the spaces between each field, with a pipe.
 
Lastly I pump it through find_largest.pl - which gives me the largest sized field
 
 

 [marcus@bree]/var/log/httpd% grep "^ [0-9]" access_log.tools
 | sed -e 's/  / /g' -e 's/ /|/g' |
 ~/Perl_Bin/find_largest.pl
 |14|1|1|21|6|5|103|9|3|5
 
 
[/example]
 
[reference]
 
[tags]Perl, Data Analysis, Perl Coding School[/tags]
 
 
Perl Doc - storable
 
Linux Man Page - sed
 
Linux Man Page - grep
 
 
[/reference]