We have a lot of code that uses Ethernet MAC addresses as identifiers. Sometime these are passed around nicely formatted as 
aa:bb:cc:dd:ee:ff

[download]
where the letter pairs represent hexadecimally-encoded octets. Sometimes however a script will be passed a Cisco-style address: 
0001.0203.0405

[download]
or even worse, something copied and pasted from a Windows dialog box: 
00-01-02-03-04-05

[download]
The one-liner below converts incoming MAC addresses in (almost) any format into the desirable form, with octets separated by colons. The MAC to be filtered is in $_.

This could of course be cleaned up, using a qr// to avoid duplicating the hex-octet identifier. 

s/([0-9a-f]{2})[^0-9a-f]*/$1:/ig;chop;

[download]

Replies are listed 'Best First'.
Re: MAC Pretty-printer
by thor (Priest) on Jan 21, 2005 at 04:37 UTC
    Whenever I see chop, wonder if something is wrong. Not saying that it's wrong to use it here, but just saying. What about this code? 
    s/[[:^xdigit:]]//g; $mac =  lc join ":", unpack "(A2)*", $_

    [download]

    thor

    Feel the white light, the light within
    Be your own disciple, fan the sparks of will
    For all of us waiting, your kingdom will come

[reply]
[d/l]
      In this case, I'm chopping the character added by the regexp, so I'm pretty sure it's safe. While your code is cleaner, my original is leaner - I went for brevity over clarity, since it sees a lot of use as a one-liner in passing MACs to non-perlish scripts (yes, I know, we have some of those.. I'm sorry ;) ).
[reply]

Back to Cool Uses for Perl