in reply to Parsing cisco log file

So you want the line printed if it does not end in "matches)"? That could be as simple as: 

while (<STDIN>)
{
    # This is backwards, see update below.
    # next unless /matches\)$/;
    # print LOG;
    print LOG unless /matches\)$/;
}

[download]
Assuming you've already opened LOG to the right file, and that you're feeding the file to the program on standard input.

Or did I misunderstand your question entirely?

Update
I forgot to mention, there's a bunch of code and information about working with cisco routers in the monastery. Follow that link, or type "cisco" in the search box at the top of the page. You may find your problem solved already.

Update 2
The logic in the original snippet was backwards. Thanks to guha for spotting it.

Replies are listed 'Best First'.
Re: Re: Parsing cisco log file
by routedude (Initiate) on Mar 06, 2003 at 20:42 UTC
    Hey thanks! Sorry for asking for help in an obviously basic area, but I'm brand spanking new to perl. Hmm..this will work, but what if I wanted to grab the number of matches (for performance tuning, stick lines that get more hits toward the top of the list)? I would assume I would need to split my array by spaces, and reference by $line but again the last $line that would hold 45454 matches for instance would be variable.
[reply]

      I'm still not sure what $line or "45454 matches" you're referring to.

      I'm guessing you do want the lines that end in "matches)", but want to do something different with those, like make a list, sorted by number of matches. In that case, the basic logic would be like this: 

      while (<STDIN>)
{
    if (/(\d+) matches\)/)
    {
        # keep this in a list we'll sort later
        push @list, [ $1, $_ ];
    }
    else
    {
        # same case as before, write it to the log
        print LOG;
    }
}
# Now print out the list we built before in sorted order
foreach (sort { $a->[0] <=> $b->[0] } @list )
{
    print $_->[1];
}

      [download]
      You can adapt the basic logic in the loop without using more advanced data structures like I just did, but that's a pretty easy way to do it. This will keep all of the file in memory, so hopefully your file is not too huge.

      The data structure that gets built up in @list is an array of arrays; in other words, each element of @list is an anonymous array that gets pushed into each time. Each of those arrays has two elements: the number of matches, and the line itself. Later, the loop works calls sort on the number matches (element 0), and the print outputs the line (element 1). To learn more about those data structures, look at perlreftut and perldsc.

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom