perl help

Steve Willoughby steve at
Wed Mar 26 22:47:38 PST 2003

> In the last episode (Mar 27), Giorgos Keramidas said:
> > On 2003-03-26 14:18, Kenzo <kenzo_chin at> wrote:
> > You don't need Perl for that.  Here's a small trick:
> > 	grep 'this' file | wc -l
> > 	grep 'that' file | wc -l
> Even better:
> 	grep -c 'this' file
> 	grep -c 'that' file

Unfortunately, that's not what he was asking for, which is to look for 
the pattern "big <foo>" where all the possible <foo>s are unknown and
report on all the <foo>s that were found.

So something like:

	while (<>) {
		while (/big\s+(\w+)/g) {

	foreach $word (sort(keys(%count))) {
		print "$word: $count{$word}\n";

ought to do the trick.  Play with $/, etc if you want to allow big and
<foo> to be across a newline from each other.
Steve Willoughby          | "The purpose of IT is to seamlessly and trans-
Intel DPG Eng. Computing  | parently provide the other nine-tenths of the
Application Development   | iceburg for people who need to work with chunks 
<steve at>  | of floating ice."           --Strata R. Chalup

More information about the freebsd-questions mailing list