blkgrep

#!/usr/bin/perl -w
#Author: James Washer washer@trlp.com
use constant VERSION => '$Revision: 1.3 $';

#bgrep:: Block based grep utility. Search for a pattern in a block of text.
#dumps entire block if match found
#sort of like csplit and grep combined into one utility.

use strict;

my @data;
my $interesting;
my $saw_seperator; #just to avoid end case where we never saw a seperator;


die "bgrep.pl ", VERSION,"\n" if (1==@ARGV && "-v" eq $ARGV[0]);
die "Usage bgrep.pl block_seperator_regex grep_search_regex\n" unless 2==@ARGV; #Must have at least 2 arguments
my ($seperator,$search)=  @ARGV;
@ARGV=();
while (<>){
        if ( /$seperator/){
                $saw_seperator++; #This is just used to handle the end-of-file case where we may never have seen the beginning seperator
                #We're about to start a new block, so process previous
                process() if $interesting;
                @data=(); #clear out our stored data to be ready for next block
                $interesting=0; #nothing interesting yet, since new block
        }
        $interesting++ if /$search/; #mark this block interesting if match
        push @data, $_;
}
process() if ($saw_seperator &&$interesting); #one last time for end of file

sub process{
	#Generally, we just print this "interesting" block
	#but feel free to munge on this block if needed
        print foreach ( @data);
}