in reply to Padding search results with context

From your original code, you can scan for duplicates and print at the same time.

my $prev = -1 - $pad;
for(@padded_overlap) {
    if ( $_ > $prev ) {
        $prev = $_;
#    print stuff here
    }
}

[download]
That eliminates one scan.

Replies are listed 'Best First'.
Re^2: Padding search results with context
by moritz (Cardinal) on Aug 14, 2007 at 06:53 UTC
    Thanks - it's obvious (of course you use the list elements where you construct them), but sometimes I need somebody to point me to the obvious ;-)

    Perl 6 in German

[reply]
      This "one pass" solution is obviously inspired by people like Grandfather, but this is a solution I can understand *^^*:

      #!/usr/bin/perl
use warnings;
use strict;

use List::Util;

package main;


my $max = 21;
my @nums = (1, 5, 6, 8, 15, 20);
my $pad = 1; # lines of context on each side

my $last = -1 - $pad;

foreach ( @nums ) {
    my $start = List::Util::max 1, $_ - $pad, $last + 1;
    my $end = List::Util::min $max - 1 , $_ + $pad;
    next if ( $start > $end );
    for ( my $i = $start; $i <= $end; $i++ ) {
    if ( $i == $_ ) {
        print "Highlighted $i\n";
    }
    else {
            print "Line $i\n";
    }
    }
    $last = $end;
}

      [download]
      The output is:

      C:\Code>perl mintest.pl
Highlighted 1
Line 2
Line 4
Highlighted 5
Line 6
Line 7
Highlighted 8
Line 9
Line 14
Highlighted 15
Line 16
Line 19
Highlighted 20

      [download]
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom