# A MR job uses counters to count the number of articles according to their 
# first letter (ignoring the case and possible diacritics).
# rm -rf step-4-out-sol; perl step-4-solution.pl /home/straka/wiki/cs-text-medium/ step-4-out-sol

package My::Mapper;
use Moose;
with 'Hadoop::Mapper';

sub map {
  my ($self, $key, $value, $context) = @_;

  $key =~ s/^\s+//;
  $key = uc substr($key, 0, 1);
  $context->counter("Articles", $key, 1);
}

package main;
use Hadoop::Runner;

my $runner = Hadoop::Runner->new(
  mapper => My::Mapper->new(),
  input_format => 'KeyValueTextInputFormat');

$runner->run();