#!/usr/local/bin/perl 

use strict;
use warnings;
use IO::File;
use Benchmark 'cmpthese';

my $count = shift || -1;
my @key_names = ('Gene number', 'Location begin', 'Location end', 'Orientation');
my $names_file = '/Users/kellert/Documents/Computing/in_progress/BatchBlast01/names01.txt';
my ($header,  @names) = process_names($names_file);
#print join "\n", @names, "\n";
my $annot_file = '/Users/kellert/Documents/Computing/in_progress/BatchBlast01/STM_Genome_Annot_short.txt';
my $fh1 = new IO::File;
my $fh2 = new IO::File;
my $fh3 = new IO::File;
cmpthese( $count,
{
    'with_string_cmp' => sub {
        my $names_join = join '|', @names; 
        my @goi;
        if ($fh1->open("< $annot_file")) {
            my @lines = <$fh1>;
		 foreach (@lines) {
				chomp;
				push @goi, $_ if $_ =~ m/($names_join)/;
            }
        } else { die "Could not get the filehandle $fh1: $!." }
        $fh1->close;
    },
        
    'with_hash' => sub {
        my %have_name = map({$_ => 1} @names);
        if ($fh2->open("< $annot_file")) {
            while(my $line = <$fh2>) {
				#print "Line2: $line\n";
				my ($name,$else) = split(/\t/, $line, 2);
				$have_name{$name} = [split(/\t/, $else)]
					or next;
            }
        } else { die "Could not get the filehandle $fh2: $!." }
        $fh2->close;
    }, 
    'with_grep' => sub {
		 if ($fh3->open("< $annot_file")) {
			my @lines =<$fh3>;
			my @index = grep {
				 my $c = $_;
				if ($_ > $c or ( grep { $lines[$c] =~ m/$_/ } @names ) ) {
					1; 	#yes, select it
				} else {
			        0;		# no, skip it
				}
			} 0..$#lines;
			my @gene_of_interest = @lines[@index]; 
        } else { die "Could not get the filehandle $fh3: $!." }
        $fh3->close;
    }
});

        

## Subs 
sub process_names {
	my $names_file = shift;
	my @names;
	open NAMES,  "< $names_file"
		or die "Cannot open $names_file.";
	chomp( my $header = <NAMES> );  	## remove header line
	while (<NAMES>) {
		chomp;
		my @line = split(/\t+/, $_);
		push @names, $line[0]
	}
	close NAMES;
	
	return $header, @names;
}

__END__
cmpthese( $count,
{
    'with_string_cmp' => sub {
        my $names_join = join '|', @names; 
        my @goi;
        if ($fh1->open("< $annot_file")) {
            my @lines = <$fh1>;
            foreach (@lines) {
                chomp;
               push @goi, $_ if $_ =~ m/($names_join)/;
            }
        } else { die "Could not get the filehandle $fh1: $!." }
        $fh1->close;
    },
        
    'with_hash' => sub {
        my %have_name = map({$_ => 1} @names);
        if ($fh2->open("< $annot_file")) {
            my $header = <$fh2>;
            while(my $line = <$fh2>) {
                #print "Line2: $line\n";
                my ($name,$else) = split(/\t/, $line, 2);
                $have_name{$name} = [split(/\t/, $else)]
                    or next;
            }
        } else { die "Could not get the filehandle $fh2: $!." }
        $fh2->close;
    }, 
    'with_grep' => sub {
        if ($fh3->open("< $annot_file")) {
            my $header = <$fh3>;
            my @lines =<$fh3>;
            my @index = grep {
                my $c = $_;
		        if ($_ > $c or ( grep { $lines[$c] =~ m/$_/ } @names ) ) {
			        1; 	#yes, select it
		        } else {
			        0;		# no, skip it
		        }
	        } 0..$#lines;
            my @gene_of_interest = @lines[@index]; 
        } else { die "Could not get the filehandle $fh3: $!." }
        $fh3->close;
    }
});