Perl: Hash operations

Abstract: introduce some useful hash data structures.

Hash is popular used in Perl programming. Performance of hash remain stable even storing millions key-values in some cases.

1. conversion between hash and text file:

#######

sub hash_to_file{

my($hash_pointer, $out_file, $sep)=@_;

my %hash=%$hash_pointer;

$sep="=" unless $sep; #default is = as split

open my($OUT), ">", $out_file or die "can't open $out_file\n";

foreach my$key(sort (keys %hash)){

print $OUT join("", $key, $sep, $hash{$key}), "\n";

}

close($OUT);

}

#########

sub file_to_hash{

my ($file, $sep, $key_index, $value_index)=@_;

$sep='=' unless $sep;

$key_index=0 unless $key_index;

$value_index=1 unless $value_index;

my %hash;

if(-f $file){

open my ($INFO), "<", $file or die;

while (<$INFO>) {#2

chomp($_);

my @items = split(/$sep/, $_); #split on the tabs

my $key=$items[$key_index];

my $value=$items[$value_index];

#print"$key:$value\n";

if ($_=~/$sep/){

if(exists $hash{$key}){

$hash{$key} .= ','.$value ;

}

else{

$hash{$key} =$value ;

}

}

}#2

close($INFO);

}

return(\%hash);

}

2 conversion between nested hash and text file

#############################

#export the hash with two levels into text file

sub hash2_to_file{

my($hash2_pointer, $statistics_file, $sep)=@_;

my %hash2=%$hash2_pointer;

$sep="\t" unless $sep;

#get keys1 and key2

my @row_names=keys %hash2;

@row_names= sort @row_names;

my @col_names=map { keys $hash2{$_} } @row_names;

@col_names=List::MoreUtils::uniq @col_names;

@col_names= sort @col_names;

open my($OUT), ">", $statistics_file or die;

print $OUT join($sep, 'names', @col_names), "\n";

foreach my $row_name(@row_names){#2

my @counting_num;

foreach my $col_name(@col_names){

my $value= (exists $hash2{$row_name}->{$col_name}) ? $hash2{$row_name}->{$col_name} : 0;

push(@counting_num, $value);

}

print $OUT join($sep, $row_name, @counting_num), "\n";

}#2

close($OUT);

#

}

#########

##########################

#read file as the hash with two levels

sub file_to_hash2{

my($file, $sep)=@_;

$sep="\t" unless $sep;

#

my %hash2;

open my($OUT), "<", $file or die;

my $header=<$OUT>;

chomp($header);

#print "####$header###\n";

my @col_names=split(/$sep/,$header);

#

while (<$OUT>){

chomp($_);

my @items=split(/$sep/, $_);

my $row_name=$items[0];

for(my $i=1;$i<@items; $i++){

my $col_name=$col_names[$i];

$hash2{$row_name}->{$col_name}=$items[$i];

#print "$row_name:$col_name=$items[$i]\n";

}

}

close($OUT);

return(\%hash2);

}