#! /usr/bin/perl 

use strict;

#Scott Doniger
#January 19, 2005

#WARNING - this file is a lot more complicated than the previous ones. Make sure you're
#comfortable with the others before looking at this. Read the section on Hashes first.

#ReadFastaFile.pl
#This is a simple program to read in a fasta file and parse each sequence with its
#name. You will hopefully see how file input and output works, how to parse pieces 
#of files that are not separated by newlines, and to use hashes.

#see the fastafile1.fa to see what the fasta format is.

#To use the program enter perl ./ReadFastaFile.pl filename
# where filename is a fasta file

use constant SEQLENGTH => 10; #We will print out the first SEQLENGTH characters of each sequence.

my %sequencehash; #this is a hash that will store sequences and their names.

if (scalar (@ARGV) != 1){ #the phrase "scalar (@arrayname)" says "treat the array as a scalar", which in this case means
                          #the number of element in the array ARGV.
  die "You did not give me a file to work with. Please include the filename\n";
}

my $fastafile = shift @ARGV; #@ARGV is the array of the input parameters. Shift grabs the first parameter

open FASTA, $fastafile; #we're now creating a file handler called FASTA which will allow us to access the lines of the file.

#Now we want to read through the file, and create a hash. The key of the hash will be the sequence name, the value stored
#will be the sequence.
my $name;
my $sequence;

$name = <FASTA>;  #pre-read the first name before we get into the loop (this makes the code in the loop easier)
$name =~ s/>//; #remove the >
$name =~ s/\n//g; #remove any newlines

while (my $line = <FASTA>){ #We know that a fasta file goes name, sequence, name, sequence, name, sequence, ...  
  if ($line =~ />/){ #we're at the start of a new fasta record, so we need to store the previous sequence
    chomp $line; #remove the ">"
    $line =~ s/\n//g; #remove newlines. The g at the end means "globally", so all newlines, not just the first are replaced
    my $firstseq = substr($sequence, 0, SEQLENGTH); #get the first SEQLENGTH characters of the sequence (counting from 0).
 
    $sequencehash{$name} = $firstseq; #so here we have a key in the hash of $name. The value that this key points to is 
                                    #$firstseq. So for every fasta entry we have now created a pair of data, the name, 
                                    #and its corresponding sequence.
    #notice that its $sequencehash here, but before it was %sequencehash. This is a syntax thing. You can simply memorize
    #that when you want the entire hash you use the %sequencehash, but when you want a specific element of the hash its
    #$sequencehash{$mykey}

    #ok we stored the previous sequence, lets start over again.
    $sequence = ""; #this is now reset to an empty string.
    $name = $line; #reset the name
    chomp $line; #remove the newline at the end of the name
    $name =~ s/>//; #substitute > with nothing
  
  }else{ #we know that this isn't a new sequence since there is now >, so we append this line to the sequence variable
    $sequence .= $line; #we grab the next line, which we know is the sequence  .= is the same as saying x = x.newstuff
  }
  
}

close FASTA;



#ok so we read the file. How do we get into the hash and get back our data?

foreach my $k (sort keys %sequencehash){ #foreach access every element of an array. key %hash returns an array that 
#contains all of the keys of the hash. sort is there to order the keys alphabetically. The tricky thing about hashes
#is that the keys are not necessarily stored in the order they were entered. The hash finds its own way to order the keys
#allowing it to efficiently access its elements.
  print "$k => ".$sequencehash{$k}."\n";
}