#!/usr/bin/perl -w
# cd /home/xiangyang/Desktop/P.stutzeri_genome/Blast_A1501/split/FASTA/ 
# perl /home/xiangyang/Desktop/P.stutzeri_genome/Blast_A1501/process.pl GCF_000282375.1_Pseudomonas.strGM50_v1.0.FASTA GCF_000474765.1_CMAA1215_1.0.FASTA


use strict;
my %HoH              = ();   # 
my %seqid_HoH        = ();   # 

my %seqid_count_hash = ();   # key:seqid, val:count
my $term             = $/;   # input record separator


my $space            = "\t"; # spacer for aligned print
my $nchar            = 0;    # nchar for phyml header.
my $nseq;                    # nseq for phyml header. Do not initiate!
my $first_name       = q{};  # First name in matrix.
my $fasta            = 0;    # Print phyml format by default
my $man              = 0;    # Manual
my $help             = 0;    # Help

my $dontprint        = 0;    # Do not print the concatenation
my $relaxed_phylip   = 0;    # Print relaxed phylip format
my $lwidth           = 60;   # default line width for fasta




my %nseq_hash        = ();   # key:infile, val:nseq
my @hash_ref_array   = ();   # array with hash references
my $nfiles           = 0;    # count number of files
my $verbose          = 0;    # Verbose
#---------------------------------------------------------------------------
#  Read all infiles to count sequences
#---------------------------------------------------------------------------
print STDERR "\nCecking sequences in infiles...\n" if ($verbose);

foreach my $arg (@ARGV) {
    my $infile       = $arg;
    my %seq_hash     = parse_fasta($infile); # key: seqid, value:sequence

    $nfiles++;

    ## Save sequences in array with hash references.
    ## Does this work for really large number of fasta files?
    my $hash_ref     = \%seq_hash;
    push(@hash_ref_array, $hash_ref);

    ## Add nseqs to global nseq_hash:
#    $nseq_hash{$infile} = scalar(keys(%seq_hash));
    $nseq_hash{$infile} = keys(%seq_hash);
    foreach (values(%seq_hash)) {
        print "$_\n";}

            }










sub parse_fasta {

    my ($infile) = @_;

    my $term     = $/; # input record separator;
    my %seq_hash = (); # key:seqid, val:seq

    open my $INFILE, "<", $infile or die "could not open infile '$infile' : $! \n"; 
    $/ = ">";
    while(<$INFILE>) {
        chomp;
        next if($_ eq '');
        my ($id, @sequencelines) = split /\n/;
        foreach my $line (@sequencelines) {
            $seq_hash{$id} .= $line;
        }
    }
    $/ = $term;

    return(%seq_hash);

} # end of parse_fasta