#!/usr/bin/perl -w

package main;

require 5.004;

use strict;
use Getopt::Std;

########## CONSTANTS  ###############

#my $ID_LIMIT = 10000 ;
my $BUFFER_LIST = 2000 ; # WRITE BUFFER Size of array accumulated before writing to file.


########### VARIABLES ####################
use vars qw($opt_h $opt_v $opt_d $opt_i $opt_o $opt_g $opt_p) ;
use vars qw($inputFile $outputFile $id $sequenceType) ;
use vars qw($number $cntr_idNumber @outputBuffer) ;

use vars qw($numRecords) ;

getopts('i:o:g:p:hvd') ;

$opt_h and die <<"QQ_HELP_QQ" ;

DESCRIPTION

     This program reads a fasta 'Genbank formatted file', retrieves
     the gb numbers by default, or gi numbers if specified
     and writes the numbers to a text file

AVAILABILITY

     Requires perl 5.004 or higher

SYNOPSIS

     makeGIList.pl -h -i [path/fileToRead] -o [path/fileToWrite]

OPTIONS

    -h                          Print help message
    -v                          Verbose mode
    -i	  [path/fileToRead]  	File to process
    -o	  [path/fileToWrite]	Destination
    -g    [gi|gb|both]          Make file with GI or Accession number or both
    -p	  [n|p]			nucleotide or protein, DEFAULT = 'n'	


EXIT STATUS

     0        Fail
     1        Successful completion


     
QQ_HELP_QQ

#undocumented in help ...
# DEBUG MODE
if ($opt_d) {
    $opt_v = 'v' ;
    $opt_i = "D:\\datasets\\ncbi\\UniGene\\Ta.seq.uniq" ;
    $opt_o = "D:\\datasets\\ncbi\\UniGene\\gbnumTest.id" ;
}


# Input path parameter
if ($opt_i) {
    $inputFile = $opt_i ;
} else {
    die "an input file with its path is required" ;
}


# output path and filename
if ($opt_o) {
    $outputFile = $opt_o ;
} else {  
    die "an ouput file with its path is required" ;
}


# id to fetch, GI, Accession, or both
if ($opt_g) {
    if (($opt_g eq 'gi') or ($opt_g eq 'GI')) {
        $id = 'gi' ;
    } elsif (($opt_g eq 'gb') or ($opt_g eq 'GB')) {
        $id = 'gi' ;
    } elsif (($opt_g eq 'both') or ($opt_g eq 'BOTH')) {
        $id = 'both' ;
    }
} else {
    $opt_g = 'gb' ;
    $id    = $opt_g ;
}


# type of file
if ($opt_p) {
    if (($opt_p eq 'n') or ($opt_p eq 'N')) {
        $sequenceType = 'n';
    } elsif (($opt_p eq 'p') or ($opt_p eq 'P')) {
        $sequenceType = 'p' ;
    }
} else {
    $opt_p = 'n' ;          #default 'n' nucleotide
    $sequenceType = $opt_p ;
}
    
# Debug mode
# print out variables mapped from input flags 
if ($opt_d) {
    print "\$opt_d: $opt_d  DEBUG mode on" ;      # undocumented 
    print "\$opt_h: $opt_h\n" ;
    print "\$opt_v: $opt_v\n" ;
    print "\$opt_i: $opt_i ==> \$inputFile:    $inputFile\n" ;
    print "\$opt_o: $opt_o ==> \$outputFile:   $outputFile\n" ;
    print "\$opt_g: $opt_g ==> \$id:           $id\n" ;
    print "\$opt_p: $opt_p\==> \$sequenceType: $sequenceType\n" ;
}
    
####################  END OF INPUT FLAGS ########################

#################################################################
####################  Main Program ##############################

#open source file
if (open (INPUT, "<$inputFile")) {
    $opt_v and print "opened file $inputFile\n" ;
} else {
    die "Can not open $inputFile: $!.\n" ;
}

#open destination file
if (open (OUTPUT, ">>$outputFile")) {
    $opt_v and print "opened file to write:  $outputFile.\n" ;
} else {
   die "Can not write to $outputFile: $!.\n" ;
}


#print first line of destination file
#specify whether accession numbers are for 'n' nucleotides or  'p' proteins.
if ($sequenceType eq 'n') {
    print OUTPUT ">nucleotide\n" ;
    close OUTPUT or die "can not close $outputFile $!\n" ;
    $opt_v and print "closed file to write after writing first line\n\n"
} elsif ($sequenceType eq 'p') {
    print OUTPUT ">protein\n" ;
    close OUTPUT or die "can not close $outputFile $!\n" ;
    $opt_v and print "closed file to write after writing first line\n\n"
}



$opt_v and print "finding $id numbers\n" ;              #verbose mode
$cntr_idNumber = 0 ;                                    #count records whiles we find the id numbers
@outputBuffer = () ;                                    #declare output buffer
while (my $line = <INPUT> ) {                           #read the input file line by line
    chomp ($line) ;
    if ($number = &getIDnumber($line,$id)) {
        $cntr_idNumber += 1 ;                           #increment id counter
        $opt_v and print "$cntr_idNumber) $number\n" ;
        push @outputBuffer, $number ;                   # add id to buffer
        if (scalar @outputBuffer == $BUFFER_LIST) {     # buffered output, print evevery $BUFFER_LIST lines
            if (open (OUTPUT, ">>$outputFile")) {
                $opt_v and print "opened file to write $BUFFER_LIST ids:  $outputFile.\n" ;
            } else {
                    die "Can not open $inputFile: $!.\n" ;
            }
            foreach my $item (@outputBuffer) { 
                    print  OUTPUT "$item\n" ;    
            }
            @outputBuffer = ();     #reset buffer
            close OUTPUT or die "can not close $outputFile after writing $BUFFER_LIST ids $!\n" ;
            $opt_v and print "Closed $outputFile after writing $BUFFER_LIST ids\n\n" ;
        }
    } 
}

# print remainder of ids
# Flush rest of the buffer
if (open (OUTPUT, ">>$outputFile")) {
    $opt_v and print "opened file to write remainder of ids:  $outputFile.\n" ;
} else {
    die "Can not open $inputFile: $!.\n" ;
}
foreach my $item (@outputBuffer) {  #print out remaining id's from buffer
    print OUTPUT "$item\n" ;
}
close OUTPUT ;          # close filehandle
@outputBuffer = () ;    #clear buffer 

$opt_v and print "Closed $outputFile after writing remainder of ids\n\n" ;
$opt_v and print "$cntr_idNumber records\n" ;
$opt_v and print "done\n" ;

exit;




####################  END of Main Program  ####################### 
################################################################## 


sub getIDnumber {
    my($line, $id) = @_ ;
    my $IDtoReturn = '' ;
    my $IDfromDef = '' ;
    if ($line =~ /^>/) {                # fasta Definition Line
        $opt_d and print "\$line: $line\n"; 
        if (($id eq 'gi') and ($line =~ /\/gi=.*?\s/)) {
            $IDtoReturn= $& ;
            chomp($IDtoReturn) ;
        } elsif (($id eq 'gb') and ($line =~ /\/gb=.*?\s/)) {
            $opt_d and print "$&:  " ;
            $IDfromDef = $&;
            $IDtoReturn = substr($IDfromDef, 4) ;
            chomp($IDtoReturn) ;
        }            
        
    } elsif ($line =~ /^$/){        # Blank line
        # do nothing
    } else {                            #sequence 
        # do nothing
    }
    return $IDtoReturn
}