// Permutation.cpp : Responsible for inserting permuted DStrings into dTable
//Coded by: Nicholas Urrea (nurrea@uci.edu) unless otherwise stated
//ICS 175A, Winter 2004
//Project: Motif detection in yeast
//Purpose: Computes all substrings for a set of genes and stores them into a customized hash table (Dtable).

#include "Permutation.h"

// count initialization & boundary checking
void Permutation::initialize()
{
	// bounds checking
	if( SUBSTR_MIN_LENGTH > SUBSTR_MAX_LENGTH || SUBSTR_MAX_LENGTH < SUBSTR_MIN_LENGTH )
		cout << "Minimum and maximum length improperly defined in Permutation.h" << endl;

	count = 0;
}

// initialize a permutation
Permutation::Permutation ()
{
	initialize();
}

// passes in a DString through the constructor
// and immediately permutes the DString (calls permute string function)
// where each substring is stored into Dtable's hash table.
Permutation::Permutation(DString str)
{
	initialize();
	permute(str);
}

// stores DString substrings of length [MIN,MAX] into dtable.
void Permutation::permute(DString str)
{
	// traverse index from 0 to n
	// hash each valid substring from cnt index to min and max length
	// #subsequences per genome = ((size*5) - 15)
	for( unsigned int i = 0; i < str.size(); i++ )
		for( unsigned int j = SUBSTR_MIN_LENGTH; j <= SUBSTR_MAX_LENGTH; j++ )
		{
			if ( i+j > str.size() ) j = SUBSTR_MAX_LENGTH+1;
			else
			{
				// add both a subsequence and its reverse complement w/ a filter
				// that will not add the reverse complement if it is the same string
				// as the non-reverse complement.
				dTable.add( str.substr(i,i+j) );
				if ( str.substr(i,i+j) == getComplement(str.substr(i, i+j) ) ) ;
				else dTable.add( getComplement(str.substr(i, i+j) ) );
				count++;
			}
		}
}

// stores dString subsequences from a DGFile in FASTA format into dTable (including subsequence lengths)
void Permutation::permute(DGFile &file)
{
	DString * string;
	CGene * cgene;

	// impt: genome sequence size needs to be passed to SequenceFinder
	// thus, dtable must store the entire genome sequence length
	for ( unsigned long i = 0; i < file.num_genes; i++ )
	{
		string = file.GetGeneData(i);
		cgene = file.GetGene(i);
		permute( string->data() );
		dTable.setSequenceRepresentedLength( dTable.getSequenceRepresentedLength() + cgene->sequence.size() );
	}
}

//returns total # of permutation sequences
int Permutation::getCount()
{
	return count;
}

//prints dTable; hash table of genomic subsequences
void Permutation::printDTable()
{
	dTable.print();
}

// PURPOSE: returns the reverse complement of a ACGT string
// EDITED BY: Nicholas Urrea, Vishakh
DString Permutation::getComplement(DString str)
{	
	int j = str.size()-1;
	DString moo = str;
	for(int i=0; i<str.size(); i++ )
	{
		if(str[i]=='A') str.setChar('T', i);
		else if(str[i]=='T') str.setChar('A',i);
		else if(str[i]=='C') str.setChar('G',i);
		else if(str[i]=='G') str.setChar('C',i);
		moo.setChar(str[i], j);
		j--;
	}
	return moo;
}

// destructor
Permutation::~Permutation()
{
	// nothing to destruct; scope removes all me variables from memory
}