/** 
** Copyright (c) 2007-2010 Illumina, Inc. 
** 
** This software is covered by the "Illumina Genome Analyzer Software
** License Agreement" and the "Illumina Source Code License Agreement",
** and certain third party copyright/licenses, and any user of this
** source file is bound by the terms therein (see accompanying files
** Illumina_Genome_Analyzer_Software_License_Agreement.pdf and
** Illumina_Source_Code_License_Agreement.pdf and third party
** copyright/license notices).
** 
** This file is part of the Consensus Assessment of Sequence And VAriation 
** (CASAVA) software package. 
** 
** @file SmallAssembler.hh
** 
** @brief Stores default params for the SmallAssembler in Grouper.
** 
** @author Tony Cox, Ole Schulz-Trieglaff
**/

#ifndef SMALLASSEMBLER_HH_
#define SMALLASSEMBLER_HH_

// CASAVA includes
#include "applications/Application.hh"
#include "common/Alignment.hh"
#include "common/SequenceUtils.hh"
#include "applications/SmallAssembler.hh"
#include "assembly/SmallAssemblerImpl.hh"       

// boost
#include <boost/tokenizer.hpp>
        
// STL                                                                                                                                                                                                    
#include <cassert>
#include <fstream>
#include <iostream>
#include <map>
#include <sstream>


//#define DEBUG_ASBL

namespace ca { namespace  applications {


/**
 * @class SmallAssembler
 *
 * @brief Performs a simple de-novo assembly using the groups of
 * shadow reads found by the indelfinder.
 *
 * Details of implementation are in c++/assembler/SmallAssemblerImpl
 *
 * We build a hash of the k-mers in the shadow reads. The most
 * frequent k-mer is used as seed and is then gradually extended.
 *
 *
 */
class SmallAssembler : public Application {

		public:

			// Tokenizer class for parsing
			typedef boost::tokenizer<boost::char_separator<char> > LineTokenizer;
			// Vector of shadow reads in export file format. Also contains the approx. read coordinates.
			typedef std::vector< std::pair< std::pair<int,int>, Alignment > > ShadowAlignmentVec;

			struct AssemblerStats {

    		AssemblerStats() :
      		numClusters(0), numContigs(0),clustLen(0),numSeeds(0),others(0) {}

    		// number of shadow read cluster
    		unsigned numClusters;
    		// number of assembled contigs
    		unsigned numContigs;
    		// contigs discarded because of length
    		unsigned clustLen;
    		// contigs discarded because of too little reads
    		unsigned numSeeds;
    		// contigs discarded because of other reasons
    		unsigned others;
			};

    	SmallAssembler(CasavaOptions &options) : Application(options){ }

    	virtual ~SmallAssembler() {}

    	/**
     	* Reads the data and starts the assembly
     	*
     	* @return status
     	*/
    	int run();

		private:

			void process_cluster(const SmallAssemblerOptions& asmOptions,
													 AssemblerStats& asmStats,
													 SmallAssemblerImpl::ShadowReadVec& shadows,
													 ShadowAlignmentVec& shadowLines, 
													 const std::string& cluster_info,
													 std::ostream& creads_os,
													 std::ostream& contigs_os);

			void updateAlignment(Alignment& al,
													 const int pos,
													 const unsigned numContigs,
													 const std::string& cseq);

};

}} // end namespace casava{ namespace { applications


#endif /* SMALLASSEMBLER_HH_ */