jpayne@68: package sketch;
jpayne@68: 
jpayne@68: import java.io.File;
jpayne@68: import java.io.PrintStream;
jpayne@68: import java.util.ArrayList;
jpayne@68: import java.util.Arrays;
jpayne@68: import java.util.Locale;
jpayne@68: import java.util.Random;
jpayne@68: 
jpayne@68: import dna.AminoAcid;
jpayne@68: import fileIO.ByteFile;
jpayne@68: import fileIO.FileFormat;
jpayne@68: import fileIO.ReadWrite;
jpayne@68: import shared.Parse;
jpayne@68: import shared.Parser;
jpayne@68: import shared.PreParser;
jpayne@68: import shared.ReadStats;
jpayne@68: import shared.Shared;
jpayne@68: import shared.Timer;
jpayne@68: import shared.Tools;
jpayne@68: import stream.ConcurrentReadInputStream;
jpayne@68: import stream.ConcurrentReadOutputStream;
jpayne@68: import stream.FASTQ;
jpayne@68: import stream.FastaReadInputStream;
jpayne@68: import stream.Read;
jpayne@68: import structures.IntMap;
jpayne@68: import structures.ListNum;
jpayne@68: 
jpayne@68: /**
jpayne@68:  * 
jpayne@68:  * @author Brian Bushnell
jpayne@68:  * @date July 30, 2018
jpayne@68:  *
jpayne@68:  */
jpayne@68: public class KmerLimit2 extends SketchObject {
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------        Initialization        ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	
jpayne@68: 	/**
jpayne@68: 	 * Code entrance from the command line.
jpayne@68: 	 * @param args Command line arguments
jpayne@68: 	 */
jpayne@68: 	public static void main(String[] args){
jpayne@68: 		//Start a timer immediately upon code entrance.
jpayne@68: 		Timer t=new Timer();
jpayne@68: 		
jpayne@68: 		//Create an instance of this class
jpayne@68: 		KmerLimit2 x=new KmerLimit2(args);
jpayne@68: 		
jpayne@68: 		//Run the object
jpayne@68: 		x.process(t);
jpayne@68: 		
jpayne@68: 		//Close the print stream if it was redirected
jpayne@68: 		Shared.closeStream(x.outstream);
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/**
jpayne@68: 	 * Constructor.
jpayne@68: 	 * @param args Command line arguments
jpayne@68: 	 */
jpayne@68: 	public KmerLimit2(String[] args){
jpayne@68: 		
jpayne@68: 		{//Preparse block for help, config files, and outstream
jpayne@68: 			PreParser pp=new PreParser(args, getClass(), false);
jpayne@68: 			args=pp.args;
jpayne@68: 			outstream=pp.outstream;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		boolean setInterleaved=false; //Whether interleaved was explicitly set.
jpayne@68: 		
jpayne@68: 		//Set shared static variables
jpayne@68: 		ReadWrite.USE_PIGZ=ReadWrite.USE_UNPIGZ=true;
jpayne@68: 		ReadWrite.MAX_ZIP_THREADS=Shared.threads();
jpayne@68: 		SketchObject.setKeyFraction(0.1);
jpayne@68: 		defaultParams.minEntropy=0;
jpayne@68: 		defaultParams.minProb=0.2f;
jpayne@68: 		
jpayne@68: 		boolean setHeapSize=false;
jpayne@68: 		int heapSize_=8091;
jpayne@68: 		long targetKmers_=0;
jpayne@68: 		int k_=32;
jpayne@68: 		int minCount_=1;
jpayne@68: 		
jpayne@68: 		//Create a parser object
jpayne@68: 		Parser parser=new Parser();
jpayne@68: 		parser.overwrite=true;
jpayne@68: 		
jpayne@68: 		//Parse each argument
jpayne@68: 		for(int i=0; i<args.length; i++){
jpayne@68: 			String arg=args[i];
jpayne@68: 			
jpayne@68: 			//Break arguments into their constituent parts, in the form of "a=b"
jpayne@68: 			String[] split=arg.split("=");
jpayne@68: 			String a=split[0].toLowerCase();
jpayne@68: 			String b=split.length>1 ? split[1] : null;
jpayne@68: 			if(b!=null && b.equalsIgnoreCase("null")){b=null;}
jpayne@68: 			
jpayne@68: 			if(a.equals("verbose")){
jpayne@68: 				verbose=Parse.parseBoolean(b);
jpayne@68: 			}else if(a.equals("ordered")){
jpayne@68: 				ordered=Parse.parseBoolean(b);
jpayne@68: 			}else if(a.equals("size") || a.equals("heapsize")){
jpayne@68: 				heapSize_=Parse.parseIntKMG(b);
jpayne@68: 				setHeapSize=true;
jpayne@68: 			}else if(a.equals("kmers") || a.equals("target") || a.equals("limit")){
jpayne@68: 				targetKmers_=Parse.parseKMG(b);
jpayne@68: 			}else if(a.equals("mincount")){
jpayne@68: 				minCount_=Parse.parseIntKMG(b);
jpayne@68: 			}else if(a.equals("maxexpandedlength") || a.equals("maxlength") || a.equals("maxlen")){
jpayne@68: 				maxExpandedLength=Parse.parseIntKMG(b);
jpayne@68: 			}else if(a.equals("seed")){
jpayne@68: 				seed=Parse.parseKMG(b);
jpayne@68: 			}else if(a.equals("trials")){
jpayne@68: 				trials=Parse.parseIntKMG(b);
jpayne@68: 			}else if(parseSketchFlags(arg, a, b)){
jpayne@68: 				parser.parse(arg, a, b);
jpayne@68: 			}else if(defaultParams.parse(arg, a, b)){
jpayne@68: 				parser.parse(arg, a, b);
jpayne@68: 			}else if(a.equals("parse_flag_goes_here")){
jpayne@68: 				long fake_variable=Parse.parseKMG(b);
jpayne@68: 				//Set a variable here
jpayne@68: 			}else if(parser.parse(arg, a, b)){//Parse standard flags in the parser
jpayne@68: 				//do nothing
jpayne@68: 			}else{
jpayne@68: 				outstream.println("Unknown parameter "+args[i]);
jpayne@68: 				assert(false) : "Unknown parameter "+args[i];
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		if(!setHeapSize && minCount_>1){heapSize_=32000;}
jpayne@68: 		heapSize=heapSize_;
jpayne@68: 		targetKmers=targetKmers_;
jpayne@68: 		k=k_;
jpayne@68: 		minCount=minCount_;
jpayne@68: 		assert(targetKmers>0) : "Must set a kmer limit.";
jpayne@68: 		assert(heapSize>0) : "Heap size must be positive.";
jpayne@68: 		assert(k>0 && k<=32) : "0<k<33; k="+k;
jpayne@68: 		postParse();
jpayne@68: 		
jpayne@68: //		if(minCount>1){
jpayne@68: //			Shared.setBufferLen(800);
jpayne@68: //		}
jpayne@68: 		
jpayne@68: 		{//Process parser fields
jpayne@68: 			Parser.processQuality();
jpayne@68: 			
jpayne@68: 			maxReads=parser.maxReads;
jpayne@68: 			
jpayne@68: 			overwrite=ReadStats.overwrite=parser.overwrite;
jpayne@68: 			append=ReadStats.append=parser.append;
jpayne@68: 			setInterleaved=parser.setInterleaved;
jpayne@68: 			
jpayne@68: 			in1=parser.in1;
jpayne@68: 			in2=parser.in2;
jpayne@68: 			qfin1=parser.qfin1;
jpayne@68: 			qfin2=parser.qfin2;
jpayne@68: 
jpayne@68: 			out1=parser.out1;
jpayne@68: 			out2=parser.out2;
jpayne@68: 			qfout1=parser.qfout1;
jpayne@68: 			qfout2=parser.qfout2;
jpayne@68: 			
jpayne@68: 			extin=parser.extin;
jpayne@68: 			extout=parser.extout;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Do input file # replacement
jpayne@68: 		if(in1!=null && in2==null && in1.indexOf('#')>-1 && !new File(in1).exists()){
jpayne@68: 			in2=in1.replace("#", "2");
jpayne@68: 			in1=in1.replace("#", "1");
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Do output file # replacement
jpayne@68: 		if(out1!=null && out2==null && out1.indexOf('#')>-1){
jpayne@68: 			out2=out1.replace("#", "2");
jpayne@68: 			out1=out1.replace("#", "1");
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Adjust interleaved detection based on the number of input files
jpayne@68: 		if(in2!=null){
jpayne@68: 			if(FASTQ.FORCE_INTERLEAVED){outstream.println("Reset INTERLEAVED to false because paired input files were specified.");}
jpayne@68: 			FASTQ.FORCE_INTERLEAVED=FASTQ.TEST_INTERLEAVED=false;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		assert(FastaReadInputStream.settingsOK());
jpayne@68: 		
jpayne@68: 		//Ensure there is an input file
jpayne@68: 		if(in1==null){throw new RuntimeException("Error - at least one input file is required.");}
jpayne@68: 		
jpayne@68: 		//Adjust the number of threads for input file reading
jpayne@68: 		if(!ByteFile.FORCE_MODE_BF1 && !ByteFile.FORCE_MODE_BF2 && Shared.threads()>2){
jpayne@68: 			ByteFile.FORCE_MODE_BF2=true;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Ensure out2 is not set without out1
jpayne@68: 		if(out1==null && out2!=null){throw new RuntimeException("Error - cannot define out2 without defining out1.");}
jpayne@68: 		
jpayne@68: 		//Adjust interleaved settings based on number of output files
jpayne@68: 		if(!setInterleaved){
jpayne@68: 			assert(in1!=null && (out1!=null || out2==null)) : "\nin1="+in1+"\nin2="+in2+"\nout1="+out1+"\nout2="+out2+"\n";
jpayne@68: 			if(in2!=null){ //If there are 2 input streams.
jpayne@68: 				FASTQ.FORCE_INTERLEAVED=FASTQ.TEST_INTERLEAVED=false;
jpayne@68: 				outstream.println("Set INTERLEAVED to "+FASTQ.FORCE_INTERLEAVED);
jpayne@68: 			}else{ //There is one input stream.
jpayne@68: 				if(out2!=null){
jpayne@68: 					FASTQ.FORCE_INTERLEAVED=true;
jpayne@68: 					FASTQ.TEST_INTERLEAVED=false;
jpayne@68: 					outstream.println("Set INTERLEAVED to "+FASTQ.FORCE_INTERLEAVED);
jpayne@68: 				}
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Ensure output files can be written
jpayne@68: 		if(!Tools.testOutputFiles(overwrite, append, false, out1, out2)){
jpayne@68: 			outstream.println((out1==null)+", "+(out2==null)+", "+out1+", "+out2);
jpayne@68: 			throw new RuntimeException("\n\noverwrite="+overwrite+"; Can't write to output files "+out1+", "+out2+"\n");
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Ensure input files can be read
jpayne@68: 		if(!Tools.testInputFiles(false, true, in1, in2)){
jpayne@68: 			throw new RuntimeException("\nCan't read some input files.\n");  
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Ensure that no file was specified multiple times
jpayne@68: 		if(!Tools.testForDuplicateFiles(true, in1, in2, out1, out2)){
jpayne@68: 			throw new RuntimeException("\nSome file names were specified multiple times.\n");
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Create output FileFormat objects
jpayne@68: 		ffout1=FileFormat.testOutput(out1, FileFormat.FASTQ, extout, true, overwrite, append, ordered);
jpayne@68: 		ffout2=FileFormat.testOutput(out2, FileFormat.FASTQ, extout, true, overwrite, append, ordered);
jpayne@68: 
jpayne@68: 		//Create input FileFormat objects
jpayne@68: 		ffin1=FileFormat.testInput(in1, FileFormat.FASTQ, extin, true, true);
jpayne@68: 		ffin2=FileFormat.testInput(in2, FileFormat.FASTQ, extin, true, true);
jpayne@68: 
jpayne@68: 		minProb=defaultParams.minProb;
jpayne@68: 		minQual=defaultParams.minQual;
jpayne@68: 		
jpayne@68: 		shift=2*k;
jpayne@68: 		shift2=shift-2;
jpayne@68: 		mask=(shift>63 ? -1L : ~((-1L)<<shift)); //Conditional allows K=32
jpayne@68: 		sharedHeap=new SketchHeap(heapSize, 0, true);
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------         Outer Methods        ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 
jpayne@68: 	/** Create read streams and process all data */
jpayne@68: 	void process(Timer t){
jpayne@68: 		
jpayne@68: 		//Turn off read validation in the input threads to increase speed
jpayne@68: 		final boolean vic=Read.VALIDATE_IN_CONSTRUCTOR;
jpayne@68: 		Read.VALIDATE_IN_CONSTRUCTOR=Shared.threads()<4;
jpayne@68: 		
jpayne@68: //		//Optionally create a read output stream
jpayne@68: //		final ConcurrentReadOutputStream ros;
jpayne@68: //		if(ffout1!=null){
jpayne@68: //			//Select output buffer size based on whether it needs to be ordered
jpayne@68: //			final int buff=(ordered ? Tools.mid(16, 128, (Shared.threads()*2)/3) : 8);
jpayne@68: //			
jpayne@68: //			//Notify user of output mode
jpayne@68: //			if(cris.paired() && out2==null && (in1!=null && !ffin1.samOrBam() && !ffout1.samOrBam())){
jpayne@68: //				outstream.println("Writing interleaved.");
jpayne@68: //			}
jpayne@68: //			
jpayne@68: //			ros=ConcurrentReadOutputStream.getStream(ffout1, ffout2, qfout1, qfout2, buff, null, false);
jpayne@68: //			ros.start(); //Start the stream
jpayne@68: //		}else{ros=null;}
jpayne@68: 		
jpayne@68: 		//Reset counters
jpayne@68: 		readsProcessed=readsOut=0;
jpayne@68: 		basesProcessed=basesOut=0;
jpayne@68: 		
jpayne@68: 		//Process the reads in separate threads
jpayne@68: 		spawnThreads0();
jpayne@68: 		
jpayne@68: //		if(verbose){outstream.println("Finished; closing streams.");}
jpayne@68: 		
jpayne@68: 		//Reset read validation
jpayne@68: 		Read.VALIDATE_IN_CONSTRUCTOR=vic;
jpayne@68: 
jpayne@68: 		
jpayne@68: 		Sketch sketch=new Sketch(sharedHeap, true, true, null);
jpayne@68: 		sketch=capLengthAtCountSum(sketch, maxExpandedLength);
jpayne@68: 		final long reads=Tools.max(1, sketch.genomeSequences);
jpayne@68: 		final long targetReads=calcTargetReads(sketch, targetKmers, minCount, trials, seed);
jpayne@68: 		final double targetRate=Tools.min(1, targetReads/(double)reads);
jpayne@68: 		final String targetRateS=String.format(Locale.ROOT, "%.4f%%",targetRate*100);
jpayne@68: 		
jpayne@68: 		//Report timing and results
jpayne@68: 		t.stop();
jpayne@68: 		outstream.println("Finished counting kmers.");
jpayne@68: 		outstream.println(Tools.timeReadsBasesProcessed(t, readsProcessed, basesProcessed, 8));
jpayne@68: 		
jpayne@68: 		String kstring0=Tools.padKM(sketch.genomeSizeEstimate(minCount), 8);
jpayne@68: 		String rstring0=Tools.padKM(targetReads, 8);
jpayne@68: 		outstream.println("Unique Kmers:       "+kstring0);
jpayne@68: 		outstream.println("Target Reads:       "+rstring0+"\t"+targetRateS);
jpayne@68: 		
jpayne@68: //		outstream.println("Reads:        \t"+reads);
jpayne@68: //		outstream.println("Unique Kmers: \t"+sketch.genomeSizeEstimate(minCount));
jpayne@68: //		outstream.println("Target Reads: \t"+targetReads);
jpayne@68: //		outstream.println("Sample Rate:  \t"+targetRateS);
jpayne@68: //		outstream.println(Tools.readsBasesOut(readsProcessed, basesProcessed, readsOut, basesOut, 8, false));
jpayne@68: 		
jpayne@68: 		t.start();
jpayne@68: 		outstream.println("\nSubsampling reads.");
jpayne@68: 		
jpayne@68: //		String kstring=Tools.padKM(sharedHeap.genomeSizeEstimate(minCount), 8);
jpayne@68: //		outstream.println("Unique Kmers Out:   "+kstring);
jpayne@68: 		
jpayne@68: 		
jpayne@68: //		ArrayList<String> args=new ArrayList<String>();
jpayne@68: //		args.add("in="+in1);
jpayne@68: //		if(in2!=null){args.add("in2="+in2);}
jpayne@68: //		args.add("out="+out1);
jpayne@68: //		if(out2!=null){args.add("out2="+out2);}
jpayne@68: //		args.add("ordered="+ordered);
jpayne@68: //		args.add("ow="+(overwrite ? "t" : "f"));
jpayne@68: //		if(targetRate<1){args.add("samplerate="+targetRateS);}
jpayne@68: //		args.add("loglogout");
jpayne@68: //		args.add("loglogk="+k);
jpayne@68: //		args.add("loglogminprob="+minProb);
jpayne@68: //		BBDukF.main(args.toArray(new String[0]));
jpayne@68: 		
jpayne@68: //		Sketch sk=new Sketch(sharedHeap, true, true, null);
jpayne@68: //		outstream.println(sk.genomeSizeEstimate());
jpayne@68: 		spawnThreads2(targetRate);
jpayne@68: 		t.stop();
jpayne@68: 		outstream.println(Tools.timeReadsBasesProcessed(t, readsProcessed, basesProcessed, 8));
jpayne@68: 		
jpayne@68: 		outstream.println(Tools.readsBasesOut(readsProcessed, basesProcessed, readsOut, basesOut, 8, false));
jpayne@68: 		String kstring=Tools.padKM(sharedHeap.genomeSizeEstimate(minCount), 8);
jpayne@68: 		outstream.println("Unique Kmers Out:   "+kstring);
jpayne@68: 		
jpayne@68: 		//Throw an exception of there was an error in a thread
jpayne@68: 		if(errorState){
jpayne@68: 			throw new RuntimeException(getClass().getName()+" terminated in an error state; the output may be corrupt.");
jpayne@68: 		}
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/** Spawn process threads */
jpayne@68: 	private void spawnThreads0(){
jpayne@68: 		
jpayne@68: 		//Create a read input stream
jpayne@68: 		final ConcurrentReadInputStream cris;
jpayne@68: 		{
jpayne@68: 			cris=ConcurrentReadInputStream.getReadInputStream(maxReads, true, ffin1, ffin2, qfin1, qfin2);
jpayne@68: 			cris.start(); //Start the stream
jpayne@68: 			if(verbose){outstream.println("Started cris");}
jpayne@68: 		}
jpayne@68: 		paired=cris.paired();
jpayne@68: 		if(!ffin1.samOrBam()){outstream.println("Input is being processed as "+(paired ? "paired" : "unpaired"));}
jpayne@68: 		
jpayne@68: 		//Determine how many threads may be used
jpayne@68: 		final int threads=Tools.min(10, Shared.threads());
jpayne@68: 		
jpayne@68: 		//Fill a list with ProcessThreads
jpayne@68: 		ArrayList<ProcessThread> alpt=new ArrayList<ProcessThread>(threads);
jpayne@68: 		for(int i=0; i<threads; i++){
jpayne@68: 			alpt.add(new ProcessThread(cris, null, i, heapSize));
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Start the threads
jpayne@68: 		for(ProcessThread pt : alpt){
jpayne@68: 			pt.start();
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Wait for completion of all threads
jpayne@68: 		boolean success=true;
jpayne@68: 		for(ProcessThread pt : alpt){
jpayne@68: 			
jpayne@68: 			//Wait until this thread has terminated
jpayne@68: 			while(pt.getState()!=Thread.State.TERMINATED){
jpayne@68: 				try {
jpayne@68: 					//Attempt a join operation
jpayne@68: 					pt.join();
jpayne@68: 				} catch (InterruptedException e) {
jpayne@68: 					//Potentially handle this, if it is expected to occur
jpayne@68: 					e.printStackTrace();
jpayne@68: 				}
jpayne@68: 			}
jpayne@68: 			
jpayne@68: 			//Accumulate per-thread statistics
jpayne@68: 			readsProcessed+=pt.readsProcessedT;
jpayne@68: 			basesProcessed+=pt.basesProcessedT;
jpayne@68: 			readsOut+=pt.readsOutT;
jpayne@68: 			basesOut+=pt.basesOutT;
jpayne@68: 			success&=pt.success;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Track whether any threads failed
jpayne@68: 		if(!success){errorState=true;}
jpayne@68: 		
jpayne@68: 		//Do anything necessary after processing
jpayne@68: 		
jpayne@68: 		//Close the read streams
jpayne@68: 		errorState|=ReadWrite.closeStreams(cris);
jpayne@68: 		
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/** Spawn process threads */
jpayne@68: 	private void spawnThreads2(double rate){
jpayne@68: 		
jpayne@68: 		//Create a read input stream
jpayne@68: 		final ConcurrentReadInputStream cris;
jpayne@68: 		{
jpayne@68: 			cris=ConcurrentReadInputStream.getReadInputStream(maxReads, true, ffin1, ffin2, qfin1, qfin2);
jpayne@68: 			cris.setSampleRate((float)rate, seed);
jpayne@68: 			cris.start(); //Start the stream
jpayne@68: 			if(verbose){outstream.println("Started cris");}
jpayne@68: 		}
jpayne@68: //		paired=cris.paired();
jpayne@68: //		if(!ffin1.samOrBam()){outstream.println("Input is being processed as "+(paired ? "paired" : "unpaired"));}
jpayne@68: 		
jpayne@68: 		//Optionally create a read output stream
jpayne@68: 		final ConcurrentReadOutputStream ros;
jpayne@68: 		if(ffout1!=null){
jpayne@68: 			//Select output buffer size based on whether it needs to be ordered
jpayne@68: 			final int buff=(ordered ? Tools.mid(16, 128, (Shared.threads()*2)/3) : 8);
jpayne@68: 			
jpayne@68: 			//Notify user of output mode
jpayne@68: 			if(cris.paired() && out2==null && (in1!=null && !ffin1.samOrBam() && !ffout1.samOrBam())){
jpayne@68: 				outstream.println("Writing interleaved.");
jpayne@68: 			}
jpayne@68: 			
jpayne@68: 			ros=ConcurrentReadOutputStream.getStream(ffout1, ffout2, qfout1, qfout2, buff, null, false);
jpayne@68: 			ros.start(); //Start the stream
jpayne@68: 		}else{ros=null;}
jpayne@68: 		
jpayne@68: 		//Determine how many threads may be used
jpayne@68: 		final int threads=Tools.min(10, Shared.threads());
jpayne@68: 		
jpayne@68: 		sharedHeap.clear();			
jpayne@68: //		readsProcessed=0;
jpayne@68: //		basesProcessed=0;
jpayne@68: 		readsOut=0;
jpayne@68: 		basesOut=0;
jpayne@68: 		
jpayne@68: 		//Fill a list with ProcessThreads
jpayne@68: 		ArrayList<ProcessThread> alpt=new ArrayList<ProcessThread>(threads);
jpayne@68: 		for(int i=0; i<threads; i++){
jpayne@68: 			alpt.add(new ProcessThread(cris, ros, i, heapSize));
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Start the threads
jpayne@68: 		for(ProcessThread pt : alpt){
jpayne@68: 			pt.start();
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Wait for completion of all threads
jpayne@68: 		boolean success=true;
jpayne@68: 		for(ProcessThread pt : alpt){
jpayne@68: 			
jpayne@68: 			//Wait until this thread has terminated
jpayne@68: 			while(pt.getState()!=Thread.State.TERMINATED){
jpayne@68: 				try {
jpayne@68: 					//Attempt a join operation
jpayne@68: 					pt.join();
jpayne@68: 				} catch (InterruptedException e) {
jpayne@68: 					//Potentially handle this, if it is expected to occur
jpayne@68: 					e.printStackTrace();
jpayne@68: 				}
jpayne@68: 			}
jpayne@68: 			
jpayne@68: 			//Accumulate per-thread statistics
jpayne@68: //			readsProcessed+=pt.readsProcessedT;
jpayne@68: //			basesProcessed+=pt.basesProcessedT;
jpayne@68: 			readsOut+=pt.readsOutT;
jpayne@68: 			basesOut+=pt.basesOutT;
jpayne@68: 			success&=pt.success;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Track whether any threads failed
jpayne@68: 		if(!success){errorState=true;}
jpayne@68: 		
jpayne@68: 		//Do anything necessary after processing
jpayne@68: 		
jpayne@68: 		//Write anything that was accumulated by ReadStats
jpayne@68: 		errorState|=ReadStats.writeAll();
jpayne@68: 		//Close the read streams
jpayne@68: 		errorState|=ReadWrite.closeStreams(cris, ros);
jpayne@68: 		
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------         Inner Methods        ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 
jpayne@68: 	public static Sketch capLengthAtCountSum(Sketch sketch0, int max) {
jpayne@68: 		int len=0;
jpayne@68: 		long sum=0;
jpayne@68: 		for(; len<sketch0.keyCounts.length; len++){
jpayne@68: 			sum=sum+sketch0.keyCounts[len];
jpayne@68: 			if(sum>max){break;}
jpayne@68: 		}
jpayne@68: 		if(len>=sketch0.length()){return sketch0;}
jpayne@68: 		
jpayne@68: 		long[] keys=Arrays.copyOf(sketch0.keys, len);
jpayne@68: 		int[] counts=Arrays.copyOf(sketch0.keyCounts, len);
jpayne@68: 		
jpayne@68: //		long[] array_, int[] counts_, int taxID_, long imgID_, long gSizeBases_, long gSizeKmers_, long gSequences_, double probCorrect_,
jpayne@68: //		String taxName_, String name0_, String fname_, ArrayList<String> meta_
jpayne@68: 		
jpayne@68: 		Sketch sk=new Sketch(keys, counts, null, null, null, -1, -1, 
jpayne@68: 				sketch0.genomeSizeBases, sketch0.genomeSizeKmers, sketch0.genomeSequences, sketch0.probCorrect,
jpayne@68: 				null, null, null, null);
jpayne@68: 		
jpayne@68: 		return sk;
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	public static long calcTargetReads(Sketch sketch, long targetKmers, int minCount, int trials, long seed){
jpayne@68: 		final int[] counts0=sketch.keyCounts;
jpayne@68: 		final int[] counts=Arrays.copyOf(counts0, counts0.length);
jpayne@68: 		final long size=sketch.genomeSizeEstimate(minCount);
jpayne@68: 		final long reads=sketch.genomeSequences;
jpayne@68: 		final double targetKmerFraction=targetKmers/(double)size;
jpayne@68: 		if(targetKmerFraction>=1){return reads;}
jpayne@68: 		
jpayne@68: 		final int targetKeys=(int)(targetKmerFraction*counts.length);
jpayne@68: 		final long countSum=Tools.sum(counts0);
jpayne@68: 		assert(countSum<Shared.MAX_ARRAY_LEN) : countSum;
jpayne@68: //		System.err.println("countsum: "+countSum);
jpayne@68: 		
jpayne@68: 		final IntMap map=new IntMap(0, counts0.length);
jpayne@68: 		final int[] expanded=new int[(int)countSum];
jpayne@68: 		
jpayne@68: 		long roundSum=0;
jpayne@68: 		final Random randy=Shared.threadLocalRandom(seed);
jpayne@68: 		for(int i=0; i<trials; i++){
jpayne@68: 			Tools.fill(counts, counts0);
jpayne@68: //			long rounds=reduceRounds(counts0, counts, minCount, targetKeys, randy);
jpayne@68: 			long rounds=reduceRoundsIM(counts0, expanded, minCount, targetKeys, randy, map);
jpayne@68: 			roundSum+=rounds;
jpayne@68: 		}
jpayne@68: 		double avgRounds=roundSum/(double)trials;
jpayne@68: //		System.err.println("avgRounds: "+avgRounds);
jpayne@68: 		double targetCountFraction=1-(avgRounds/countSum);
jpayne@68: //		System.err.println("targetFraction: "+targetCountFraction);
jpayne@68: 		return (long)(targetCountFraction*reads);
jpayne@68: 	}
jpayne@68: 	
jpayne@68: //	public static int reduceRoundsOld(final int[] counts, final int minCount, final int targetKeys, final Random randy){
jpayne@68: //		assert(minCount>=0) : minCount;
jpayne@68: //		int rounds=0;
jpayne@68: //		int valid=0;
jpayne@68: //		for(int x : counts){
jpayne@68: //			if(x>=minCount){valid++;}
jpayne@68: //		}
jpayne@68: //		
jpayne@68: //		int len=counts.length;
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts)+", "+Arrays.toString(counts));
jpayne@68: //		for(; valid>targetKeys; rounds++){
jpayne@68: //			int pos=randy.nextInt(len);
jpayne@68: ////			assert(counts[pos]>0) : pos+"/"+len+": "+targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Arrays.toString(counts);
jpayne@68: //			if(counts[pos]==minCount){valid--;}
jpayne@68: //			counts[pos]--;
jpayne@68: //			if(counts[pos]==0){
jpayne@68: //				len--;//shrink the array
jpayne@68: //				System.err.println("len="+len+", counts[len]="+counts[len]);
jpayne@68: //				System.err.println("pos="+pos+", counts[pos]="+counts[pos]);
jpayne@68: //				counts[pos]=counts[len];//move the last element to the empty slot 
jpayne@68: //				counts[len]=0;
jpayne@68: //				if(pos!=len && len>0){
jpayne@68: //					assert(counts[pos]>0) : pos+"/"+len+": "+targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Arrays.toString(counts);
jpayne@68: //				}
jpayne@68: //			}
jpayne@68: //			System.err.println(len+", "+pos+": "+Arrays.toString(counts));
jpayne@68: //		}
jpayne@68: //		
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts));
jpayne@68: //		
jpayne@68: //		return rounds;
jpayne@68: //	}
jpayne@68: 	
jpayne@68: 	//This can be done faster with bins.
jpayne@68: 	//Each bin contains all kmers with count x.  When a bin is hit, one kmer moves to the next bin lower.
jpayne@68: 	//Alternately, expand the array into one physical kmer per count.  Store the current counts in an IntMap. Remove key each time.
jpayne@68: 	public static long reduceRounds(final int[] counts0, final int[] counts, final int minCount, final int targetKeys, final Random randy){
jpayne@68: 		assert(minCount>=0) : minCount;
jpayne@68: 		long rounds=0;
jpayne@68: 		int valid=0;
jpayne@68: 		for(int x : counts){
jpayne@68: 			if(x>=minCount){valid++;}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		int len=counts.length;
jpayne@68: 		final long sum0=Tools.sum(counts);
jpayne@68: 		long sum=sum0;
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts)+", "+Arrays.toString(counts));
jpayne@68: 		for(; valid>targetKeys; rounds++){
jpayne@68: 			long posNum=(Long.MAX_VALUE&randy.nextLong())%sum;
jpayne@68: 			long sum2=0;
jpayne@68: 			int pos=0;
jpayne@68: 			
jpayne@68: 			for(int i=0; i<counts.length; i++){
jpayne@68: 				int x=counts[i];
jpayne@68: 				if(x>0){
jpayne@68: 					sum2+=x;
jpayne@68: 					if(sum2>=posNum){
jpayne@68: 						pos=i;
jpayne@68: 						break;
jpayne@68: 					}
jpayne@68: 				}
jpayne@68: 			}
jpayne@68: 			
jpayne@68: //			for(int i=0; i<counts0.length; i++){
jpayne@68: //				int x=counts0[i];
jpayne@68: //				if(x>0){
jpayne@68: //					sum2+=x;
jpayne@68: //					if(sum2>=posNum){
jpayne@68: //						pos=i;
jpayne@68: //						break;
jpayne@68: //					}
jpayne@68: //				}
jpayne@68: //			}
jpayne@68: 			
jpayne@68: 			sum--;
jpayne@68: 			
jpayne@68: 			assert(counts[pos]>0) : pos+"/"+len+": "+targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Arrays.toString(counts);
jpayne@68: 			if(counts[pos]==minCount){valid--;}
jpayne@68: 			counts[pos]--;
jpayne@68: 			if(counts[pos]==0){
jpayne@68: 				len--;//shrink the array
jpayne@68: 			}
jpayne@68: //			System.err.println(len+", "+pos+": "+Arrays.toString(counts));
jpayne@68: 		}
jpayne@68: 		
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts));
jpayne@68: 		
jpayne@68: 		return rounds;
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	//This can be done faster with bins.
jpayne@68: 	//Each bin contains all kmers with count x.  When a bin is hit, one kmer moves to the next bin lower.
jpayne@68: 	//Alternately, expand the array into one physical kmer per count.  Store the current counts in an IntMap. Remove key each time.
jpayne@68: 	public static long reduceRoundsIM(final int[] counts0, final int[] expanded, final int minCount, final int targetKeys, final Random randy, final IntMap map){
jpayne@68: 		assert(minCount>=0) : minCount;
jpayne@68: 		long rounds=0;
jpayne@68: 		int valid=0;
jpayne@68: 		map.clear();
jpayne@68: 		for(int i=0, k=0; i<counts0.length; i++){
jpayne@68: 			int x=counts0[i];
jpayne@68: //			counts[i]=counts0[i];
jpayne@68: 			if(x>=minCount){valid++;}
jpayne@68: 			map.put(i, x);
jpayne@68: 			for(int j=0; j<x; j++, k++){
jpayne@68: 				expanded[k]=i;
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 		assert(expanded.length==Tools.sum(counts0));
jpayne@68: 		
jpayne@68: 		int len=expanded.length;
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts)+", "+Arrays.toString(counts));
jpayne@68: 		for(; valid>targetKeys; rounds++){
jpayne@68: 			final int pos=randy.nextInt(len);
jpayne@68: 			final int key=expanded[pos];
jpayne@68: 			final int x=map.get(key);
jpayne@68: 			assert(x>0);
jpayne@68: 			
jpayne@68: 			
jpayne@68: 			if(x==minCount){valid--;}
jpayne@68: 			map.put(key, x-1);
jpayne@68: 			
jpayne@68: 			len--;//shrink the array
jpayne@68: 			//				System.err.println("len="+len+", counts[len]="+counts[len]);
jpayne@68: 			//				System.err.println("pos="+pos+", counts[pos]="+counts[pos]);
jpayne@68: 			expanded[pos]=expanded[len];//move the last element to the empty slot 
jpayne@68: 			expanded[len]=0;
jpayne@68: 			
jpayne@68: //			System.err.println(len+", "+pos+": "+Arrays.toString(counts));
jpayne@68: 		}
jpayne@68: 		
jpayne@68: //		System.err.println(targetKeys+", "+counts.length+", "+valid+", "+len+", "+rounds+", "+Tools.sum(counts));
jpayne@68: 		
jpayne@68: 		return rounds;
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------         Inner Classes        ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	
jpayne@68: 	/** This class is static to prevent accidental writing to shared variables.
jpayne@68: 	 * It is safe to remove the static modifier. */
jpayne@68: 	private class ProcessThread extends Thread {
jpayne@68: 		
jpayne@68: 		//Constructor
jpayne@68: 		ProcessThread(final ConcurrentReadInputStream cris_, final ConcurrentReadOutputStream ros_, final int tid_, final int size){
jpayne@68: 			cris=cris_;
jpayne@68: 			ros=ros_;
jpayne@68: 			tid=tid_;
jpayne@68: 			localHeap=new SketchHeap(size, 0, true);
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		//Called by start()
jpayne@68: 		@Override
jpayne@68: 		public void run(){
jpayne@68: 			//Do anything necessary prior to processing
jpayne@68: 			
jpayne@68: 			//Process the reads
jpayne@68: 			processInner();
jpayne@68: 			
jpayne@68: 			//Do anything necessary after processing
jpayne@68: 			dumpHeap();
jpayne@68: 			
jpayne@68: 			//Indicate successful exit status
jpayne@68: 			success=true;
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		/** Iterate through the reads */
jpayne@68: 		void processInner(){
jpayne@68: 			
jpayne@68: 			//Grab the first ListNum of reads
jpayne@68: 			ListNum<Read> ln=cris.nextList();
jpayne@68: 			//Grab the actual read list from the ListNum
jpayne@68: 			ArrayList<Read> reads=(ln!=null ? ln.list : null);
jpayne@68: 
jpayne@68: 			//Check to ensure pairing is as expected
jpayne@68: 			if(reads!=null && !reads.isEmpty()){
jpayne@68: 				Read r=reads.get(0);
jpayne@68: //				assert(ffin1.samOrBam() || (r.mate!=null)==cris.paired()); //Disabled due to non-static access
jpayne@68: 			}
jpayne@68: 
jpayne@68: 			//As long as there is a nonempty read list...
jpayne@68: 			while(ln!=null && reads!=null && reads.size()>0){//ln!=null prevents a compiler potential null access warning
jpayne@68: //				if(verbose){outstream.println("Fetched "+reads.size()+" reads.");} //Disabled due to non-static access
jpayne@68: 
jpayne@68: 				//Loop through each read in the list
jpayne@68: 				for(int idx=0; idx<reads.size(); idx++){
jpayne@68: 					final Read r1=reads.get(idx);
jpayne@68: 					final Read r2=r1.mate;
jpayne@68: 					
jpayne@68: 					//Validate reads in worker threads
jpayne@68: 					if(!r1.validated()){r1.validate(true);}
jpayne@68: 					if(r2!=null && !r2.validated()){r2.validate(true);}
jpayne@68: 
jpayne@68: 					//Track the initial length for statistics
jpayne@68: 					final int initialLength1=r1.length();
jpayne@68: 					final int initialLength2=r1.mateLength();
jpayne@68: 
jpayne@68: 					//Increment counters
jpayne@68: 					readsProcessedT+=r1.pairCount();
jpayne@68: 					basesProcessedT+=initialLength1+initialLength2;
jpayne@68: 					
jpayne@68: 					//Reads are processed in this block.
jpayne@68: 					processReadPair(r1, r2);
jpayne@68: 				}
jpayne@68: 				
jpayne@68: 				if(ros!=null){
jpayne@68: 					for(Read r1 : reads){
jpayne@68: 						readsOutT+=r1.pairCount();
jpayne@68: 						basesOutT+=r1.pairLength();
jpayne@68: 					}
jpayne@68: 
jpayne@68: 					//Output reads to the output stream
jpayne@68: 					if(ros!=null){ros.add(reads, ln.id);}
jpayne@68: 				}
jpayne@68: 
jpayne@68: 				//Notify the input stream that the list was used
jpayne@68: 				cris.returnList(ln);
jpayne@68: //				if(verbose){outstream.println("Returned a list.");} //Disabled due to non-static access
jpayne@68: 
jpayne@68: 				//Fetch a new list
jpayne@68: 				ln=cris.nextList();
jpayne@68: 				reads=(ln!=null ? ln.list : null);
jpayne@68: 			}
jpayne@68: 
jpayne@68: 			//Notify the input stream that the final list was used
jpayne@68: 			if(ln!=null){
jpayne@68: 				if(ln.list!=null){ln.list.clear();}
jpayne@68: 				cris.returnList(ln.id, true);
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		/**
jpayne@68: 		 * Process a read or a read pair.
jpayne@68: 		 * @param r1 Read 1
jpayne@68: 		 * @param r2 Read 2 (may be null)
jpayne@68: 		 */
jpayne@68: 		void processReadPair(final Read r1, final Read r2){
jpayne@68: 			processReadNucleotide(r1);
jpayne@68: 			if(r2!=null){processReadNucleotide(r2);}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		void processReadNucleotide(final Read r){
jpayne@68: 			final byte[] bases=r.bases;
jpayne@68: 			final byte[] quals=r.quality;
jpayne@68: 			long kmer=0;
jpayne@68: 			long rkmer=0;
jpayne@68: 			int len=0;
jpayne@68: 			assert(!r.aminoacid());
jpayne@68: 			
jpayne@68: 			final long min=minHashValue;
jpayne@68: 			localHeap.genomeSizeBases+=r.length();
jpayne@68: 			localHeap.genomeSequences++;
jpayne@68: 			
jpayne@68: 			if(quals==null || (minProb<=0 && minQual<2)){
jpayne@68: 				for(int i=0; i<bases.length; i++){
jpayne@68: 					byte b=bases[i];
jpayne@68: 					long x=AminoAcid.baseToNumber[b];
jpayne@68: 					long x2=AminoAcid.baseToComplementNumber[b];
jpayne@68: 					
jpayne@68: 					kmer=((kmer<<2)|x)&mask;
jpayne@68: 					rkmer=((rkmer>>>2)|(x2<<shift2))&mask;
jpayne@68: 					
jpayne@68: 					if(x<0){len=0; rkmer=0;}else{len++;}
jpayne@68: 					if(len>=k){
jpayne@68: 						localHeap.genomeSizeKmers++;
jpayne@68: 						final long hashcode=hash(kmer, rkmer);
jpayne@68: 						if(hashcode>min){localHeap.add(hashcode);}
jpayne@68: 					}
jpayne@68: 				}
jpayne@68: 			}else{
jpayne@68: 				float prob=1;
jpayne@68: 				for(int i=0; i<bases.length; i++){
jpayne@68: 					final byte b=bases[i];
jpayne@68: 					final long x=AminoAcid.baseToNumber[b];
jpayne@68: 					final long x2=AminoAcid.baseToComplementNumber[b];
jpayne@68: 					
jpayne@68: 					{//Quality-related stuff
jpayne@68: 						final byte q=quals[i];
jpayne@68: 						assert(q>=0) : Arrays.toString(quals)+"\n"+minProb+", "+minQual;
jpayne@68: 						prob=prob*align2.QualityTools.PROB_CORRECT[q];
jpayne@68: 						if(len>k){
jpayne@68: 							byte oldq=quals[i-k];
jpayne@68: 							prob=prob*align2.QualityTools.PROB_CORRECT_INVERSE[oldq];
jpayne@68: 						}
jpayne@68: 						if(x<0 || q<minQual){
jpayne@68: 							len=0;
jpayne@68: 							kmer=rkmer=0;
jpayne@68: 							prob=1;
jpayne@68: 						}else{
jpayne@68: 							len++;
jpayne@68: 						}
jpayne@68: 					}
jpayne@68: 					
jpayne@68: 					kmer=((kmer<<2)|x)&mask;
jpayne@68: 					rkmer=((rkmer>>>2)|(x2<<shift2))&mask;
jpayne@68: 					
jpayne@68: 					if(len>=k && prob>=minProb){
jpayne@68: 						localHeap.genomeSizeKmers++;
jpayne@68: 						localHeap.probSum+=prob;
jpayne@68: 						final long hashcode=hash(kmer, rkmer);
jpayne@68: 						if(hashcode>min){localHeap.checkAndAdd(hashcode);}
jpayne@68: 					}
jpayne@68: 				}
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 		
jpayne@68: 		private void dumpHeap(){
jpayne@68: 			synchronized(sharedHeap){
jpayne@68: 				sharedHeap.add(localHeap);
jpayne@68: 			}
jpayne@68: 		}
jpayne@68: 
jpayne@68: 		/** Number of reads processed by this thread */
jpayne@68: 		protected long readsProcessedT=0;
jpayne@68: 		/** Number of bases processed by this thread */
jpayne@68: 		protected long basesProcessedT=0;
jpayne@68: 		
jpayne@68: 		/** Number of reads retained by this thread */
jpayne@68: 		protected long readsOutT=0;
jpayne@68: 		/** Number of bases retained by this thread */
jpayne@68: 		protected long basesOutT=0;
jpayne@68: 		
jpayne@68: 		/** True only if this thread has completed successfully */
jpayne@68: 		boolean success=false;
jpayne@68: 
jpayne@68: 		/** Shared input stream */
jpayne@68: 		private final ConcurrentReadInputStream cris;
jpayne@68: 		/** Shared output stream */
jpayne@68: 		private final ConcurrentReadOutputStream ros;
jpayne@68: 		/** Thread ID */
jpayne@68: 		final int tid;
jpayne@68: 		
jpayne@68: 		final SketchHeap localHeap;
jpayne@68: 	}
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------            Fields            ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 
jpayne@68: 	/** Primary input file path */
jpayne@68: 	private String in1=null;
jpayne@68: 	/** Secondary input file path */
jpayne@68: 	private String in2=null;
jpayne@68: 	
jpayne@68: 	private String qfin1=null;
jpayne@68: 	private String qfin2=null;
jpayne@68: 
jpayne@68: 	/** Primary output file path */
jpayne@68: 	private String out1=null;
jpayne@68: 	/** Secondary output file path */
jpayne@68: 	private String out2=null;
jpayne@68: 
jpayne@68: 	private String qfout1=null;
jpayne@68: 	private String qfout2=null;
jpayne@68: 	
jpayne@68: 	/** Override input file extension */
jpayne@68: 	private String extin=null;
jpayne@68: 	/** Override output file extension */
jpayne@68: 	private String extout=null;
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 
jpayne@68: 	/** Number of reads processed */
jpayne@68: 	protected long readsProcessed=0;
jpayne@68: 	/** Number of bases processed */
jpayne@68: 	protected long basesProcessed=0;
jpayne@68: 
jpayne@68: 	/** Number of reads retained */
jpayne@68: 	protected long readsOut=0;
jpayne@68: 	/** Number of bases retained */
jpayne@68: 	protected long basesOut=0;
jpayne@68: 
jpayne@68: 	/** Quit after processing this many input reads; -1 means no limit */
jpayne@68: 	private long maxReads=-1;
jpayne@68: 	
jpayne@68: 	private boolean paired=false;
jpayne@68: 	private int trials=25;
jpayne@68: 	private long seed=-1;
jpayne@68: 	private int maxExpandedLength=50000000;
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------         Final Fields         ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 
jpayne@68: 	/** Primary input file */
jpayne@68: 	private final FileFormat ffin1;
jpayne@68: 	/** Secondary input file */
jpayne@68: 	private final FileFormat ffin2;
jpayne@68: 	
jpayne@68: 	/** Primary output file */
jpayne@68: 	private final FileFormat ffout1;
jpayne@68: 	/** Secondary output file */
jpayne@68: 	private final FileFormat ffout2;
jpayne@68: 	
jpayne@68: 	private final SketchHeap sharedHeap;
jpayne@68: 	private final int heapSize;
jpayne@68: 	private final long targetKmers;
jpayne@68: 	private final int minCount;
jpayne@68: 
jpayne@68: 	final int shift;
jpayne@68: 	final int shift2;
jpayne@68: 	final long mask;
jpayne@68: 	
jpayne@68: 	final float minProb;
jpayne@68: 	final byte minQual;
jpayne@68: 	
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	/*----------------        Common Fields         ----------------*/
jpayne@68: 	/*--------------------------------------------------------------*/
jpayne@68: 	
jpayne@68: 	/** Print status messages to this output stream */
jpayne@68: 	private PrintStream outstream=System.err;
jpayne@68: 	/** Print verbose messages */
jpayne@68: 	public static boolean verbose=false;
jpayne@68: 	/** True if an error was encountered */
jpayne@68: 	public boolean errorState=false;
jpayne@68: 	/** Overwrite existing output files */
jpayne@68: 	private boolean overwrite=true;
jpayne@68: 	/** Append to existing output files */
jpayne@68: 	private boolean append=false;
jpayne@68: 	/** Reads are output in input order (not enabled) */
jpayne@68: 	private boolean ordered=true;
jpayne@68: 	
jpayne@68: }