/*
 * @(#)SequenceUtil.java 1.0 September 2009 Copyright (c) 2009 Peter Troshin
 * Jalview Web Services version: 2.0 This library is free software; you can
 * redistribute it and/or modify it under the terms of the Apache License
 * version 2 as published by the Apache Software Foundation This library is
 * distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
 * without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
 * PARTICULAR PURPOSE. See the Apache License for more details. A copy of the
 * license is in apache_license.txt. It is also available here: see:
 * http://www.apache.org/licenses/LICENSE-2.0.txt Any republication or derived
 * work distributed in source code form must include this copyright and license
 * notice.
 */

package compbio.data.sequence;

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.Closeable;
import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.util.ArrayList;
import java.util.List;
import java.util.Scanner;
import java.util.logging.Level;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * Utility class for operations on sequences
 * 
 * @author Petr Troshin
 * @version 1.0
 */
public final class SequenceUtil {

	/**
	 * A whitespace character: [\t\n\x0B\f\r]
	 */
	public static final Pattern WHITE_SPACE = Pattern.compile("\\s");

	/**
	 * A digit
	 */
	public static final Pattern DIGIT = Pattern.compile("\\d");

	/**
	 * Non word
	 */
	public static final Pattern NONWORD = Pattern.compile("\\W");

	/**
	 * Valid Amino acids
	 */
	public static final Pattern AA = Pattern.compile("[ARNDCQEGHILKMFPSTWYV]+",
			Pattern.CASE_INSENSITIVE);

	/**
	 * inversion of AA pattern
	 */
	public static final Pattern NON_AA = Pattern.compile(
			"[^ARNDCQEGHILKMFPSTWYV]+", Pattern.CASE_INSENSITIVE);

	/**
	 * Same as AA pattern but with two additional letters - XU
	 */
	public static final Pattern AMBIGUOUS_AA = Pattern.compile(
			"[ARNDCQEGHILKMFPSTWYVXU]+", Pattern.CASE_INSENSITIVE);

	/**
	 * Nucleotides a, t, g, c, u
	 */
	public static final Pattern NUCLEOTIDE = Pattern.compile("[AGTCU]+",
			Pattern.CASE_INSENSITIVE);

	/**
	 * Ambiguous nucleotide
	 */
	public static final Pattern AMBIGUOUS_NUCLEOTIDE = Pattern.compile(
			"[AGTCRYMKSWHBVDNU]+", Pattern.CASE_INSENSITIVE); // see IUPAC
	/**
	 * Non nucleotide
	 */
	public static final Pattern NON_NUCLEOTIDE = Pattern.compile("[^AGTCU]+",
			Pattern.CASE_INSENSITIVE);

	private SequenceUtil() {
	} // utility class, no instantiation

	/*
	 * public static void write_PirSeq(OutputStream os, FastaSequence seq)
	 * throws IOException { BufferedWriter pir_out = new BufferedWriter(new
	 * OutputStreamWriter(os)); pir_out.write(">P1;" + seq.getId() +
	 * SysPrefs.newlinechar); pir_out.write(seq.getSequence() +
	 * SysPrefs.newlinechar); pir_out.close(); } public static void
	 * write_FastaSeq(OutputStream os, FastaSequence seq) throws IOException {
	 * BufferedWriter fasta_out = new BufferedWriter( new
	 * OutputStreamWriter(os)); fasta_out.write(">" + seq.getId() +
	 * SysPrefs.newlinechar); fasta_out.write(seq.getSequence() +
	 * SysPrefs.newlinechar); fasta_out.close(); }
	 */

	/**
	 * @return true is the sequence contains only letters a,c, t, g, u
	 */
	public static boolean isNucleotideSequence(final FastaSequence s) {
		return SequenceUtil.isNonAmbNucleotideSequence(s.getSequence());
	}

	/**
	 * Ambiguous DNA chars : AGTCRYMKSWHBVDN // differs from protein in only one
	 * (!) - B char
	 */
	public static boolean isNonAmbNucleotideSequence(String sequence) {
		sequence = SequenceUtil.cleanSequence(sequence);
		if (SequenceUtil.DIGIT.matcher(sequence).find()) {
			return false;
		}
		if (SequenceUtil.NON_NUCLEOTIDE.matcher(sequence).find()) {
			return false;
			/*
			 * System.out.format("I found the text starting at " +
			 * "index %d and ending at index %d.%n", nonDNAmatcher .start(),
			 * nonDNAmatcher.end());
			 */
		}
		final Matcher DNAmatcher = SequenceUtil.NUCLEOTIDE.matcher(sequence);
		return DNAmatcher.find();
	}

	/**
	 * Removes all whitespace chars in the sequence string
	 * 
	 * @param sequence
	 * @return cleaned up sequence
	 */
	public static String cleanSequence(String sequence) {
		assert sequence != null;
		final Matcher m = SequenceUtil.WHITE_SPACE.matcher(sequence);
		sequence = m.replaceAll("").toUpperCase();
		return sequence;
	}

	/**
	 * Removes all special characters and digits as well as whitespace chars
	 * from the sequence
	 * 
	 * @param sequence
	 * @return cleaned up sequence
	 */
	public static String deepCleanSequence(String sequence) {
		sequence = SequenceUtil.cleanSequence(sequence);
		sequence = SequenceUtil.DIGIT.matcher(sequence).replaceAll("");
		sequence = SequenceUtil.NONWORD.matcher(sequence).replaceAll("");
		final Pattern othernonSeqChars = Pattern.compile("[_-]+");
		sequence = othernonSeqChars.matcher(sequence).replaceAll("");
		return sequence;
	}

	/**
	 * @param sequence
	 * @return true is the sequence is a protein sequence, false overwise
	 */
	public static boolean isProteinSequence(String sequence) {
		sequence = SequenceUtil.cleanSequence(sequence);
		if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {
			return false;
		}
		if (SequenceUtil.DIGIT.matcher(sequence).find()) {
			return false;
		}
		if (SequenceUtil.NON_AA.matcher(sequence).find()) {
			return false;
		}
		final Matcher protmatcher = SequenceUtil.AA.matcher(sequence);
		return protmatcher.find();
	}

	/**
	 * Check whether the sequence confirms to amboguous protein sequence
	 * 
	 * @param sequence
	 * @return return true only if the sequence if ambiguous protein sequence
	 *         Return false otherwise. e.g. if the sequence is non-ambiguous
	 *         protein or DNA
	 */
	public static boolean isAmbiguosProtein(String sequence) {
		sequence = SequenceUtil.cleanSequence(sequence);
		if (SequenceUtil.isNonAmbNucleotideSequence(sequence)) {
			return false;
		}
		if (SequenceUtil.DIGIT.matcher(sequence).find()) {
			return false;
		}
		if (SequenceUtil.NON_AA.matcher(sequence).find()) {
			return false;
		}
		if (SequenceUtil.AA.matcher(sequence).find()) {
			return false;
		}
		final Matcher amb_prot = SequenceUtil.AMBIGUOUS_AA.matcher(sequence);
		return amb_prot.find();
	}

	/**
	 * Writes list of FastaSequeces into the outstream formatting the sequence
	 * so that it contains width chars on each line
	 * 
	 * @param outstream
	 * @param sequences
	 * @param width
	 *            - the maximum number of characters to write in one line
	 * @throws IOException
	 */
	public static void writeFasta(final OutputStream outstream,
			final List<FastaSequence> sequences, final int width)
			throws IOException {
		writeFastaKeepTheStream(outstream, sequences, width);
		outstream.close();
	}

	public static void writeFastaKeepTheStream(final OutputStream outstream,
			final List<FastaSequence> sequences, final int width)
			throws IOException {
		final OutputStreamWriter writer = new OutputStreamWriter(outstream);
		final BufferedWriter fastawriter = new BufferedWriter(writer);
		for (final FastaSequence fs : sequences) {
			fastawriter.write(">" + fs.getId() + "\n");
			fastawriter.write(fs.getFormatedSequence(width));
			fastawriter.write("\n");
		}
		fastawriter.flush();
		writer.flush();
	}

	/**
	 * Reads fasta sequences from inStream into the list of FastaSequence
	 * objects
	 * 
	 * @param inStream
	 *            from
	 * @return list of FastaSequence objects
	 * @throws IOException
	 */
	public static List<FastaSequence> readFasta(final InputStream inStream)
			throws IOException {
		final List<FastaSequence> seqs = new ArrayList<FastaSequence>();

		final BufferedReader infasta = new BufferedReader(
				new InputStreamReader(inStream, "UTF8"), 16000);
		final Pattern pattern = Pattern.compile("//s+");

		String line;
		String sname = "", seqstr = null;
		do {
			line = infasta.readLine();
			if ((line == null) || line.startsWith(">")) {
				if (seqstr != null) {
					seqs.add(new FastaSequence(sname.substring(1), seqstr));
				}
				sname = line; // remove >
				seqstr = "";
			} else {
				final String subseq = pattern.matcher(line).replaceAll("");
				seqstr += subseq;
			}
		} while (line != null);

		infasta.close();
		return seqs;
	}

	/**
	 * Writes FastaSequence in the file, each sequence will take one line only
	 * 
	 * @param os
	 * @param sequences
	 * @throws IOException
	 */
	public static void writeFasta(final OutputStream os,
			final List<FastaSequence> sequences) throws IOException {
		final OutputStreamWriter outWriter = new OutputStreamWriter(os);
		final BufferedWriter fasta_out = new BufferedWriter(outWriter);
		for (final FastaSequence fs : sequences) {
			fasta_out.write(fs.getOnelineFasta());
		}
		fasta_out.close();
		outWriter.close();
	}

	public static List<AnnotatedSequence> readJRonn(final File result)
			throws IOException, UnknownFileFormatException {
		InputStream input = new FileInputStream(result);
		List<AnnotatedSequence> sequences = readJRonn(input);
		input.close();
		return sequences;
	}

	/**
	 * Reader for JRonn horizontal file format >Foobar M G D T T A G 0.48 0.42
	 * 0.42 0.48 0.52 0.53 0.54 All values are tab delimited
	 * 
	 * @param inStream
	 * @return
	 * @throws IOException
	 * @throws UnknownFileFormatException
	 */
	public static List<AnnotatedSequence> readJRonn(final InputStream inStream)
			throws IOException, UnknownFileFormatException {
		final List<AnnotatedSequence> seqs = new ArrayList<AnnotatedSequence>();

		final BufferedReader infasta = new BufferedReader(
				new InputStreamReader(inStream, "UTF8"), 16000);

		String line;
		String sname = "";
		do {
			line = infasta.readLine();
			if (line == null || line.isEmpty()) {
				// skip empty lines
				continue;
			}
			if (line.startsWith(">")) {
				// read name
				sname = line.trim().substring(1);
				// read sequence line
				line = infasta.readLine();
				final String sequence = line.replace("\t", "");
				// read annotation line
				line = infasta.readLine();
				String[] annotValues = line.split("\t");
				float[] annotation = convertToNumber(annotValues);
				if (annotation.length != sequence.length()) {
					throw new UnknownFileFormatException(
							"File does not look like Jronn horizontally formatted output file!\n"
									+ JRONN_WRONG_FORMAT_MESSAGE);
				}
				seqs.add(new AnnotatedSequence(sname, sequence, annotation));
			}
		} while (line != null);

		infasta.close();
		return seqs;
	}

	private static float[] convertToNumber(String[] annotValues)
			throws UnknownFileFormatException {
		float[] annotation = new float[annotValues.length];
		try {
			for (int i = 0; i < annotation.length; i++) {
				annotation[i] = Float.parseFloat(annotValues[i]);
			}
		} catch (NumberFormatException e) {
			throw new UnknownFileFormatException(JRONN_WRONG_FORMAT_MESSAGE,
					e.getCause());
		}
		return annotation;
	}

	private static final String JRONN_WRONG_FORMAT_MESSAGE = "Jronn file must be in the following format:\n"
			+ ">sequence_name\n "
			+ "M	V	S\n"
			+ "0.43	0.22	0.65\n"
			+ "Where first line is the sequence name,\n"
			+ "second line is the tab delimited sequence,\n"
			+ "third line contains tab delimited disorder prediction values.\n"
			+ "No lines are allowed between these three. Additionally, the number of  "
			+ "sequence residues must be equal to the number of the disorder values.";

	/**
	 * Closes the Closable and logs the exception if any
	 * 
	 * @param log
	 * @param stream
	 */
	public final static void closeSilently(java.util.logging.Logger log,
			Closeable stream) {
		if (stream != null) {
			try {
				stream.close();
			} catch (IOException e) {
				log.log(Level.WARNING, e.getLocalizedMessage(), e.getCause());
			}
		}
	}

	/**
	 * 
	 * TODO complete!
	 * 
	 * # RESIDUE COILS REM465 HOTLOOPS M 0.86010 0.88512 0.37094 T 0.79983
	 * 0.85864 0.44331 .... # RESIDUE COILS REM465 HOTLOOPS M 0.86010 0.88512
	 * 0.37094
	 * 
	 * @param input
	 * @return
	 * @throws IOException
	 * @throws UnknownFileFormatException
	 */
	public static List<MultiAnnotatedSequence<DisemblResultAnnot>> readDisembl(
			final InputStream input) throws IOException,
			UnknownFileFormatException {
		Scanner scan = new Scanner(input);
		scan.useDelimiter("# RESIDUE COILS REM465 HOTLOOPS\n");
		if (!scan.hasNext()) {
			throw new UnknownFileFormatException(
					"In Disembl score format each seqeunce score is expected to start from the line: "
							+ "'# RESIDUE COILS REM465 HOTLOOPS\\n'."
							+ " No such line was found!");
		}

		List<MultiAnnotatedSequence<DisemblResultAnnot>> results = new ArrayList<MultiAnnotatedSequence<DisemblResultAnnot>>();
		int seqCounter = 0;
		while (scan.hasNext()) {
			seqCounter++;
			String singleSeq = scan.next();
			Scanner scansingle = new Scanner(singleSeq);
			StringBuffer seqbuffer = new StringBuffer();
			List<Float> coils = new ArrayList<Float>();
			List<Float> rem = new ArrayList<Float>();
			List<Float> hotloops = new ArrayList<Float>();

			MultiAnnotatedSequence<DisemblResultAnnot> disemblRes = new MultiAnnotatedSequence<DisemblResultAnnot>(
					DisemblResultAnnot.class);

			while (scansingle.hasNextLine()) {
				String valueLine = scansingle.nextLine();
				Scanner values = new Scanner(valueLine);
				seqbuffer.append(values.next());
				coils.add(values.nextFloat());
				rem.add(values.nextFloat());
				hotloops.add(values.nextFloat());
				values.close();
			}
			disemblRes.addAnnotation(DisemblResultAnnot.COILS, coils);
			disemblRes.addAnnotation(DisemblResultAnnot.REM465, rem);
			disemblRes.addAnnotation(DisemblResultAnnot.HOTLOOPS, hotloops);
			// TODO
			// disemblRes.sequence = seqbuffer.toString();
			scansingle.close();
			results.add(disemblRes);
		}

		input.close();
		return results;
	}

}