/*
 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
 * Copyright (C) $$Year-Rel$$ The Jalview Authors
 * 
 * This file is part of Jalview.
 * 
 * Jalview is free software: you can redistribute it and/or
 * modify it under the terms of the GNU General Public License 
 * as published by the Free Software Foundation, either version 3
 * of the License, or (at your option) any later version.
 *  
 * Jalview 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 GNU General Public License for more details.
 * 
 * You should have received a copy of the GNU General Public License
 * along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
 * The Jalview Authors are detailed in the 'AUTHORS' file.
 */
package jalview.analysis.scoremodels;

import jalview.math.Matrix;
import jalview.math.MatrixI;

import java.util.Arrays;

public class ScoreMatrix implements PairwiseScoreModelI
{
  public static final short UNMAPPED = (short) -1;

  private static final String BAD_ASCII_ERROR = "Unexpected character %s in getPairwiseScore";

  private static final int MAX_ASCII = 127;

  /*
   * the name of the model as shown in menus
   */
  private String name;

  /*
   * the characters that the model provides scores for
   */
  private char[] symbols;

  /*
   * the score matrix; both dimensions must equal the number of symbols
   * matrix[i][j] is the substitution score for replacing symbols[i] with symbols[j]
   */
  private float[][] matrix;

  /*
   * quick lookup to convert from an ascii character value to the index
   * of the corresponding symbol in the score matrix 
   */
  private short[] symbolIndex;

  /*
   * true for Protein Score matrix, false for dna score matrix
   */
  private boolean peptide;

  /**
   * Constructor given a name, symbol alphabet, and matrix of scores for pairs
   * of symbols. The matrix should be square and of the same size as the
   * alphabet, for example 20x20 for a 20 symbol alphabet.
   * 
   * @param name
   *          Unique, human readable name for the matrix
   * @param alphabet
   *          the symbols to which scores apply
   * @param matrix
   *          Pairwise scores indexed according to the symbol alphabet
   */
  public ScoreMatrix(String name, char[] alphabet, float[][] matrix)
  {
    if (alphabet.length != matrix.length)
    {
      throw new IllegalArgumentException(
              "score matrix size must match alphabet size");
    }
    for (float[] row : matrix)
    {
      if (row.length != alphabet.length)
      {
        throw new IllegalArgumentException(
                "score matrix size must be square");
      }
    }

    this.matrix = matrix;
    this.name = name;
    this.symbols = alphabet;

    symbolIndex = buildSymbolIndex(alphabet);

    /*
     * crude heuristic for now...
     */
    peptide = alphabet.length >= 20;
  }

  /**
   * Returns an array A where A[i] is the position in the alphabet array of the
   * character whose value is i. For example if the alphabet is { 'A', 'D', 'X'
   * } then A['D'] = A[68] = 1.
   * <p>
   * Unmapped characters (not in the alphabet) get an index of -1.
   * <p>
   * Mappings are added automatically for lower case symbols (for non case
   * sensitive scoring), unless they are explicitly present in the alphabet (are
   * scored separately in the score matrix).
   * 
   * @param alphabet
   * @return
   */
  static short[] buildSymbolIndex(char[] alphabet)
  {
    short[] index = new short[MAX_ASCII + 1];
    Arrays.fill(index, UNMAPPED);
    short pos = 0;
    for (char c : alphabet)
    {
      if (c <= MAX_ASCII)
      {
        index[c] = pos;
      }

      /*
       * also map lower-case character (unless separately mapped)
       */
      if (c >= 'A' && c <= 'Z')
      {
        short lowerCase = (short) (c + ('a' - 'A'));
        if (index[lowerCase] == UNMAPPED)
        {
          index[lowerCase] = pos;
        }
      }
      pos++;
    }
    return index;
  }

  @Override
  public String getName()
  {
    return name;
  }

  @Override
  public boolean isDNA()
  {
    return !peptide;
  }

  @Override
  public boolean isProtein()
  {
    return peptide;
  }

  /**
   * Returns the score matrix as used in getPairwiseScore. If using this matrix
   * directly, callers <em>must</em> also call <code>getMatrixIndex</code> in
   * order to get the matrix index for each character (symbol).
   * 
   * @return
   * @see #getMatrixIndex(char)
   */
  public float[][] getMatrix()
  {
    return matrix;
  }

  /**
   * Answers the matrix index for a given character, or -1 if unmapped in the
   * matrix. Use this method only if using <code>getMatrix</code> in order to
   * compute scores directly (without symbol lookup) for efficiency.
   * 
   * @param c
   * @return
   * @see #getMatrix()
   */
  public int getMatrixIndex(char c)
  {
    if (c < symbolIndex.length)
    {
      return symbolIndex[c];
    }
    else
    {
      return UNMAPPED;
    }
  }

  /**
   * Returns the pairwise score for substituting c with d, or zero if c or d is
   * an unscored or unexpected character
   */
  @Override
  public float getPairwiseScore(char c, char d)
  {
    if (c >= symbolIndex.length)
    {
      System.err.println(String.format(BAD_ASCII_ERROR, c));
      return 0;
    }
    if (d >= symbolIndex.length)
    {
      System.err.println(String.format(BAD_ASCII_ERROR, d));
      return 0;
    }

    int cIndex = symbolIndex[c];
    int dIndex = symbolIndex[d];
    if (cIndex != UNMAPPED && dIndex != UNMAPPED)
    {
      return matrix[cIndex][dIndex];
    }
    return 0;
  }

  /**
   * pretty print the matrix
   */
  @Override
  public String toString()
  {
    return outputMatrix(false);
  }

  /**
   * Print the score matrix, optionally formatted as html, with the alphabet symbols as column headings and at the start of each row
   * @param html
   * @return
   */
  public String outputMatrix(boolean html)
  {
    StringBuilder sb = new StringBuilder(512);

    /*
     * heading row with alphabet
     */
    if (html)
    {
      sb.append("<table border=\"1\">");
      sb.append(html ? "<tr><th></th>" : "");
    }
    for (char sym : symbols)
    {
      if (html)
      {
        sb.append("<th>&nbsp;").append(sym).append("&nbsp;</th>");
      }
      else
      {
        sb.append("\t").append(sym);
      }
    }
    sb.append(html ? "</tr>\n" : "\n");

    /*
     * table of scores
     */
    for (char c1 : symbols)
    {
      if (html)
      {
        sb.append("<tr><td>");
      }
      sb.append(c1).append(html ? "</td>" : "");
      for (char c2 : symbols)
      {
        sb.append(html ? "<td>" : "\t")
                .append(matrix[symbolIndex[c1]][symbolIndex[c2]])
                .append(html ? "</td>" : "");
      }
      sb.append(html ? "</tr>\n" : "\n");
    }
    if (html)
    {
      sb.append("</table>");
    }
    return sb.toString();
  }

  /**
   * Answers the number of symbols coded for (also equal to the number of rows
   * and columns of the score matrix)
   * 
   * @return
   */
  public int getSize()
  {
    return symbols.length;
  }

  /**
   * Computes an NxN matrix where N is the number of sequences, and entry [i, j]
   * is sequence[i] pairwise multiplied with sequence[j], as a sum of scores
   * computed using the current score matrix. For example
   * <ul>
   * <li>Sequences:</li>
   * <li>FKL</li>
   * <li>R-D</li>
   * <li>QIA</li>
   * <li>GWC</li>
   * <li>Score matrix is BLOSUM62</li>
   * <li>Gaps treated same as X (unknown)</li>
   * <li>product [0, 0] = F.F + K.K + L.L = 6 + 5 + 4 = 15</li>
   * <li>product [1, 1] = R.R + -.- + D.D = 5 + -1 + 6 = 10</li>
   * <li>product [2, 2] = Q.Q + I.I + A.A = 5 + 4 + 4 = 13</li>
   * <li>product [3, 3] = G.G + W.W + C.C = 6 + 11 + 9 = 26</li>
   * <li>product[0, 1] = F.R + K.- + L.D = -3 + -1 + -3 = -8
   * <li>and so on</li>
   * </ul>
   */
  public MatrixI computePairwiseScores(String[] seqs)
  {
    double[][] values = new double[seqs.length][];
    for (int row = 0; row < seqs.length; row++)
    {
      values[row] = new double[seqs.length];
      for (int col = 0; col < seqs.length; col++)
      {
        int total = 0;
        int width = Math.min(seqs[row].length(), seqs[col].length());
        for (int i = 0; i < width; i++)
        {
          char c1 = seqs[row].charAt(i);
          char c2 = seqs[col].charAt(i);
          float score = getPairwiseScore(c1, c2);
          total += score;
        }
        values[row][col] = total;
      }
    }
    return new Matrix(values);
  }
}