| 
001002
 003
 004
 005
 006
 007
 008
 009
 010
 011
 012
 013
 014
 015
 016
 017
 018
 019
 020
 021
 022
 023
 024
 025
 026
 027
 028
 029
 030
 031
 032
 033
 034
 035
 036
 037
 038
 039
 040
 041
 042
 043
 044
 045
 046
 047
 048
 049
 050
 051
 052
 053
 054
 055
 056
 057
 058
 059
 060
 061
 062
 063
 064
 065
 066
 067
 068
 069
 070
 071
 072
 073
 074
 075
 076
 077
 078
 079
 080
 081
 082
 083
 084
 085
 086
 087
 088
 089
 090
 091
 092
 093
 094
 095
 096
 097
 098
 099
 100
 101
 102
 103
 104
 105
 106
 107
 108
 109
 110
 111
 112
 113
 114
 115
 116
 117
 118
 119
 120
 121
 122
 123
 124
 125
 126
 127
 128
 129
 130
 131
 132
 133
 134
 135
 136
 137
 138
 139
 140
 141
 142
 143
 144
 145
 146
 147
 148
 149
 150
 151
 152
 153
 154
 155
 156
 157
 158
 159
 160
 161
 162
 163
 164
 165
 166
 167
 168
 169
 170
 171
 172
 173
 174
 175
 176
 
 | package algs63; // section 6.3
import stdlib.*;
/* ***********************************************************************
 *  Compilation:  javac Manber.java
 *  Execution:    java Manber < text.txt
 *  Dependencies: In.java
 *
 *  Reads a text corpus from stdin and suffix sorts it in subquadratic
 *  time using a variant of XManber's algorithm.
 *
 *  NOTE: I THINK THIS IS CYCLIC SUFFIX SORTING
 *
 *************************************************************************/
public class XManber {
  private final int N;               // length of input string
  private final String text;         // input text
  private final int[] index;         // offset of ith string in order
  private final int[] rank;          // rank of ith string
  private final int[] newrank;       // rank of ith string (temporary)
  private int offset;
  public XManber(String s) {
    N    = s.length();
    text = s;
    index   = new int[N+1];
    rank    = new int[N+1];
    newrank = new int[N+1];
    // sentinels
    index[N] = N;
    rank[N] = -1;
    msd();
    doit();
  }
  // do one pass of msd sorting by rank at given offset
  private void doit() {
    for (offset = 1; offset < N; offset += offset) {
      StdOut.println("offset = " + offset);
      int count = 0;
      for (int i = 1; i <= N; i++) {
        if (rank[index[i]] == rank[index[i-1]]) count++;
        else if (count > 0) {
          // sort
          int left = i-1-count;
          int right = i-1;
          quicksort(left, right);
          // now fix up ranks
          int r = rank[index[left]];
          for (int j = left + 1; j <= right; j++) {
            if (less(index[j-1], index[j]))  {
              r = rank[index[left]] + j - left;
            }
            newrank[index[j]] = r;
          }
          // copy back - note can't update rank too eagerly
          for (int j = left + 1; j <= right; j++) {
            rank[index[j]] = newrank[index[j]];
          }
          count = 0;
        }
      }
    }
  }
  // sort by leading char, assumes extended ASCII (256 values)
  private void msd() {
    // calculate frequencies
    int[] freq = new int[256];
    for (int i = 0; i < N; i++)
      freq[text.charAt(i)]++;
    // calculate cumulative frequencies
    int[] cumm = new int[256];
    for (int i = 1; i < 256; i++)
      cumm[i] = cumm[i-1] + freq[i-1];
    // compute ranks
    for (int i = 0; i < N; i++)
      rank[i] = cumm[text.charAt(i)];
    // sort by first char
    for (int i = 0; i < N; i++)
      index[cumm[text.charAt(i)]++] = i;
  }
  // for debugging
  public void show() {
    String texttext = text + text;  // make cyclic
    StdOut.println("j, rank[index[j]], index[j]");
    for (int i = 0; i < N; i++) {
      String s = texttext.substring(index[i], index[i] +  Math.min(40, N));
      StdOut.println(s + " " + i + " " + rank[index[i]] + " " + index[i]);
    }
    StdOut.println();
  }
  // test client
  public static void main(String[] args) {
    String s = StdIn.readAll();
    XManber m = new XManber(s);
    m.show();
  }
  /* ********************************************************************
   *  Helper functions for comparing suffixes.
   **********************************************************************/
  /* ********************************************************************
   * Is the substring text[v..N] lexicographically less than the
   * substring text[w..N] ?
   **********************************************************************/
  private boolean less(int v, int w) {
    if (v + offset >= N) v -= N;
    if (w + offset >= N) w -= N;
    return rank[v + offset] < rank[w + offset];
  }
  /* ***********************************************************************
   *  Quicksort code from Sedgewick 7.1, 7.2.
   *************************************************************************/
  // swap pointer sort indices
  private void exch(int i, int j) {
    int swap = index[i];
    index[i] = index[j];
    index[j] = swap;
  }
  // SUGGEST REPLACING WITH 3-WAY QUICKSORT SINCE ELEMENTS ARE
  // RANKS AND THERE MAY BE DUPLICATES
  void quicksort(int l, int r) {
    if (r <= l) return;
    int i = partition(l, r);
    quicksort(l, i-1);
    quicksort(i+1, r);
  }
  int partition(int l, int r) {
    int i = l-1, j = r;
    int v = index[r];
    while (true) {
      // find item on left to swap
      while (less(index[++i], v))
        ;
      // find item on right to swap
      while (less(v, index[--j]))
        if (j == l) break;
      // check if pointers cross
      if (i >= j) break;
      exch(i, j);
    }
    // swap with partition element
    exch(i, r);
    return i;
  }
}
 |