Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
105 changes: 69 additions & 36 deletions StringAlgorithm/StringAlgorithm.java
Original file line number Diff line number Diff line change
@@ -1,62 +1,95 @@
import java.util.Arrays;

class StringAlgorithm {
private static int[] saNaive(int[] s) {
int n = s.length;
Integer[] _sa = new Integer[n];
for (int i = 0; i < n; i++) {
_sa[i] = i;
int[] sa = new int[n];
for(int i = 0;i < n;i++){
sa[i] = i;
}
java.util.Arrays.sort(_sa, (l, r) -> {
insertionsortUsingComparator(sa, (l, r) -> {
while (l < n && r < n) {
if (s[l] != s[r]) return s[l] - s[r];
l++;
r++;
}
return -(l - r);
});
int[] sa = new int[n];
for (int i = 0; i < n; i++) {
sa[i] = _sa[i];
}
return sa;
}

private static int[] saDoubling(int[] s) {
int n = s.length;
Integer[] _sa = new Integer[n];
for (int i = 0; i < n; i++) {
_sa[i] = i;
int[] sa = new int[n];
for(int i = 0;i < n;i++){
sa[i] = i;
}
int[] rnk = s;
int[] tmp = new int[n];

for (int k = 1; k < n; k *= 2) {
final int _k = k;
final int[] _rnk = rnk;
java.util.Comparator<Integer> cmp = (x, y) -> {
java.util.function.IntBinaryOperator cmp = (x, y) -> {
if (_rnk[x] != _rnk[y]) return _rnk[x] - _rnk[y];
int rx = x + _k < n ? _rnk[x + _k] : -1;
int ry = y + _k < n ? _rnk[y + _k] : -1;
return rx - ry;
};
java.util.Arrays.sort(_sa, cmp);
tmp[_sa[0]] = 0;
mergesortUsingComparator(sa, cmp);
tmp[sa[0]] = 0;
for (int i = 1; i < n; i++) {
tmp[_sa[i]] = tmp[_sa[i - 1]] + (cmp.compare(_sa[i - 1], _sa[i]) < 0 ? 1 : 0);
tmp[sa[i]] = tmp[sa[i - 1]] + (cmp.applyAsInt(sa[i - 1], sa[i]) < 0 ? 1 : 0);
}
int[] buf = tmp;
tmp = rnk;
rnk = buf;
}
return sa;
}

int[] sa = new int[n];
for (int i = 0; i < n; i++) {
sa[i] = _sa[i];
private static void insertionsortUsingComparator(int[] a, java.util.function.IntBinaryOperator comparator) {
final int n = a.length;
for (int i = 1; i < n; i++) {
final int tmp = a[i];
if (comparator.applyAsInt(a[i - 1], tmp) > 0) {
int j = i;
do {
a[j] = a[j - 1];
j--;
} while (j > 0 && comparator.applyAsInt(a[j - 1], tmp) > 0);
a[j] = tmp;
}
}
return sa;
}

private static final int THRESHOLD_NAIVE = 10;
private static final int THRESHOLD_DOUBLING = 40;
private static void mergesortUsingComparator(int[] a, java.util.function.IntBinaryOperator comparator) {
final int n = a.length;
final int[] work = new int[n];
for (int block = 1; block <= n; block <<= 1) {
final int block2 = block << 1;
for (int l = 0, max = n - block; l < max; l += block2) {
int m = l + block;
int r = Math.min(l + block2, n);
System.arraycopy(a, l, work, 0, block);
for (int i = l, wi = 0, ti = m;; i++) {
if (ti == r) {
System.arraycopy(work, wi, a, i, block - wi);
break;
}
if (comparator.applyAsInt(work[wi], a[ti]) > 0) {
a[i] = a[ti++];
} else {
a[i] = work[wi++];
if (wi == block) break;
}
}
}
}
}

private static final int THRESHOLD_NAIVE = 50;
private static final int THRESHOLD_DOUBLING = 0;

private static int[] sais(int[] s, int upper) {
int n = s.length;
Expand All @@ -72,9 +105,9 @@ private static int[] sais(int[] s, int upper) {
if (n < THRESHOLD_NAIVE) {
return saNaive(s);
}
if (n < THRESHOLD_DOUBLING) {
return saDoubling(s);
}
// if (n < THRESHOLD_DOUBLING) {
// return saDoubling(s);
// }

int[] sa = new int[n];
boolean[] ls = new boolean[n];
Expand Down Expand Up @@ -195,22 +228,22 @@ public static int[] suffixArray(int[] s, int upper) {
return sais(s, upper);
}

public static int[] suffixArray(int[] s) {
public static int[] suffixArray(int[] s)
{
int n = s.length;
Integer[] idx = new Integer[n];
for (int i = 0; i < n; i++) {
idx[i] = i;
int[] vals = Arrays.copyOf(s, n);
java.util.Arrays.sort(vals);
int p = 1;
for(int i = 1;i < n;i++){
if(vals[i] != vals[i-1]){
vals[p++] = vals[i];
}
}
java.util.Arrays.sort(idx, (l, r) -> s[l] - s[r]);
int[] s2 = new int[n];
int now = 0;
for (int i = 0; i < n; i++) {
if (i > 0 && s[idx[i - 1]] != s[idx[i]]) {
now++;
}
s2[idx[i]] = now;
for(int i = 0;i < n;i++){
s2[i] = java.util.Arrays.binarySearch(vals, 0, p, s[i]);
}
return sais(s2, now);
return sais(s2, p);
}

public static int[] suffixArray(char[] s) {
Expand Down