|
| 1 | +package com.jwetherell.algorithms.data_structures; |
| 2 | + |
| 3 | +import java.util.ArrayList; |
| 4 | + |
| 5 | +/** |
| 6 | + * In computer science, the longest common prefix array (LCP array) is an auxiliary |
| 7 | + * data structure to the suffix array. It stores the lengths of the longest common |
| 8 | + * prefixes (LCPs) between all pairs of consecutive suffixes in a sorted suffix array. |
| 9 | + * |
| 10 | + * https://en.wikipedia.org/wiki/LCP_array |
| 11 | + * |
| 12 | + * @author Jakub Szarawarski <kubaszarawarski@gmail.com> |
| 13 | + */ |
| 14 | +public class LCPArray { |
| 15 | + private static final char DEFAULT_END_SEQ_CHAR = '$'; |
| 16 | + private char END_SEQ_CHAR; |
| 17 | + |
| 18 | + private SuffixArray suffixArrayBuilder; |
| 19 | + |
| 20 | + private ArrayList<Integer> LCP; |
| 21 | + |
| 22 | + public LCPArray(CharSequence sequence){ |
| 23 | + this(sequence, DEFAULT_END_SEQ_CHAR); |
| 24 | + } |
| 25 | + |
| 26 | + public LCPArray(CharSequence sequence, char endChar) { |
| 27 | + END_SEQ_CHAR = endChar; |
| 28 | + suffixArrayBuilder = new SuffixArray(sequence, endChar); |
| 29 | + } |
| 30 | + |
| 31 | + public ArrayList<Integer> getLCPArray() { |
| 32 | + if(LCP == null){ |
| 33 | + LCPAlgorithm(); |
| 34 | + } |
| 35 | + return LCP; |
| 36 | + } |
| 37 | + |
| 38 | + private void LCPAlgorithm() { |
| 39 | + ArrayList<Integer> LCPR = getLCPR(); |
| 40 | + getLCPfromLCPR(LCPR); |
| 41 | + } |
| 42 | + |
| 43 | + private ArrayList<Integer> getLCPR() { |
| 44 | + ArrayList<Integer> KMRArray = suffixArrayBuilder.getKMRarray(); |
| 45 | + ArrayList<Integer> suffixArray = suffixArrayBuilder.getSuffixArray(); |
| 46 | + String string = suffixArrayBuilder.getString(); |
| 47 | + int length = KMRArray.size(); |
| 48 | + |
| 49 | + ArrayList<Integer> LCPR = new ArrayList<Integer>(); // helper array, LCP[i] = LCPR[suffixArray[i]] |
| 50 | + |
| 51 | + int startingValue = 0; |
| 52 | + |
| 53 | + for(int i=0; i<length; i++){ |
| 54 | + if(KMRArray.get(i).equals(0)){ |
| 55 | + LCPR.add(0); |
| 56 | + startingValue = 0; |
| 57 | + } |
| 58 | + else{ |
| 59 | + int LCPRValue = startingValue; |
| 60 | + int predecessor = suffixArray.get(KMRArray.get(i)-1); |
| 61 | + |
| 62 | + while(string.charAt(i+LCPRValue) == string.charAt(predecessor+LCPRValue)) |
| 63 | + LCPRValue++; |
| 64 | + |
| 65 | + LCPR.add(LCPRValue); |
| 66 | + startingValue = LCPRValue-1 > 0 ? LCPRValue-1 : 0; |
| 67 | + } |
| 68 | + } |
| 69 | + |
| 70 | + return LCPR; |
| 71 | + } |
| 72 | + |
| 73 | + private void getLCPfromLCPR(ArrayList<Integer> LCPR) { |
| 74 | + ArrayList<Integer> suffixArray = suffixArrayBuilder.getSuffixArray(); |
| 75 | + int length = suffixArray.size(); |
| 76 | + |
| 77 | + LCP = new ArrayList<Integer>(); |
| 78 | + LCP.add(null); //no value for LCP[0] |
| 79 | + |
| 80 | + for(int i=1; i<length; i++){ |
| 81 | + LCP.add(LCPR.get(suffixArray.get(i))); |
| 82 | + } |
| 83 | + } |
| 84 | +} |
0 commit comments