1 | // Copyright 2011 Google Inc. All Rights Reserved. |
2 | // |
3 | // Licensed under the Apache License, Version 2.0 (the "License"); |
4 | // you may not use this file except in compliance with the License. |
5 | // You may obtain a copy of the License at |
6 | // |
7 | // http://www.apache.org/licenses/LICENSE-2.0 |
8 | // |
9 | // Unless required by applicable law or agreed to in writing, software |
10 | // distributed under the License is distributed on an "AS IS" BASIS, |
11 | // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
12 | // See the License for the specific language governing permissions and |
13 | // limitations under the License. |
14 | |
15 | #include "edit_distance.h" |
16 | |
17 | #include <algorithm> |
18 | #include <vector> |
19 | |
20 | using namespace std; |
21 | |
22 | int EditDistance(const StringPiece& s1, |
23 | const StringPiece& s2, |
24 | bool allow_replacements, |
25 | int max_edit_distance) { |
26 | // The algorithm implemented below is the "classic" |
27 | // dynamic-programming algorithm for computing the Levenshtein |
28 | // distance, which is described here: |
29 | // |
30 | // http://en.wikipedia.org/wiki/Levenshtein_distance |
31 | // |
32 | // Although the algorithm is typically described using an m x n |
33 | // array, only one row plus one element are used at a time, so this |
34 | // implementation just keeps one vector for the row. To update one entry, |
35 | // only the entries to the left, top, and top-left are needed. The left |
36 | // entry is in row[x-1], the top entry is what's in row[x] from the last |
37 | // iteration, and the top-left entry is stored in previous. |
38 | int m = s1.len_; |
39 | int n = s2.len_; |
40 | |
41 | vector<int> row(n + 1); |
42 | for (int i = 1; i <= n; ++i) |
43 | row[i] = i; |
44 | |
45 | for (int y = 1; y <= m; ++y) { |
46 | row[0] = y; |
47 | int best_this_row = row[0]; |
48 | |
49 | int previous = y - 1; |
50 | for (int x = 1; x <= n; ++x) { |
51 | int old_row = row[x]; |
52 | if (allow_replacements) { |
53 | row[x] = min(previous + (s1.str_[y - 1] == s2.str_[x - 1] ? 0 : 1), |
54 | min(row[x - 1], row[x]) + 1); |
55 | } |
56 | else { |
57 | if (s1.str_[y - 1] == s2.str_[x - 1]) |
58 | row[x] = previous; |
59 | else |
60 | row[x] = min(row[x - 1], row[x]) + 1; |
61 | } |
62 | previous = old_row; |
63 | best_this_row = min(best_this_row, row[x]); |
64 | } |
65 | |
66 | if (max_edit_distance && best_this_row > max_edit_distance) |
67 | return max_edit_distance + 1; |
68 | } |
69 | |
70 | return row[n]; |
71 | } |
72 | |