1 /*
2 * Copyright (C) 2018 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 * http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
16
17 #include "utils/codepoint-range.h"
18
19 #include <algorithm>
20
21 namespace libtextclassifier3 {
22
23 // Returns a sorted list of the codepoint ranges.
SortCodepointRanges(const std::vector<const CodepointRange * > & codepoint_ranges,std::vector<CodepointRangeStruct> * sorted_codepoint_ranges)24 void SortCodepointRanges(
25 const std::vector<const CodepointRange*>& codepoint_ranges,
26 std::vector<CodepointRangeStruct>* sorted_codepoint_ranges) {
27 sorted_codepoint_ranges->clear();
28 sorted_codepoint_ranges->reserve(codepoint_ranges.size());
29 for (const CodepointRange* range : codepoint_ranges) {
30 sorted_codepoint_ranges->push_back(
31 CodepointRangeStruct(range->start(), range->end()));
32 }
33
34 std::stable_sort(
35 sorted_codepoint_ranges->begin(), sorted_codepoint_ranges->end(),
36 [](const CodepointRangeStruct& a, const CodepointRangeStruct& b) {
37 return a.start < b.start;
38 });
39 }
40
41 // Returns true if given codepoint is covered by the given sorted vector of
42 // codepoint ranges.
IsCodepointInRanges(int codepoint,const std::vector<CodepointRangeStruct> & codepoint_ranges)43 bool IsCodepointInRanges(
44 int codepoint, const std::vector<CodepointRangeStruct>& codepoint_ranges) {
45 auto it = std::lower_bound(
46 codepoint_ranges.begin(), codepoint_ranges.end(), codepoint,
47 [](const CodepointRangeStruct& range, int codepoint) {
48 // This function compares range with the
49 // codepoint for the purpose of finding the first
50 // greater or equal range. Because of the use of
51 // std::lower_bound it needs to return true when
52 // range < codepoint; the first time it will
53 // return false the lower bound is found and
54 // returned.
55 //
56 // It might seem weird that the condition is
57 // range.end <= codepoint here but when codepoint
58 // == range.end it means it's actually just
59 // outside of the range, thus the range is less
60 // than the codepoint.
61 return range.end <= codepoint;
62 });
63 if (it != codepoint_ranges.end() && it->start <= codepoint &&
64 it->end > codepoint) {
65 return true;
66 } else {
67 return false;
68 }
69 }
70
71 } // namespace libtextclassifier3
72