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/strings/utf8.h"
18 
19 namespace libtextclassifier3 {
IsValidUTF8(const char * src,int size)20 bool IsValidUTF8(const char *src, int size) {
21   for (int i = 0; i < size;) {
22     const int char_length = ValidUTF8CharLength(src + i, size - i);
23     if (char_length <= 0) {
24       return false;
25     }
26     i += char_length;
27   }
28   return true;
29 }
30 
ValidUTF8CharLength(const char * src,int size)31 int ValidUTF8CharLength(const char *src, int size) {
32   // Unexpected trail byte.
33   if (IsTrailByte(src[0])) {
34     return -1;
35   }
36 
37   const int num_codepoint_bytes = GetNumBytesForUTF8Char(&src[0]);
38   if (num_codepoint_bytes <= 0 || num_codepoint_bytes > size) {
39     return -1;
40   }
41 
42   // Check that remaining bytes in the codepoint are trailing bytes.
43   for (int k = 1; k < num_codepoint_bytes; k++) {
44     if (!IsTrailByte(src[k])) {
45       return -1;
46     }
47   }
48 
49   return num_codepoint_bytes;
50 }
51 
52 }  // namespace libtextclassifier3
53