1 /*
2 * Copyright (c) Huawei Technologies Co., Ltd. 2023. All rights reserved.
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
16 #include "codec_cov.h"
17
18 #include <memory>
19 #ifdef _WIN32
20 #include <windows.h>
21 #endif
22
23 namespace SysTuning {
24 namespace base {
PreNum(unsigned char byte)25 int32_t PreNum(unsigned char byte)
26 {
27 constexpr uint32_t BITS = 8;
28 unsigned char mask = 0x80;
29 int32_t num = 0;
30 for (uint32_t i = 0; i < BITS; i++) {
31 if ((byte & mask) == mask) {
32 mask = mask >> 1;
33 num++;
34 } else {
35 break;
36 }
37 }
38 return num;
39 }
40
IsUTF8(const uint8_t * data,int32_t len)41 bool IsUTF8(const uint8_t *data, int32_t len)
42 {
43 constexpr uint8_t mask = 0x80;
44 constexpr uint8_t firstByte = 0xc0;
45 constexpr int32_t target = 2;
46 int32_t num = 0;
47 int32_t i = 0;
48 while (i < len) {
49 if ((data[i] & mask) == 0x00) {
50 i++;
51 continue;
52 }
53 if ((num = PreNum(data[i])) <= target) {
54 return false;
55 }
56 i++;
57 for (int32_t j = 0; j < num - 1; j++) {
58 if ((data[i] & firstByte) != mask) {
59 return false;
60 }
61 i++;
62 }
63 }
64 return true;
65 }
66
IsGBK(const uint8_t * data,int32_t len)67 bool IsGBK(const uint8_t *data, int32_t len)
68 {
69 constexpr int32_t step = 2;
70 constexpr uint8_t asciiEnd = 0x7f;
71 constexpr uint8_t firstByte = 0x81;
72 constexpr uint8_t firstByteEnd = 0xfe;
73 constexpr uint8_t secondByteOne = 0x40;
74 constexpr uint8_t secondByteTwoEnd = 0xfe;
75 constexpr uint8_t gbkMask = 0xf7;
76 int32_t i = 0;
77 while (i < len) {
78 if (data[i] <= asciiEnd) {
79 i++;
80 continue;
81 } else {
82 if (data[i] >= firstByte && data[i] <= firstByteEnd && data[i + 1] >= secondByteOne &&
83 data[i + 1] <= secondByteTwoEnd && data[i + 1] != gbkMask) {
84 i += step;
85 continue;
86 } else {
87 return false;
88 }
89 }
90 }
91 return true;
92 }
93
GetCoding(const uint8_t * data,int32_t len)94 CODING GetCoding(const uint8_t *data, int32_t len)
95 {
96 CODING coding;
97 if (IsUTF8(data, len)) {
98 coding = UTF8;
99 } else if (IsGBK(data, len)) {
100 coding = GBK;
101 } else {
102 coding = UNKOWN;
103 }
104 return coding;
105 }
106
107 #ifdef _WIN32
GbkToUtf8(const char * srcStr)108 std::string GbkToUtf8(const char *srcStr)
109 {
110 int32_t len = MultiByteToWideChar(CP_ACP, 0, srcStr, -1, NULL, 0);
111 std::unique_ptr<wchar_t[]> wstr = std::make_unique<wchar_t[]>(len + 1);
112 MultiByteToWideChar(CP_ACP, 0, srcStr, -1, wstr.get(), len);
113 len = WideCharToMultiByte(CP_UTF8, 0, wstr.get(), -1, NULL, 0, NULL, NULL);
114 std::unique_ptr<char[]> str = std::make_unique<char[]>(len + 1);
115 WideCharToMultiByte(CP_UTF8, 0, wstr.get(), -1, str.get(), len, NULL, NULL);
116 return std::string(str.get());
117 }
118 #endif
119 } // namespace base
120 } // namespace SysTuning
121