1 | // LAF Base Library |
2 | // Copyright (c) 2020-2022 Igara Studio S.A. |
3 | // Copyright (c) 2001-2016 David Capello |
4 | // |
5 | // This file is released under the terms of the MIT license. |
6 | // Read LICENSE.txt for more information. |
7 | |
8 | #ifdef HAVE_CONFIG_H |
9 | #include "config.h" |
10 | #endif |
11 | |
12 | #include "base/debug.h" |
13 | #include "base/string.h" |
14 | #include "base/utf8_decode.h" |
15 | |
16 | #include <cctype> |
17 | #include <vector> |
18 | |
19 | #ifdef LAF_WINDOWS |
20 | #include <windows.h> |
21 | #endif |
22 | |
23 | namespace base { |
24 | |
25 | std::string string_printf(const char* format, ...) |
26 | { |
27 | std::va_list ap; |
28 | va_start(ap, format); |
29 | std::string result = string_vprintf(format, ap); |
30 | va_end(ap); |
31 | return result; |
32 | } |
33 | |
34 | std::string string_vprintf(const char* format, va_list ap) |
35 | { |
36 | std::vector<char> buf(1, 0); |
37 | std::va_list ap2; |
38 | va_copy(ap2, ap); |
39 | size_t required_size = std::vsnprintf(nullptr, 0, format, ap); |
40 | if (required_size) { |
41 | buf.resize(required_size+1); |
42 | std::vsnprintf(&buf[0], buf.size(), format, ap2); |
43 | } |
44 | va_end(ap2); |
45 | return std::string(&buf[0]); |
46 | } |
47 | |
48 | std::string string_to_lower(const std::string& original) |
49 | { |
50 | std::wstring result(from_utf8(original)); |
51 | auto it(result.begin()); |
52 | auto end(result.end()); |
53 | while (it != end) { |
54 | *it = std::tolower(*it); |
55 | ++it; |
56 | } |
57 | return to_utf8(result); |
58 | } |
59 | |
60 | std::string string_to_upper(const std::string& original) |
61 | { |
62 | std::wstring result(from_utf8(original)); |
63 | auto it(result.begin()); |
64 | auto end(result.end()); |
65 | while (it != end) { |
66 | *it = std::toupper(*it); |
67 | ++it; |
68 | } |
69 | return to_utf8(result); |
70 | } |
71 | |
72 | #ifdef LAF_WINDOWS |
73 | |
74 | std::string to_utf8(const wchar_t* src, const int n) |
75 | { |
76 | int required_size = |
77 | ::WideCharToMultiByte(CP_UTF8, 0, |
78 | src, (int)n, |
79 | NULL, 0, NULL, NULL); |
80 | |
81 | if (required_size == 0) |
82 | return std::string(); |
83 | |
84 | std::vector<char> buf(++required_size); |
85 | |
86 | ::WideCharToMultiByte(CP_UTF8, 0, |
87 | src, (int)n, |
88 | &buf[0], required_size, |
89 | NULL, NULL); |
90 | |
91 | return std::string(&buf[0]); |
92 | } |
93 | |
94 | std::wstring from_utf8(const std::string& src) |
95 | { |
96 | int required_size = |
97 | MultiByteToWideChar(CP_UTF8, 0, |
98 | src.c_str(), (int)src.size(), |
99 | NULL, 0); |
100 | |
101 | if (required_size == 0) |
102 | return std::wstring(); |
103 | |
104 | std::vector<wchar_t> buf(++required_size); |
105 | |
106 | ::MultiByteToWideChar(CP_UTF8, 0, |
107 | src.c_str(), (int)src.size(), |
108 | &buf[0], required_size); |
109 | |
110 | return std::wstring(&buf[0]); |
111 | } |
112 | |
113 | #else |
114 | |
115 | // Based on Allegro Unicode code (allegro/src/unicode.c) |
116 | static std::size_t insert_utf8_char(std::string* result, wchar_t chr) |
117 | { |
118 | int size, bits, b, i; |
119 | |
120 | if (chr < 128) { |
121 | if (result) |
122 | result->push_back(chr); |
123 | return 1; |
124 | } |
125 | |
126 | bits = 7; |
127 | while (chr >= (1<<bits)) |
128 | bits++; |
129 | |
130 | size = 2; |
131 | b = 11; |
132 | |
133 | while (b < bits) { |
134 | size++; |
135 | b += 5; |
136 | } |
137 | |
138 | if (result) { |
139 | b -= (7-size); |
140 | int firstbyte = chr>>b; |
141 | for (i=0; i<size; i++) |
142 | firstbyte |= (0x80>>i); |
143 | |
144 | result->push_back(firstbyte); |
145 | |
146 | for (i=1; i<size; i++) { |
147 | b -= 6; |
148 | result->push_back(0x80 | ((chr>>b)&0x3F)); |
149 | } |
150 | } |
151 | |
152 | return size; |
153 | } |
154 | |
155 | std::string to_utf8(const wchar_t* src, const int n) |
156 | { |
157 | // Get required size to reserve a string so string::push_back() |
158 | // doesn't need to reallocate its data. |
159 | std::size_t required_size = 0; |
160 | auto p = src; |
161 | for (int i=0; i<n; ++i, ++p) |
162 | required_size += insert_utf8_char(nullptr, *p); |
163 | if (!required_size) |
164 | return "" ; |
165 | |
166 | std::string result; |
167 | result.reserve(++required_size); |
168 | p = src; |
169 | for (int i=0; i<n; ++i, ++p) |
170 | insert_utf8_char(&result, *p); |
171 | return result; |
172 | } |
173 | |
174 | std::wstring from_utf8(const std::string& src) |
175 | { |
176 | int required_size = utf8_length(src); |
177 | std::vector<wchar_t> buf(++required_size); |
178 | std::vector<wchar_t>::iterator buf_it = buf.begin(); |
179 | #ifdef _DEBUG |
180 | std::vector<wchar_t>::iterator buf_end = buf.end(); |
181 | #endif |
182 | utf8_decode decode(src); |
183 | |
184 | while (int chr = decode.next()) { |
185 | ASSERT(buf_it != buf_end); |
186 | *buf_it = chr; |
187 | ++buf_it; |
188 | } |
189 | |
190 | return std::wstring(&buf[0]); |
191 | } |
192 | |
193 | #endif |
194 | |
195 | int utf8_length(const std::string& utf8string) |
196 | { |
197 | utf8_decode decode(utf8string); |
198 | int c = 0; |
199 | |
200 | while (decode.next()) |
201 | ++c; |
202 | |
203 | return c; |
204 | } |
205 | |
206 | int utf8_icmp(const std::string& a, const std::string& b, int n) |
207 | { |
208 | utf8_decode a_decode(a); |
209 | utf8_decode b_decode(b); |
210 | int i = 0; |
211 | |
212 | for (; (n == 0 || i < n) |
213 | && !a_decode.is_end() |
214 | && !b_decode.is_end(); ++i) { |
215 | int a_chr = a_decode.next(); |
216 | if (!a_chr) |
217 | break; |
218 | |
219 | int b_chr = b_decode.next(); |
220 | if (!b_chr) |
221 | break; |
222 | |
223 | a_chr = std::tolower(a_chr); |
224 | b_chr = std::tolower(b_chr); |
225 | |
226 | if (a_chr < b_chr) |
227 | return -1; |
228 | else if (a_chr > b_chr) |
229 | return 1; |
230 | } |
231 | |
232 | if (n > 0 && i == n) |
233 | return 0; |
234 | else if (a_decode.is_end() && b_decode.is_end()) |
235 | return 0; |
236 | else if (a_decode.is_end()) |
237 | return -1; |
238 | else |
239 | return 1; |
240 | } |
241 | |
242 | } // namespace base |
243 | |