projects / mirror_edk2.git / blame
| Commit | Line | Data |
|---|---|---|
| a7a8363d | 1 | /** @file\r |
| 450ea6d5 | 2 | Copyright (c) 2016, Daryl McDaniel. All rights reserved.<BR>\r |
| a7a8363d | 3 | Copyright (c) 2012, Intel Corporation. All rights reserved.<BR>\r |
| 4 | This program and the accompanying materials\r | |
| 5 | are licensed and made available under the terms and conditions of the BSD License\r | |
| 6 | which accompanies this distribution. The full text of the license may be found at\r | |
| 7 | http://opensource.org/licenses/bsd-license.php\r | |
| 8 | \r | |
| 9 | THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,\r | |
| 10 | WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.\r | |
| 11 | **/\r | |
| 12 | #include <assert.h>\r | |
| 13 | #include <string.h>\r | |
| 14 | #include <errno.h>\r | |
| 15 | #include <stdlib.h>\r | |
| 16 | #include <wchar.h>\r | |
| 17 | #include <sys/types.h>\r | |
| 05c7d5f6 | 18 | #include <limits.h>\r |
| a7a8363d | 19 | \r |
| 450ea6d5 | 20 | typedef int ch_UCS4;\r |
| a7a8363d | 21 | \r |
| c42c9cac | 22 | static mbstate_t LocalConvState = {0};\r |
| a7a8363d | 23 | \r |
| 24 | /** Map a UTF-8 encoded prefix byte to a sequence length.\r | |
| 25 | Zero means illegal prefix, but valid surrogate if < 0xC0.\r | |
| 26 | One indicates an ASCII-7 equivalent character.\r | |
| 27 | Two, three, and four are the first byte for 2, 3, and 4 byte sequences, respectively.\r | |
| 28 | See RFC 3629 for details.\r | |
| 29 | \r | |
| 30 | TABLE ENCODING:\r | |
| 31 | Low Nibble decodes the first byte into the number of bytes in the sequence.\r | |
| 32 | A value of zero indicates an invalid byte.\r | |
| 33 | The High Nibble encodes a bit mask to be used to match against the high nibble of the second byte.\r | |
| 34 | \r | |
| 35 | example:\r | |
| 36 | SequenceLength = code[c0] & 0x0F;\r | |
| 37 | Mask = 0x80 | code[c0];\r | |
| 38 | \r | |
| 39 | Surrogate bytes are valid if: code[cX] & Mask > 0x80;\r | |
| 40 | \r | |
| 41 | */\r | |
| 42 | static\r | |
| 43 | UINT8 utf8_code_length[256] = {\r | |
| 44 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, /* 00-0F */\r | |
| 45 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 46 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 47 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 48 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 49 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 50 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,\r | |
| 51 | 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, /* 70-7F */\r | |
| 52 | 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, /* 80-8F */\r | |
| 53 | 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, 0xA0, /* 90-9F */\r | |
| 54 | 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, /* A0-AF */\r | |
| 55 | 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, 0xC0, /* B0-BF */\r | |
| 56 | 0x00, 0x00, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, /* C0-C1 + C2-CF */\r | |
| 57 | 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, /* D0-DF */\r | |
| 58 | 0x43, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x73, 0x33, 0x73, 0x73, /* E0-EF */\r | |
| 59 | 0x64, 0x74, 0x74, 0x74, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 /* F0-F4 + F5-FF */\r | |
| 60 | };\r | |
| 61 | \r | |
| 62 | /** Process one byte of a multibyte character.\r | |
| 63 | \r | |
| c42c9cac | 64 | @param[in] ch One byte of a multibyte character.\r |
| 65 | @param[in,out] ps Pointer to a conversion state object.\r | |
| a7a8363d | 66 | \r |
| c42c9cac | 67 | @retval -2 ch is an incomplete but potentially valid character.\r |
| 68 | @retval -1 ch is not valid in this context.\r | |
| 69 | @retval 1:4 The length, in bytes, of the character ch just completed.\r | |
| a7a8363d | 70 | **/\r |
| 71 | static\r | |
| 72 | int\r | |
| 73 | ProcessOneByte(unsigned char ch, mbstate_t *ps)\r | |
| 74 | {\r | |
| 75 | UINT32 Mask;\r | |
| 76 | UINT32 Length;\r | |
| 77 | int RetVal = 0;\r | |
| 78 | \r | |
| 79 | if(ps->A > 3) {\r | |
| 80 | // We are in an invalid state\r | |
| 81 | ps->A = 0; // Initial State\r | |
| 82 | }\r | |
| 450ea6d5 | 83 | ps->C[ps->A] = ch; // Save the current byte\r |
| a7a8363d | 84 | Mask = utf8_code_length[ch];\r |
| 85 | \r | |
| 86 | if(ps->A == 0) { // Initial State. First byte of sequence.\r | |
| 87 | ps->E = Mask | 0x80;\r | |
| 88 | Length = Mask & 0xF;\r | |
| 89 | switch(Length) {\r | |
| 90 | case 0: // State 0, Code 0\r | |
| 91 | errno = EILSEQ;\r | |
| 92 | RetVal = -1;\r | |
| 450ea6d5 | 93 | ps->E = 1; // Consume this byte\r |
| a7a8363d | 94 | break;\r |
| 95 | case 1: // State 0, Code 1\r | |
| 96 | // ASCII-7 Character\r | |
| 97 | ps->B = ps->D[0] = ch;\r | |
| 98 | RetVal = 1;\r | |
| 99 | break;\r | |
| 100 | default: // State 0, Code 2, 3, 4\r | |
| 101 | ps->A = 1; // Next state is State-1\r | |
| 102 | RetVal = -2; // Incomplete but potentially valid character\r | |
| 103 | break;\r | |
| 104 | }\r | |
| 105 | }\r | |
| 106 | else {\r | |
| 107 | // We are in state 1, 2, or 3 and processing a surrogate byte\r | |
| 108 | Length = ps->E & 0xF;\r | |
| 109 | if((Mask & ps->E) > 0x80) {\r | |
| 110 | // This byte is valid\r | |
| 111 | switch(ps->A) { // Process based upon our current state\r | |
| 112 | case 1: // Second byte of the sequence.\r | |
| 113 | if(Length == 2) { // State 1, Code 2\r | |
| 114 | Length = ((ps->C[0] & 0x1f) << 6) + (ps->C[1] & 0x3f);\r | |
| 115 | assert ((Length > 0x007F) && (Length <= 0x07FF));\r | |
| 116 | ps->B = ps->D[0] = (UINT16)Length;\r | |
| 117 | ps->A = 0; // Next state is State-0\r | |
| 118 | RetVal = 2;\r | |
| 119 | }\r | |
| 450ea6d5 | 120 | else { // This isn't the last byte, get more. State 1, Code 3 or 4\r |
| a7a8363d | 121 | ps->A = 2;\r |
| 122 | RetVal = -2;\r | |
| 123 | }\r | |
| 124 | break;\r | |
| 125 | case 2: // Third byte of the sequence\r | |
| 126 | if(Length == 3) {\r | |
| 127 | Length = ((ps->C[0] & 0x0f) << 12) + ((ps->C[1] & 0x3f) << 6) + (ps->C[2] & 0x3f);\r | |
| 128 | assert ((Length > 0x07FF) && (Length <= 0xFFFF));\r | |
| 129 | ps->B = ps->D[0] = (UINT16)Length;\r | |
| 130 | ps->A = 0; // Next state is State-0\r | |
| 131 | RetVal = 3;\r | |
| 132 | }\r | |
| 133 | else {\r | |
| 134 | ps->A = 3;\r | |
| 135 | RetVal = -2;\r | |
| 136 | }\r | |
| 137 | break;\r | |
| 138 | case 3: // Fourth byte of the sequence\r | |
| 139 | if(Length == 4) {\r | |
| 140 | Length = ((ps->C[0] & 0x7) << 18) + ((ps->C[1] & 0x3f) << 12) +\r | |
| 141 | ((ps->C[2] & 0x3f) << 6) + (ps->C[3] & 0x3f);\r | |
| 142 | ps->B = Length;\r | |
| 143 | assert ((Length > 0xFFFF) && (Length <= 0x10ffff));\r | |
| 144 | \r | |
| 145 | /* compute and append the two surrogates: */\r | |
| 146 | \r | |
| 147 | /* translate from 10000..10FFFF to 0..FFFF */\r | |
| 148 | Length -= 0x10000;\r | |
| 149 | \r | |
| 150 | /* high surrogate = top 10 bits added to D800 */\r | |
| 151 | ps->D[0] = (UINT16)(0xD800 + (Length >> 10));\r | |
| 152 | \r | |
| 153 | /* low surrogate = bottom 10 bits added to DC00 */\r | |
| 154 | ps->D[1] = (UINT16)(0xDC00 + (Length & 0x03FF));\r | |
| 155 | ps->A = 0; // Next state is State-0\r | |
| 156 | RetVal = 4;\r | |
| 157 | }\r | |
| 158 | else {\r | |
| 159 | errno = EILSEQ;\r | |
| 160 | ps->A = 0;\r | |
| 161 | RetVal = -1;\r | |
| 450ea6d5 | 162 | ps->E = 4; // Can't happen, but consume this byte anyway\r |
| a7a8363d | 163 | }\r |
| 164 | break;\r | |
| 165 | }\r | |
| 166 | }\r | |
| 450ea6d5 | 167 | else { // Invalid surrogate byte\r |
| a7a8363d | 168 | errno = EILSEQ;\r |
| 169 | ps->A = 0; // Next is State-0\r | |
| 170 | RetVal = -1;\r | |
| 171 | ps->E = 0; // Don't Consume, it may be an initial byte\r | |
| 172 | }\r | |
| 173 | }\r | |
| 174 | return RetVal;\r | |
| 175 | }\r | |
| 176 | \r | |
| 177 | /** Convert one Multibyte sequence.\r | |
| 178 | \r | |
| c42c9cac | 179 | @param[out] Dest Pointer to output location, or NULL\r |
| 180 | @param[in] Src Multibyte Source (UTF8)\r | |
| 181 | @param[in] Len Max Number of bytes to convert\r | |
| 182 | @param[in] pS Pointer to State struct., or NULL\r | |
| a7a8363d | 183 | \r |
| 184 | @retval -2 Bytes processed comprise an incomplete, but potentially valid, character.\r | |
| 185 | @retval -1 An encoding error was encountered. ps->E indicates the number of bytes consumed.\r | |
| 186 | @retval 0 Either Src is NULL or it points to a NUL character.\r | |
| 187 | @retval 1:N N bytes were consumed producing a valid wide character.\r | |
| 188 | **/\r | |
| 189 | int\r | |
| 190 | DecodeOneStateful(\r | |
| 191 | wchar_t *Dest, // Pointer to output location, or NULL\r | |
| 192 | const char *Src, // Multibyte Source (UTF8)\r | |
| 193 | ssize_t Len, // Max Number of bytes to convert\r | |
| 194 | mbstate_t *pS // Pointer to State struct., or NULL\r | |
| 195 | )\r | |
| 196 | {\r | |
| 197 | const char *SrcEnd;\r | |
| 198 | int NumConv;\r | |
| 199 | unsigned char ch;\r | |
| 200 | \r | |
| a7a8363d | 201 | if(pS == NULL) {\r |
| 202 | pS = &LocalConvState;\r | |
| 203 | }\r | |
| a7a8363d | 204 | NumConv = 0;\r |
| 5a522112 | 205 | if(Src != NULL) {\r |
| 206 | if(*Src != 0) {\r | |
| 207 | SrcEnd = Src + Len;\r | |
| 208 | while(Src < SrcEnd) {\r | |
| 209 | ch = (unsigned char)*Src++;\r | |
| 210 | NumConv = ProcessOneByte(ch, pS);\r | |
| 211 | if(NumConv != -2) {\r | |
| 212 | break;\r | |
| 213 | }\r | |
| 214 | }\r | |
| 215 | }\r | |
| 216 | else if(Dest != NULL) {\r | |
| 217 | *Dest = 0;\r | |
| 218 | }\r | |
| a7a8363d | 219 | }\r |
| 220 | if((NumConv > 0) && (Dest != NULL)) {\r | |
| 221 | Dest[0] = pS->D[0];\r | |
| 222 | if(NumConv == 4) {\r | |
| 223 | Dest[1] = pS->D[1];\r | |
| 224 | }\r | |
| 225 | }\r | |
| 226 | return NumConv;\r | |
| 227 | }\r | |
| 228 | \r | |
| c42c9cac | 229 | /* Determine the number of bytes needed to represent a Wide character\r |
| 230 | as a MBCS character.\r | |
| 231 | \r | |
| 232 | A single wide character may convert into a one, two, three, or four byte\r | |
| 233 | narrow (MBCS or UTF-8) character. The number of MBCS bytes can be determined\r | |
| 234 | as follows.\r | |
| 235 | \r | |
| 236 | If WCS char < 0x00000080 One Byte\r | |
| 237 | Else if WCS char < 0x0000D800 Two Bytes\r | |
| 238 | Else Three Bytes\r | |
| 239 | \r | |
| 240 | Since UEFI only supports the Unicode Base Multilingual Plane (BMP),\r | |
| 241 | Four-byte characters are not supported.\r | |
| 242 | \r | |
| 243 | @param[in] InCh Wide character to test.\r | |
| 244 | \r | |
| 245 | @retval -1 Improperly formed character\r | |
| 246 | @retval 0 InCh is 0x0000\r | |
| 247 | @retval >0 Number of bytes needed for the MBCS character\r | |
| 248 | */\r | |
| 249 | int\r | |
| 250 | EFIAPI\r | |
| 251 | OneWcToMcLen(const wchar_t InCh)\r | |
| 252 | {\r | |
| 253 | ssize_t NumBytes;\r | |
| 254 | \r | |
| 255 | if(InCh == 0) { // Is this a NUL, 0x0000 ?\r | |
| 256 | NumBytes = 0;\r | |
| 257 | }\r | |
| 258 | else if(InCh < 0x0080) { // Is this a 1-byte character?\r | |
| 259 | NumBytes = 1;\r | |
| 260 | }\r | |
| 261 | else if(InCh < 0x0800) { // Is this a 2-byte character?\r | |
| 262 | NumBytes = 2;\r | |
| 263 | }\r | |
| 264 | else if((InCh >= 0xD800) && (InCh < 0xE000)) { // Is this a surrogate?\r | |
| 265 | NumBytes = -1;\r | |
| 266 | }\r | |
| 267 | else {\r | |
| 268 | NumBytes = 3; // Otherwise, it must be a 3-byte character.\r | |
| 269 | }\r | |
| 270 | return (int)NumBytes; // Return extimate of required bytes.\r | |
| 271 | }\r | |
| 272 | \r | |
| 273 | /* Determine the number of bytes needed to represent a Wide character string\r | |
| 274 | as a MBCS string of given maximum length. Will optionally return the number\r | |
| 275 | of wide characters that would be consumed.\r | |
| 276 | \r | |
| 277 | A single wide character may convert into a one, two, three, or four byte\r | |
| 278 | narrow (MBCS or UTF-8) character. The number of MBCS bytes can be determined\r | |
| 279 | as follows.\r | |
| 280 | \r | |
| 281 | If WCS char < 0x00000080 One Byte\r | |
| 282 | Else if WCS char < 0x00000800 Two Bytes\r | |
| 283 | Else if WCS char < 0x00010000 Three Bytes\r | |
| 284 | Else Four Bytes\r | |
| 285 | \r | |
| 286 | Since UEFI only supports the Unicode Base Multilingual Plane (BMP),\r | |
| 287 | Four-byte characters should not be encountered.\r | |
| 288 | \r | |
| 289 | @param[in] Src Pointer to a wide character string.\r | |
| 290 | @param[in] Limit Maximum number of bytes the converted string may occupy.\r | |
| 450ea6d5 DM |
291 | @param[out] NumChar Pointer to where to store the number of wide characters\r |
| 292 | consumed, or NULL.\r | |
| c42c9cac | 293 | \r |
| 294 | @return The number of bytes required to convert Src to MBCS,\r | |
| 295 | not including the terminating NUL. If NumChar is not NULL, the number\r | |
| 296 | of characters represented by the return value will be written to\r | |
| 297 | where it points.\r | |
| 298 | */\r | |
| 299 | size_t\r | |
| 300 | EFIAPI\r | |
| 301 | EstimateWtoM(const wchar_t * Src, size_t Limit, size_t *NumChar)\r | |
| 302 | {\r | |
| 303 | ssize_t Estimate;\r | |
| 304 | size_t CharCount;\r | |
| 305 | ssize_t NumBytes;\r | |
| 306 | wchar_t EChar;\r | |
| 307 | \r | |
| 308 | Estimate = 0;\r | |
| 309 | CharCount = 0;\r | |
| 310 | EChar = *Src++; // Get the initial character and point to next\r | |
| 311 | while(((NumBytes = OneWcToMcLen(EChar)) > 0) &&\r | |
| 312 | ((size_t)(Estimate + NumBytes) < Limit))\r | |
| 313 | { // Until one of the source characters is NUL\r | |
| 314 | ++CharCount; // Count this character.\r | |
| 315 | Estimate += NumBytes; // Count the Bytes for this character\r | |
| 316 | EChar = *Src++; // Get the next source character and point to the next.\r | |
| 317 | }\r | |
| 318 | if(NumChar != NULL) {\r | |
| 319 | *NumChar = CharCount;\r | |
| 320 | }\r | |
| 321 | return (size_t)Estimate; // Return esimate of required bytes.\r | |
| 322 | }\r | |
| 323 | \r | |
| 324 | /* Determine the number of characters in a MBCS string.\r | |
| 325 | MBCS characters are one to four bytes long. By examining the first byte\r | |
| 326 | of a MBCS character, one can determine the number of bytes comprising the\r | |
| 327 | character.\r | |
| 328 | \r | |
| 329 | 0x00 - 0x7F One\r | |
| 330 | 0xC0 - 0xDF Two\r | |
| 331 | 0xE0 - 0xEF Three\r | |
| 332 | 0xF0 - 0xF7 Four\r | |
| 333 | \r | |
| 334 | Since UEFI only supports the Unicode Base Multilingual Plane (BMP),\r | |
| 335 | Four-byte characters should not be encountered.\r | |
| 336 | \r | |
| 337 | @param[in] Src The string to examine\r | |
| 338 | \r | |
| 339 | @return The number of characters represented by the MBCS string.\r | |
| 340 | **/\r | |
| 341 | size_t\r | |
| 342 | EFIAPI\r | |
| 343 | CountMbcsChars(const char *Src)\r | |
| 344 | {\r | |
| 345 | size_t Count;\r | |
| 346 | char EChar;\r | |
| 347 | \r | |
| 348 | Count = 0;\r | |
| 349 | EChar = *Src++;\r | |
| 350 | while(EChar != 0) {\r | |
| 351 | if(EChar < 0x80) {\r | |
| 352 | ++Count;\r | |
| 353 | }\r | |
| 354 | else if(EChar < 0xE0) {\r | |
| 355 | Count += 2;\r | |
| 356 | ++Src;\r | |
| 357 | }\r | |
| 358 | else if(EChar < 0xF0) {\r | |
| 359 | Count += 3;\r | |
| 360 | Src += 2;\r | |
| 361 | }\r | |
| 362 | else {\r | |
| 363 | // Ill-formed character\r | |
| 364 | break;\r | |
| 365 | }\r | |
| 366 | }\r | |
| 367 | return Count;\r | |
| 368 | }\r | |
| 369 | \r | |
| 370 | /** Convert a wide character (UTF16) into a multibyte character (UTF8)\r | |
| 371 | \r | |
| 372 | Converts a wide character into a corresponding multibyte character that\r | |
| 373 | begins in the conversion state described by the object pointed to by ps.\r | |
| 374 | If dst is not a null pointer, the converted character is then stored into\r | |
| 375 | the array pointed to by dst.\r | |
| 376 | \r | |
| 377 | It is the caller's responsibility to ensure that Dest is large enough to\r | |
| 378 | hold the resulting MBCS sequence.\r | |
| a7a8363d | 379 | \r |
| 380 | @param s Pointer to the wide-character string to convert\r | |
| c42c9cac | 381 | @param Dest Pointer to the buffer in which to place the converted sequence, or NULL.\r |
| a7a8363d | 382 | \r |
| c42c9cac | 383 | @retval -1 An error occurred. The error reason is in errno.\r |
| 384 | @retval >=0 The number of bytes stored into Dest.\r | |
| a7a8363d | 385 | **/\r |
| 386 | ssize_t\r | |
| c42c9cac | 387 | EncodeUtf8(char *Dest, wchar_t ch)\r |
| a7a8363d | 388 | {\r |
| 389 | char *p; /* next free byte in build buffer */\r | |
| a7a8363d | 390 | int NumInBuff; // number of bytes in Buff\r |
| 391 | char Buff[4]; // Buffer into which each character is built\r | |
| 392 | \r | |
| a7a8363d | 393 | p = Buff;\r |
| 394 | \r | |
| c42c9cac | 395 | NumInBuff = 0;\r |
| 396 | if (ch < 0x80) {\r | |
| 397 | /* Encode ASCII -- One Byte */\r | |
| 398 | *p++ = (char) ch;\r | |
| 399 | NumInBuff = 1;\r | |
| 400 | }\r | |
| 401 | else if (ch < 0x0800) {\r | |
| 402 | /* Encode Latin-1 -- Two Byte */\r | |
| 403 | *p++ = (char)(0xc0 | (ch >> 6));\r | |
| 404 | *p++ = (char)(0x80 | (ch & 0x3f));\r | |
| 405 | NumInBuff = 2;\r | |
| 406 | }\r | |
| 407 | else {\r | |
| a7a8363d | 408 | /* Encode UCS2 Unicode ordinals -- Three Byte */\r |
| c42c9cac | 409 | /* Special case: check for surrogate -- Shouldn't happen in UEFI */\r |
| 410 | if (0xD800 <= ch && ch < 0xE000) {\r | |
| 411 | errno = EILSEQ;\r | |
| 412 | return -1;\r | |
| a7a8363d | 413 | }\r |
| c42c9cac | 414 | else {\r |
| a7a8363d | 415 | *p++ = (char)(0xe0 | (ch >> 12));\r |
| 416 | *p++ = (char)(0x80 | ((ch >> 6) & 0x3f));\r | |
| 417 | *p++ = (char)(0x80 | (ch & 0x3f));\r | |
| c42c9cac | 418 | NumInBuff = 3;\r |
| a7a8363d | 419 | }\r |
| a7a8363d | 420 | }\r |
| c42c9cac | 421 | /* At this point, Buff holds the converted character which is NumInBuff bytes long.\r |
| 422 | NumInBuff is the value 1, 2, 3, or 4\r | |
| 423 | */\r | |
| 424 | if(Dest != NULL) { // Save character if Dest is not NULL\r | |
| 425 | memcpy(Dest, Buff, NumInBuff);\r | |
| a7a8363d | 426 | }\r |
| c42c9cac | 427 | return NumInBuff; // Tell the caller\r |
| a7a8363d | 428 | }\r |
| 429 | \r | |
| 430 | // ######################## Narrow to Wide Conversions #######################\r | |
| 431 | \r | |
| 432 | /** If ps is not a null pointer, the mbsinit function determines whether the\r | |
| 433 | pointed-to mbstate_t object describes an initial conversion state.\r | |
| 434 | \r | |
| c42c9cac | 435 | @param[in] ps Pointer to the conversion state object to test.\r |
| 436 | \r | |
| a7a8363d | 437 | @return The mbsinit function returns nonzero if ps is a null pointer\r |
| 438 | or if the pointed-to object describes an initial conversion\r | |
| 439 | state; otherwise, it returns zero.\r | |
| 440 | \r | |
| 441 | Declared in: wchar.h\r | |
| 442 | **/\r | |
| 443 | int\r | |
| 444 | mbsinit(const mbstate_t *ps)\r | |
| 445 | {\r | |
| 446 | if((ps == NULL) || (ps->A == 0)) {\r | |
| 447 | return 1;\r | |
| 448 | }\r | |
| 449 | return 0;\r | |
| 450 | }\r | |
| 451 | \r | |
| 452 | /** The mbrlen function is equivalent to the call:<BR>\r | |
| 453 | @verbatim\r | |
| 454 | mbrtowc(NULL, s, n, ps != NULL ? ps : &internal)\r | |
| 455 | @endverbatim\r | |
| 456 | where internal is the mbstate_t object for the mbrlen function, except that\r | |
| 457 | the expression designated by ps is evaluated only once.\r | |
| 458 | \r | |
| c42c9cac | 459 | @param[in] s Pointer to a multibyte character sequence.\r |
| 460 | @param[in] n Maximum number of bytes to examine.\r | |
| 461 | @param[in] pS Pointer to the conversion state object.\r | |
| 462 | \r | |
| 463 | @retval 0 The next n or fewer characters complete a NUL.\r | |
| 464 | @retval 1..n The number of bytes that complete the multibyte character.\r | |
| 465 | @retval -2 The next n bytes contribute to an incomplete (but potentially valid) multibyte character.\r | |
| 466 | @retval -1 An encoding error occurred.\r | |
| a7a8363d | 467 | \r |
| 468 | Declared in: wchar.h\r | |
| 469 | **/\r | |
| 470 | size_t\r | |
| 471 | mbrlen(\r | |
| 472 | const char *s,\r | |
| 473 | size_t n,\r | |
| c42c9cac | 474 | mbstate_t *pS\r |
| a7a8363d | 475 | )\r |
| 476 | {\r | |
| c42c9cac | 477 | return mbrtowc(NULL, s, n, pS);\r |
| a7a8363d | 478 | }\r |
| 479 | \r | |
| 480 | /** Determine the number of bytes comprising a multibyte character.\r | |
| 481 | \r | |
| 482 | If S is not a null pointer, the mblen function determines the number of bytes\r | |
| 483 | contained in the multibyte character pointed to by S. Except that the\r | |
| 484 | conversion state of the mbtowc function is not affected, it is equivalent to\r | |
| 485 | mbtowc((wchar_t *)0, S, N);\r | |
| 486 | \r | |
| 487 | @param[in] S NULL to query whether multibyte characters have\r | |
| 488 | state-dependent encodings. Otherwise, points to a\r | |
| 489 | multibyte character.\r | |
| 490 | @param[in] N The maximum number of bytes in a multibyte character.\r | |
| 491 | \r | |
| 492 | @return If S is a null pointer, the mblen function returns a nonzero or\r | |
| 493 | zero value, if multibyte character encodings, respectively, do\r | |
| 494 | or do not have state-dependent encodings. If S is not a null\r | |
| 495 | pointer, the mblen function either returns 0 (if S points to the\r | |
| 496 | null character), or returns the number of bytes that are contained\r | |
| 497 | in the multibyte character (if the next N or fewer bytes form a\r | |
| 498 | valid multibyte character), or returns -1 (if they do not form a\r | |
| 499 | valid multibyte character).\r | |
| 500 | \r | |
| 501 | Declared in: stdlib.h\r | |
| 502 | **/\r | |
| 503 | int\r | |
| 504 | mblen(\r | |
| 505 | const char *s,\r | |
| 506 | size_t n\r | |
| 507 | )\r | |
| 508 | {\r | |
| 509 | return (int)mbrlen(s, n, NULL);\r | |
| 510 | }\r | |
| 511 | \r | |
| 512 | /**\r | |
| 513 | If S is a null pointer, the mbrtowc function is equivalent to the call:<BR>\r | |
| 514 | @verbatim\r | |
| 515 | mbrtowc(NULL, "", 1, ps)\r | |
| 516 | @endverbatim\r | |
| 517 | \r | |
| 518 | In this case, the values of the parameters pwc and n are ignored.\r | |
| 519 | \r | |
| 520 | If S is not a null pointer, the mbrtowc function inspects at most n bytes beginning with\r | |
| 521 | the byte pointed to by S to determine the number of bytes needed to complete the next\r | |
| 522 | multibyte character (including any shift sequences). If the function determines that the\r | |
| 523 | next multibyte character is complete and valid, it determines the value of the\r | |
| 524 | corresponding wide character and then, if pwc is not a null pointer, stores that value in\r | |
| 525 | the object pointed to by pwc. If the corresponding wide character is the null wide\r | |
| 526 | character, the resulting state described is the initial conversion state.\r | |
| 527 | \r | |
| c42c9cac | 528 | @param[out] pwc Pointer to where the resulting wide character is to be stored.\r |
| 529 | @param[in] s Pointer to a multibyte character "string".\r | |
| 530 | @param[in] n The maximum number of bytes to inspect.\r | |
| 531 | @param[in] ps Pointer to a conversion state object.\r | |
| 532 | \r | |
| a7a8363d | 533 | @retval 0 if the next n or fewer bytes complete the multibyte\r |
| 534 | character that corresponds to the null wide\r | |
| 535 | character (which is the value stored).\r | |
| 536 | @retval between_1_and_n_inclusive if the next n or fewer bytes complete\r | |
| 537 | a valid multibyte character (which is the value\r | |
| 538 | stored); the value returned is the number of bytes\r | |
| 539 | that complete the multibyte character.\r | |
| 540 | @retval (size_t)(-2) if the next n bytes contribute to an incomplete\r | |
| 541 | (but potentially valid) multibyte character, and\r | |
| 542 | all n bytes have been processed (no value is stored).\r | |
| 543 | @retval (size_t)(-1) if an encoding error occurs, in which case the next\r | |
| 544 | n or fewer bytes do not contribute to a complete and\r | |
| 545 | valid multibyte character (no value is stored); the\r | |
| 546 | value of the macro EILSEQ is stored in errno, and\r | |
| 547 | the conversion state is unspecified.\r | |
| 548 | \r | |
| 549 | Declared in: wchar.h\r | |
| 550 | **/\r | |
| 551 | size_t\r | |
| 552 | mbrtowc(\r | |
| 553 | wchar_t *pwc,\r | |
| 554 | const char *s,\r | |
| 555 | size_t n,\r | |
| 556 | mbstate_t *ps\r | |
| 557 | )\r | |
| 558 | {\r | |
| 559 | int RetVal;\r | |
| 560 | \r | |
| 561 | RetVal = DecodeOneStateful(pwc, s, (ssize_t)n, ps);\r | |
| 562 | return (size_t)RetVal;\r | |
| 563 | }\r | |
| 564 | \r | |
| 565 | /** Convert a multibyte character into a wide character.\r | |
| 566 | \r | |
| 567 | If S is not a null pointer, the mbtowc function inspects at most N bytes\r | |
| 568 | beginning with the byte pointed to by S to determine the number of bytes\r | |
| 569 | needed to complete the next multibyte character (including any shift\r | |
| 570 | sequences). If the function determines that the next multibyte character\r | |
| 571 | is complete and valid, it determines the value of the corresponding wide\r | |
| 572 | character and then, if Pwc is not a null pointer, stores that value in\r | |
| 573 | the object pointed to by Pwc. If the corresponding wide character is the\r | |
| 574 | null wide character, the function is left in the initial conversion state.\r | |
| 575 | \r | |
| 576 | @param[out] Pwc Pointer to a wide-character object to receive the converted character.\r | |
| 577 | @param[in] S Pointer to a multibyte character to convert.\r | |
| 578 | @param[in] N Maximum number of bytes in a multibyte character.\r | |
| 579 | \r | |
| 580 | @return If S is a null pointer, the mbtowc function returns a nonzero or\r | |
| 581 | zero value, if multibyte character encodings, respectively, do\r | |
| 582 | or do not have state-dependent encodings. If S is not a null\r | |
| 583 | pointer, the mbtowc function either returns 0 (if S points to\r | |
| 584 | the null character), or returns the number of bytes that are\r | |
| 585 | contained in the converted multibyte character (if the next N or\r | |
| 586 | fewer bytes form a valid multibyte character), or returns -1\r | |
| 587 | (if they do not form a valid multibyte character).\r | |
| 588 | \r | |
| 589 | In no case will the value returned be greater than N or the value\r | |
| 590 | of the MB_CUR_MAX macro.\r | |
| 591 | \r | |
| 592 | Declared in: stdlib.h\r | |
| 593 | **/\r | |
| 594 | int\r | |
| 595 | mbtowc(\r | |
| 596 | wchar_t *pwc,\r | |
| 597 | const char *s,\r | |
| 598 | size_t n\r | |
| 599 | )\r | |
| 600 | {\r | |
| 601 | return (int)mbrtowc(pwc, s, n, NULL);\r | |
| 602 | }\r | |
| 603 | \r | |
| 604 | /**\r | |
| 605 | The mbsrtowcs function converts a sequence of multibyte characters that begins in the\r | |
| 606 | conversion state described by the object pointed to by ps, from the array indirectly\r | |
| 607 | pointed to by src into a sequence of corresponding wide characters. If dst is not a null\r | |
| 608 | pointer, the converted characters are stored into the array pointed to by dst. Conversion\r | |
| 609 | continues up to and including a terminating null character, which is also stored.\r | |
| 610 | Conversion stops earlier in two cases: when a sequence of bytes is encountered that does\r | |
| 611 | not form a valid multibyte character, or (if dst is not a null pointer) when len wide\r | |
| 612 | characters have been stored into the array pointed to by dst. Each conversion takes\r | |
| 613 | place as if by a call to the mbrtowc function.\r | |
| 614 | \r | |
| 615 | If dst is not a null pointer, the pointer object pointed to by src is assigned either a null\r | |
| 616 | pointer (if conversion stopped due to reaching a terminating null character) or the address\r | |
| 617 | just past the last multibyte character converted (if any). If conversion stopped due to\r | |
| 618 | reaching a terminating null character and if dst is not a null pointer, the resulting state\r | |
| 619 | described is the initial conversion state.\r | |
| 620 | \r | |
| c42c9cac | 621 | @param[out] dst Pointer to where the resulting wide character sequence is stored.\r |
| 622 | @param[in] src Pointer to a pointer to the multibyte character sequence to convert.\r | |
| 623 | @param[in] len Maximum number of wide characters to be stored into dst.\r | |
| 624 | @param[in] ps Pointer to a conversion state object.\r | |
| 625 | \r | |
| a7a8363d | 626 | @return If the input conversion encounters a sequence of bytes that do\r |
| 627 | not form a valid multibyte character, an encoding error occurs:\r | |
| 628 | the mbsrtowcs function stores the value of the macro EILSEQ in\r | |
| 629 | errno and returns (size_t)(-1); the conversion state is\r | |
| 630 | unspecified. Otherwise, it returns the number of multibyte\r | |
| 631 | characters successfully converted, not including the terminating\r | |
| 632 | null character (if any).\r | |
| 633 | \r | |
| 634 | Declared in: wchar.h\r | |
| 635 | **/\r | |
| 636 | size_t\r | |
| 637 | mbsrtowcs(\r | |
| 638 | wchar_t *dst,\r | |
| 639 | const char **src,\r | |
| 640 | size_t len,\r | |
| 641 | mbstate_t *ps\r | |
| 642 | )\r | |
| 643 | {\r | |
| 644 | int x;\r | |
| 645 | size_t RetVal = 0;\r | |
| 646 | const char *MySrc;\r | |
| 647 | \r | |
| 5a522112 | 648 | if((src == NULL) || (*src == NULL)) {\r |
| a7a8363d | 649 | return 0;\r |
| 650 | }\r | |
| 651 | \r | |
| 652 | MySrc = *src;\r | |
| 653 | for(x = 1 ; (len != 0) && (x > 0); --len) {\r | |
| 654 | x = DecodeOneStateful(dst, MySrc, MB_LEN_MAX, ps);\r | |
| 655 | switch(x) {\r | |
| 656 | case -2: // Incomplete character\r | |
| 657 | case -1: // Encoding error\r | |
| 658 | RetVal = (size_t)x;\r | |
| 659 | break;\r | |
| 660 | case 0: // Encountered NUL character: done.\r | |
| 661 | if(dst != NULL) {\r | |
| 662 | *dst = 0;\r | |
| 663 | *src = NULL;\r | |
| 664 | }\r | |
| 665 | break;\r | |
| 666 | default: // Successfully decoded a character, continue with next\r | |
| 667 | MySrc += x;\r | |
| 668 | if(dst != NULL) {\r | |
| 669 | ++dst;\r | |
| 670 | if(x == 4) {\r | |
| 671 | ++dst;\r | |
| 672 | }\r | |
| 673 | *src = MySrc;\r | |
| 674 | }\r | |
| 675 | ++RetVal;\r | |
| 676 | break;\r | |
| 677 | }\r | |
| 678 | }\r | |
| 679 | return RetVal;\r | |
| 680 | }\r | |
| 681 | \r | |
| 682 | /** Convert a multibyte character string into a wide-character string.\r | |
| 683 | \r | |
| 684 | The mbstowcs function converts a sequence of multibyte characters that\r | |
| 685 | begins in the initial shift state from the array pointed to by Src into\r | |
| 686 | a sequence of corresponding wide characters and stores not more than limit\r | |
| 687 | wide characters into the array pointed to by Dest. No multibyte\r | |
| 688 | characters that follow a null character (which is converted into a null\r | |
| 689 | wide character) will be examined or converted. Each multibyte character\r | |
| 690 | is converted as if by a call to the mbtowc function, except that the\r | |
| 691 | conversion state of the mbtowc function is not affected.\r | |
| 692 | \r | |
| 693 | No more than Limit elements will be modified in the array pointed to by Dest.\r | |
| 694 | If copying takes place between objects that overlap,\r | |
| 695 | the behavior is undefined.\r | |
| 696 | \r | |
| 697 | @param[out] Dest Pointer to the array to receive the converted string.\r | |
| 698 | @param[in] Src Pointer to the string to be converted.\r | |
| 699 | @param[in] Limit Maximum number of elements to be written to Dest.\r | |
| 700 | \r | |
| 701 | @return If an invalid multibyte character is encountered, the mbstowcs\r | |
| 702 | function returns (size_t)(-1). Otherwise, the mbstowcs function\r | |
| 703 | returns the number of array elements modified, not including a\r | |
| 704 | terminating null wide character, if any.\r | |
| 705 | \r | |
| 706 | Declared in: stdlib.h\r | |
| 707 | **/\r | |
| 708 | size_t\r | |
| 709 | mbstowcs(\r | |
| c42c9cac | 710 | wchar_t *Dest,\r |
| 711 | const char *Src,\r | |
| 712 | size_t Limit\r | |
| a7a8363d | 713 | )\r |
| 714 | {\r | |
| 715 | \r | |
| c42c9cac | 716 | /* Dest may be NULL */\r |
| 717 | /* Src may be NULL */\r | |
| a7a8363d | 718 | \r |
| c42c9cac | 719 | return mbsrtowcs(Dest, &Src, Limit, NULL);\r |
| a7a8363d | 720 | }\r |
| 721 | \r | |
| 722 | /** The btowc function determines whether C constitutes a valid single-byte\r | |
| 723 | character in the initial shift state.\r | |
| 724 | \r | |
| c42c9cac | 725 | @param[in] C A narrow character to test or convert to wide.\r |
| 726 | \r | |
| a7a8363d | 727 | @return The btowc function returns WEOF if c has the value EOF or if\r |
| 728 | (unsigned char)C does not constitute a valid single-byte\r | |
| 729 | character in the initial shift state. Otherwise, it returns the\r | |
| 730 | wide character representation of that character.\r | |
| 731 | \r | |
| 732 | Declared in: wchar.h\r | |
| 733 | **/\r | |
| 734 | wint_t\r | |
| 735 | btowc(int c)\r | |
| 736 | {\r | |
| 737 | int x;\r | |
| 738 | wchar_t Dest;\r | |
| 739 | wint_t RetVal = WEOF;\r | |
| 740 | \r | |
| 741 | if (c == EOF)\r | |
| 742 | return WEOF;\r | |
| 743 | x = DecodeOneStateful(&Dest, (const char *)&c, 1, NULL);\r | |
| 744 | if(x == 0) {\r | |
| 745 | RetVal = 0;\r | |
| 746 | }\r | |
| 747 | else if(x == 1) {\r | |
| 748 | RetVal = (wint_t)Dest;\r | |
| 749 | }\r | |
| 750 | return RetVal;\r | |
| 751 | }\r | |
| 752 | \r | |
| 753 | // ######################## Wide to Narrow Conversions #######################\r | |
| 754 | \r | |
| 755 | /**\r | |
| 756 | If S is a null pointer, the wcrtomb function is equivalent to the call:<BR>\r | |
| 757 | @verbatim\r | |
| 758 | wcrtomb(buf, L'\0', ps)\r | |
| 759 | @endverbatim\r | |
| 760 | where buf is an internal buffer.\r | |
| 761 | \r | |
| 762 | If S is not a null pointer, the wcrtomb function determines the number of bytes needed\r | |
| 763 | to represent the multibyte character that corresponds to the wide character given by wc\r | |
| 764 | (including any shift sequences), and stores the multibyte character representation in the\r | |
| 765 | array whose first element is pointed to by S. At most MB_CUR_MAX bytes are stored. If\r | |
| 766 | wc is a null wide character, a null byte is stored, preceded by any shift sequence needed\r | |
| 767 | to restore the initial shift state; the resulting state described is the initial conversion state.\r | |
| 768 | \r | |
| c42c9cac | 769 | @param[out] Dest Pointer to the location in which to store the resulting\r |
| 770 | multibyte character. Otherwise, NULL to reset the\r | |
| 771 | conversion state.\r | |
| 772 | @param[in] wchar The wide character to convert.\r | |
| 773 | @param[in,out] pS Pointer to a conversion state object, or NULL.\r | |
| 774 | \r | |
| a7a8363d | 775 | @return The wcrtomb function returns the number of bytes stored in the\r |
| 776 | array object (including any shift sequences). When wc is not a\r | |
| 777 | valid wide character, an encoding error occurs: the function\r | |
| 778 | stores the value of the macro EILSEQ in errno and\r | |
| 779 | returns (size_t)(-1); the conversion state is unspecified.\r | |
| 780 | \r | |
| 781 | Declared in: wchar.h\r | |
| 782 | **/\r | |
| 783 | size_t\r | |
| 784 | wcrtomb(\r | |
| c42c9cac | 785 | char *Dest,\r |
| a7a8363d | 786 | wchar_t wchar,\r |
| c42c9cac | 787 | mbstate_t *pS\r |
| a7a8363d | 788 | )\r |
| 789 | {\r | |
| 790 | size_t RetVal;\r | |
| 791 | \r | |
| c42c9cac | 792 | /* Dest may be NULL */\r |
| 793 | if (Dest == NULL) {\r | |
| a7a8363d | 794 | RetVal = 1;\r |
| 795 | }\r | |
| 796 | else {\r | |
| 797 | if (wchar == L'\0') {\r | |
| c42c9cac | 798 | *Dest = '\0';\r |
| a7a8363d | 799 | RetVal = 1;\r |
| 800 | }\r | |
| 801 | else {\r | |
| c42c9cac | 802 | RetVal = EncodeUtf8(Dest, wchar);\r |
| a7a8363d | 803 | }\r |
| 804 | }\r | |
| c42c9cac | 805 | if(pS == NULL) {\r |
| 806 | pS = &LocalConvState;\r | |
| 807 | }\r | |
| 808 | pS->A = 0; // Set ps to the initial conversion state\r | |
| 809 | \r | |
| a7a8363d | 810 | return RetVal;\r |
| 811 | }\r | |
| 812 | \r | |
| 813 | /** Convert a wide character into a multibyte character.\r | |
| 814 | \r | |
| 815 | The wctomb function determines the number of bytes needed to represent the\r | |
| 816 | multibyte character corresponding to the wide character given by WC\r | |
| 817 | (including any shift sequences), and stores the multibyte character\r | |
| 818 | representation in the array whose first element is pointed to by S (if S is\r | |
| 819 | not a null pointer). At most MB_CUR_MAX characters are stored. If WC is a\r | |
| 820 | null wide character, a null byte is stored, preceded by any shift sequence\r | |
| 821 | needed to restore the initial shift state, and the function is left in the\r | |
| 822 | initial conversion state.\r | |
| 823 | \r | |
| 824 | @param[out] S Pointer to the object to receive the converted multibyte character.\r | |
| 825 | @param[in] WC Wide character to be converted.\r | |
| 826 | \r | |
| 827 | @return If S is a null pointer, the wctomb function returns a nonzero or\r | |
| 828 | zero value, if multibyte character encodings, respectively, do or\r | |
| 829 | do not have state-dependent encodings. If S is not a null pointer,\r | |
| 830 | the wctomb function returns -1 if the value of WC does not\r | |
| 831 | correspond to a valid multibyte character, or returns the number\r | |
| 832 | of bytes that are contained in the multibyte character\r | |
| 833 | corresponding to the value of WC.\r | |
| 834 | \r | |
| 835 | In no case will the value returned be greater than the value of\r | |
| 836 | the MB_CUR_MAX macro.\r | |
| 837 | \r | |
| 838 | Declared in: stdlib.h\r | |
| 839 | **/\r | |
| 840 | int\r | |
| 841 | wctomb(\r | |
| 842 | char *s,\r | |
| 843 | wchar_t wchar\r | |
| 844 | )\r | |
| 845 | {\r | |
| 846 | /*\r | |
| 847 | If s is NULL just return whether MB Characters have state\r | |
| 848 | dependent encodings -- they don't.\r | |
| 849 | */\r | |
| 850 | if (s == NULL)\r | |
| 851 | return 0;\r | |
| 852 | \r | |
| 853 | return (int)wcrtomb(s, wchar, NULL);\r | |
| 854 | }\r | |
| 855 | \r | |
| 856 | /** The wcsrtombs function converts a sequence of wide characters from the array\r | |
| 5a522112 | 857 | indirectly pointed to by Src into a sequence of corresponding multibyte\r |
| a7a8363d | 858 | characters that begins in the conversion state described by the object\r |
| 859 | pointed to by ps.\r | |
| 860 | \r | |
| c42c9cac | 861 | If Dest is not a null pointer, the converted characters are stored into the\r |
| 862 | array pointed to by Dest. Conversion continues up to and including a\r | |
| 863 | terminating null wide character, which is also stored. Conversion stops\r | |
| 864 | earlier in two cases: when a wide character is reached that does not\r | |
| 865 | correspond to a valid multibyte character, or (if Dest is not a null\r | |
| 866 | pointer) when the next multibyte character would exceed the limit of Limit\r | |
| 867 | total bytes to be stored into the array pointed to by Dest. Each conversion\r | |
| 868 | takes place as if by a call to the wcrtomb function.)\r | |
| 869 | \r | |
| 870 | If Dest is not a null pointer, the pointer object pointed to by Src is\r | |
| a7a8363d | 871 | assigned either a null pointer (if conversion stopped due to reaching\r |
| 872 | a terminating null wide character) or the address just past the last wide\r | |
| 873 | character converted (if any). If conversion stopped due to reaching a\r | |
| 874 | terminating null wide character, the resulting state described is the\r | |
| 875 | initial conversion state.\r | |
| 876 | \r | |
| c42c9cac | 877 | @param[in] Dest\r |
| 878 | @param[in,out] Src\r | |
| 879 | @param[in] Limit Max number of bytes to store in Dest.\r | |
| 880 | @param[in,out] ps\r | |
| 881 | \r | |
| a7a8363d | 882 | @return If conversion stops because a wide character is reached that\r |
| 883 | does not correspond to a valid multibyte character, an\r | |
| 884 | encoding error occurs: the wcsrtombs function stores the\r | |
| 885 | value of the macro EILSEQ in errno and returns (size_t)(-1);\r | |
| 886 | the conversion state is unspecified. Otherwise, it returns\r | |
| 887 | the number of bytes in the resulting multibyte character\r | |
| 888 | sequence, not including the terminating null character (if any).\r | |
| 889 | \r | |
| 890 | Declared in: wchar.h\r | |
| 891 | **/\r | |
| 892 | size_t\r | |
| 893 | wcsrtombs(\r | |
| c42c9cac | 894 | char *Dest,\r |
| 895 | const wchar_t **Src,\r | |
| 896 | size_t Limit,\r | |
| 897 | mbstate_t *ps\r | |
| a7a8363d | 898 | )\r |
| 899 | {\r | |
| c42c9cac | 900 | size_t NumStored;\r |
| 901 | ssize_t MaxBytes;\r | |
| 902 | int count;\r | |
| 903 | wchar_t InCh;\r | |
| a7a8363d | 904 | \r |
| c42c9cac | 905 | NumStored = 0;\r |
| 906 | MaxBytes = (ssize_t)Limit;\r | |
| 907 | \r | |
| 908 | /* Dest may be NULL */\r | |
| 909 | /* Src may be NULL */\r | |
| a7a8363d | 910 | /* ps appears to be unused */\r |
| 911 | \r | |
| c42c9cac | 912 | if (Src == NULL || *Src == NULL)\r |
| a7a8363d | 913 | return (0);\r |
| 914 | \r | |
| c42c9cac | 915 | if (Dest == NULL) {\r |
| 5a522112 | 916 | NumStored = EstimateWtoM(*Src, ASCII_STRING_MAX, NULL);\r |
| a7a8363d | 917 | }\r |
| c42c9cac | 918 | else {\r |
| 5a522112 | 919 | if((MaxBytes < 0) || (MaxBytes > ASCII_STRING_MAX)) {\r |
| 920 | MaxBytes = ASCII_STRING_MAX;\r | |
| 921 | }\r | |
| 922 | while ((MaxBytes > 0) && (OneWcToMcLen(InCh = *(*Src)++) <= MaxBytes)) {\r | |
| c42c9cac | 923 | if(InCh == 0) {\r |
| 924 | *Src = NULL;\r | |
| 5a522112 | 925 | *Dest = 0; // NUL terminate Dest string, but don't count the NUL\r |
| a7a8363d | 926 | break;\r |
| 927 | }\r | |
| c42c9cac | 928 | count = (int)wcrtomb(Dest, InCh, NULL);\r |
| 929 | if(count >= 0) {\r | |
| 930 | Dest += count;\r | |
| 931 | MaxBytes -= count;\r | |
| 932 | NumStored += count;\r | |
| 933 | }\r | |
| 934 | else {\r | |
| 935 | NumStored = (size_t)(-1);\r | |
| 936 | }\r | |
| 937 | }\r | |
| a7a8363d | 938 | }\r |
| 939 | \r | |
| c42c9cac | 940 | \r |
| 941 | return NumStored;\r | |
| a7a8363d | 942 | }\r |
| 943 | \r | |
| 944 | /** Convert a wide-character string into a multibyte character string.\r | |
| 945 | \r | |
| 946 | The wcstombs function converts a sequence of wide characters from the\r | |
| 947 | array pointed to by Src into a sequence of corresponding multibyte\r | |
| 948 | characters that begins in the initial shift state, and stores these\r | |
| 949 | multibyte characters into the array pointed to by Dest, stopping if a\r | |
| 950 | multibyte character would exceed the limit of Limit total bytes or if a\r | |
| 951 | null character is stored. Each wide character is converted as if by\r | |
| 952 | a call to the wctomb function, except that the conversion state of\r | |
| 953 | the wctomb function is not affected.\r | |
| 954 | \r | |
| 955 | No more than Limit bytes will be modified in the array pointed to by Dest.\r | |
| 956 | If copying takes place between objects that overlap,\r | |
| 957 | the behavior is undefined.\r | |
| 958 | \r | |
| 959 | @param[out] Dest Pointer to the array to receive the converted string.\r | |
| 960 | @param[in] Src Pointer to the string to be converted.\r | |
| 961 | @param[in] Limit Maximum number of elements to be written to Dest.\r | |
| 962 | \r | |
| 963 | @return If a wide character is encountered that does not correspond to a\r | |
| 964 | valid multibyte character, the wcstombs function returns\r | |
| 965 | (size_t)(-1). Otherwise, the wcstombs function returns the number\r | |
| 450ea6d5 DM |
966 | of bytes in the resulting multibyte character sequence,\r |
| 967 | not including the terminating null character (if any).\r | |
| a7a8363d | 968 | \r |
| 969 | Declared in: stdlib.h\r | |
| 970 | **/\r | |
| 971 | size_t\r | |
| 972 | wcstombs(\r | |
| c42c9cac | 973 | char *Dest,\r |
| 974 | const wchar_t *Src,\r | |
| 975 | size_t Limit\r | |
| a7a8363d | 976 | )\r |
| 977 | {\r | |
| c42c9cac | 978 | /* Dest may be NULL */\r |
| 979 | return wcsrtombs(Dest, &Src, Limit, NULL);\r | |
| a7a8363d | 980 | }\r |
| 981 | \r | |
| 982 | /** The wctob function determines whether C corresponds to a member of the extended\r | |
| 983 | character set whose multibyte character representation is a single byte when in the initial\r | |
| 984 | shift state.\r | |
| 985 | \r | |
| c42c9cac | 986 | wctob needs to be consistent with wcrtomb.\r |
| 987 | If wcrtomb says that a character is representable in 1 byte,\r | |
| 988 | then wctob needs to also represent the character as 1 byte.\r | |
| 989 | \r | |
| a7a8363d | 990 | @return The wctob function returns EOF if C does not correspond to a multibyte\r |
| 991 | character with length one in the initial shift state. Otherwise, it\r | |
| 992 | returns the single-byte representation of that character as an\r | |
| 993 | unsigned char converted to an int.\r | |
| 994 | \r | |
| 995 | Declared in: wchar.h\r | |
| 996 | **/\r | |
| 997 | int\r | |
| 998 | wctob(wint_t c)\r | |
| 999 | {\r | |
| c42c9cac | 1000 | int RetVal;\r |
| 1001 | \r | |
| 1002 | RetVal = EOF;\r | |
| 1003 | if(c == 0) {\r | |
| 1004 | RetVal = 0;\r | |
| 1005 | }\r | |
| 1006 | else if (OneWcToMcLen((const wchar_t)c) == 1) {\r | |
| 1007 | RetVal = (int)(c & 0xFF);\r | |
| a7a8363d | 1008 | }\r |
| c42c9cac | 1009 | return RetVal;\r |
| a7a8363d | 1010 | }\r |