1 /* 2 * Unit tests for code page to/from unicode translations 3 * 4 * Copyright (c) 2002 Dmitry Timoshkov 5 * Copyright (c) 2008 Colin Finck 6 * 7 * This library is free software; you can redistribute it and/or 8 * modify it under the terms of the GNU Lesser General Public 9 * License as published by the Free Software Foundation; either 10 * version 2.1 of the License, or (at your option) any later version. 11 * 12 * This library is distributed in the hope that it will be useful, 13 * but WITHOUT ANY WARRANTY; without even the implied warranty of 14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 15 * Lesser General Public License for more details. 16 * 17 * You should have received a copy of the GNU Lesser General Public 18 * License along with this library; if not, write to the Free Software 19 * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA 20 */ 21 22 #include "precomp.h" 23 24 static const char foobarA[] = "foobar"; 25 static const WCHAR foobarW[] = {'f','o','o','b','a','r',0}; 26 27 static void test_destination_buffer(void) 28 { 29 LPSTR buffer; 30 INT maxsize; 31 INT needed; 32 INT len; 33 34 SetLastError(0xdeadbeef); 35 needed = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL); 36 ok( (needed > 0), "returned %d with %u (expected '> 0')\n", 37 needed, GetLastError()); 38 39 maxsize = needed*2; 40 buffer = HeapAlloc(GetProcessHeap(), 0, maxsize); 41 if (buffer == NULL) return; 42 43 maxsize--; 44 memset(buffer, 'x', maxsize); 45 buffer[maxsize] = '\0'; 46 SetLastError(0xdeadbeef); 47 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed+1, NULL, NULL); 48 ok( (len > 0), "returned %d with %u and '%s' (expected '> 0')\n", 49 len, GetLastError(), buffer); 50 51 memset(buffer, 'x', maxsize); 52 buffer[maxsize] = '\0'; 53 SetLastError(0xdeadbeef); 54 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed, NULL, NULL); 55 ok( (len > 0), "returned %d with %u and '%s' (expected '> 0')\n", 56 len, GetLastError(), buffer); 57 58 memset(buffer, 'x', maxsize); 59 buffer[maxsize] = '\0'; 60 SetLastError(0xdeadbeef); 61 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed-1, NULL, NULL); 62 ok( !len && (GetLastError() == ERROR_INSUFFICIENT_BUFFER), 63 "returned %d with %u and '%s' (expected '0' with " 64 "ERROR_INSUFFICIENT_BUFFER)\n", len, GetLastError(), buffer); 65 66 memset(buffer, 'x', maxsize); 67 buffer[maxsize] = '\0'; 68 SetLastError(0xdeadbeef); 69 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, 1, NULL, NULL); 70 ok( !len && (GetLastError() == ERROR_INSUFFICIENT_BUFFER), 71 "returned %d with %u and '%s' (expected '0' with " 72 "ERROR_INSUFFICIENT_BUFFER)\n", len, GetLastError(), buffer); 73 74 SetLastError(0xdeadbeef); 75 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, 0, NULL, NULL); 76 ok( (len > 0), "returned %d with %u (expected '> 0')\n", 77 len, GetLastError()); 78 79 SetLastError(0xdeadbeef); 80 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, needed, NULL, NULL); 81 ok( !len && (GetLastError() == ERROR_INVALID_PARAMETER), 82 "returned %d with %u (expected '0' with " 83 "ERROR_INVALID_PARAMETER)\n", len, GetLastError()); 84 85 HeapFree(GetProcessHeap(), 0, buffer); 86 } 87 88 89 static void test_null_source(void) 90 { 91 int len; 92 DWORD GLE; 93 94 SetLastError(0); 95 len = WideCharToMultiByte(CP_ACP, 0, NULL, 0, NULL, 0, NULL, NULL); 96 GLE = GetLastError(); 97 ok(!len && GLE == ERROR_INVALID_PARAMETER, 98 "WideCharToMultiByte returned %d with GLE=%u (expected 0 with ERROR_INVALID_PARAMETER)\n", 99 len, GLE); 100 101 SetLastError(0); 102 len = WideCharToMultiByte(CP_ACP, 0, NULL, -1, NULL, 0, NULL, NULL); 103 GLE = GetLastError(); 104 ok(!len && GLE == ERROR_INVALID_PARAMETER, 105 "WideCharToMultiByte returned %d with GLE=%u (expected 0 with ERROR_INVALID_PARAMETER)\n", 106 len, GLE); 107 } 108 109 static void test_negative_source_length(void) 110 { 111 int len; 112 char buf[10]; 113 WCHAR bufW[10]; 114 115 /* Test, whether any negative source length works as strlen() + 1 */ 116 SetLastError( 0xdeadbeef ); 117 memset(buf,'x',sizeof(buf)); 118 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -2002, buf, 10, NULL, NULL); 119 ok(len == 7 && GetLastError() == 0xdeadbeef, 120 "WideCharToMultiByte(-2002): len=%d error=%u\n", len, GetLastError()); 121 ok(!lstrcmpA(buf, "foobar"), 122 "WideCharToMultiByte(-2002): expected \"foobar\" got \"%s\"\n", buf); 123 124 SetLastError( 0xdeadbeef ); 125 memset(bufW,'x',sizeof(bufW)); 126 len = MultiByteToWideChar(CP_ACP, 0, "foobar", -2002, bufW, 10); 127 ok(len == 7 && !lstrcmpW(bufW, foobarW) && GetLastError() == 0xdeadbeef, 128 "MultiByteToWideChar(-2002): len=%d error=%u\n", len, GetLastError()); 129 130 SetLastError(0xdeadbeef); 131 memset(bufW, 'x', sizeof(bufW)); 132 len = MultiByteToWideChar(CP_ACP, 0, "foobar", -1, bufW, 6); 133 ok(len == 0 && GetLastError() == ERROR_INSUFFICIENT_BUFFER, 134 "MultiByteToWideChar(-1): len=%d error=%u\n", len, GetLastError()); 135 } 136 137 #define LONGBUFLEN 100000 138 static void test_negative_dest_length(void) 139 { 140 int len, i; 141 static WCHAR bufW[LONGBUFLEN]; 142 static char bufA[LONGBUFLEN]; 143 static WCHAR originalW[LONGBUFLEN]; 144 static char originalA[LONGBUFLEN]; 145 DWORD theError; 146 147 /* Test return on -1 dest length */ 148 SetLastError( 0xdeadbeef ); 149 memset(bufA,'x',sizeof(bufA)); 150 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, -1, NULL, NULL); 151 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, 152 "WideCharToMultiByte(destlen -1): len=%d error=%x\n", len, GetLastError()); 153 154 SetLastError( 0xdeadbeef ); 155 memset(bufW,'x',sizeof(bufW)); 156 len = MultiByteToWideChar(CP_ACP, 0, foobarA, -1, bufW, -1); 157 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, 158 "MultiByteToWideChar(destlen -1): len=%d error=%x\n", len, GetLastError()); 159 160 /* Test return on -1000 dest length */ 161 SetLastError( 0xdeadbeef ); 162 memset(bufA,'x',sizeof(bufA)); 163 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, -1000, NULL, NULL); 164 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, 165 "WideCharToMultiByte(destlen -1000): len=%d error=%x\n", len, GetLastError()); 166 167 SetLastError( 0xdeadbeef ); 168 memset(bufW,'x',sizeof(bufW)); 169 len = MultiByteToWideChar(CP_ACP, 0, foobarA, -1000, bufW, -1); 170 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, 171 "MultiByteToWideChar(destlen -1000): len=%d error=%x\n", len, GetLastError()); 172 173 /* Test return on INT_MAX dest length */ 174 SetLastError( 0xdeadbeef ); 175 memset(bufA,'x',sizeof(bufA)); 176 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, INT_MAX, NULL, NULL); 177 ok(len == 7 && !lstrcmpA(bufA, "foobar") && GetLastError() == 0xdeadbeef, 178 "WideCharToMultiByte(destlen INT_MAX): len=%d error=%x\n", len, GetLastError()); 179 180 /* Test return on INT_MAX dest length and very long input */ 181 SetLastError( 0xdeadbeef ); 182 memset(bufA,'x',sizeof(bufA)); 183 for (i=0; i < LONGBUFLEN - 1; i++) { 184 originalW[i] = 'Q'; 185 originalA[i] = 'Q'; 186 } 187 originalW[LONGBUFLEN-1] = 0; 188 originalA[LONGBUFLEN-1] = 0; 189 len = WideCharToMultiByte(CP_ACP, 0, originalW, -1, bufA, INT_MAX, NULL, NULL); 190 theError = GetLastError(); 191 ok(len == LONGBUFLEN && !lstrcmpA(bufA, originalA) && theError == 0xdeadbeef, 192 "WideCharToMultiByte(srclen %d, destlen INT_MAX): len %d error=%x\n", LONGBUFLEN, len, theError); 193 194 } 195 196 static void test_other_invalid_parameters(void) 197 { 198 char c_string[] = "Hello World"; 199 size_t c_string_len = sizeof(c_string); 200 WCHAR w_string[] = {'H','e','l','l','o',' ','W','o','r','l','d',0}; 201 size_t w_string_len = sizeof(w_string) / sizeof(WCHAR); 202 BOOL used; 203 INT len; 204 205 /* Unrecognized flag => ERROR_INVALID_FLAGS */ 206 SetLastError(0xdeadbeef); 207 len = WideCharToMultiByte(CP_ACP, 0x100, w_string, -1, c_string, c_string_len, NULL, NULL); 208 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%x\n", len, GetLastError()); 209 210 SetLastError(0xdeadbeef); 211 len = WideCharToMultiByte(CP_ACP, 0x800, w_string, -1, c_string, c_string_len, NULL, NULL); 212 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%x\n", len, GetLastError()); 213 214 SetLastError(0xdeadbeef); 215 len = MultiByteToWideChar(CP_ACP, 0x10, c_string, -1, w_string, w_string_len); 216 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%x\n", len, GetLastError()); 217 218 219 /* Unrecognized flag and invalid codepage => ERROR_INVALID_PARAMETER */ 220 SetLastError(0xdeadbeef); 221 len = WideCharToMultiByte(0xdeadbeef, 0x100, w_string, w_string_len, c_string, c_string_len, NULL, NULL); 222 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 223 224 SetLastError(0xdeadbeef); 225 len = MultiByteToWideChar(0xdeadbeef, 0x10, c_string, c_string_len, w_string, w_string_len); 226 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 227 228 229 /* Unrecognized flag and src is NULL => ERROR_INVALID_PARAMETER */ 230 SetLastError(0xdeadbeef); 231 len = WideCharToMultiByte(CP_ACP, 0x100, NULL, -1, c_string, c_string_len, NULL, NULL); 232 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 233 234 SetLastError(0xdeadbeef); 235 len = MultiByteToWideChar(CP_ACP, 0x10, NULL, -1, w_string, w_string_len); 236 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 237 238 239 /* srclen=0 => ERROR_INVALID_PARAMETER */ 240 SetLastError(0xdeadbeef); 241 len = WideCharToMultiByte(CP_ACP, 0, w_string, 0, c_string, c_string_len, NULL, NULL); 242 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 243 244 SetLastError(0xdeadbeef); 245 len = MultiByteToWideChar(CP_ACP, 0, c_string, 0, w_string, w_string_len); 246 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 247 248 249 /* dst=NULL but dstlen not 0 => ERROR_INVALID_PARAMETER */ 250 SetLastError(0xdeadbeef); 251 len = WideCharToMultiByte(CP_ACP, 0, w_string, w_string_len, NULL, c_string_len, NULL, NULL); 252 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 253 254 SetLastError(0xdeadbeef); 255 len = MultiByteToWideChar(CP_ACP, 0, c_string, c_string_len, NULL, w_string_len); 256 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 257 258 259 /* CP_UTF7, CP_UTF8, or CP_SYMBOL and defchar not NULL => ERROR_INVALID_PARAMETER */ 260 /* CP_SYMBOL's behavior here is undocumented */ 261 SetLastError(0xdeadbeef); 262 len = WideCharToMultiByte(CP_UTF7, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL); 263 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 264 265 SetLastError(0xdeadbeef); 266 len = WideCharToMultiByte(CP_UTF8, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL); 267 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 268 269 SetLastError(0xdeadbeef); 270 len = WideCharToMultiByte(CP_SYMBOL, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL); 271 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 272 273 274 /* CP_UTF7, CP_UTF8, or CP_SYMBOL and used not NULL => ERROR_INVALID_PARAMETER */ 275 /* CP_SYMBOL's behavior here is undocumented */ 276 SetLastError(0xdeadbeef); 277 len = WideCharToMultiByte(CP_UTF7, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used); 278 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 279 280 SetLastError(0xdeadbeef); 281 len = WideCharToMultiByte(CP_UTF8, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used); 282 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 283 284 SetLastError(0xdeadbeef); 285 len = WideCharToMultiByte(CP_SYMBOL, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used); 286 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 287 288 289 /* CP_UTF7, flags not 0 and used not NULL => ERROR_INVALID_PARAMETER */ 290 /* (tests precedence of ERROR_INVALID_PARAMETER over ERROR_INVALID_FLAGS) */ 291 /* The same test with CP_SYMBOL instead of CP_UTF7 gives ERROR_INVALID_FLAGS 292 instead except on Windows NT4 */ 293 SetLastError(0xdeadbeef); 294 len = WideCharToMultiByte(CP_UTF7, 1, w_string, w_string_len, c_string, c_string_len, NULL, &used); 295 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 296 297 /* CP_UTF8, unrecognized flag and used not NULL => ERROR_INVALID_PARAMETER */ 298 SetLastError(0xdeadbeef); 299 len = WideCharToMultiByte(CP_UTF8, 0x100, w_string, w_string_len, c_string, c_string_len, NULL, &used); 300 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%x\n", len, GetLastError()); 301 } 302 303 static void test_overlapped_buffers(void) 304 { 305 static const WCHAR strW[] = {'j','u','s','t',' ','a',' ','t','e','s','t',0}; 306 static const char strA[] = "just a test"; 307 char buf[256]; 308 int ret; 309 310 SetLastError(0xdeadbeef); 311 memcpy(buf + 1, strW, sizeof(strW)); 312 ret = WideCharToMultiByte(CP_ACP, 0, (WCHAR *)(buf + 1), -1, buf, sizeof(buf), NULL, NULL); 313 ok(ret == sizeof(strA), "unexpected ret %d\n", ret); 314 ok(!memcmp(buf, strA, sizeof(strA)), "conversion failed: %s\n", buf); 315 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 316 } 317 318 static void test_string_conversion(LPBOOL bUsedDefaultChar) 319 { 320 char mbc; 321 char mbs[15]; 322 int ret; 323 WCHAR wc1 = 228; /* Western Windows-1252 character */ 324 WCHAR wc2 = 1088; /* Russian Windows-1251 character not displayable for Windows-1252 */ 325 static const WCHAR wcs[] = {'T', 'h', 1088, 'i', 0}; /* String with ASCII characters and a Russian character */ 326 static const WCHAR dbwcs[] = {28953, 25152, 0}; /* String with Chinese (codepage 950) characters */ 327 static const WCHAR dbwcs2[] = {0x7bb8, 0x3d, 0xc813, 0xac00, 0xb77d, 0}; 328 static const char default_char[] = {0xa3, 0xbf, 0}; 329 330 SetLastError(0xdeadbeef); 331 ret = WideCharToMultiByte(1252, 0, &wc1, 1, &mbc, 1, NULL, bUsedDefaultChar); 332 ok(ret == 1, "ret is %d\n", ret); 333 ok(mbc == '\xe4', "mbc is %d\n", mbc); 334 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 335 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 336 337 SetLastError(0xdeadbeef); 338 ret = WideCharToMultiByte(1252, 0, &wc2, 1, &mbc, 1, NULL, bUsedDefaultChar); 339 ok(ret == 1, "ret is %d\n", ret); 340 ok(mbc == 63, "mbc is %d\n", mbc); 341 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 342 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 343 344 if (IsValidCodePage(1251)) 345 { 346 SetLastError(0xdeadbeef); 347 ret = WideCharToMultiByte(1251, 0, &wc2, 1, &mbc, 1, NULL, bUsedDefaultChar); 348 ok(ret == 1, "ret is %d\n", ret); 349 ok(mbc == '\xf0', "mbc is %d\n", mbc); 350 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 351 ok(GetLastError() == 0xdeadbeef || 352 broken(GetLastError() == 0), /* win95 */ 353 "GetLastError() is %u\n", GetLastError()); 354 355 SetLastError(0xdeadbeef); 356 ret = WideCharToMultiByte(1251, 0, &wc1, 1, &mbc, 1, NULL, bUsedDefaultChar); 357 ok(ret == 1, "ret is %d\n", ret); 358 ok(mbc == 97, "mbc is %d\n", mbc); 359 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 360 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 361 } 362 else 363 skip("Codepage 1251 not available\n"); 364 365 /* This call triggers the last Win32 error */ 366 SetLastError(0xdeadbeef); 367 ret = WideCharToMultiByte(1252, 0, wcs, -1, &mbc, 1, NULL, bUsedDefaultChar); 368 ok(ret == 0, "ret is %d\n", ret); 369 ok(mbc == 84, "mbc is %d\n", mbc); 370 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 371 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %u\n", GetLastError()); 372 373 SetLastError(0xdeadbeef); 374 ret = WideCharToMultiByte(1252, 0, wcs, -1, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 375 ok(ret == 5, "ret is %d\n", ret); 376 ok(!strcmp(mbs, "Th?i"), "mbs is %s\n", mbs); 377 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 378 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 379 mbs[0] = 0; 380 381 /* WideCharToMultiByte mustn't add any null character automatically. 382 So in this case, we should get the same string again, even if we only copied the first three bytes. */ 383 SetLastError(0xdeadbeef); 384 ret = WideCharToMultiByte(1252, 0, wcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 385 ok(ret == 3, "ret is %d\n", ret); 386 ok(!strcmp(mbs, "Th?i"), "mbs is %s\n", mbs); 387 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 388 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 389 ZeroMemory(mbs, 5); 390 391 /* Now this shouldn't be the case like above as we zeroed the complete string buffer. */ 392 SetLastError(0xdeadbeef); 393 ret = WideCharToMultiByte(1252, 0, wcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 394 ok(ret == 3, "ret is %d\n", ret); 395 ok(!strcmp(mbs, "Th?"), "mbs is %s\n", mbs); 396 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 397 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 398 399 /* Double-byte tests */ 400 ret = WideCharToMultiByte(1252, 0, dbwcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 401 ok(ret == 3, "ret is %d\n", ret); 402 ok(!strcmp(mbs, "??"), "mbs is %s\n", mbs); 403 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 404 405 ret = WideCharToMultiByte(936, WC_COMPOSITECHECK, dbwcs2, -1, mbs, sizeof(mbs), (const char *)default_char, bUsedDefaultChar); 406 ok(ret == 10, "ret is %d\n", ret); 407 ok(!strcmp(mbs, "\xf3\xe7\x3d\xa3\xbf\xa3\xbf\xa3\xbf"), "mbs is %s\n", mbs); 408 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 409 410 /* Length-only tests */ 411 SetLastError(0xdeadbeef); 412 ret = WideCharToMultiByte(1252, 0, &wc2, 1, NULL, 0, NULL, bUsedDefaultChar); 413 ok(ret == 1, "ret is %d\n", ret); 414 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 415 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 416 417 SetLastError(0xdeadbeef); 418 ret = WideCharToMultiByte(1252, 0, wcs, -1, NULL, 0, NULL, bUsedDefaultChar); 419 ok(ret == 5, "ret is %d\n", ret); 420 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 421 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 422 423 if (!IsValidCodePage(950)) 424 { 425 skip("Codepage 950 not available\n"); 426 return; 427 } 428 429 /* Double-byte tests */ 430 SetLastError(0xdeadbeef); 431 ret = WideCharToMultiByte(950, 0, dbwcs, -1, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 432 ok(ret == 5, "ret is %d\n", ret); 433 ok(!strcmp(mbs, "\xb5H\xa9\xd2"), "mbs is %s\n", mbs); 434 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 435 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 436 437 SetLastError(0xdeadbeef); 438 ret = WideCharToMultiByte(950, 0, dbwcs, 1, &mbc, 1, NULL, bUsedDefaultChar); 439 ok(ret == 0, "ret is %d\n", ret); 440 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 441 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %u\n", GetLastError()); 442 ZeroMemory(mbs, 5); 443 444 SetLastError(0xdeadbeef); 445 ret = WideCharToMultiByte(950, 0, dbwcs, 1, mbs, sizeof(mbs), NULL, bUsedDefaultChar); 446 ok(ret == 2, "ret is %d\n", ret); 447 ok(!strcmp(mbs, "\xb5H"), "mbs is %s\n", mbs); 448 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 449 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 450 451 /* Length-only tests */ 452 SetLastError(0xdeadbeef); 453 ret = WideCharToMultiByte(950, 0, dbwcs, 1, NULL, 0, NULL, bUsedDefaultChar); 454 ok(ret == 2, "ret is %d\n", ret); 455 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 456 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 457 458 SetLastError(0xdeadbeef); 459 ret = WideCharToMultiByte(950, 0, dbwcs, -1, NULL, 0, NULL, bUsedDefaultChar); 460 ok(ret == 5, "ret is %d\n", ret); 461 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar); 462 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %u\n", GetLastError()); 463 } 464 465 static void test_utf7_encoding(void) 466 { 467 WCHAR input[16]; 468 char output[16], expected[16]; 469 int i, len, expected_len; 470 471 static const BOOL directly_encodable_table[] = 472 { 473 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, /* 0x00 - 0x0F */ 474 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0x10 - 0x1F */ 475 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, /* 0x20 - 0x2F */ 476 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 0x30 - 0x3F */ 477 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 0x40 - 0x4F */ 478 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 0x50 - 0x5F */ 479 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 0x60 - 0x6F */ 480 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 /* 0x70 - 0x7F */ 481 }; 482 static const char base64_encoding_table[] = 483 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; 484 485 const struct 486 { 487 /* inputs */ 488 WCHAR src[16]; 489 int srclen; 490 char *dst; 491 int dstlen; 492 /* expected outputs */ 493 char expected_dst[16]; 494 int chars_written; 495 int len; 496 } 497 tests[] = 498 { 499 /* tests string conversion with srclen=-1 */ 500 { 501 {0x4F60,0x597D,0x5417,0}, -1, output, sizeof(output) - 1, 502 "+T2BZfVQX-", 11, 11 503 }, 504 /* tests string conversion with srclen=-2 */ 505 { 506 {0x4F60,0x597D,0x5417,0}, -2, output, sizeof(output) - 1, 507 "+T2BZfVQX-", 11, 11 508 }, 509 /* tests string conversion with dstlen=strlen(expected_dst) */ 510 { 511 {0x4F60,0x597D,0x5417,0}, -1, output, 10, 512 "+T2BZfVQX-", 10, 0 513 }, 514 /* tests string conversion with dstlen=strlen(expected_dst)+1 */ 515 { 516 {0x4F60,0x597D,0x5417,0}, -1, output, 11, 517 "+T2BZfVQX-", 11, 11 518 }, 519 /* tests string conversion with dstlen=strlen(expected_dst)+2 */ 520 { 521 {0x4F60,0x597D,0x5417,0}, -1, output, 12, 522 "+T2BZfVQX-", 11, 11 523 }, 524 /* tests dry run with dst=NULL and dstlen=0 */ 525 { 526 {0x4F60,0x597D,0x5417,0}, -1, NULL, 0, 527 {0}, 0, 11 528 }, 529 /* tests dry run with dst!=NULL and dstlen=0 */ 530 { 531 {0x4F60,0x597D,0x5417,0}, -1, output, 0, 532 {0}, 0, 11 533 }, 534 /* tests srclen < strlenW(src) with directly encodable chars */ 535 { 536 {'h','e','l','l','o',0}, 2, output, sizeof(output) - 1, 537 "he", 2, 2 538 }, 539 /* tests srclen < strlenW(src) with non-directly encodable chars */ 540 { 541 {0x4F60,0x597D,0x5417,0}, 2, output, sizeof(output) - 1, 542 "+T2BZfQ-", 8, 8 543 }, 544 /* tests a single null char */ 545 { 546 {0}, -1, output, sizeof(output) - 1, 547 "", 1, 1 548 }, 549 /* tests a buffer that runs out while not encoding a UTF-7 sequence */ 550 { 551 {'h','e','l','l','o',0}, -1, output, 2, 552 "he", 2, 0 553 }, 554 /* tests a buffer that runs out after writing 1 base64 character */ 555 { 556 {0x4F60,0x0001,0}, -1, output, 2, 557 "+T", 2, 0 558 }, 559 /* tests a buffer that runs out after writing 2 base64 characters */ 560 { 561 {0x4F60,0x0001,0}, -1, output, 3, 562 "+T2", 3, 0 563 }, 564 /* tests a buffer that runs out after writing 3 base64 characters */ 565 { 566 {0x4F60,0x0001,0}, -1, output, 4, 567 "+T2A", 4, 0 568 }, 569 /* tests a buffer that runs out just after writing the + sign */ 570 { 571 {0x4F60,0}, -1, output, 1, 572 "+", 1, 0 573 }, 574 /* tests a buffer that runs out just before writing the - sign 575 * the number of bits to encode here is evenly divisible by 6 */ 576 { 577 {0x4F60,0x597D,0x5417,0}, -1, output, 9, 578 "+T2BZfVQX", 9, 0 579 }, 580 /* tests a buffer that runs out just before writing the - sign 581 * the number of bits to encode here is NOT evenly divisible by 6 */ 582 { 583 {0x4F60,0}, -1, output, 4, 584 "+T2", 3, 0 585 }, 586 /* tests a buffer that runs out in the middle of escaping a + sign */ 587 { 588 {'+',0}, -1, output, 1, 589 "+", 1, 0 590 } 591 }; 592 593 /* test which characters are encoded if surrounded by non-encoded characters */ 594 for (i = 0; i <= 0xFFFF; i++) 595 { 596 input[0] = ' '; 597 input[1] = i; 598 input[2] = ' '; 599 input[3] = 0; 600 601 memset(output, '#', sizeof(output) - 1); 602 output[sizeof(output) - 1] = 0; 603 604 len = WideCharToMultiByte(CP_UTF7, 0, input, 4, output, sizeof(output) - 1, NULL, NULL); 605 606 if (i == '+') 607 { 608 /* '+' is a special case and is encoded as "+-" */ 609 expected_len = 5; 610 strcpy(expected, " +- "); 611 } 612 else if (i <= 0x7F && directly_encodable_table[i]) 613 { 614 /* encodes directly */ 615 expected_len = 4; 616 sprintf(expected, " %c ", i); 617 } 618 else 619 { 620 /* base64-encodes */ 621 expected_len = 8; 622 sprintf(expected, " +%c%c%c- ", 623 base64_encoding_table[(i & 0xFC00) >> 10], 624 base64_encoding_table[(i & 0x03F0) >> 4], 625 base64_encoding_table[(i & 0x000F) << 2]); 626 } 627 628 ok(len == expected_len, "i=0x%04x: expected len=%i, got len=%i\n", i, expected_len, len); 629 ok(memcmp(output, expected, expected_len) == 0, 630 "i=0x%04x: expected output='%s', got output='%s'\n", i, expected, output); 631 ok(output[expected_len] == '#', "i=0x%04x: expected output[%i]='#', got output[%i]=%i\n", 632 i, expected_len, expected_len, output[expected_len]); 633 } 634 635 /* test which one-byte characters are absorbed into surrounding base64 blocks 636 * (Windows always ends the base64 block when it encounters a directly encodable character) */ 637 for (i = 0; i <= 0xFFFF; i++) 638 { 639 input[0] = 0x2672; 640 input[1] = i; 641 input[2] = 0x2672; 642 input[3] = 0; 643 644 memset(output, '#', sizeof(output) - 1); 645 output[sizeof(output) - 1] = 0; 646 647 len = WideCharToMultiByte(CP_UTF7, 0, input, 4, output, sizeof(output) - 1, NULL, NULL); 648 649 if (i == '+') 650 { 651 /* '+' is a special case and is encoded as "+-" */ 652 expected_len = 13; 653 strcpy(expected, "+JnI-+-+JnI-"); 654 } 655 else if (i <= 0x7F && directly_encodable_table[i]) 656 { 657 /* encodes directly */ 658 expected_len = 12; 659 sprintf(expected, "+JnI-%c+JnI-", i); 660 } 661 else 662 { 663 /* base64-encodes */ 664 expected_len = 11; 665 sprintf(expected, "+Jn%c%c%c%cZy-", 666 base64_encoding_table[8 | ((i & 0xC000) >> 14)], 667 base64_encoding_table[(i & 0x3F00) >> 8], 668 base64_encoding_table[(i & 0x00FC) >> 2], 669 base64_encoding_table[((i & 0x0003) << 4) | 2]); 670 } 671 672 ok(len == expected_len, "i=0x%04x: expected len=%i, got len=%i\n", i, expected_len, len); 673 ok(memcmp(output, expected, expected_len) == 0, 674 "i=0x%04x: expected output='%s', got output='%s'\n", i, expected, output); 675 ok(output[expected_len] == '#', "i=0x%04x: expected output[%i]='#', got output[%i]=%i\n", 676 i, expected_len, expected_len, output[expected_len]); 677 } 678 679 for (i = 0; i < sizeof(tests) / sizeof(tests[0]); i++) 680 { 681 memset(output, '#', sizeof(output) - 1); 682 output[sizeof(output) - 1] = 0; 683 SetLastError(0xdeadbeef); 684 685 len = WideCharToMultiByte(CP_UTF7, 0, tests[i].src, tests[i].srclen, 686 tests[i].dst, tests[i].dstlen, NULL, NULL); 687 688 if (!tests[i].len) 689 { 690 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, 691 "tests[%i]: expected error=0x%x, got error=0x%x\n", 692 i, ERROR_INSUFFICIENT_BUFFER, GetLastError()); 693 } 694 ok(len == tests[i].len, "tests[%i]: expected len=%i, got len=%i\n", i, tests[i].len, len); 695 696 if (tests[i].dst) 697 { 698 ok(memcmp(tests[i].dst, tests[i].expected_dst, tests[i].chars_written) == 0, 699 "tests[%i]: expected dst='%s', got dst='%s'\n", 700 i, tests[i].expected_dst, tests[i].dst); 701 ok(tests[i].dst[tests[i].chars_written] == '#', 702 "tests[%i]: expected dst[%i]='#', got dst[%i]=%i\n", 703 i, tests[i].chars_written, tests[i].chars_written, tests[i].dst[tests[i].chars_written]); 704 } 705 } 706 } 707 708 static void test_utf7_decoding(void) 709 { 710 char input[32]; 711 WCHAR output[32], expected[32]; 712 int i, len, expected_len; 713 714 static const signed char base64_decoding_table[] = 715 { 716 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0x00-0x0F */ 717 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0x10-0x1F */ 718 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1, -1, 63, /* 0x20-0x2F */ 719 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, /* 0x30-0x3F */ 720 -1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 0x40-0x4F */ 721 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, -1, /* 0x50-0x5F */ 722 -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, /* 0x60-0x6F */ 723 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1 /* 0x70-0x7F */ 724 }; 725 726 struct 727 { 728 /* inputs */ 729 char src[32]; 730 int srclen; 731 WCHAR *dst; 732 int dstlen; 733 /* expected outputs */ 734 WCHAR expected_dst[32]; 735 int chars_written; 736 int len; 737 } 738 tests[] = 739 { 740 /* tests string conversion with srclen=-1 */ 741 { 742 "+T2BZfQ-", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 743 {0x4F60,0x597D,0}, 3, 3 744 }, 745 /* tests string conversion with srclen=-2 */ 746 { 747 "+T2BZfQ-", -2, output, sizeof(output) / sizeof(WCHAR) - 1, 748 {0x4F60,0x597D,0}, 3, 3 749 }, 750 /* tests string conversion with dstlen=strlen(expected_dst) */ 751 { 752 "+T2BZfQ-", -1, output, 2, 753 {0x4F60,0x597D}, 2, 0 754 }, 755 /* tests string conversion with dstlen=strlen(expected_dst)+1 */ 756 { 757 "+T2BZfQ-", -1, output, 3, 758 {0x4F60,0x597D,0}, 3, 3 759 }, 760 /* tests string conversion with dstlen=strlen(expected_dst)+2 */ 761 { 762 "+T2BZfQ-", -1, output, 4, 763 {0x4F60,0x597D,0}, 3, 3 764 }, 765 /* tests dry run with dst=NULL and dstlen=0 */ 766 { 767 "+T2BZfQ-", -1, NULL, 0, 768 {0}, 0, 3 769 }, 770 /* tests dry run with dst!=NULL and dstlen=0 */ 771 { 772 "+T2BZfQ-", -1, output, 0, 773 {0}, 0, 3 774 }, 775 /* tests ill-formed UTF-7: 6 bits, not enough for a byte pair */ 776 { 777 "+T-+T-+T-hello", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 778 {'h','e','l','l','o',0}, 6, 6 779 }, 780 /* tests ill-formed UTF-7: 12 bits, not enough for a byte pair */ 781 { 782 "+T2-+T2-+T2-hello", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 783 {'h','e','l','l','o',0}, 6, 6 784 }, 785 /* tests ill-formed UTF-7: 18 bits, not a multiple of 16 and the last bit is a 1 */ 786 { 787 "+T2B-+T2B-+T2B-hello", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 788 {0x4F60,0x4F60,0x4F60,'h','e','l','l','o',0}, 9, 9 789 }, 790 /* tests ill-formed UTF-7: 24 bits, a multiple of 8 but not a multiple of 16 */ 791 { 792 "+T2BZ-+T2BZ-+T2BZ-hello", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 793 {0x4F60,0x4F60,0x4F60,'h','e','l','l','o',0}, 9, 9 794 }, 795 /* tests UTF-7 followed by characters that should be encoded but aren't */ 796 { 797 "+T2BZ-\x82\xFE", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 798 {0x4F60,0x0082,0x00FE,0}, 4, 4 799 }, 800 /* tests srclen > strlen(src) */ 801 { 802 "a\0b", 4, output, sizeof(output) / sizeof(WCHAR) - 1, 803 {'a',0,'b',0}, 4, 4 804 }, 805 /* tests srclen < strlen(src) outside of a UTF-7 sequence */ 806 { 807 "hello", 2, output, sizeof(output) / sizeof(WCHAR) - 1, 808 {'h','e'}, 2, 2 809 }, 810 /* tests srclen < strlen(src) inside of a UTF-7 sequence */ 811 { 812 "+T2BZfQ-", 4, output, sizeof(output) / sizeof(WCHAR) - 1, 813 {0x4F60}, 1, 1 814 }, 815 /* tests srclen < strlen(src) right at the beginning of a UTF-7 sequence */ 816 { 817 "hi+T2A-", 3, output, sizeof(output) / sizeof(WCHAR) - 1, 818 {'h','i'}, 2, 2 819 }, 820 /* tests srclen < strlen(src) right at the end of a UTF-7 sequence */ 821 { 822 "+T2A-hi", 5, output, sizeof(output) / sizeof(WCHAR) - 1, 823 {0x4F60}, 1, 1 824 }, 825 /* tests srclen < strlen(src) at the beginning of an escaped + sign */ 826 { 827 "hi+-", 3, output, sizeof(output) / sizeof(WCHAR) - 1, 828 {'h','i'}, 2, 2 829 }, 830 /* tests srclen < strlen(src) at the end of an escaped + sign */ 831 { 832 "+-hi", 2, output, sizeof(output) / sizeof(WCHAR) - 1, 833 {'+'}, 1, 1 834 }, 835 /* tests len=0 but no error */ 836 { 837 "+", 1, output, sizeof(output) / sizeof(WCHAR) - 1, 838 {0}, 0, 0 839 }, 840 /* tests a single null char */ 841 { 842 "", -1, output, sizeof(output) / sizeof(WCHAR) - 1, 843 {0}, 1, 1 844 }, 845 /* tests a buffer that runs out while not decoding a UTF-7 sequence */ 846 { 847 "hello", -1, output, 2, 848 {'h','e'}, 2, 0 849 }, 850 /* tests a buffer that runs out in the middle of decoding a UTF-7 sequence */ 851 { 852 "+T2BZfQ-", -1, output, 1, 853 {0x4F60}, 1, 0 854 } 855 }; 856 857 /* test which one-byte characters remove stray + signs */ 858 for (i = 0; i < 256; i++) 859 { 860 sprintf(input, "+%c+AAA", i); 861 862 memset(output, 0x23, sizeof(output) - sizeof(WCHAR)); 863 output[sizeof(output) / sizeof(WCHAR) - 1] = 0; 864 865 len = MultiByteToWideChar(CP_UTF7, 0, input, 7, output, sizeof(output) / sizeof(WCHAR) - 1); 866 867 if (i == '-') 868 { 869 /* removes the - sign */ 870 expected_len = 3; 871 expected[0] = 0x002B; 872 expected[1] = 0; 873 expected[2] = 0; 874 } 875 else if (i <= 0x7F && base64_decoding_table[i] != -1) 876 { 877 /* absorbs the character into the base64 sequence */ 878 expected_len = 2; 879 expected[0] = (base64_decoding_table[i] << 10) | 0x03E0; 880 expected[1] = 0; 881 } 882 else 883 { 884 /* removes the + sign */ 885 expected_len = 3; 886 expected[0] = i; 887 expected[1] = 0; 888 expected[2] = 0; 889 } 890 expected[expected_len] = 0x2323; 891 892 ok(len == expected_len, "i=0x%02x: expected len=%i, got len=%i\n", i, expected_len, len); 893 ok(memcmp(output, expected, (expected_len + 1) * sizeof(WCHAR)) == 0, 894 "i=0x%02x: expected output=%s, got output=%s\n", 895 i, wine_dbgstr_wn(expected, expected_len + 1), wine_dbgstr_wn(output, expected_len + 1)); 896 } 897 898 /* test which one-byte characters terminate a sequence 899 * also test whether the unfinished byte pair is discarded or not */ 900 for (i = 0; i < 256; i++) 901 { 902 sprintf(input, "+B%c+AAA", i); 903 904 memset(output, 0x23, sizeof(output) - sizeof(WCHAR)); 905 output[sizeof(output) / sizeof(WCHAR) - 1] = 0; 906 907 len = MultiByteToWideChar(CP_UTF7, 0, input, 8, output, sizeof(output) / sizeof(WCHAR) - 1); 908 909 if (i == '-') 910 { 911 /* explicitly terminates */ 912 expected_len = 2; 913 expected[0] = 0; 914 expected[1] = 0; 915 } 916 else if (i <= 0x7F) 917 { 918 if (base64_decoding_table[i] != -1) 919 { 920 /* absorbs the character into the base64 sequence */ 921 expected_len = 3; 922 expected[0] = 0x0400 | (base64_decoding_table[i] << 4) | 0x000F; 923 expected[1] = 0x8000; 924 expected[2] = 0; 925 } 926 else 927 { 928 /* implicitly terminates and discards the unfinished byte pair */ 929 expected_len = 3; 930 expected[0] = i; 931 expected[1] = 0; 932 expected[2] = 0; 933 } 934 } 935 else 936 { 937 /* implicitly terminates but does not the discard unfinished byte pair */ 938 expected_len = 3; 939 expected[0] = i; 940 expected[1] = 0x0400; 941 expected[2] = 0; 942 } 943 expected[expected_len] = 0x2323; 944 945 ok(len == expected_len, "i=0x%02x: expected len=%i, got len=%i\n", i, expected_len, len); 946 ok(memcmp(output, expected, (expected_len + 1) * sizeof(WCHAR)) == 0, 947 "i=0x%02x: expected output=%s, got output=%s\n", 948 i, wine_dbgstr_wn(expected, expected_len + 1), wine_dbgstr_wn(output, expected_len + 1)); 949 } 950 951 for (i = 0; i < sizeof(tests) / sizeof(tests[0]); i++) 952 { 953 memset(output, 0x23, sizeof(output) - sizeof(WCHAR)); 954 output[sizeof(output) / sizeof(WCHAR) - 1] = 0; 955 SetLastError(0xdeadbeef); 956 957 len = MultiByteToWideChar(CP_UTF7, 0, tests[i].src, tests[i].srclen, 958 tests[i].dst, tests[i].dstlen); 959 960 tests[i].expected_dst[tests[i].chars_written] = 0x2323; 961 962 if (!tests[i].len && tests[i].chars_written) 963 { 964 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, 965 "tests[%i]: expected error=0x%x, got error=0x%x\n", 966 i, ERROR_INSUFFICIENT_BUFFER, GetLastError()); 967 } 968 ok(len == tests[i].len, "tests[%i]: expected len=%i, got len=%i\n", i, tests[i].len, len); 969 970 if (tests[i].dst) 971 { 972 ok(memcmp(tests[i].dst, tests[i].expected_dst, (tests[i].chars_written + 1) * sizeof(WCHAR)) == 0, 973 "tests[%i]: expected dst=%s, got dst=%s\n", 974 i, wine_dbgstr_wn(tests[i].expected_dst, tests[i].chars_written + 1), 975 wine_dbgstr_wn(tests[i].dst, tests[i].chars_written + 1)); 976 } 977 } 978 } 979 980 static void test_undefined_byte_char(void) 981 { 982 static const struct tag_testset { 983 INT codepage; 984 LPCSTR str; 985 BOOL is_error; 986 } testset[] = { 987 { 874, "\xdd", TRUE }, 988 { 932, "\xfe", TRUE }, 989 { 932, "\x80", FALSE }, 990 { 936, "\xff", TRUE }, 991 { 949, "\xff", TRUE }, 992 { 950, "\xff", TRUE }, 993 { 1252, "\x90", FALSE }, 994 { 1253, "\xaa", TRUE }, 995 { 1255, "\xff", TRUE }, 996 { 1257, "\xa5", TRUE }, 997 }; 998 INT i, ret; 999 1000 for (i = 0; i < (sizeof(testset) / sizeof(testset[0])); i++) { 1001 if (! IsValidCodePage(testset[i].codepage)) 1002 { 1003 skip("Codepage %d not available\n", testset[i].codepage); 1004 continue; 1005 } 1006 1007 SetLastError(0xdeadbeef); 1008 ret = MultiByteToWideChar(testset[i].codepage, MB_ERR_INVALID_CHARS, 1009 testset[i].str, -1, NULL, 0); 1010 if (testset[i].is_error) { 1011 ok(ret == 0 && GetLastError() == ERROR_NO_UNICODE_TRANSLATION, 1012 "ret is %d, GetLastError is %u (cp %d)\n", 1013 ret, GetLastError(), testset[i].codepage); 1014 } 1015 else { 1016 ok(ret == strlen(testset[i].str)+1 && GetLastError() == 0xdeadbeef, 1017 "ret is %d, GetLastError is %u (cp %d)\n", 1018 ret, GetLastError(), testset[i].codepage); 1019 } 1020 1021 SetLastError(0xdeadbeef); 1022 ret = MultiByteToWideChar(testset[i].codepage, 0, 1023 testset[i].str, -1, NULL, 0); 1024 ok(ret == strlen(testset[i].str)+1 && GetLastError() == 0xdeadbeef, 1025 "ret is %d, GetLastError is %u (cp %d)\n", 1026 ret, GetLastError(), testset[i].codepage); 1027 } 1028 } 1029 1030 static void test_threadcp(void) 1031 { 1032 static const LCID ENGLISH = MAKELCID(MAKELANGID(LANG_ENGLISH, SUBLANG_ENGLISH_US), SORT_DEFAULT); 1033 static const LCID HINDI = MAKELCID(MAKELANGID(LANG_HINDI, SUBLANG_HINDI_INDIA), SORT_DEFAULT); 1034 static const LCID GEORGIAN = MAKELCID(MAKELANGID(LANG_GEORGIAN, SUBLANG_GEORGIAN_GEORGIA), SORT_DEFAULT); 1035 static const LCID RUSSIAN = MAKELCID(MAKELANGID(LANG_RUSSIAN, SUBLANG_RUSSIAN_RUSSIA), SORT_DEFAULT); 1036 static const LCID JAPANESE = MAKELCID(MAKELANGID(LANG_JAPANESE, SUBLANG_JAPANESE_JAPAN), SORT_DEFAULT); 1037 static const LCID CHINESE = MAKELCID(MAKELANGID(LANG_CHINESE, SUBLANG_CHINESE_SIMPLIFIED), SORT_DEFAULT); 1038 1039 BOOL islead, islead_acp; 1040 CPINFOEXA cpi; 1041 UINT cp, acp; 1042 int i, num; 1043 LCID last; 1044 BOOL ret; 1045 1046 struct test { 1047 LCID lcid; 1048 UINT threadcp; 1049 } lcids[] = { 1050 { HINDI, 0 }, 1051 { GEORGIAN, 0 }, 1052 { ENGLISH, 1252 }, 1053 { RUSSIAN, 1251 }, 1054 { JAPANESE, 932 }, 1055 { CHINESE, 936 } 1056 }; 1057 1058 struct test_islead_nocp { 1059 LCID lcid; 1060 BYTE testchar; 1061 } isleads_nocp[] = { 1062 { HINDI, 0x00 }, 1063 { HINDI, 0x81 }, 1064 { HINDI, 0xa0 }, 1065 { HINDI, 0xe0 }, 1066 1067 { GEORGIAN, 0x00 }, 1068 { GEORGIAN, 0x81 }, 1069 { GEORGIAN, 0xa0 }, 1070 { GEORGIAN, 0xe0 }, 1071 }; 1072 1073 struct test_islead { 1074 LCID lcid; 1075 BYTE testchar; 1076 BOOL islead; 1077 } isleads[] = { 1078 { ENGLISH, 0x00, FALSE }, 1079 { ENGLISH, 0x81, FALSE }, 1080 { ENGLISH, 0xa0, FALSE }, 1081 { ENGLISH, 0xe0, FALSE }, 1082 1083 { RUSSIAN, 0x00, FALSE }, 1084 { RUSSIAN, 0x81, FALSE }, 1085 { RUSSIAN, 0xa0, FALSE }, 1086 { RUSSIAN, 0xe0, FALSE }, 1087 1088 { JAPANESE, 0x00, FALSE }, 1089 { JAPANESE, 0x81, TRUE }, 1090 { JAPANESE, 0xa0, FALSE }, 1091 { JAPANESE, 0xe0, TRUE }, 1092 1093 { CHINESE, 0x00, FALSE }, 1094 { CHINESE, 0x81, TRUE }, 1095 { CHINESE, 0xa0, TRUE }, 1096 { CHINESE, 0xe0, TRUE }, 1097 }; 1098 1099 last = GetThreadLocale(); 1100 acp = GetACP(); 1101 1102 for (i = 0; i < sizeof(lcids)/sizeof(lcids[0]); i++) 1103 { 1104 SetThreadLocale(lcids[i].lcid); 1105 1106 cp = 0xdeadbeef; 1107 GetLocaleInfoA(lcids[i].lcid, LOCALE_IDEFAULTANSICODEPAGE|LOCALE_RETURN_NUMBER, (LPSTR)&cp, sizeof(cp)); 1108 ok(cp == lcids[i].threadcp, "wrong codepage %u for lcid %04x, should be %u\n", cp, lcids[i].threadcp, cp); 1109 1110 /* GetCPInfoEx/GetCPInfo - CP_ACP */ 1111 SetLastError(0xdeadbeef); 1112 memset(&cpi, 0, sizeof(cpi)); 1113 ret = GetCPInfoExA(CP_ACP, 0, &cpi); 1114 ok(ret, "GetCPInfoExA failed for lcid %04x, error %d\n", lcids[i].lcid, GetLastError()); 1115 ok(cpi.CodePage == acp, "wrong codepage %u for lcid %04x, should be %u\n", cpi.CodePage, lcids[i].lcid, acp); 1116 1117 /* WideCharToMultiByte - CP_ACP */ 1118 num = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL); 1119 ok(num == 7, "ret is %d (%04x)\n", num, lcids[i].lcid); 1120 1121 /* MultiByteToWideChar - CP_ACP */ 1122 num = MultiByteToWideChar(CP_ACP, 0, "foobar", -1, NULL, 0); 1123 ok(num == 7, "ret is %d (%04x)\n", num, lcids[i].lcid); 1124 1125 /* GetCPInfoEx/GetCPInfo - CP_THREAD_ACP */ 1126 SetLastError(0xdeadbeef); 1127 memset(&cpi, 0, sizeof(cpi)); 1128 ret = GetCPInfoExA(CP_THREAD_ACP, 0, &cpi); 1129 ok(ret, "GetCPInfoExA failed for lcid %04x, error %d\n", lcids[i].lcid, GetLastError()); 1130 if (lcids[i].threadcp) 1131 ok(cpi.CodePage == lcids[i].threadcp, "wrong codepage %u for lcid %04x, should be %u\n", 1132 cpi.CodePage, lcids[i].lcid, lcids[i].threadcp); 1133 else 1134 ok(cpi.CodePage == acp, "wrong codepage %u for lcid %04x, should be %u\n", 1135 cpi.CodePage, lcids[i].lcid, acp); 1136 1137 /* WideCharToMultiByte - CP_THREAD_ACP */ 1138 num = WideCharToMultiByte(CP_THREAD_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL); 1139 ok(num == 7, "ret is %d (%04x)\n", num, lcids[i].lcid); 1140 1141 /* MultiByteToWideChar - CP_THREAD_ACP */ 1142 num = MultiByteToWideChar(CP_THREAD_ACP, 0, "foobar", -1, NULL, 0); 1143 ok(num == 7, "ret is %d (%04x)\n", num, lcids[i].lcid); 1144 } 1145 1146 /* IsDBCSLeadByteEx - locales without codepage */ 1147 for (i = 0; i < sizeof(isleads_nocp)/sizeof(isleads_nocp[0]); i++) 1148 { 1149 SetThreadLocale(isleads_nocp[i].lcid); 1150 1151 islead_acp = IsDBCSLeadByteEx(CP_ACP, isleads_nocp[i].testchar); 1152 islead = IsDBCSLeadByteEx(CP_THREAD_ACP, isleads_nocp[i].testchar); 1153 1154 ok(islead == islead_acp, "wrong islead %i for test char %x in lcid %04x. should be %i\n", 1155 islead, isleads_nocp[i].testchar, isleads_nocp[i].lcid, islead_acp); 1156 } 1157 1158 /* IsDBCSLeadByteEx - locales with codepage */ 1159 for (i = 0; i < sizeof(isleads)/sizeof(isleads[0]); i++) 1160 { 1161 SetThreadLocale(isleads[i].lcid); 1162 1163 islead = IsDBCSLeadByteEx(CP_THREAD_ACP, isleads[i].testchar); 1164 ok(islead == isleads[i].islead, "wrong islead %i for test char %x in lcid %04x. should be %i\n", 1165 islead, isleads[i].testchar, isleads[i].lcid, isleads[i].islead); 1166 } 1167 1168 SetThreadLocale(last); 1169 } 1170 1171 static void test_dbcs_to_widechar(void) 1172 { 1173 int i, count, count2; 1174 WCHAR wbuf[5]; 1175 unsigned char buf[] = {0xbf, 0xb4, 0xc7, '\0', 'x'}; 1176 static const DWORD flags[] = { 1177 MB_PRECOMPOSED, 1178 MB_COMPOSITE, 1179 1180 MB_PRECOMPOSED|MB_USEGLYPHCHARS, 1181 MB_COMPOSITE |MB_USEGLYPHCHARS, 1182 1183 MB_PRECOMPOSED|MB_ERR_INVALID_CHARS, 1184 MB_COMPOSITE |MB_ERR_INVALID_CHARS, 1185 1186 MB_PRECOMPOSED|MB_ERR_INVALID_CHARS|MB_USEGLYPHCHARS, 1187 MB_COMPOSITE |MB_ERR_INVALID_CHARS|MB_USEGLYPHCHARS, 1188 }; 1189 1190 for (i = 0; i < sizeof(flags)/sizeof(DWORD); ++i) 1191 { 1192 memset(wbuf, 0xff, sizeof(wbuf)); 1193 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 2, NULL, 0); 1194 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 2, wbuf, count); 1195 1196 ok(count == 1, "%04x: returned %d (expected 1)\n", flags[i], count); 1197 ok(count2 == 1, "%04x: returned %d (expected 1)\n", flags[i], count2); 1198 ok(wbuf[0] == 0x770b, "%04x: returned %04x (expected 770b)\n", flags[i], wbuf[0]); 1199 ok(wbuf[1] == 0xffff, "%04x: returned %04x (expected ffff)\n", flags[i], wbuf[1]); 1200 } 1201 1202 for (i = 0; i < sizeof(flags)/sizeof(DWORD); ++i) 1203 { 1204 memset(wbuf, 0xff, sizeof(wbuf)); 1205 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 3, NULL, 0); 1206 SetLastError( 0xdeadbeef ); 1207 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 3, wbuf, count); 1208 1209 if (flags[i] & MB_ERR_INVALID_CHARS) 1210 { 1211 ok(count == 0, "%04x: returned %d (expected 0)\n", flags[i], count); 1212 ok(count2 == 0, "%04x: returned %d (expected 0)\n", flags[i], count2); 1213 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04x: returned %d (expected %d)\n", 1214 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION); 1215 } 1216 else 1217 { 1218 ok(count == 2, "%04x: returned %d (expected 2)\n", flags[i], count); 1219 ok(count2 == 2, "%04x: returned %d (expected 2)\n", flags[i], count2); 1220 ok(wbuf[0] == 0x770b, "%04x: returned %04x (expected 770b)\n", flags[i], wbuf[0]); 1221 ok(wbuf[1] == 0x003f || broken(wbuf[1] == 0), /*windows xp*/ 1222 "%04x: wrong wide char: %04x\n", flags[i], wbuf[1]); 1223 ok(wbuf[2] == 0xffff, "%04x: returned %04x (expected ffff)\n", flags[i], wbuf[2]); 1224 } 1225 } 1226 1227 /* src ends with null character */ 1228 for (i = 0; i < sizeof(flags)/sizeof(DWORD); ++i) 1229 { 1230 memset(wbuf, 0xff, sizeof(wbuf)); 1231 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 4, NULL, 0); 1232 SetLastError( 0xdeadbeef ); 1233 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 4, wbuf, count); 1234 ok(count == count2, "%04x: returned %d (expected %d)\n", flags[i], count2, count); 1235 1236 if (flags[i] & MB_ERR_INVALID_CHARS) 1237 { 1238 ok(count == 0, "%04x: returned %d (expected 0)\n", flags[i], count); 1239 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04x: returned %d (expected %d)\n", 1240 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION); 1241 } 1242 else 1243 { 1244 WCHAR wbuf_ok[] = { 0x770b, 0x003f, '\0', 0xffff }; 1245 WCHAR wbuf_broken[] = { 0x770b, '\0', 0xffff, 0xffff }; 1246 ok(count == 3 || broken(count == 2 /*windows xp*/), 1247 "%04x: returned %d (expected 3)\n", flags[i], count); 1248 ok(!memcmp(wbuf, wbuf_ok, sizeof(wbuf_ok)) 1249 || broken(!memcmp(wbuf, wbuf_broken, sizeof(wbuf_broken))), 1250 "%04x: returned %04x %04x %04x %04x (expected %04x %04x %04x %04x)\n", 1251 flags[i], wbuf[0], wbuf[1], wbuf[2], wbuf[3], 1252 wbuf_ok[0], wbuf_ok[1], wbuf_ok[2], wbuf_ok[3]); 1253 } 1254 } 1255 1256 /* src has null character, but not ends with it */ 1257 for (i = 0; i < sizeof(flags)/sizeof(DWORD); ++i) 1258 { 1259 memset(wbuf, 0xff, sizeof(wbuf)); 1260 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 5, NULL, 0); 1261 SetLastError( 0xdeadbeef ); 1262 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 5, wbuf, count); 1263 ok(count == count2, "%04x: returned %d (expected %d)\n", flags[i], count2, count); 1264 1265 if (flags[i] & MB_ERR_INVALID_CHARS) 1266 { 1267 ok(count == 0, "%04x: returned %d (expected 0)\n", flags[i], count); 1268 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04x: returned %d (expected %d)\n", 1269 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION); 1270 } 1271 else 1272 { 1273 WCHAR wbuf_ok[] = { 0x770b, 0x003f, '\0', 'x', 0xffff }; 1274 WCHAR wbuf_broken[] = { 0x770b, '\0', 'x', 0xffff, 0xffff }; 1275 ok(count == 4 || broken(count == 3), 1276 "%04x: returned %d (expected 4)\n", flags[i], count); 1277 ok(!memcmp(wbuf, wbuf_ok, sizeof(wbuf_ok)) 1278 || broken(!memcmp(wbuf, wbuf_broken, sizeof(wbuf_broken))), 1279 "%04x: returned %04x %04x %04x %04x %04x (expected %04x %04x %04x %04x %04x)\n", 1280 flags[i], wbuf[0], wbuf[1], wbuf[2], wbuf[3], wbuf[4], 1281 wbuf_ok[0], wbuf_ok[1], wbuf_ok[2], wbuf_ok[3], wbuf_ok[4]); 1282 } 1283 } 1284 } 1285 1286 START_TEST(codepage) 1287 { 1288 BOOL bUsedDefaultChar; 1289 1290 test_destination_buffer(); 1291 test_null_source(); 1292 test_negative_source_length(); 1293 test_negative_dest_length(); 1294 test_other_invalid_parameters(); 1295 test_overlapped_buffers(); 1296 1297 /* WideCharToMultiByte has two code paths, test both here */ 1298 test_string_conversion(NULL); 1299 test_string_conversion(&bUsedDefaultChar); 1300 1301 test_utf7_encoding(); 1302 test_utf7_decoding(); 1303 1304 test_undefined_byte_char(); 1305 test_threadcp(); 1306 1307 test_dbcs_to_widechar(); 1308 } 1309