huwcampbell.com
Reactos
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 <stdarg.h>
23#include <stdio.h>
24#include <limits.h>
25
26#include "wine/test.h"
27#include "windef.h"
28#include "winbase.h"
29#include "winnls.h"
30
31static const char foobarA[] = "foobar";
32static const WCHAR foobarW[] = {'f','o','o','b','a','r',0};
33
34static void test_destination_buffer(void)
35{
36 LPSTR buffer;
37 INT maxsize;
38 INT needed;
39 INT len;
40
41 SetLastError(0xdeadbeef);
42 needed = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL);
43 ok( (needed > 0), "returned %d with %lu (expected '> 0')\n",
44 needed, GetLastError());
45
46 maxsize = needed*2;
47 buffer = HeapAlloc(GetProcessHeap(), 0, maxsize);
48 if (buffer == NULL) return;
49
50 maxsize--;
51 memset(buffer, 'x', maxsize);
52 buffer[maxsize] = '\0';
53 SetLastError(0xdeadbeef);
54 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed+1, NULL, NULL);
55 ok( (len > 0), "returned %d with %lu and '%s' (expected '> 0')\n",
56 len, GetLastError(), buffer);
57 ok(!lstrcmpA(buffer, "foobar"), "expected \"foobar\" got \"%s\"\n", buffer);
58
59 memset(buffer, 'x', maxsize);
60 buffer[maxsize] = '\0';
61 SetLastError(0xdeadbeef);
62 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed, NULL, NULL);
63 ok( (len > 0), "returned %d with %lu and '%s' (expected '> 0')\n",
64 len, GetLastError(), buffer);
65 ok(!lstrcmpA(buffer, "foobar"), "expected \"foobar\" got \"%s\"\n", buffer);
66
67 memset(buffer, 'x', maxsize);
68 buffer[maxsize] = '\0';
69 SetLastError(0xdeadbeef);
70 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, needed-1, NULL, NULL);
71 ok( !len && (GetLastError() == ERROR_INSUFFICIENT_BUFFER),
72 "returned %d with %lu and '%s' (expected '0' with "
73 "ERROR_INSUFFICIENT_BUFFER)\n", len, GetLastError(), buffer);
74 ok(!strncmp(buffer, "foobar", 6), "expected \"foobar\" got \"%s\"\n", buffer);
75 ok(buffer[6] == 'x', "expected buf[5]=='x', got \"%s\"\n", buffer);
76
77 memset(buffer, 'x', maxsize);
78 buffer[maxsize] = '\0';
79 SetLastError(0xdeadbeef);
80 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, 1, NULL, NULL);
81 ok( !len && (GetLastError() == ERROR_INSUFFICIENT_BUFFER),
82 "returned %d with %lu and '%s' (expected '0' with "
83 "ERROR_INSUFFICIENT_BUFFER)\n", len, GetLastError(), buffer);
84 ok(buffer[0] == 'f', "expected buf[1]=='f', got \"%s\"\n", buffer);
85 ok(buffer[1] == 'x', "expected buf[1]=='x', got \"%s\"\n", buffer);
86
87 SetLastError(0xdeadbeef);
88 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, buffer, 0, NULL, NULL);
89 ok( (len > 0), "returned %d with %lu (expected '> 0')\n",
90 len, GetLastError());
91
92 SetLastError(0xdeadbeef);
93 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, needed, NULL, NULL);
94 ok( !len && (GetLastError() == ERROR_INVALID_PARAMETER),
95 "returned %d with %lu (expected '0' with "
96 "ERROR_INVALID_PARAMETER)\n", len, GetLastError());
97
98 HeapFree(GetProcessHeap(), 0, buffer);
99}
100
101
102static void test_null_source(void)
103{
104 int len;
105 DWORD GLE;
106
107 SetLastError(0);
108 len = WideCharToMultiByte(CP_ACP, 0, NULL, 0, NULL, 0, NULL, NULL);
109 GLE = GetLastError();
110 ok(!len && GLE == ERROR_INVALID_PARAMETER,
111 "WideCharToMultiByte returned %d with GLE=%lu (expected 0 with ERROR_INVALID_PARAMETER)\n",
112 len, GLE);
113
114 SetLastError(0);
115 len = WideCharToMultiByte(CP_ACP, 0, NULL, -1, NULL, 0, NULL, NULL);
116 GLE = GetLastError();
117 ok(!len && GLE == ERROR_INVALID_PARAMETER,
118 "WideCharToMultiByte returned %d with GLE=%lu (expected 0 with ERROR_INVALID_PARAMETER)\n",
119 len, GLE);
120}
121
122static void test_negative_source_length(void)
123{
124 int len;
125 char buf[10];
126 WCHAR bufW[10];
127
128 /* Test, whether any negative source length works as strlen() + 1 */
129 SetLastError( 0xdeadbeef );
130 memset(buf,'x',sizeof(buf));
131 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -2002, buf, 10, NULL, NULL);
132 ok(len == 7 && GetLastError() == 0xdeadbeef,
133 "WideCharToMultiByte(-2002): len=%d error=%lu\n", len, GetLastError());
134 ok(!lstrcmpA(buf, "foobar"),
135 "WideCharToMultiByte(-2002): expected \"foobar\" got \"%s\"\n", buf);
136
137 SetLastError( 0xdeadbeef );
138 memset(bufW,'x',sizeof(bufW));
139 len = MultiByteToWideChar(CP_ACP, 0, "foobar", -2002, bufW, 10);
140 ok(len == 7 && !lstrcmpW(bufW, foobarW) && GetLastError() == 0xdeadbeef,
141 "MultiByteToWideChar(-2002): len=%d error=%lu\n", len, GetLastError());
142
143 SetLastError(0xdeadbeef);
144 memset(bufW, 'x', sizeof(bufW));
145 len = MultiByteToWideChar(CP_ACP, 0, "foobar", -1, bufW, 6);
146 ok(len == 0 && GetLastError() == ERROR_INSUFFICIENT_BUFFER,
147 "MultiByteToWideChar(-1): len=%d error=%lu\n", len, GetLastError());
148}
149
150#define LONGBUFLEN 100000
151static void test_negative_dest_length(void)
152{
153 int len, i;
154 static WCHAR bufW[LONGBUFLEN];
155 static char bufA[LONGBUFLEN];
156 static WCHAR originalW[LONGBUFLEN];
157 static char originalA[LONGBUFLEN];
158 DWORD theError;
159
160 /* Test return on -1 dest length */
161 SetLastError( 0xdeadbeef );
162 memset(bufA,'x',sizeof(bufA));
163 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, -1, NULL, NULL);
164 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER,
165 "WideCharToMultiByte(destlen -1): len=%d error=%lx\n", len, GetLastError());
166
167 SetLastError( 0xdeadbeef );
168 memset(bufW,'x',sizeof(bufW));
169 len = MultiByteToWideChar(CP_ACP, 0, foobarA, -1, bufW, -1);
170 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER,
171 "MultiByteToWideChar(destlen -1): len=%d error=%lx\n", len, GetLastError());
172
173 /* Test return on -1000 dest length */
174 SetLastError( 0xdeadbeef );
175 memset(bufA,'x',sizeof(bufA));
176 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, -1000, NULL, NULL);
177 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER,
178 "WideCharToMultiByte(destlen -1000): len=%d error=%lx\n", len, GetLastError());
179
180 SetLastError( 0xdeadbeef );
181 memset(bufW,'x',sizeof(bufW));
182 len = MultiByteToWideChar(CP_ACP, 0, foobarA, -1000, bufW, -1);
183 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER,
184 "MultiByteToWideChar(destlen -1000): len=%d error=%lx\n", len, GetLastError());
185
186 /* Test return on INT_MAX dest length */
187 SetLastError( 0xdeadbeef );
188 memset(bufA,'x',sizeof(bufA));
189 len = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, bufA, INT_MAX, NULL, NULL);
190 ok(len == 7 && !lstrcmpA(bufA, "foobar") && GetLastError() == 0xdeadbeef,
191 "WideCharToMultiByte(destlen INT_MAX): len=%d error=%lx\n", len, GetLastError());
192
193 /* Test return on INT_MAX dest length and very long input */
194 SetLastError( 0xdeadbeef );
195 memset(bufA,'x',sizeof(bufA));
196 for (i=0; i < LONGBUFLEN - 1; i++) {
197 originalW[i] = 'Q';
198 originalA[i] = 'Q';
199 }
200 originalW[LONGBUFLEN-1] = 0;
201 originalA[LONGBUFLEN-1] = 0;
202 len = WideCharToMultiByte(CP_ACP, 0, originalW, -1, bufA, INT_MAX, NULL, NULL);
203 theError = GetLastError();
204 ok(len == LONGBUFLEN && !lstrcmpA(bufA, originalA) && theError == 0xdeadbeef,
205 "WideCharToMultiByte(srclen %d, destlen INT_MAX): len %d error=%lx\n", LONGBUFLEN, len, theError);
206
207}
208
209static void test_other_invalid_parameters(void)
210{
211 char c_string[] = "Hello World";
212 size_t c_string_len = sizeof(c_string);
213 WCHAR w_string[] = {'H','e','l','l','o',' ','W','o','r','l','d',0};
214 size_t w_string_len = ARRAY_SIZE(w_string);
215 BOOL used;
216 INT len;
217
218 /* Unrecognized flag => ERROR_INVALID_FLAGS */
219 SetLastError(0xdeadbeef);
220 len = WideCharToMultiByte(CP_ACP, 0x100, w_string, -1, c_string, c_string_len, NULL, NULL);
221 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%lx\n", len, GetLastError());
222
223 SetLastError(0xdeadbeef);
224 len = WideCharToMultiByte(CP_ACP, 0x800, w_string, -1, c_string, c_string_len, NULL, NULL);
225 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%lx\n", len, GetLastError());
226
227 SetLastError(0xdeadbeef);
228 len = MultiByteToWideChar(CP_ACP, 0x10, c_string, -1, w_string, w_string_len);
229 ok(len == 0 && GetLastError() == ERROR_INVALID_FLAGS, "len=%d error=%lx\n", len, GetLastError());
230
231
232 /* Unrecognized flag and invalid codepage => ERROR_INVALID_PARAMETER */
233 SetLastError(0xdeadbeef);
234 len = WideCharToMultiByte(0xdeadbeef, 0x100, w_string, w_string_len, c_string, c_string_len, NULL, NULL);
235 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
236
237 SetLastError(0xdeadbeef);
238 len = MultiByteToWideChar(0xdeadbeef, 0x10, c_string, c_string_len, w_string, w_string_len);
239 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
240
241
242 /* Unrecognized flag and src is NULL => ERROR_INVALID_PARAMETER */
243 SetLastError(0xdeadbeef);
244 len = WideCharToMultiByte(CP_ACP, 0x100, NULL, -1, c_string, c_string_len, NULL, NULL);
245 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
246
247 SetLastError(0xdeadbeef);
248 len = MultiByteToWideChar(CP_ACP, 0x10, NULL, -1, w_string, w_string_len);
249 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
250
251
252 /* srclen=0 => ERROR_INVALID_PARAMETER */
253 SetLastError(0xdeadbeef);
254 len = WideCharToMultiByte(CP_ACP, 0, w_string, 0, c_string, c_string_len, NULL, NULL);
255 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
256
257 SetLastError(0xdeadbeef);
258 len = MultiByteToWideChar(CP_ACP, 0, c_string, 0, w_string, w_string_len);
259 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
260
261
262 /* dst=NULL but dstlen not 0 => ERROR_INVALID_PARAMETER */
263 SetLastError(0xdeadbeef);
264 len = WideCharToMultiByte(CP_ACP, 0, w_string, w_string_len, NULL, c_string_len, NULL, NULL);
265 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
266
267 SetLastError(0xdeadbeef);
268 len = MultiByteToWideChar(CP_ACP, 0, c_string, c_string_len, NULL, w_string_len);
269 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
270
271
272 /* CP_UTF7 or CP_SYMBOL and defchar not NULL => ERROR_INVALID_PARAMETER */
273 /* CP_UTF8 allows it since win10 1709 */
274 SetLastError(0xdeadbeef);
275 len = WideCharToMultiByte(CP_UTF7, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL);
276 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
277
278 SetLastError(0xdeadbeef);
279 len = WideCharToMultiByte(CP_UTF8, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL);
280 ok(len == 12 || broken(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER),
281 "len=%d error=%lx\n", len, GetLastError());
282
283 SetLastError(0xdeadbeef);
284 len = WideCharToMultiByte(CP_SYMBOL, 0, w_string, w_string_len, c_string, c_string_len, c_string, NULL);
285 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
286
287
288 /* CP_UTF7 or CP_SYMBOL and used not NULL => ERROR_INVALID_PARAMETER */
289 /* CP_UTF8 allows it since win10 1709 */
290 SetLastError(0xdeadbeef);
291 len = WideCharToMultiByte(CP_UTF7, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used);
292 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
293
294 SetLastError(0xdeadbeef);
295 len = WideCharToMultiByte(CP_UTF8, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used);
296 ok(len == 12 || broken(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER),
297 "len=%d error=%lx\n", len, GetLastError());
298
299 SetLastError(0xdeadbeef);
300 len = WideCharToMultiByte(CP_SYMBOL, 0, w_string, w_string_len, c_string, c_string_len, NULL, &used);
301 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
302
303
304 /* CP_UTF7, flags not 0 and used not NULL => ERROR_INVALID_PARAMETER */
305 /* (tests precedence of ERROR_INVALID_PARAMETER over ERROR_INVALID_FLAGS) */
306 /* The same test with CP_SYMBOL instead of CP_UTF7 gives ERROR_INVALID_FLAGS
307 instead except on Windows NT4 */
308 SetLastError(0xdeadbeef);
309 len = WideCharToMultiByte(CP_UTF7, 1, w_string, w_string_len, c_string, c_string_len, NULL, &used);
310 ok(len == 0 && GetLastError() == ERROR_INVALID_PARAMETER, "len=%d error=%lx\n", len, GetLastError());
311
312 /* CP_UTF8, unrecognized flag and used not NULL => ERROR_INVALID_PARAMETER */
313 SetLastError(0xdeadbeef);
314 len = WideCharToMultiByte(CP_UTF8, 0x100, w_string, w_string_len, c_string, c_string_len, NULL, &used);
315 ok(len == 0, "wrong ret %d\n", len);
316 ok(GetLastError() == ERROR_INVALID_PARAMETER
317 || GetLastError() == ERROR_INVALID_FLAGS /* Win10 1709+ */, "wrong error %lu\n", GetLastError());
318}
319
320static void test_overlapped_buffers(void)
321{
322 static const WCHAR strW[] = {'j','u','s','t',' ','a',' ','t','e','s','t',0};
323 static const char strA[] = "just a test";
324 char buf[256];
325 int ret;
326
327 /* limit such that strA's NUL terminator overlaps strW's NUL */
328 size_t overlap_limit = (sizeof(strW)-sizeof(strA)) - (sizeof(strW[0])-sizeof(strA[0]));
329
330 SetLastError(0xdeadbeef);
331 memcpy(buf + 1, strW, sizeof(strW));
332 ret = WideCharToMultiByte(CP_ACP, 0, (WCHAR *)(buf + 1), -1, buf, sizeof(buf), NULL, NULL);
333 ok(ret == sizeof(strA), "unexpected ret %d\n", ret);
334 ok(!memcmp(buf, strA, sizeof(strA)), "conversion failed: %s\n", buf);
335 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
336
337 SetLastError(0xdeadbeef);
338 memcpy(buf + overlap_limit, strA, sizeof(strA));
339 ret = MultiByteToWideChar(CP_ACP, 0, buf + overlap_limit, -1, (WCHAR *)buf, sizeof(buf) / sizeof(WCHAR));
340 ok(ret == ARRAY_SIZE(strW), "unexpected ret %d\n", ret);
341 ok(!memcmp(buf, strW, sizeof(strW)), "conversion failed: %s\n", wine_dbgstr_wn((WCHAR *)buf, ARRAY_SIZE(strW)));
342 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
343}
344
345static void test_string_conversion(LPBOOL bUsedDefaultChar)
346{
347 char mbc;
348 char mbs[15];
349 int ret;
350 WCHAR wc1 = 228; /* Western Windows-1252 character */
351 WCHAR wc2 = 1088; /* Russian Windows-1251 character not displayable for Windows-1252 */
352 static const WCHAR wcs[] = {'T', 'h', 1088, 'i', 0}; /* String with ASCII characters and a Russian character */
353 static const WCHAR dbwcs[] = {28953, 25152, 0}; /* String with Chinese (codepage 950) characters */
354 static const WCHAR dbwcs2[] = {0x7bb8, 0x3d, 0xc813, 0xac00, 0xb77d, 0};
355 static const char default_char[] = {0xa3, 0xbf, 0};
356
357 SetLastError(0xdeadbeef);
358 ret = WideCharToMultiByte(1252, 0, &wc1, 1, &mbc, 1, NULL, bUsedDefaultChar);
359 ok(ret == 1, "ret is %d\n", ret);
360 ok(mbc == '\xe4', "mbc is %d\n", mbc);
361 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
362 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
363
364 SetLastError(0xdeadbeef);
365 ret = WideCharToMultiByte(1252, 0, &wc2, 1, &mbc, 1, NULL, bUsedDefaultChar);
366 ok(ret == 1, "ret is %d\n", ret);
367 ok(mbc == 63, "mbc is %d\n", mbc);
368 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
369 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
370
371 if (IsValidCodePage(1251))
372 {
373 SetLastError(0xdeadbeef);
374 ret = WideCharToMultiByte(1251, 0, &wc2, 1, &mbc, 1, NULL, bUsedDefaultChar);
375 ok(ret == 1, "ret is %d\n", ret);
376 ok(mbc == '\xf0', "mbc is %d\n", mbc);
377 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
378 ok(GetLastError() == 0xdeadbeef ||
379 broken(GetLastError() == 0), /* win95 */
380 "GetLastError() is %lu\n", GetLastError());
381
382 SetLastError(0xdeadbeef);
383 ret = WideCharToMultiByte(1251, 0, &wc1, 1, &mbc, 1, NULL, bUsedDefaultChar);
384 ok(ret == 1, "ret is %d\n", ret);
385 ok(mbc == 97, "mbc is %d\n", mbc);
386 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
387 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
388 }
389 else
390 skip("Codepage 1251 not available\n");
391
392 /* This call triggers the last Win32 error */
393 SetLastError(0xdeadbeef);
394 ret = WideCharToMultiByte(1252, 0, wcs, -1, &mbc, 1, NULL, bUsedDefaultChar);
395 ok(ret == 0, "ret is %d\n", ret);
396 ok(mbc == 84, "mbc is %d\n", mbc);
397 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
398 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %lu\n", GetLastError());
399
400 SetLastError(0xdeadbeef);
401 ret = WideCharToMultiByte(1252, 0, wcs, -1, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
402 ok(ret == 5, "ret is %d\n", ret);
403 ok(!strcmp(mbs, "Th?i"), "mbs is %s\n", mbs);
404 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
405 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
406 mbs[0] = 0;
407
408 /* WideCharToMultiByte mustn't add any null character automatically.
409 So in this case, we should get the same string again, even if we only copied the first three bytes. */
410 SetLastError(0xdeadbeef);
411 ret = WideCharToMultiByte(1252, 0, wcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
412 ok(ret == 3, "ret is %d\n", ret);
413 ok(!strcmp(mbs, "Th?i"), "mbs is %s\n", mbs);
414 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
415 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
416 ZeroMemory(mbs, 5);
417
418 /* Now this shouldn't be the case like above as we zeroed the complete string buffer. */
419 SetLastError(0xdeadbeef);
420 ret = WideCharToMultiByte(1252, 0, wcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
421 ok(ret == 3, "ret is %d\n", ret);
422 ok(!strcmp(mbs, "Th?"), "mbs is %s\n", mbs);
423 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
424 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
425
426 /* Double-byte tests */
427 ret = WideCharToMultiByte(1252, 0, dbwcs, 3, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
428 ok(ret == 3, "ret is %d\n", ret);
429 ok(!strcmp(mbs, "??"), "mbs is %s\n", mbs);
430 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
431
432 ret = WideCharToMultiByte(936, WC_COMPOSITECHECK, dbwcs2, -1, mbs, sizeof(mbs), (const char *)default_char, bUsedDefaultChar);
433 ok(ret == 10, "ret is %d\n", ret);
434 ok(!strcmp(mbs, "\xf3\xe7\x3d\xa3\xbf\xa3\xbf\xa3\xbf"), "mbs is %s\n", mbs);
435 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
436
437 /* Show that characters are not truncated */
438 ZeroMemory(mbs, 5);
439 ret = WideCharToMultiByte(936, 0, dbwcs2, -1, mbs, 1, (const char *)default_char, bUsedDefaultChar);
440 ok(!ret, "ret is %d\n", ret);
441 ok(mbs[0] == '\0', "mbs is %s\n", mbs);
442 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
443 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %lu\n", GetLastError());
444
445 /* And destination is not null-terminated even when too short */
446 FillMemory(mbs, 5, 'x');
447 mbs[5] = '\0';
448 ret = WideCharToMultiByte(936, 0, dbwcs2, -1, mbs, 2, (const char *)default_char, bUsedDefaultChar);
449 ok(!ret, "ret is %d\n", ret);
450 ok(!strcmp(mbs, "\xf3\xe7""xxx"), "mbs is %s\n", mbs);
451 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
452 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %lu\n", GetLastError());
453
454 /* Length-only tests */
455 SetLastError(0xdeadbeef);
456 ret = WideCharToMultiByte(1252, 0, &wc2, 1, NULL, 0, NULL, bUsedDefaultChar);
457 ok(ret == 1, "ret is %d\n", ret);
458 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
459 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
460
461 SetLastError(0xdeadbeef);
462 ret = WideCharToMultiByte(1252, 0, wcs, -1, NULL, 0, NULL, bUsedDefaultChar);
463 ok(ret == 5, "ret is %d\n", ret);
464 if(bUsedDefaultChar) ok(*bUsedDefaultChar == TRUE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
465 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
466
467 if (!IsValidCodePage(950))
468 {
469 skip("Codepage 950 not available\n");
470 return;
471 }
472
473 /* Double-byte tests */
474 SetLastError(0xdeadbeef);
475 ret = WideCharToMultiByte(950, 0, dbwcs, -1, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
476 ok(ret == 5, "ret is %d\n", ret);
477 ok(!strcmp(mbs, "\xb5H\xa9\xd2"), "mbs is %s\n", mbs);
478 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
479 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
480
481 SetLastError(0xdeadbeef);
482 ret = WideCharToMultiByte(950, 0, dbwcs, 1, &mbc, 1, NULL, bUsedDefaultChar);
483 ok(ret == 0, "ret is %d\n", ret);
484 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
485 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER, "GetLastError() is %lu\n", GetLastError());
486 ZeroMemory(mbs, 5);
487
488 SetLastError(0xdeadbeef);
489 ret = WideCharToMultiByte(950, 0, dbwcs, 1, mbs, sizeof(mbs), NULL, bUsedDefaultChar);
490 ok(ret == 2, "ret is %d\n", ret);
491 ok(!strcmp(mbs, "\xb5H"), "mbs is %s\n", mbs);
492 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
493 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
494
495 /* Length-only tests */
496 SetLastError(0xdeadbeef);
497 ret = WideCharToMultiByte(950, 0, dbwcs, 1, NULL, 0, NULL, bUsedDefaultChar);
498 ok(ret == 2, "ret is %d\n", ret);
499 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
500 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
501
502 SetLastError(0xdeadbeef);
503 ret = WideCharToMultiByte(950, 0, dbwcs, -1, NULL, 0, NULL, bUsedDefaultChar);
504 ok(ret == 5, "ret is %d\n", ret);
505 if(bUsedDefaultChar) ok(*bUsedDefaultChar == FALSE, "bUsedDefaultChar is %d\n", *bUsedDefaultChar);
506 ok(GetLastError() == 0xdeadbeef, "GetLastError() is %lu\n", GetLastError());
507}
508
509static void test_utf7_encoding(void)
510{
511 WCHAR input[16];
512 char output[16], expected[16];
513 int i, len, expected_len;
514
515 static const BOOL directly_encodable_table[] =
516 {
517 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, /* 0x00 - 0x0F */
518 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0x10 - 0x1F */
519 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, /* 0x20 - 0x2F */
520 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 0x30 - 0x3F */
521 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 0x40 - 0x4F */
522 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 0x50 - 0x5F */
523 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 0x60 - 0x6F */
524 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 /* 0x70 - 0x7F */
525 };
526 static const char base64_encoding_table[] =
527 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
528
529 const struct
530 {
531 /* inputs */
532 WCHAR src[16];
533 int srclen;
534 char *dst;
535 int dstlen;
536 /* expected outputs */
537 char expected_dst[16];
538 int chars_written;
539 int len;
540 }
541 tests[] =
542 {
543 /* tests string conversion with srclen=-1 */
544 {
545 {0x4F60,0x597D,0x5417,0}, -1, output, sizeof(output) - 1,
546 "+T2BZfVQX-", 11, 11
547 },
548 /* tests string conversion with srclen=-2 */
549 {
550 {0x4F60,0x597D,0x5417,0}, -2, output, sizeof(output) - 1,
551 "+T2BZfVQX-", 11, 11
552 },
553 /* tests string conversion with dstlen=strlen(expected_dst) */
554 {
555 {0x4F60,0x597D,0x5417,0}, -1, output, 10,
556 "+T2BZfVQX-", 10, 0
557 },
558 /* tests string conversion with dstlen=strlen(expected_dst)+1 */
559 {
560 {0x4F60,0x597D,0x5417,0}, -1, output, 11,
561 "+T2BZfVQX-", 11, 11
562 },
563 /* tests string conversion with dstlen=strlen(expected_dst)+2 */
564 {
565 {0x4F60,0x597D,0x5417,0}, -1, output, 12,
566 "+T2BZfVQX-", 11, 11
567 },
568 /* tests dry run with dst=NULL and dstlen=0 */
569 {
570 {0x4F60,0x597D,0x5417,0}, -1, NULL, 0,
571#if defined(__REACTOS__) && defined(_MSC_VER)
572 {0}, 0, 11
573#else
574 {}, 0, 11
575#endif
576 },
577 /* tests dry run with dst!=NULL and dstlen=0 */
578 {
579 {0x4F60,0x597D,0x5417,0}, -1, output, 0,
580#if defined(__REACTOS__) && defined(_MSC_VER)
581 {0}, 0, 11
582#else
583 {}, 0, 11
584#endif
585 },
586 /* tests srclen < strlenW(src) with directly encodable chars */
587 {
588 {'h','e','l','l','o',0}, 2, output, sizeof(output) - 1,
589 "he", 2, 2
590 },
591 /* tests srclen < strlenW(src) with non-directly encodable chars */
592 {
593 {0x4F60,0x597D,0x5417,0}, 2, output, sizeof(output) - 1,
594 "+T2BZfQ-", 8, 8
595 },
596 /* tests a single null char */
597 {
598 {0}, -1, output, sizeof(output) - 1,
599 "", 1, 1
600 },
601 /* tests a buffer that runs out while not encoding a UTF-7 sequence */
602 {
603 {'h','e','l','l','o',0}, -1, output, 2,
604 "he", 2, 0
605 },
606 /* tests a buffer that runs out after writing 1 base64 character */
607 {
608 {0x4F60,0x0001,0}, -1, output, 2,
609 "+T", 2, 0
610 },
611 /* tests a buffer that runs out after writing 2 base64 characters */
612 {
613 {0x4F60,0x0001,0}, -1, output, 3,
614 "+T2", 3, 0
615 },
616 /* tests a buffer that runs out after writing 3 base64 characters */
617 {
618 {0x4F60,0x0001,0}, -1, output, 4,
619 "+T2A", 4, 0
620 },
621 /* tests a buffer that runs out just after writing the + sign */
622 {
623 {0x4F60,0}, -1, output, 1,
624 "+", 1, 0
625 },
626 /* tests a buffer that runs out just before writing the - sign
627 * the number of bits to encode here is evenly divisible by 6 */
628 {
629 {0x4F60,0x597D,0x5417,0}, -1, output, 9,
630 "+T2BZfVQX", 9, 0
631 },
632 /* tests a buffer that runs out just before writing the - sign
633 * the number of bits to encode here is NOT evenly divisible by 6 */
634 {
635 {0x4F60,0}, -1, output, 4,
636 "+T2", 3, 0
637 },
638 /* tests a buffer that runs out in the middle of escaping a + sign */
639 {
640 {'+',0}, -1, output, 1,
641 "+", 1, 0
642 }
643 };
644
645 /* test which characters are encoded if surrounded by non-encoded characters */
646 for (i = 0; i <= 0xFFFF; i++)
647 {
648 input[0] = ' ';
649 input[1] = i;
650 input[2] = ' ';
651 input[3] = 0;
652
653 memset(output, '#', sizeof(output) - 1);
654 output[sizeof(output) - 1] = 0;
655
656 len = WideCharToMultiByte(CP_UTF7, 0, input, 4, output, sizeof(output) - 1, NULL, NULL);
657
658 if (i == '+')
659 {
660 /* '+' is a special case and is encoded as "+-" */
661 expected_len = 5;
662 strcpy(expected, " +- ");
663 }
664 else if (i <= 0x7F && directly_encodable_table[i])
665 {
666 /* encodes directly */
667 expected_len = 4;
668 sprintf(expected, " %c ", i);
669 }
670 else
671 {
672 /* base64-encodes */
673 expected_len = 8;
674 sprintf(expected, " +%c%c%c- ",
675 base64_encoding_table[(i & 0xFC00) >> 10],
676 base64_encoding_table[(i & 0x03F0) >> 4],
677 base64_encoding_table[(i & 0x000F) << 2]);
678 }
679
680 ok(len == expected_len, "i=0x%04x: expected len=%i, got len=%i\n", i, expected_len, len);
681 ok(memcmp(output, expected, expected_len) == 0,
682 "i=0x%04x: expected output='%s', got output='%s'\n", i, expected, output);
683 ok(output[expected_len] == '#', "i=0x%04x: expected output[%i]='#', got output[%i]=%i\n",
684 i, expected_len, expected_len, output[expected_len]);
685 }
686
687 /* test which one-byte characters are absorbed into surrounding base64 blocks
688 * (Windows always ends the base64 block when it encounters a directly encodable character) */
689 for (i = 0; i <= 0xFFFF; i++)
690 {
691 input[0] = 0x2672;
692 input[1] = i;
693 input[2] = 0x2672;
694 input[3] = 0;
695
696 memset(output, '#', sizeof(output) - 1);
697 output[sizeof(output) - 1] = 0;
698
699 len = WideCharToMultiByte(CP_UTF7, 0, input, 4, output, sizeof(output) - 1, NULL, NULL);
700
701 if (i == '+')
702 {
703 /* '+' is a special case and is encoded as "+-" */
704 expected_len = 13;
705 strcpy(expected, "+JnI-+-+JnI-");
706 }
707 else if (i <= 0x7F && directly_encodable_table[i])
708 {
709 /* encodes directly */
710 expected_len = 12;
711 sprintf(expected, "+JnI-%c+JnI-", i);
712 }
713 else
714 {
715 /* base64-encodes */
716 expected_len = 11;
717 sprintf(expected, "+Jn%c%c%c%cZy-",
718 base64_encoding_table[8 | ((i & 0xC000) >> 14)],
719 base64_encoding_table[(i & 0x3F00) >> 8],
720 base64_encoding_table[(i & 0x00FC) >> 2],
721 base64_encoding_table[((i & 0x0003) << 4) | 2]);
722 }
723
724 ok(len == expected_len, "i=0x%04x: expected len=%i, got len=%i\n", i, expected_len, len);
725 ok(memcmp(output, expected, expected_len) == 0,
726 "i=0x%04x: expected output='%s', got output='%s'\n", i, expected, output);
727 ok(output[expected_len] == '#', "i=0x%04x: expected output[%i]='#', got output[%i]=%i\n",
728 i, expected_len, expected_len, output[expected_len]);
729 }
730
731 for (i = 0; i < ARRAY_SIZE(tests); i++)
732 {
733 memset(output, '#', sizeof(output) - 1);
734 output[sizeof(output) - 1] = 0;
735 SetLastError(0xdeadbeef);
736
737 len = WideCharToMultiByte(CP_UTF7, 0, tests[i].src, tests[i].srclen,
738 tests[i].dst, tests[i].dstlen, NULL, NULL);
739
740 if (!tests[i].len)
741 {
742 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER,
743 "tests[%i]: expected error=0x%x, got error=0x%lx\n",
744 i, ERROR_INSUFFICIENT_BUFFER, GetLastError());
745 }
746 ok(len == tests[i].len, "tests[%i]: expected len=%i, got len=%i\n", i, tests[i].len, len);
747
748 if (tests[i].dst)
749 {
750 ok(memcmp(tests[i].dst, tests[i].expected_dst, tests[i].chars_written) == 0,
751 "tests[%i]: expected dst='%s', got dst='%s'\n",
752 i, tests[i].expected_dst, tests[i].dst);
753 ok(tests[i].dst[tests[i].chars_written] == '#',
754 "tests[%i]: expected dst[%i]='#', got dst[%i]=%i\n",
755 i, tests[i].chars_written, tests[i].chars_written, tests[i].dst[tests[i].chars_written]);
756 }
757 }
758}
759
760static void test_utf7_decoding(void)
761{
762 char input[32];
763 WCHAR output[32], expected[32];
764 int i, len, expected_len;
765
766 static const signed char base64_decoding_table[] =
767 {
768 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0x00-0x0F */
769 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0x10-0x1F */
770 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1, -1, 63, /* 0x20-0x2F */
771 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, /* 0x30-0x3F */
772 -1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 0x40-0x4F */
773 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, -1, /* 0x50-0x5F */
774 -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, /* 0x60-0x6F */
775 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1 /* 0x70-0x7F */
776 };
777
778 struct
779 {
780 /* inputs */
781 char src[32];
782 int srclen;
783 WCHAR *dst;
784 int dstlen;
785 /* expected outputs */
786 WCHAR expected_dst[32];
787 int chars_written;
788 int len;
789 }
790 tests[] =
791 {
792 /* tests string conversion with srclen=-1 */
793 {
794 "+T2BZfQ-", -1, output, ARRAY_SIZE(output) - 1,
795 {0x4F60,0x597D,0}, 3, 3
796 },
797 /* tests string conversion with srclen=-2 */
798 {
799 "+T2BZfQ-", -2, output, ARRAY_SIZE(output) - 1,
800 {0x4F60,0x597D,0}, 3, 3
801 },
802 /* tests string conversion with dstlen=strlen(expected_dst) */
803 {
804 "+T2BZfQ-", -1, output, 2,
805 {0x4F60,0x597D}, 2, 0
806 },
807 /* tests string conversion with dstlen=strlen(expected_dst)+1 */
808 {
809 "+T2BZfQ-", -1, output, 3,
810 {0x4F60,0x597D,0}, 3, 3
811 },
812 /* tests string conversion with dstlen=strlen(expected_dst)+2 */
813 {
814 "+T2BZfQ-", -1, output, 4,
815 {0x4F60,0x597D,0}, 3, 3
816 },
817 /* tests dry run with dst=NULL and dstlen=0 */
818 {
819 "+T2BZfQ-", -1, NULL, 0,
820#if defined(__REACTOS__) && defined(_MSC_VER)
821 {0}, 0, 3
822#else
823 {}, 0, 3
824#endif
825 },
826 /* tests dry run with dst!=NULL and dstlen=0 */
827 {
828 "+T2BZfQ-", -1, output, 0,
829#if defined(__REACTOS__) && defined(_MSC_VER)
830 {0}, 0, 3
831#else
832 {}, 0, 3
833#endif
834 },
835 /* tests ill-formed UTF-7: 6 bits, not enough for a byte pair */
836 {
837 "+T-+T-+T-hello", -1, output, ARRAY_SIZE(output) - 1,
838 {'h','e','l','l','o',0}, 6, 6
839 },
840 /* tests ill-formed UTF-7: 12 bits, not enough for a byte pair */
841 {
842 "+T2-+T2-+T2-hello", -1, output, ARRAY_SIZE(output) - 1,
843 {'h','e','l','l','o',0}, 6, 6
844 },
845 /* tests ill-formed UTF-7: 18 bits, not a multiple of 16 and the last bit is a 1 */
846 {
847 "+T2B-+T2B-+T2B-hello", -1, output, ARRAY_SIZE(output) - 1,
848 {0x4F60,0x4F60,0x4F60,'h','e','l','l','o',0}, 9, 9
849 },
850 /* tests ill-formed UTF-7: 24 bits, a multiple of 8 but not a multiple of 16 */
851 {
852 "+T2BZ-+T2BZ-+T2BZ-hello", -1, output, ARRAY_SIZE(output) - 1,
853 {0x4F60,0x4F60,0x4F60,'h','e','l','l','o',0}, 9, 9
854 },
855 /* tests UTF-7 followed by characters that should be encoded but aren't */
856 {
857 "+T2BZ-\x82\xFE", -1, output, ARRAY_SIZE(output) - 1,
858 {0x4F60,0x0082,0x00FE,0}, 4, 4
859 },
860 /* tests srclen > strlen(src) */
861 {
862 "a\0b", 4, output, ARRAY_SIZE(output) - 1,
863 {'a',0,'b',0}, 4, 4
864 },
865 /* tests srclen < strlen(src) outside of a UTF-7 sequence */
866 {
867 "hello", 2, output, ARRAY_SIZE(output) - 1,
868 {'h','e'}, 2, 2
869 },
870 /* tests srclen < strlen(src) inside of a UTF-7 sequence */
871 {
872 "+T2BZfQ-", 4, output, ARRAY_SIZE(output) - 1,
873 {0x4F60}, 1, 1
874 },
875 /* tests srclen < strlen(src) right at the beginning of a UTF-7 sequence */
876 {
877 "hi+T2A-", 3, output, ARRAY_SIZE(output) - 1,
878 {'h','i'}, 2, 2
879 },
880 /* tests srclen < strlen(src) right at the end of a UTF-7 sequence */
881 {
882 "+T2A-hi", 5, output, ARRAY_SIZE(output) - 1,
883 {0x4F60}, 1, 1
884 },
885 /* tests srclen < strlen(src) at the beginning of an escaped + sign */
886 {
887 "hi+-", 3, output, ARRAY_SIZE(output) - 1,
888 {'h','i'}, 2, 2
889 },
890 /* tests srclen < strlen(src) at the end of an escaped + sign */
891 {
892 "+-hi", 2, output, ARRAY_SIZE(output) - 1,
893 {'+'}, 1, 1
894 },
895 /* tests len=0 but no error */
896 {
897 "+", 1, output, ARRAY_SIZE(output) - 1,
898#if defined(__REACTOS__) && defined(_MSC_VER)
899 {0}, 0, 0
900#else
901 {}, 0, 0
902#endif
903 },
904 /* tests a single null char */
905 {
906 "", -1, output, ARRAY_SIZE(output) - 1,
907 {0}, 1, 1
908 },
909 /* tests a buffer that runs out while not decoding a UTF-7 sequence */
910 {
911 "hello", -1, output, 2,
912 {'h','e'}, 2, 0
913 },
914 /* tests a buffer that runs out in the middle of decoding a UTF-7 sequence */
915 {
916 "+T2BZfQ-", -1, output, 1,
917 {0x4F60}, 1, 0
918 }
919 };
920
921 /* test which one-byte characters remove stray + signs */
922 for (i = 0; i < 256; i++)
923 {
924 sprintf(input, "+%c+AAA", i);
925
926 memset(output, 0x23, sizeof(output) - sizeof(WCHAR));
927 output[ARRAY_SIZE(output) - 1] = 0;
928
929 len = MultiByteToWideChar(CP_UTF7, 0, input, 7, output, ARRAY_SIZE(output) - 1);
930
931 if (i == '-')
932 {
933 /* removes the - sign */
934 expected_len = 3;
935 expected[0] = 0x002B;
936 expected[1] = 0;
937 expected[2] = 0;
938 }
939 else if (i <= 0x7F && base64_decoding_table[i] != -1)
940 {
941 /* absorbs the character into the base64 sequence */
942 expected_len = 2;
943 expected[0] = (base64_decoding_table[i] << 10) | 0x03E0;
944 expected[1] = 0;
945 }
946 else
947 {
948 /* removes the + sign */
949 expected_len = 3;
950 expected[0] = i;
951 expected[1] = 0;
952 expected[2] = 0;
953 }
954 expected[expected_len] = 0x2323;
955
956 ok(len == expected_len, "i=0x%02x: expected len=%i, got len=%i\n", i, expected_len, len);
957 ok(memcmp(output, expected, (expected_len + 1) * sizeof(WCHAR)) == 0,
958 "i=0x%02x: expected output=%s, got output=%s\n",
959 i, wine_dbgstr_wn(expected, expected_len + 1), wine_dbgstr_wn(output, expected_len + 1));
960 }
961
962 /* test which one-byte characters terminate a sequence
963 * also test whether the unfinished byte pair is discarded or not */
964 for (i = 0; i < 256; i++)
965 {
966 sprintf(input, "+B%c+AAA", i);
967
968 memset(output, 0x23, sizeof(output) - sizeof(WCHAR));
969 output[ARRAY_SIZE(output) - 1] = 0;
970
971 len = MultiByteToWideChar(CP_UTF7, 0, input, 8, output, ARRAY_SIZE(output) - 1);
972
973 if (i == '-')
974 {
975 /* explicitly terminates */
976 expected_len = 2;
977 expected[0] = 0;
978 expected[1] = 0;
979 }
980 else if (i <= 0x7F)
981 {
982 if (base64_decoding_table[i] != -1)
983 {
984 /* absorbs the character into the base64 sequence */
985 expected_len = 3;
986 expected[0] = 0x0400 | (base64_decoding_table[i] << 4) | 0x000F;
987 expected[1] = 0x8000;
988 expected[2] = 0;
989 }
990 else
991 {
992 /* implicitly terminates and discards the unfinished byte pair */
993 expected_len = 3;
994 expected[0] = i;
995 expected[1] = 0;
996 expected[2] = 0;
997 }
998 }
999 else
1000 {
1001 /* implicitly terminates but does not the discard unfinished byte pair */
1002 expected_len = 3;
1003 expected[0] = i;
1004 expected[1] = 0x0400;
1005 expected[2] = 0;
1006 }
1007 expected[expected_len] = 0x2323;
1008
1009 ok(len == expected_len, "i=0x%02x: expected len=%i, got len=%i\n", i, expected_len, len);
1010 ok(memcmp(output, expected, (expected_len + 1) * sizeof(WCHAR)) == 0,
1011 "i=0x%02x: expected output=%s, got output=%s\n",
1012 i, wine_dbgstr_wn(expected, expected_len + 1), wine_dbgstr_wn(output, expected_len + 1));
1013 }
1014
1015 for (i = 0; i < ARRAY_SIZE(tests); i++)
1016 {
1017 memset(output, 0x23, sizeof(output) - sizeof(WCHAR));
1018 output[ARRAY_SIZE(output) - 1] = 0;
1019 SetLastError(0xdeadbeef);
1020
1021 len = MultiByteToWideChar(CP_UTF7, 0, tests[i].src, tests[i].srclen,
1022 tests[i].dst, tests[i].dstlen);
1023
1024 tests[i].expected_dst[tests[i].chars_written] = 0x2323;
1025
1026 if (!tests[i].len && tests[i].chars_written)
1027 {
1028 ok(GetLastError() == ERROR_INSUFFICIENT_BUFFER,
1029 "tests[%i]: expected error=0x%x, got error=0x%lx\n",
1030 i, ERROR_INSUFFICIENT_BUFFER, GetLastError());
1031 }
1032 ok(len == tests[i].len, "tests[%i]: expected len=%i, got len=%i\n", i, tests[i].len, len);
1033
1034 if (tests[i].dst)
1035 {
1036 ok(memcmp(tests[i].dst, tests[i].expected_dst, (tests[i].chars_written + 1) * sizeof(WCHAR)) == 0,
1037 "tests[%i]: expected dst=%s, got dst=%s\n",
1038 i, wine_dbgstr_wn(tests[i].expected_dst, tests[i].chars_written + 1),
1039 wine_dbgstr_wn(tests[i].dst, tests[i].chars_written + 1));
1040 }
1041 }
1042}
1043
1044static void test_undefined_byte_char(void)
1045{
1046 static const struct tag_testset {
1047 UINT codepage;
1048 LPCSTR str;
1049 BOOL is_error;
1050 } testset[] = {
1051 { 37, "\x6f", FALSE },
1052 { 874, "\xdd", TRUE },
1053 { 932, "\xfe", TRUE },
1054 { 932, "\x80", FALSE },
1055 { 932, "\x81\x45", FALSE },
1056 { 936, "\xff", TRUE },
1057 { 949, "\xff", TRUE },
1058 { 950, "\xff", TRUE },
1059 { 1252, "?", FALSE },
1060 { 1252, "\x90", FALSE },
1061 { 1253, "\xaa", TRUE },
1062 { 1255, "\xff", TRUE },
1063 { 1257, "\xa5", TRUE },
1064 };
1065 INT i, ret;
1066 DWORD err;
1067
1068 for (i = 0; i < ARRAY_SIZE(testset); i++) {
1069 if (! IsValidCodePage(testset[i].codepage))
1070 {
1071 skip("Codepage %d not available\n", testset[i].codepage);
1072 continue;
1073 }
1074
1075 SetLastError(0xdeadbeef);
1076 ret = MultiByteToWideChar(testset[i].codepage, MB_ERR_INVALID_CHARS,
1077 testset[i].str, -1, NULL, 0);
1078 err = GetLastError();
1079 if (testset[i].is_error) {
1080 ok(err == ERROR_NO_UNICODE_TRANSLATION, "Test %u: err is %lu\n", i, err);
1081 ok(ret == 0, "Test %u: ret is %d\n", i, ret);
1082 }
1083 else {
1084 ok(err == 0xdeadbeef, "Test %u: err is %lu\n", i, err);
1085 ok(ret == 2, "Test %u: ret is %d\n", i, ret);
1086 }
1087
1088 SetLastError(0xdeadbeef);
1089 ret = MultiByteToWideChar(testset[i].codepage, 0,
1090 testset[i].str, -1, NULL, 0);
1091 err = GetLastError();
1092 ok(err == 0xdeadbeef, "Test %u: err is %lu\n", i, err);
1093 ok(ret == 2, "Test %u: ret is %d\n", i, ret);
1094 }
1095}
1096
1097static void test_threadcp(void)
1098{
1099 static const LCID ENGLISH = MAKELCID(MAKELANGID(LANG_ENGLISH, SUBLANG_ENGLISH_US), SORT_DEFAULT);
1100 static const LCID HINDI = MAKELCID(MAKELANGID(LANG_HINDI, SUBLANG_HINDI_INDIA), SORT_DEFAULT);
1101 static const LCID GEORGIAN = MAKELCID(MAKELANGID(LANG_GEORGIAN, SUBLANG_GEORGIAN_GEORGIA), SORT_DEFAULT);
1102 static const LCID RUSSIAN = MAKELCID(MAKELANGID(LANG_RUSSIAN, SUBLANG_RUSSIAN_RUSSIA), SORT_DEFAULT);
1103 static const LCID JAPANESE = MAKELCID(MAKELANGID(LANG_JAPANESE, SUBLANG_JAPANESE_JAPAN), SORT_DEFAULT);
1104 static const LCID CHINESE = MAKELCID(MAKELANGID(LANG_CHINESE, SUBLANG_CHINESE_SIMPLIFIED), SORT_DEFAULT);
1105
1106 BOOL islead, islead_default;
1107 CPINFOEXA cpi;
1108 UINT cp, acp;
1109 int i, num;
1110 LCID last;
1111 BOOL ret;
1112
1113 struct test {
1114 LCID lcid;
1115 UINT threadcp;
1116 } lcids[] = {
1117 { HINDI, 0 },
1118 { GEORGIAN, 0 },
1119 { ENGLISH, 1252 },
1120 { RUSSIAN, 1251 },
1121 { JAPANESE, 932 },
1122 { CHINESE, 936 }
1123 };
1124
1125 struct test_islead_nocp {
1126 LCID lcid;
1127 BYTE testchar;
1128 } isleads_nocp[] = {
1129 { HINDI, 0x00 },
1130 { HINDI, 0x81 },
1131 { HINDI, 0xa0 },
1132 { HINDI, 0xe0 },
1133
1134 { GEORGIAN, 0x00 },
1135 { GEORGIAN, 0x81 },
1136 { GEORGIAN, 0xa0 },
1137 { GEORGIAN, 0xe0 },
1138 };
1139
1140 struct test_islead {
1141 LCID lcid;
1142 BYTE testchar;
1143 BOOL islead;
1144 } isleads[] = {
1145 { ENGLISH, 0x00, FALSE },
1146 { ENGLISH, 0x81, FALSE },
1147 { ENGLISH, 0xa0, FALSE },
1148 { ENGLISH, 0xe0, FALSE },
1149
1150 { RUSSIAN, 0x00, FALSE },
1151 { RUSSIAN, 0x81, FALSE },
1152 { RUSSIAN, 0xa0, FALSE },
1153 { RUSSIAN, 0xe0, FALSE },
1154
1155 { JAPANESE, 0x00, FALSE },
1156 { JAPANESE, 0x81, TRUE },
1157 { JAPANESE, 0xa0, FALSE },
1158 { JAPANESE, 0xe0, TRUE },
1159
1160 { CHINESE, 0x00, FALSE },
1161 { CHINESE, 0x81, TRUE },
1162 { CHINESE, 0xa0, TRUE },
1163 { CHINESE, 0xe0, TRUE },
1164 };
1165
1166 last = GetThreadLocale();
1167 acp = GetACP();
1168
1169 for (i = 0; i < ARRAY_SIZE(lcids); i++)
1170 {
1171 SetThreadLocale(lcids[i].lcid);
1172
1173 cp = 0xdeadbeef;
1174 GetLocaleInfoA(lcids[i].lcid, LOCALE_IDEFAULTANSICODEPAGE|LOCALE_RETURN_NUMBER, (LPSTR)&cp, sizeof(cp));
1175 ok(cp == lcids[i].threadcp, "wrong codepage %u for lcid %04lx, should be %u\n", cp, lcids[i].lcid, lcids[i].threadcp);
1176
1177 /* GetCPInfoEx/GetCPInfo - CP_ACP */
1178 SetLastError(0xdeadbeef);
1179 memset(&cpi, 0, sizeof(cpi));
1180 ret = GetCPInfoExA(CP_ACP, 0, &cpi);
1181 ok(ret, "GetCPInfoExA failed for lcid %04lx, error %ld\n", lcids[i].lcid, GetLastError());
1182 ok(cpi.CodePage == acp, "wrong codepage %u for lcid %04lx, should be %u\n", cpi.CodePage, lcids[i].lcid, acp);
1183
1184 /* WideCharToMultiByte - CP_ACP */
1185 num = WideCharToMultiByte(CP_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL);
1186 ok(num == 7, "ret is %d (%04lx)\n", num, lcids[i].lcid);
1187
1188 /* MultiByteToWideChar - CP_ACP */
1189 num = MultiByteToWideChar(CP_ACP, 0, "foobar", -1, NULL, 0);
1190 ok(num == 7, "ret is %d (%04lx)\n", num, lcids[i].lcid);
1191
1192 /* GetCPInfoEx/GetCPInfo - CP_THREAD_ACP */
1193 SetLastError(0xdeadbeef);
1194 memset(&cpi, 0, sizeof(cpi));
1195 ret = GetCPInfoExA(CP_THREAD_ACP, 0, &cpi);
1196 ok(ret, "GetCPInfoExA failed for lcid %04lx, error %ld\n", lcids[i].lcid, GetLastError());
1197 if (lcids[i].threadcp)
1198 ok(cpi.CodePage == lcids[i].threadcp || cpi.CodePage == CP_UTF8 /* Win10 1809+ */,
1199 "wrong codepage %u for lcid %04lx, should be %u\n",
1200 cpi.CodePage, lcids[i].lcid, lcids[i].threadcp);
1201 else
1202 ok(cpi.CodePage == acp || cpi.CodePage == CP_UTF8 /* Win10 1809+ */,
1203 "wrong codepage %u for lcid %04lx, should be %u\n", cpi.CodePage, lcids[i].lcid, acp);
1204
1205 /* WideCharToMultiByte - CP_THREAD_ACP */
1206 num = WideCharToMultiByte(CP_THREAD_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL);
1207 ok(num == 7, "ret is %d (%04lx)\n", num, lcids[i].lcid);
1208
1209 /* MultiByteToWideChar - CP_THREAD_ACP */
1210 num = MultiByteToWideChar(CP_THREAD_ACP, 0, "foobar", -1, NULL, 0);
1211 ok(num == 7, "ret is %d (%04lx)\n", num, lcids[i].lcid);
1212 }
1213
1214 /* IsDBCSLeadByteEx - locales without codepage */
1215 for (i = 0; i < ARRAY_SIZE(isleads_nocp); i++)
1216 {
1217 SetThreadLocale(isleads_nocp[i].lcid);
1218
1219 GetCPInfoExA(CP_THREAD_ACP, 0, &cpi);
1220 islead_default = IsDBCSLeadByteEx(cpi.CodePage, isleads_nocp[i].testchar);
1221 islead = IsDBCSLeadByteEx(CP_THREAD_ACP, isleads_nocp[i].testchar);
1222
1223 ok(islead == islead_default, "wrong islead %i for test char %x in lcid %04lx. should be %i\n",
1224 islead, isleads_nocp[i].testchar, isleads_nocp[i].lcid, islead_default);
1225 }
1226
1227 /* IsDBCSLeadByteEx - locales with codepage */
1228 for (i = 0; i < ARRAY_SIZE(isleads); i++)
1229 {
1230 SetThreadLocale(isleads[i].lcid);
1231
1232 GetCPInfoExA(CP_THREAD_ACP, 0, &cpi);
1233 islead = IsDBCSLeadByteEx(CP_THREAD_ACP, isleads[i].testchar);
1234 ok(islead == isleads[i].islead || (cpi.CodePage == CP_UTF8 && !islead),
1235 "wrong islead %i for test char %x in lcid %04lx. should be %i\n",
1236 islead, isleads[i].testchar, isleads[i].lcid, isleads[i].islead);
1237 }
1238
1239 SetThreadLocale(last);
1240}
1241
1242static void test_dbcs_to_widechar(void)
1243{
1244 int i, count, count2;
1245 WCHAR wbuf[5];
1246 unsigned char buf[] = {0xbf, 0xb4, 0xc7, '\0', 'x'};
1247 static const DWORD flags[] = {
1248 MB_PRECOMPOSED,
1249 MB_COMPOSITE,
1250
1251 MB_PRECOMPOSED|MB_USEGLYPHCHARS,
1252 MB_COMPOSITE |MB_USEGLYPHCHARS,
1253
1254 MB_PRECOMPOSED|MB_ERR_INVALID_CHARS,
1255 MB_COMPOSITE |MB_ERR_INVALID_CHARS,
1256
1257 MB_PRECOMPOSED|MB_ERR_INVALID_CHARS|MB_USEGLYPHCHARS,
1258 MB_COMPOSITE |MB_ERR_INVALID_CHARS|MB_USEGLYPHCHARS,
1259 };
1260
1261 for (i = 0; i < ARRAY_SIZE(flags); ++i)
1262 {
1263 memset(wbuf, 0xff, sizeof(wbuf));
1264 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 2, NULL, 0);
1265 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 2, wbuf, count);
1266
1267 ok(count == 1, "%04lx: returned %d (expected 1)\n", flags[i], count);
1268 ok(count2 == 1, "%04lx: returned %d (expected 1)\n", flags[i], count2);
1269 ok(wbuf[0] == 0x770b, "%04lx: returned %04x (expected 770b)\n", flags[i], wbuf[0]);
1270 ok(wbuf[1] == 0xffff, "%04lx: returned %04x (expected ffff)\n", flags[i], wbuf[1]);
1271 }
1272
1273 for (i = 0; i < ARRAY_SIZE(flags); ++i)
1274 {
1275 memset(wbuf, 0xff, sizeof(wbuf));
1276 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 3, NULL, 0);
1277 SetLastError( 0xdeadbeef );
1278 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 3, wbuf, count);
1279
1280 if (flags[i] & MB_ERR_INVALID_CHARS)
1281 {
1282 ok(count == 0, "%04lx: returned %d (expected 0)\n", flags[i], count);
1283 ok(count2 == 0, "%04lx: returned %d (expected 0)\n", flags[i], count2);
1284 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04lx: returned %ld (expected %d)\n",
1285 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION);
1286 }
1287 else
1288 {
1289 ok(count == 2, "%04lx: returned %d (expected 2)\n", flags[i], count);
1290 ok(count2 == 2, "%04lx: returned %d (expected 2)\n", flags[i], count2);
1291 ok(wbuf[0] == 0x770b, "%04lx: returned %04x (expected 770b)\n", flags[i], wbuf[0]);
1292 ok(wbuf[1] == 0x003f || broken(wbuf[1] == 0), /*windows xp*/
1293 "%04lx: wrong wide char: %04x\n", flags[i], wbuf[1]);
1294 ok(wbuf[2] == 0xffff, "%04lx: returned %04x (expected ffff)\n", flags[i], wbuf[2]);
1295 }
1296 }
1297
1298 /* src ends with null character */
1299 for (i = 0; i < ARRAY_SIZE(flags); ++i)
1300 {
1301 memset(wbuf, 0xff, sizeof(wbuf));
1302 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 4, NULL, 0);
1303 SetLastError( 0xdeadbeef );
1304 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 4, wbuf, count);
1305 ok(count == count2, "%04lx: returned %d (expected %d)\n", flags[i], count2, count);
1306
1307 if (flags[i] & MB_ERR_INVALID_CHARS)
1308 {
1309 ok(count == 0, "%04lx: returned %d (expected 0)\n", flags[i], count);
1310 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04lx: returned %ld (expected %d)\n",
1311 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION);
1312 }
1313 else
1314 {
1315 WCHAR wbuf_ok[] = { 0x770b, 0x003f, '\0', 0xffff };
1316 WCHAR wbuf_broken[] = { 0x770b, '\0', 0xffff, 0xffff };
1317 ok(count == 3 || broken(count == 2 /*windows xp*/),
1318 "%04lx: returned %d (expected 3)\n", flags[i], count);
1319 ok(!memcmp(wbuf, wbuf_ok, sizeof(wbuf_ok))
1320 || broken(!memcmp(wbuf, wbuf_broken, sizeof(wbuf_broken))),
1321 "%04lx: returned %04x %04x %04x %04x (expected %04x %04x %04x %04x)\n",
1322 flags[i], wbuf[0], wbuf[1], wbuf[2], wbuf[3],
1323 wbuf_ok[0], wbuf_ok[1], wbuf_ok[2], wbuf_ok[3]);
1324 }
1325 }
1326
1327 /* src has null character, but not ends with it */
1328 for (i = 0; i < ARRAY_SIZE(flags); ++i)
1329 {
1330 memset(wbuf, 0xff, sizeof(wbuf));
1331 count = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 5, NULL, 0);
1332 SetLastError( 0xdeadbeef );
1333 count2 = MultiByteToWideChar(936, flags[i], (char*)&buf[0], 5, wbuf, count);
1334 ok(count == count2, "%04lx: returned %d (expected %d)\n", flags[i], count2, count);
1335
1336 if (flags[i] & MB_ERR_INVALID_CHARS)
1337 {
1338 ok(count == 0, "%04lx: returned %d (expected 0)\n", flags[i], count);
1339 ok(GetLastError() == ERROR_NO_UNICODE_TRANSLATION, "%04lx: returned %ld (expected %d)\n",
1340 flags[i], GetLastError(), ERROR_NO_UNICODE_TRANSLATION);
1341 }
1342 else
1343 {
1344 WCHAR wbuf_ok[] = { 0x770b, 0x003f, '\0', 'x', 0xffff };
1345 WCHAR wbuf_broken[] = { 0x770b, '\0', 'x', 0xffff, 0xffff };
1346 ok(count == 4 || broken(count == 3),
1347 "%04lx: returned %d (expected 4)\n", flags[i], count);
1348 ok(!memcmp(wbuf, wbuf_ok, sizeof(wbuf_ok))
1349 || broken(!memcmp(wbuf, wbuf_broken, sizeof(wbuf_broken))),
1350 "%04lx: returned %04x %04x %04x %04x %04x (expected %04x %04x %04x %04x %04x)\n",
1351 flags[i], wbuf[0], wbuf[1], wbuf[2], wbuf[3], wbuf[4],
1352 wbuf_ok[0], wbuf_ok[1], wbuf_ok[2], wbuf_ok[3], wbuf_ok[4]);
1353 }
1354 }
1355}
1356
1357START_TEST(codepage)
1358{
1359 BOOL bUsedDefaultChar;
1360
1361 test_destination_buffer();
1362 test_null_source();
1363 test_negative_source_length();
1364 test_negative_dest_length();
1365 test_other_invalid_parameters();
1366 test_overlapped_buffers();
1367
1368 /* WideCharToMultiByte has two code paths, test both here */
1369 test_string_conversion(NULL);
1370 test_string_conversion(&bUsedDefaultChar);
1371
1372 test_utf7_encoding();
1373 test_utf7_decoding();
1374
1375 test_undefined_byte_char();
1376 test_threadcp();
1377
1378 test_dbcs_to_widechar();
1379}