VirtualBox

source: vbox/trunk/src/VBox/Runtime/testcase/tstUtf8.cpp@ 85678

Last change on this file since 85678 was 83886, checked in by vboxsync, 5 years ago

IPRT: Implement RTStrEnd in assembly & C++, drop the inlining. memchr from VC++ 14.1 cannot deal wtih RTSTR_MAX on unaligned string, it may step into an electric fence page after the end of the string. bugref:8489
  • Property svn:eol-style set to native
  • Property svn:keywords set to Id Revision
File size: 57.0 KB
Line 
1/* $Id: tstUtf8.cpp 83886 2020-04-21 11:14:20Z vboxsync $ */
2/** @file
3 * IPRT Testcase - UTF-8 and UTF-16 string conversions.
4 */
5
6/*
7 * Copyright (C) 2006-2020 Oracle Corporation
8 *
9 * This file is part of VirtualBox Open Source Edition (OSE), as
10 * available from http://www.virtualbox.org. This file is free software;
11 * you can redistribute it and/or modify it under the terms of the GNU
12 * General Public License (GPL) as published by the Free Software
13 * Foundation, in version 2 as it comes in the "COPYING" file of the
14 * VirtualBox OSE distribution. VirtualBox OSE is distributed in the
15 * hope that it will be useful, but WITHOUT ANY WARRANTY of any kind.
16 *
17 * The contents of this file may alternatively be used under the terms
18 * of the Common Development and Distribution License Version 1.0
19 * (CDDL) only, as it comes in the "COPYING.CDDL" file of the
20 * VirtualBox OSE distribution, in which case the provisions of the
21 * CDDL are applicable instead of those of the GPL.
22 *
23 * You may elect to license modified versions of this file under the
24 * terms and conditions of either the GPL or the CDDL or both.
25 */
26
27
28/*********************************************************************************************************************************
29* Header Files *
30*********************************************************************************************************************************/
31#include <iprt/string.h>
32#include <iprt/latin1.h>
33#include <iprt/utf16.h>
34
35#include <iprt/alloc.h>
36#include <iprt/assert.h>
37#include <iprt/env.h>
38#include <iprt/err.h>
39#include <iprt/rand.h>
40#include <iprt/stream.h>
41#include <iprt/test.h>
42#include <iprt/time.h>
43#include <iprt/uni.h>
44#include <iprt/uuid.h>
45
46
47
48/**
49 * Generate a random codepoint for simple UTF-16 encoding.
50 */
51static RTUTF16 GetRandUtf16(void)
52{
53 RTUTF16 wc;
54 do
55 {
56 wc = (RTUTF16)RTRandU32Ex(1, 0xfffd);
57 } while (wc >= 0xd800 && wc <= 0xdfff);
58 return wc;
59}
60
61
62/**
63 *
64 */
65static void test1(RTTEST hTest)
66{
67 static const char s_szBadString1[] = "Bad \xe0\x13\x0";
68 static const char s_szBadString2[] = "Bad \xef\xbf\xc3";
69 int rc;
70 char *pszUtf8;
71 char *pszCurrent;
72 PRTUTF16 pwsz;
73 PRTUTF16 pwszRand;
74
75 /*
76 * Invalid UTF-8 to UCS-2 test.
77 */
78 RTTestSub(hTest, "Feeding bad UTF-8 to RTStrToUtf16");
79 rc = RTStrToUtf16(s_szBadString1, &pwsz);
80 RTTEST_CHECK_MSG(hTest, rc == VERR_NO_TRANSLATION || rc == VERR_INVALID_UTF8_ENCODING,
81 (hTest, "Conversion of first bad UTF-8 string to UTF-16 apparently succeeded. It shouldn't. rc=%Rrc\n", rc));
82 rc = RTStrToUtf16(s_szBadString2, &pwsz);
83 RTTEST_CHECK_MSG(hTest, rc == VERR_NO_TRANSLATION || rc == VERR_INVALID_UTF8_ENCODING,
84 (hTest, "Conversion of second bad UTF-8 strings to UTF-16 apparently succeeded. It shouldn't. rc=%Rrc\n", rc));
85
86 /*
87 * Test current CP conversion.
88 */
89 RTTestSub(hTest, "Rand UTF-16 -> UTF-8 -> CP -> UTF-8");
90 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
91 for (int i = 0; i < 30; i++)
92 pwszRand[i] = GetRandUtf16();
93 pwszRand[30] = 0;
94
95 rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
96 if (rc == VINF_SUCCESS)
97 {
98 rc = RTStrUtf8ToCurrentCP(&pszCurrent, pszUtf8);
99 if (rc == VINF_SUCCESS)
100 {
101 rc = RTStrCurrentCPToUtf8(&pszUtf8, pszCurrent);
102 if (rc == VINF_SUCCESS)
103 RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> Current -> UTF-8 successful.\n");
104 else
105 RTTestFailed(hTest, "%d: The third part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
106 __LINE__, rc);
107 }
108 else if (rc == VERR_NO_TRANSLATION)
109 RTTestPassed(hTest, "The second part of random UTF-16 -> UTF-8 -> Current -> UTF-8 returned VERR_NO_TRANSLATION. This is probably as it should be.\n");
110 else if (rc == VWRN_NO_TRANSLATION)
111 RTTestPassed(hTest, "The second part of random UTF-16 -> UTF-8 -> Current -> UTF-8 returned VWRN_NO_TRANSLATION. This is probably as it should be.\n");
112 else
113 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
114 __LINE__, rc);
115 }
116 else
117 RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
118 __LINE__, rc);
119
120 /*
121 * Generate a new random string.
122 */
123 RTTestSub(hTest, "Random UTF-16 -> UTF-8 -> UTF-16");
124 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
125 for (int i = 0; i < 30; i++)
126 pwszRand[i] = GetRandUtf16();
127 pwszRand[30] = 0;
128 rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
129 if (rc == VINF_SUCCESS)
130 {
131 rc = RTStrToUtf16(pszUtf8, &pwsz);
132 if (rc == VINF_SUCCESS)
133 {
134 int i;
135 for (i = 0; pwszRand[i] == pwsz[i] && pwsz[i] != 0; i++)
136 /* nothing */;
137 if (pwszRand[i] == pwsz[i] && pwsz[i] == 0)
138 RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> UTF-16 successful.\n");
139 else
140 {
141 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> UTF-16 failed.", __LINE__);
142 RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz[i]);
143 }
144 }
145 else
146 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> UTF-16 failed with return value %Rrc.",
147 __LINE__, rc);
148 }
149 else
150 RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> UTF-16 failed with return value %Rrc.",
151 __LINE__, rc);
152
153 /*
154 * Generate yet another random string and convert it to a buffer.
155 */
156 RTTestSub(hTest, "Random RTUtf16ToUtf8Ex + RTStrToUtf16");
157 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
158 for (int i = 0; i < 30; i++)
159 pwszRand[i] = GetRandUtf16();
160 pwszRand[30] = 0;
161
162 char szUtf8Array[120];
163 char *pszUtf8Array = szUtf8Array;
164 rc = RTUtf16ToUtf8Ex(pwszRand, RTSTR_MAX, &pszUtf8Array, 120, NULL);
165 if (rc == 0)
166 {
167 rc = RTStrToUtf16(pszUtf8Array, &pwsz);
168 if (rc == 0)
169 {
170 int i;
171 for (i = 0; pwszRand[i] == pwsz[i] && pwsz[i] != 0; i++)
172 ;
173 if (pwsz[i] == 0 && i >= 8)
174 RTTestPassed(hTest, "Random UTF-16 -> fixed length UTF-8 -> UTF-16 successful.\n");
175 else
176 {
177 RTTestFailed(hTest, "%d: Incorrect conversion of UTF-16 -> fixed length UTF-8 -> UTF-16.\n", __LINE__);
178 RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz[i]);
179 }
180 }
181 else
182 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> fixed length UTF-8 -> UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
183 }
184 else
185 RTTestFailed(hTest, "%d: The first part of random UTF-16 -> fixed length UTF-8 -> UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
186
187 /*
188 * And again.
189 */
190 RTTestSub(hTest, "Random RTUtf16ToUtf8 + RTStrToUtf16Ex");
191 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
192 for (int i = 0; i < 30; i++)
193 pwszRand[i] = GetRandUtf16();
194 pwszRand[30] = 0;
195
196 RTUTF16 wszBuf[70];
197 PRTUTF16 pwsz2Buf = wszBuf;
198 rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
199 if (rc == 0)
200 {
201 rc = RTStrToUtf16Ex(pszUtf8, RTSTR_MAX, &pwsz2Buf, 70, NULL);
202 if (rc == 0)
203 {
204 int i;
205 for (i = 0; pwszRand[i] == pwsz2Buf[i] && pwsz2Buf[i] != 0; i++)
206 ;
207 if (pwszRand[i] == 0 && pwsz2Buf[i] == 0)
208 RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> fixed length UTF-16 successful.\n");
209 else
210 {
211 RTTestFailed(hTest, "%d: Incorrect conversion of random UTF-16 -> UTF-8 -> fixed length UTF-16.\n", __LINE__);
212 RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz2Buf[i]);
213 }
214 }
215 else
216 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
217 }
218 else
219 RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n",
220 __LINE__, rc);
221 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
222 for (int i = 0; i < 30; i++)
223 pwszRand[i] = GetRandUtf16();
224 pwszRand[30] = 0;
225
226 rc = RTUtf16ToUtf8Ex(pwszRand, RTSTR_MAX, &pszUtf8Array, 20, NULL);
227 if (rc == VERR_BUFFER_OVERFLOW)
228 RTTestPassed(hTest, "Random UTF-16 -> fixed length UTF-8 with too short buffer successfully rejected.\n");
229 else
230 RTTestFailed(hTest, "%d: Random UTF-16 -> fixed length UTF-8 with too small buffer returned value %d instead of VERR_BUFFER_OVERFLOW.\n",
231 __LINE__, rc);
232
233 /*
234 * last time...
235 */
236 RTTestSub(hTest, "Random RTUtf16ToUtf8 + RTStrToUtf16Ex");
237 pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
238 for (int i = 0; i < 30; i++)
239 pwszRand[i] = GetRandUtf16();
240 pwszRand[30] = 0;
241
242 rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
243 if (rc == VINF_SUCCESS)
244 {
245 rc = RTStrToUtf16Ex(pszUtf8, RTSTR_MAX, &pwsz2Buf, 20, NULL);
246 if (rc == VERR_BUFFER_OVERFLOW)
247 RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> fixed length UTF-16 with too short buffer successfully rejected.\n");
248 else
249 RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> fixed length UTF-16 with too short buffer returned value %Rrc instead of VERR_BUFFER_OVERFLOW.\n",
250 __LINE__, rc);
251 }
252 else
253 RTTestFailed(hTest, "%d:The first part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n",
254 __LINE__, rc);
255
256
257 RTTestSubDone(hTest);
258}
259
260
261static RTUNICP g_uszAll[0x110000 - 1 - 0x800 - 2 + 1];
262static RTUTF16 g_wszAll[0xfffe - (0xe000 - 0xd800) + (0x110000 - 0x10000) * 2];
263static char g_szAll[0x7f + (0x800 - 0x80) * 2 + (0xfffe - 0x800 - (0xe000 - 0xd800))* 3 + (0x110000 - 0x10000) * 4 + 1];
264
265static void whereami(int cBits, size_t off)
266{
267 if (cBits == 8)
268 {
269 if (off < 0x7f)
270 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", off + 1);
271 else if (off < 0xf7f)
272 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x7f) / 2 + 0x80);
273 else if (off < 0x27f7f)
274 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0xf7f) / 3 + 0x800);
275 else if (off < 0x2df79)
276 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x27f7f) / 3 + 0xe000);
277 else if (off < 0x42df79)
278 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x2df79) / 4 + 0x10000);
279 else
280 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 ???\n");
281 }
282 else if (cBits == 16)
283 {
284 if (off < 0xd7ff*2)
285 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", off / 2 + 1);
286 else if (off < 0xf7fd*2)
287 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", (off - 0xd7ff*2) / 2 + 0xe000);
288 else if (off < 0x20f7fd)
289 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", (off - 0xf7fd*2) / 4 + 0x10000);
290 else
291 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 ???\n");
292 }
293 else
294 {
295 if (off < (0xd800 - 1) * sizeof(RTUNICP))
296 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 1);
297 else if (off < (0xfffe - 0x800 - 1) * sizeof(RTUNICP))
298 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 0x800 + 1);
299 else
300 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 0x800 + 1 + 2);
301 }
302}
303
304int mymemcmp(const void *pv1, const void *pv2, size_t cb, int cBits)
305{
306 const uint8_t *pb1 = (const uint8_t *)pv1;
307 const uint8_t *pb2 = (const uint8_t *)pv2;
308 for (size_t off = 0; off < cb; off++)
309 {
310 if (pb1[off] != pb2[off])
311 {
312 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "mismatch at %#x: ", off);
313 whereami(cBits, off);
314 if (off > 0)
315 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, " %#x: %02x != %02x!\n", off-1, pb1[off-1], pb2[off-1]);
316 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "*%#x: %02x != %02x!\n", off, pb1[off], pb2[off]);
317 for (size_t i = 1; i < 10; i++)
318 if (off + i < cb)
319 RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, " %#x: %02x != %02x!\n", off+i, pb1[off+i], pb2[off+i]);
320 return 1;
321 }
322 }
323 return 0;
324}
325
326
327void InitStrings()
328{
329 /*
330 * Generate unicode string containing all the legal UTF-16 codepoints, both UTF-16 and UTF-8 version.
331 */
332 /* the simple code point array first */
333 unsigned i = 0;
334 RTUNICP uc = 1;
335 while (uc < 0xd800)
336 g_uszAll[i++] = uc++;
337 uc = 0xe000;
338 while (uc < 0xfffe)
339 g_uszAll[i++] = uc++;
340 uc = 0x10000;
341 while (uc < 0x110000)
342 g_uszAll[i++] = uc++;
343 g_uszAll[i++] = 0;
344 Assert(RT_ELEMENTS(g_uszAll) == i);
345
346 /* the utf-16 one */
347 i = 0;
348 uc = 1;
349 //RTPrintf("tstUtf8: %#x=%#x", i, uc);
350 while (uc < 0xd800)
351 g_wszAll[i++] = uc++;
352 uc = 0xe000;
353 //RTPrintf(" %#x=%#x", i, uc);
354 while (uc < 0xfffe)
355 g_wszAll[i++] = uc++;
356 uc = 0x10000;
357 //RTPrintf(" %#x=%#x", i, uc);
358 while (uc < 0x110000)
359 {
360 g_wszAll[i++] = 0xd800 | ((uc - 0x10000) >> 10);
361 g_wszAll[i++] = 0xdc00 | ((uc - 0x10000) & 0x3ff);
362 uc++;
363 }
364 //RTPrintf(" %#x=%#x\n", i, uc);
365 g_wszAll[i++] = '\0';
366 Assert(RT_ELEMENTS(g_wszAll) == i);
367
368 /*
369 * The utf-8 one
370 */
371 i = 0;
372 uc = 1;
373 //RTPrintf("tstUtf8: %#x=%#x", i, uc);
374 while (uc < 0x80)
375 g_szAll[i++] = uc++;
376 //RTPrintf(" %#x=%#x", i, uc);
377 while (uc < 0x800)
378 {
379 g_szAll[i++] = 0xc0 | (uc >> 6);
380 g_szAll[i++] = 0x80 | (uc & 0x3f);
381 Assert(!((uc >> 6) & ~0x1f));
382 uc++;
383 }
384 //RTPrintf(" %#x=%#x", i, uc);
385 while (uc < 0xd800)
386 {
387 g_szAll[i++] = 0xe0 | (uc >> 12);
388 g_szAll[i++] = 0x80 | ((uc >> 6) & 0x3f);
389 g_szAll[i++] = 0x80 | (uc & 0x3f);
390 Assert(!((uc >> 12) & ~0xf));
391 uc++;
392 }
393 uc = 0xe000;
394 //RTPrintf(" %#x=%#x", i, uc);
395 while (uc < 0xfffe)
396 {
397 g_szAll[i++] = 0xe0 | (uc >> 12);
398 g_szAll[i++] = 0x80 | ((uc >> 6) & 0x3f);
399 g_szAll[i++] = 0x80 | (uc & 0x3f);
400 Assert(!((uc >> 12) & ~0xf));
401 uc++;
402 }
403 uc = 0x10000;
404 //RTPrintf(" %#x=%#x", i, uc);
405 while (uc < 0x110000)
406 {
407 g_szAll[i++] = 0xf0 | (uc >> 18);
408 g_szAll[i++] = 0x80 | ((uc >> 12) & 0x3f);
409 g_szAll[i++] = 0x80 | ((uc >> 6) & 0x3f);
410 g_szAll[i++] = 0x80 | (uc & 0x3f);
411 Assert(!((uc >> 18) & ~0x7));
412 uc++;
413 }
414 //RTPrintf(" %#x=%#x\n", i, uc);
415 g_szAll[i++] = '\0';
416 Assert(RT_ELEMENTS(g_szAll) == i);
417}
418
419
420void test2(RTTEST hTest)
421{
422 /*
423 * Convert to UTF-8 and back.
424 */
425 RTTestSub(hTest, "UTF-16 -> UTF-8 -> UTF-16");
426 char *pszUtf8;
427 int rc = RTUtf16ToUtf8(&g_wszAll[0], &pszUtf8);
428 if (rc == VINF_SUCCESS)
429 {
430 pszUtf8[0] = 1;
431 if (mymemcmp(pszUtf8, g_szAll, sizeof(g_szAll), 8))
432 RTTestFailed(hTest, "UTF-16 -> UTF-8 mismatch!");
433
434 PRTUTF16 pwszUtf16;
435 rc = RTStrToUtf16(pszUtf8, &pwszUtf16);
436 if (rc == VINF_SUCCESS)
437 {
438 if (mymemcmp(pwszUtf16, g_wszAll, sizeof(g_wszAll), 16))
439 RTTestFailed(hTest, "UTF-8 -> UTF-16 failed compare!");
440 RTUtf16Free(pwszUtf16);
441 }
442 else
443 RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
444 RTStrFree(pszUtf8);
445 }
446 else
447 RTTestFailed(hTest, "UTF-16 -> UTF-8 failed, rc=%Rrc.", rc);
448
449
450 /*
451 * Convert to UTF-16 and back. (just in case the above test fails)
452 */
453 RTTestSub(hTest, "UTF-8 -> UTF-16 -> UTF-8");
454 PRTUTF16 pwszUtf16;
455 rc = RTStrToUtf16(&g_szAll[0], &pwszUtf16);
456 if (rc == VINF_SUCCESS)
457 {
458 if (mymemcmp(pwszUtf16, g_wszAll, sizeof(g_wszAll), 16))
459 RTTestFailed(hTest, "UTF-8 -> UTF-16 failed compare!");
460
461 rc = RTUtf16ToUtf8(pwszUtf16, &pszUtf8);
462 if (rc == VINF_SUCCESS)
463 {
464 if (mymemcmp(pszUtf8, g_szAll, sizeof(g_szAll), 8))
465 RTTestFailed(hTest, "UTF-16 -> UTF-8 failed compare!");
466 RTStrFree(pszUtf8);
467 }
468 else
469 RTTestFailed(hTest, "UTF-16 -> UTF-8 failed, rc=%Rrc.", rc);
470 RTUtf16Free(pwszUtf16);
471 }
472 else
473 RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
474
475 /*
476 * Convert UTF-8 to CPs.
477 */
478 RTTestSub(hTest, "UTF-8 -> UNI -> UTF-8");
479 PRTUNICP paCps;
480 rc = RTStrToUni(g_szAll, &paCps);
481 if (rc == VINF_SUCCESS)
482 {
483 if (mymemcmp(paCps, g_uszAll, sizeof(g_uszAll), 32))
484 RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
485
486 size_t cCps;
487 rc = RTStrToUniEx(g_szAll, RTSTR_MAX, &paCps, RT_ELEMENTS(g_uszAll), &cCps);
488 if (rc == VINF_SUCCESS)
489 {
490 if (cCps != RT_ELEMENTS(g_uszAll) - 1)
491 RTTestFailed(hTest, "wrong Code Point count %zu, expected %zu\n", cCps, RT_ELEMENTS(g_uszAll) - 1);
492 }
493 else
494 RTTestFailed(hTest, "UTF-8 -> Code Points failed, rc=%Rrc.\n", rc);
495
496 /** @todo RTCpsToUtf8 or something. */
497 }
498 else
499 RTTestFailed(hTest, "UTF-8 -> Code Points failed, rc=%Rrc.\n", rc);
500
501 /*
502 * Check the various string lengths.
503 */
504 RTTestSub(hTest, "Lengths");
505 size_t cuc1 = RTStrCalcUtf16Len(g_szAll);
506 size_t cuc2 = RTUtf16Len(g_wszAll);
507 if (cuc1 != cuc2)
508 RTTestFailed(hTest, "cuc1=%zu != cuc2=%zu\n", cuc1, cuc2);
509 //size_t cuc3 = RTUniLen(g_uszAll);
510
511
512 /*
513 * Enumerate the strings.
514 */
515 RTTestSub(hTest, "Code Point Getters and Putters");
516 char *pszPut1Base = (char *)RTMemAlloc(sizeof(g_szAll));
517 AssertRelease(pszPut1Base);
518 char *pszPut1 = pszPut1Base;
519 PRTUTF16 pwszPut2Base = (PRTUTF16)RTMemAlloc(sizeof(g_wszAll));
520 AssertRelease(pwszPut2Base);
521 PRTUTF16 pwszPut2 = pwszPut2Base;
522 const char *psz1 = g_szAll;
523 const char *psz2 = g_szAll;
524 PCRTUTF16 pwsz3 = g_wszAll;
525 PCRTUTF16 pwsz4 = g_wszAll;
526 for (;;)
527 {
528 /*
529 * getters
530 */
531 RTUNICP uc1;
532 rc = RTStrGetCpEx(&psz1, &uc1);
533 if (RT_FAILURE(rc))
534 {
535 RTTestFailed(hTest, "RTStrGetCpEx failed with rc=%Rrc at %.10Rhxs", rc, psz2);
536 whereami(8, psz2 - &g_szAll[0]);
537 break;
538 }
539 char *pszPrev1 = RTStrPrevCp(g_szAll, psz1);
540 if (pszPrev1 != psz2)
541 {
542 RTTestFailed(hTest, "RTStrPrevCp returned %p expected %p!", pszPrev1, psz2);
543 whereami(8, psz2 - &g_szAll[0]);
544 break;
545 }
546 RTUNICP uc2 = RTStrGetCp(psz2);
547 if (uc2 != uc1)
548 {
549 RTTestFailed(hTest, "RTStrGetCpEx and RTStrGetCp returned different CPs: %RTunicp != %RTunicp", uc2, uc1);
550 whereami(8, psz2 - &g_szAll[0]);
551 break;
552 }
553 psz2 = RTStrNextCp(psz2);
554 if (psz2 != psz1)
555 {
556 RTTestFailed(hTest, "RTStrGetCpEx and RTStrGetNext returned different next pointer!");
557 whereami(8, psz2 - &g_szAll[0]);
558 break;
559 }
560
561 RTUNICP uc3;
562 rc = RTUtf16GetCpEx(&pwsz3, &uc3);
563 if (RT_FAILURE(rc))
564 {
565 RTTestFailed(hTest, "RTUtf16GetCpEx failed with rc=%Rrc at %.10Rhxs", rc, pwsz4);
566 whereami(16, pwsz4 - &g_wszAll[0]);
567 break;
568 }
569 if (uc3 != uc2)
570 {
571 RTTestFailed(hTest, "RTUtf16GetCpEx and RTStrGetCp returned different CPs: %RTunicp != %RTunicp", uc3, uc2);
572 whereami(16, pwsz4 - &g_wszAll[0]);
573 break;
574 }
575 RTUNICP uc4 = RTUtf16GetCp(pwsz4);
576 if (uc3 != uc4)
577 {
578 RTTestFailed(hTest, "RTUtf16GetCpEx and RTUtf16GetCp returned different CPs: %RTunicp != %RTunicp", uc3, uc4);
579 whereami(16, pwsz4 - &g_wszAll[0]);
580 break;
581 }
582 pwsz4 = RTUtf16NextCp(pwsz4);
583 if (pwsz4 != pwsz3)
584 {
585 RTTestFailed(hTest, "RTUtf16GetCpEx and RTUtf16GetNext returned different next pointer!");
586 whereami(8, pwsz4 - &g_wszAll[0]);
587 break;
588 }
589
590
591 /*
592 * putters
593 */
594 pszPut1 = RTStrPutCp(pszPut1, uc1);
595 if (pszPut1 - pszPut1Base != psz1 - &g_szAll[0])
596 {
597 RTTestFailed(hTest, "RTStrPutCp is not at the same offset! %p != %p",
598 pszPut1 - pszPut1Base, psz1 - &g_szAll[0]);
599 whereami(8, psz2 - &g_szAll[0]);
600 break;
601 }
602
603 pwszPut2 = RTUtf16PutCp(pwszPut2, uc3);
604 if (pwszPut2 - pwszPut2Base != pwsz3 - &g_wszAll[0])
605 {
606 RTTestFailed(hTest, "RTStrPutCp is not at the same offset! %p != %p",
607 pwszPut2 - pwszPut2Base, pwsz3 - &g_wszAll[0]);
608 whereami(8, pwsz4 - &g_wszAll[0]);
609 break;
610 }
611
612
613 /* the end? */
614 if (!uc1)
615 break;
616 }
617
618 /* check output if we seems to have made it thru it all. */
619 if (psz2 == &g_szAll[sizeof(g_szAll)])
620 {
621 if (mymemcmp(pszPut1Base, g_szAll, sizeof(g_szAll), 8))
622 RTTestFailed(hTest, "RTStrPutCp encoded the string incorrectly.");
623 if (mymemcmp(pwszPut2Base, g_wszAll, sizeof(g_wszAll), 16))
624 RTTestFailed(hTest, "RTUtf16PutCp encoded the string incorrectly.");
625 }
626
627 RTMemFree(pszPut1Base);
628 RTMemFree(pwszPut2Base);
629
630 RTTestSubDone(hTest);
631}
632
633
634/**
635 * Check case insensitivity.
636 */
637void test3(RTTEST hTest)
638{
639 RTTestSub(hTest, "Case Sensitivity");
640
641 if ( RTUniCpToLower('a') != 'a'
642 || RTUniCpToLower('A') != 'a'
643 || RTUniCpToLower('b') != 'b'
644 || RTUniCpToLower('B') != 'b'
645 || RTUniCpToLower('Z') != 'z'
646 || RTUniCpToLower('z') != 'z'
647 || RTUniCpToUpper('c') != 'C'
648 || RTUniCpToUpper('C') != 'C'
649 || RTUniCpToUpper('z') != 'Z'
650 || RTUniCpToUpper('Z') != 'Z')
651 RTTestFailed(hTest, "RTUniToUpper/Lower failed basic tests.\n");
652
653 if (RTUtf16ICmp(g_wszAll, g_wszAll))
654 RTTestFailed(hTest, "RTUtf16ICmp failed the basic test.\n");
655
656 if (RTUtf16Cmp(g_wszAll, g_wszAll))
657 RTTestFailed(hTest, "RTUtf16Cmp failed the basic test.\n");
658
659 static RTUTF16 s_wszTst1a[] = { 'a', 'B', 'c', 'D', 'E', 'f', 'g', 'h', 'i', 'j', 'K', 'L', 'm', 'N', 'o', 'P', 'q', 'r', 'S', 't', 'u', 'V', 'w', 'x', 'Y', 'Z', 0xc5, 0xc6, 0xf8, 0 };
660 static RTUTF16 s_wszTst1b[] = { 'A', 'B', 'c', 'd', 'e', 'F', 'G', 'h', 'i', 'J', 'k', 'l', 'M', 'n', 'O', 'p', 'Q', 'R', 's', 't', 'U', 'v', 'w', 'X', 'y', 'z', 0xe5, 0xe6, 0xd8, 0 };
661 if ( RTUtf16ICmp(s_wszTst1b, s_wszTst1b)
662 || RTUtf16ICmp(s_wszTst1a, s_wszTst1a)
663 || RTUtf16ICmp(s_wszTst1a, s_wszTst1b)
664 || RTUtf16ICmp(s_wszTst1b, s_wszTst1a)
665 )
666 RTTestFailed(hTest, "RTUtf16ICmp failed the alphabet test.\n");
667
668 if ( RTUtf16Cmp(s_wszTst1b, s_wszTst1b)
669 || RTUtf16Cmp(s_wszTst1a, s_wszTst1a)
670 || !RTUtf16Cmp(s_wszTst1a, s_wszTst1b)
671 || !RTUtf16Cmp(s_wszTst1b, s_wszTst1a)
672 )
673 RTTestFailed(hTest, "RTUtf16Cmp failed the alphabet test.\n");
674
675 RTTestSubDone(hTest);
676}
677
678
679/**
680 * Test the RTStr*Cmp functions.
681 */
682void TstRTStrXCmp(RTTEST hTest)
683{
684#define CHECK_DIFF(expr, op) \
685 do \
686 { \
687 int iDiff = expr; \
688 if (!(iDiff op 0)) \
689 RTTestFailed(hTest, "%d: %d " #op " 0: %s\n", __LINE__, iDiff, #expr); \
690 } while (0)
691
692/** @todo test the non-ascii bits. */
693
694 RTTestSub(hTest, "RTStrCmp");
695 CHECK_DIFF(RTStrCmp(NULL, NULL), == );
696 CHECK_DIFF(RTStrCmp(NULL, ""), < );
697 CHECK_DIFF(RTStrCmp("", NULL), > );
698 CHECK_DIFF(RTStrCmp("", ""), == );
699 CHECK_DIFF(RTStrCmp("abcdef", "abcdef"), == );
700 CHECK_DIFF(RTStrCmp("abcdef", "abcde"), > );
701 CHECK_DIFF(RTStrCmp("abcde", "abcdef"), < );
702 CHECK_DIFF(RTStrCmp("abcdeg", "abcdef"), > );
703 CHECK_DIFF(RTStrCmp("abcdef", "abcdeg"), < );
704 CHECK_DIFF(RTStrCmp("abcdeF", "abcdef"), < );
705 CHECK_DIFF(RTStrCmp("abcdef", "abcdeF"), > );
706
707
708 RTTestSub(hTest, "RTStrNCmp");
709 CHECK_DIFF(RTStrNCmp(NULL, NULL, RTSTR_MAX), == );
710 CHECK_DIFF(RTStrNCmp(NULL, "", RTSTR_MAX), < );
711 CHECK_DIFF(RTStrNCmp("", NULL, RTSTR_MAX), > );
712 CHECK_DIFF(RTStrNCmp("", "", RTSTR_MAX), == );
713 CHECK_DIFF(RTStrNCmp("abcdef", "abcdef", RTSTR_MAX), == );
714 CHECK_DIFF(RTStrNCmp("abcdef", "abcde", RTSTR_MAX), > );
715 CHECK_DIFF(RTStrNCmp("abcde", "abcdef", RTSTR_MAX), < );
716 CHECK_DIFF(RTStrNCmp("abcdeg", "abcdef", RTSTR_MAX), > );
717 CHECK_DIFF(RTStrNCmp("abcdef", "abcdeg", RTSTR_MAX), < );
718 CHECK_DIFF(RTStrNCmp("abcdeF", "abcdef", RTSTR_MAX), < );
719 CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", RTSTR_MAX), > );
720
721 CHECK_DIFF(RTStrNCmp("abcdef", "fedcba", 0), ==);
722 CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", 5), ==);
723 CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", 6), > );
724
725
726 RTTestSub(hTest, "RTStrICmp");
727 CHECK_DIFF(RTStrICmp(NULL, NULL), == );
728 CHECK_DIFF(RTStrICmp(NULL, ""), < );
729 CHECK_DIFF(RTStrICmp("", NULL), > );
730 CHECK_DIFF(RTStrICmp("", ""), == );
731 CHECK_DIFF(RTStrICmp("abcdef", "abcdef"), == );
732 CHECK_DIFF(RTStrICmp("abcdef", "abcde"), > );
733 CHECK_DIFF(RTStrICmp("abcde", "abcdef"), < );
734 CHECK_DIFF(RTStrICmp("abcdeg", "abcdef"), > );
735 CHECK_DIFF(RTStrICmp("abcdef", "abcdeg"), < );
736
737 CHECK_DIFF(RTStrICmp("abcdeF", "abcdef"), ==);
738 CHECK_DIFF(RTStrICmp("abcdef", "abcdeF"), ==);
739 CHECK_DIFF(RTStrICmp("ABCDEF", "abcdef"), ==);
740 CHECK_DIFF(RTStrICmp("abcdef", "ABCDEF"), ==);
741 CHECK_DIFF(RTStrICmp("AbCdEf", "aBcDeF"), ==);
742 CHECK_DIFF(RTStrICmp("AbCdEg", "aBcDeF"), > );
743 CHECK_DIFF(RTStrICmp("AbCdEG", "aBcDef"), > ); /* diff performed on the lower case cp. */
744
745
746 RTTestSub(hTest, "RTStrICmpAscii");
747 CHECK_DIFF(RTStrICmpAscii(NULL, NULL), == );
748 CHECK_DIFF(RTStrICmpAscii(NULL, ""), < );
749 CHECK_DIFF(RTStrICmpAscii("", NULL), > );
750 CHECK_DIFF(RTStrICmpAscii("", ""), == );
751 CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdef"), == );
752 CHECK_DIFF(RTStrICmpAscii("abcdef", "abcde"), > );
753 CHECK_DIFF(RTStrICmpAscii("abcde", "abcdef"), < );
754 CHECK_DIFF(RTStrICmpAscii("abcdeg", "abcdef"), > );
755 CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdeg"), < );
756
757 CHECK_DIFF(RTStrICmpAscii("abcdeF", "abcdef"), ==);
758 CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdeF"), ==);
759 CHECK_DIFF(RTStrICmpAscii("ABCDEF", "abcdef"), ==);
760 CHECK_DIFF(RTStrICmpAscii("abcdef", "ABCDEF"), ==);
761 CHECK_DIFF(RTStrICmpAscii("AbCdEf", "aBcDeF"), ==);
762 CHECK_DIFF(RTStrICmpAscii("AbCdEg", "aBcDeF"), > );
763 CHECK_DIFF(RTStrICmpAscii("AbCdEG", "aBcDef"), > ); /* diff performed on the lower case cp. */
764
765
766 RTTestSub(hTest, "RTStrNICmp");
767 CHECK_DIFF(RTStrNICmp(NULL, NULL, RTSTR_MAX), == );
768 CHECK_DIFF(RTStrNICmp(NULL, "", RTSTR_MAX), < );
769 CHECK_DIFF(RTStrNICmp("", NULL, RTSTR_MAX), > );
770 CHECK_DIFF(RTStrNICmp("", "", RTSTR_MAX), == );
771 CHECK_DIFF(RTStrNICmp(NULL, NULL, 0), == );
772 CHECK_DIFF(RTStrNICmp(NULL, "", 0), == );
773 CHECK_DIFF(RTStrNICmp("", NULL, 0), == );
774 CHECK_DIFF(RTStrNICmp("", "", 0), == );
775 CHECK_DIFF(RTStrNICmp("abcdef", "abcdef", RTSTR_MAX), == );
776 CHECK_DIFF(RTStrNICmp("abcdef", "abcde", RTSTR_MAX), > );
777 CHECK_DIFF(RTStrNICmp("abcde", "abcdef", RTSTR_MAX), < );
778 CHECK_DIFF(RTStrNICmp("abcdeg", "abcdef", RTSTR_MAX), > );
779 CHECK_DIFF(RTStrNICmp("abcdef", "abcdeg", RTSTR_MAX), < );
780
781 CHECK_DIFF(RTStrNICmp("abcdeF", "abcdef", RTSTR_MAX), ==);
782 CHECK_DIFF(RTStrNICmp("abcdef", "abcdeF", RTSTR_MAX), ==);
783 CHECK_DIFF(RTStrNICmp("ABCDEF", "abcdef", RTSTR_MAX), ==);
784 CHECK_DIFF(RTStrNICmp("abcdef", "ABCDEF", RTSTR_MAX), ==);
785 CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDeF", RTSTR_MAX), ==);
786 CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", RTSTR_MAX), > );
787 CHECK_DIFF(RTStrNICmp("AbCdEG", "aBcDef", RTSTR_MAX), > ); /* diff performed on the lower case cp. */
788
789 CHECK_DIFF(RTStrNICmp("ABCDEF", "fedcba", 0), ==);
790 CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", 5), ==);
791 CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDeF", 5), ==);
792 CHECK_DIFF(RTStrNICmp("AbCdE", "aBcDe", 5), ==);
793 CHECK_DIFF(RTStrNICmp("AbCdE", "aBcDeF", 5), ==);
794 CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDe", 5), ==);
795 CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", 6), > );
796 CHECK_DIFF(RTStrNICmp("AbCdEG", "aBcDef", 6), > ); /* diff performed on the lower case cp. */
797 /* We should continue using byte comparison when we hit the invalid CP. Will assert in debug builds. */
798 // CHECK_DIFF(RTStrNICmp("AbCd\xff""eg", "aBcD\xff""eF", 6), ==);
799
800 RTTestSubDone(hTest);
801}
802
803
804
805/**
806 * Check UTF-8 encoding purging.
807 */
808void TstRTStrPurgeEncoding(RTTEST hTest)
809{
810 RTTestSub(hTest, "RTStrPurgeEncoding");
811
812 /*
813 * Test some good strings.
814 */
815 char sz1[] = "1234567890wertyuiopsdfghjklzxcvbnm";
816 char sz1Copy[sizeof(sz1)];
817 memcpy(sz1Copy, sz1, sizeof(sz1));
818
819 RTTESTI_CHECK_RETV(RTStrPurgeEncoding(sz1) == 0);
820 RTTESTI_CHECK_RETV(!memcmp(sz1, sz1Copy, sizeof(sz1)));
821
822 char *pszAll = RTStrDup(g_szAll);
823 if (pszAll)
824 {
825 RTTESTI_CHECK(RTStrPurgeEncoding(pszAll) == 0);
826 RTTESTI_CHECK(!memcmp(pszAll, g_szAll, sizeof(g_szAll)));
827 RTStrFree(pszAll);
828 }
829
830 /*
831 * Test some bad stuff.
832 */
833 struct
834 {
835 size_t cErrors;
836 unsigned char szIn[5];
837 const char *pszExpect;
838 } aTests[] =
839 {
840 { 0, { '1', '2', '3', '4', '\0' }, "1234" },
841 { 1, { 0x80, '2', '3', '4', '\0' }, "?234" },
842 { 1, { '1', 0x80, '3', '4', '\0' }, "1?34" },
843 { 1, { '1', '2', 0x80, '4', '\0' }, "12?4" },
844 { 1, { '1', '2', '3', 0x80, '\0' }, "123?" },
845 { 2, { 0x80, 0x81, '3', '4', '\0' }, "??34" },
846 { 2, { '1', 0x80, 0x81, '4', '\0' }, "1??4" },
847 { 2, { '1', '2', 0x80, 0x81, '\0' }, "12??" },
848 };
849 for (size_t i = 0; i < RT_ELEMENTS(aTests); i++)
850 {
851 size_t cErrors = RTStrPurgeEncoding((char *)aTests[i].szIn);
852 if (cErrors != aTests[i].cErrors)
853 RTTestFailed(hTest, "#%u: cErrors=%u expected %u\n", i, cErrors, aTests[i].cErrors);
854 else if (strcmp((char *)aTests[i].szIn, aTests[i].pszExpect))
855 RTTestFailed(hTest, "#%u: %.5Rhxs expected %.5Rhxs (%s)\n", i, aTests[i].szIn, aTests[i].pszExpect, aTests[i].pszExpect);
856 }
857
858 RTTestSubDone(hTest);
859}
860
861
862/**
863 * Check string sanitising.
864 */
865void TstRTStrPurgeComplementSet(RTTEST hTest)
866{
867 RTTestSub(hTest, "RTStrPurgeComplementSet");
868 RTUNICP aCpSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
869 '\0' };
870 RTUNICP aCpBadSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
871 '7', '\0' }; /* Contains an incomplete pair. */
872 struct
873 {
874 const char *pcszIn;
875 const char *pcszOut;
876 PCRTUNICP pcCpSet;
877 char chReplacement;
878 ssize_t cExpected;
879 }
880 aTests[] =
881 {
882 { "1234werttrew4321", "1234werttrew4321", aCpSet, '_', 0 },
883 { "123654wert\xc2\xa2trew\xe2\x82\xac""4321",
884 "123_54wert__trew___4321", aCpSet, '_', 3 },
885 { "hjhj8766", "????????", aCpSet, '?', 8 },
886 { "123\xf0\xa4\xad\xa2""4", "123____4", aCpSet, '_', 1 },
887 { "\xff", "\xff", aCpSet, '_', -1 },
888 { "____", "____", aCpBadSet, '_', -1 }
889 };
890 enum { MAX_IN_STRING = 256 };
891
892 for (unsigned i = 0; i < RT_ELEMENTS(aTests); ++i)
893 {
894 char szCopy[MAX_IN_STRING];
895 ssize_t cReplacements;
896 AssertRC(RTStrCopy(szCopy, RT_ELEMENTS(szCopy), aTests[i].pcszIn));
897 RTTestDisableAssertions(hTest);
898 cReplacements = RTStrPurgeComplementSet(szCopy, aTests[i].pcCpSet, aTests[i].chReplacement);
899 RTTestRestoreAssertions(hTest);
900 if (cReplacements != aTests[i].cExpected)
901 RTTestFailed(hTest, "#%u: expected %lld, actual %lld\n", i,
902 (long long) aTests[i].cExpected,
903 (long long) cReplacements);
904 if (strcmp(aTests[i].pcszOut, szCopy))
905 RTTestFailed(hTest, "#%u: expected %s, actual %s\n", i,
906 aTests[i].pcszOut, szCopy);
907 }
908}
909
910
911/**
912 * Check string sanitising.
913 */
914void TstRTUtf16PurgeComplementSet(RTTEST hTest)
915{
916 RTTestSub(hTest, "RTUtf16PurgeComplementSet");
917 RTUNICP aCpSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
918 '\0' };
919 RTUNICP aCpBadSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
920 '7', '\0' }; /* Contains an incomplete pair. */
921 struct
922 {
923 const char *pcszIn;
924 const char *pcszOut;
925 size_t cwc; /* Zero means the strings are Utf-8. */
926 PCRTUNICP pcCpSet;
927 char chReplacement;
928 ssize_t cExpected;
929 }
930 aTests[] =
931 {
932 { "1234werttrew4321", "1234werttrew4321", 0, aCpSet, '_', 0 },
933 { "123654wert\xc2\xa2trew\xe2\x82\xac""4321",
934 "123_54wert_trew_4321", 0, aCpSet, '_', 3 },
935 { "hjhj8766", "????????", 0, aCpSet, '?', 8 },
936 { "123\xf0\xa4\xad\xa2""4", "123__4", 0, aCpSet, '_', 1 },
937 { "\xff\xff\0", "\xff\xff\0", 2, aCpSet, '_', -1 },
938 { "\xff\xff\0", "\xff\xff\0", 2, aCpSet, '_', -1 },
939 { "____", "____", 0, aCpBadSet, '_', -1 }
940 };
941 enum { MAX_IN_STRING = 256 };
942
943 for (unsigned i = 0; i < RT_ELEMENTS(aTests); ++i)
944 {
945 RTUTF16 wszInCopy[MAX_IN_STRING], *pwszInCopy = wszInCopy;
946 RTUTF16 wszOutCopy[MAX_IN_STRING], *pwszOutCopy = wszOutCopy;
947 ssize_t cReplacements;
948 if (!aTests[i].cwc)
949 {
950 AssertRC(RTStrToUtf16Ex(aTests[i].pcszIn, RTSTR_MAX, &pwszInCopy,
951 RT_ELEMENTS(wszInCopy), NULL));
952 AssertRC(RTStrToUtf16Ex(aTests[i].pcszOut, RTSTR_MAX, &pwszOutCopy,
953 RT_ELEMENTS(wszOutCopy), NULL));
954 }
955 else
956 {
957 Assert(aTests[i].cwc <= RT_ELEMENTS(wszInCopy));
958 memcpy(wszInCopy, aTests[i].pcszIn, aTests[i].cwc * 2);
959 memcpy(wszOutCopy, aTests[i].pcszOut, aTests[i].cwc * 2);
960 }
961
962 RTTestDisableAssertions(hTest);
963 cReplacements = RTUtf16PurgeComplementSet(wszInCopy, aTests[i].pcCpSet, aTests[i].chReplacement);
964 RTTestRestoreAssertions(hTest);
965
966 if (cReplacements != aTests[i].cExpected)
967 RTTestFailed(hTest, "#%u: expected %lld, actual %lld\n", i,
968 (long long) aTests[i].cExpected,
969 (long long) cReplacements);
970 if (RTUtf16Cmp(wszInCopy, wszOutCopy))
971 RTTestFailed(hTest, "#%u: expected %ls, actual %ls\n", i,
972 wszOutCopy, wszInCopy);
973 }
974}
975
976
977/**
978 * Benchmark stuff.
979 */
980void Benchmarks(RTTEST hTest)
981{
982 static union
983 {
984 RTUTF16 wszBuf[sizeof(g_wszAll)];
985 char szBuf[sizeof(g_szAll)];
986 } s_Buf;
987
988 RTTestSub(hTest, "Benchmarks");
989/** @todo add RTTest* methods for reporting benchmark results. */
990 RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "Benchmarking RTStrToUtf16Ex: "); /** @todo figure this stuff into the test framework. */
991 PRTUTF16 pwsz = &s_Buf.wszBuf[0];
992 int rc = RTStrToUtf16Ex(&g_szAll[0], RTSTR_MAX, &pwsz, RT_ELEMENTS(s_Buf.wszBuf), NULL);
993 if (RT_SUCCESS(rc))
994 {
995 int i;
996 uint64_t u64Start = RTTimeNanoTS();
997 for (i = 0; i < 100; i++)
998 {
999 rc = RTStrToUtf16Ex(&g_szAll[0], RTSTR_MAX, &pwsz, RT_ELEMENTS(s_Buf.wszBuf), NULL);
1000 if (RT_FAILURE(rc))
1001 {
1002 RTTestFailed(hTest, "UTF-8 -> UTF-16 benchmark failed at i=%d, rc=%Rrc\n", i, rc);
1003 break;
1004 }
1005 }
1006 uint64_t u64Elapsed = RTTimeNanoTS() - u64Start;
1007 RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "%d in %'RI64 ns\n", i, u64Elapsed);
1008 }
1009
1010 RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "Benchmarking RTUtf16ToUtf8Ex: ");
1011 char *psz = &s_Buf.szBuf[0];
1012 rc = RTUtf16ToUtf8Ex(&g_wszAll[0], RTSTR_MAX, &psz, RT_ELEMENTS(s_Buf.szBuf), NULL);
1013 if (RT_SUCCESS(rc))
1014 {
1015 int i;
1016 uint64_t u64Start = RTTimeNanoTS();
1017 for (i = 0; i < 100; i++)
1018 {
1019 rc = RTUtf16ToUtf8Ex(&g_wszAll[0], RTSTR_MAX, &psz, RT_ELEMENTS(s_Buf.szBuf), NULL);
1020 if (RT_FAILURE(rc))
1021 {
1022 RTTestFailed(hTest, "UTF-16 -> UTF-8 benchmark failed at i=%d, rc=%Rrc\n", i, rc);
1023 break;
1024 }
1025 }
1026 uint64_t u64Elapsed = RTTimeNanoTS() - u64Start;
1027 RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "%d in %'RI64 ns\n", i, u64Elapsed);
1028 }
1029
1030 RTTestSubDone(hTest);
1031}
1032
1033
1034/**
1035 * Tests RTStrEnd
1036 */
1037static void testStrEnd(RTTEST hTest)
1038{
1039 RTTestSub(hTest, "RTStrEnd");
1040
1041 static char const s_szEmpty[1] = "";
1042 RTTESTI_CHECK(RTStrEnd(s_szEmpty, 0) == NULL);
1043 RTTESTI_CHECK(RTStrEnd(s_szEmpty, 1) == &s_szEmpty[0]);
1044 for (size_t i = 0; i < _1M; i++)
1045 RTTESTI_CHECK(RTStrEnd(s_szEmpty, ~i) == &s_szEmpty[0]);
1046
1047 /* Check the implementation won't ever overshoot the '\0' in the input in
1048 anyway that may lead to a SIGSEV. (VC++ 14.1 does this) */
1049 size_t const cchStr = 1023;
1050 char *pszStr = (char *)RTTestGuardedAllocTail(hTest, cchStr + 1);
1051 memset(pszStr, ' ', cchStr);
1052 char * const pszStrEnd = &pszStr[cchStr];
1053 *pszStrEnd = '\0';
1054 RTTEST_CHECK_RETV(hTest, strlen(pszStr) == cchStr);
1055
1056 for (size_t off = 0; off <= cchStr; off++)
1057 {
1058 RTTEST_CHECK(hTest, RTStrEnd(&pszStr[off], cchStr + 1 - off) == pszStrEnd);
1059 RTTEST_CHECK(hTest, RTStrEnd(&pszStr[off], RTSTR_MAX) == pszStrEnd);
1060
1061 RTTEST_CHECK(hTest, memchr(&pszStr[off], '\0', cchStr + 1 - off) == pszStrEnd);
1062 RTTEST_CHECK(hTest, strchr(&pszStr[off], '\0') == pszStrEnd);
1063 RTTEST_CHECK(hTest, strchr(&pszStr[off], '?') == NULL);
1064
1065 size_t cchMax = 0;
1066 for (; cchMax <= cchStr - off; cchMax++)
1067 {
1068 const char *pszRet = RTStrEnd(&pszStr[off], cchMax);
1069 if (pszRet != NULL)
1070 {
1071 RTTestFailed(hTest, "off=%zu cchMax=%zu: %p, expected NULL\n", off, cchMax, pszRet);
1072 break;
1073 }
1074 }
1075 for (; cchMax <= _8K; cchMax++)
1076 {
1077 const char *pszRet = RTStrEnd(&pszStr[off], cchMax);
1078 if (pszRet != pszStrEnd)
1079 {
1080 RTTestFailed(hTest, "off=%zu cchMax=%zu: off by %p\n", off, cchMax, pszRet);
1081 break;
1082 }
1083 }
1084 }
1085 RTTestGuardedFree(hTest, pszStr);
1086}
1087
1088
1089/**
1090 * Tests RTStrStr and RTStrIStr.
1091 */
1092static void testStrStr(RTTEST hTest)
1093{
1094#define CHECK_NULL(expr) \
1095 do { \
1096 const char *pszRet = expr; \
1097 if (pszRet != NULL) \
1098 RTTestFailed(hTest, "%d: %#x -> %s expected NULL", __LINE__, #expr, pszRet); \
1099 } while (0)
1100
1101#define CHECK(expr, expect) \
1102 do { \
1103 const char *pszRet = expr; \
1104 const char *pszExpect = (expect); \
1105 if ( (pszRet != NULL && pszExpect == NULL) \
1106 || (pszRet == NULL && pszExpect != NULL) \
1107 || strcmp(pszRet, pszExpect) \
1108 ) \
1109 RTTestFailed(hTest, "%d: %#x -> %s expected %s", __LINE__, #expr, pszRet, pszExpect); \
1110 } while (0)
1111
1112
1113 RTTestSub(hTest, "RTStrStr");
1114 CHECK(RTStrStr("abcdef", ""), "abcdef");
1115 CHECK_NULL(RTStrStr("abcdef", NULL));
1116 CHECK_NULL(RTStrStr(NULL, ""));
1117 CHECK_NULL(RTStrStr(NULL, NULL));
1118 CHECK(RTStrStr("abcdef", "abcdef"), "abcdef");
1119 CHECK(RTStrStr("abcdef", "b"), "bcdef");
1120 CHECK(RTStrStr("abcdef", "bcdef"), "bcdef");
1121 CHECK(RTStrStr("abcdef", "cdef"), "cdef");
1122 CHECK(RTStrStr("abcdef", "cde"), "cdef");
1123 CHECK(RTStrStr("abcdef", "cd"), "cdef");
1124 CHECK(RTStrStr("abcdef", "c"), "cdef");
1125 CHECK(RTStrStr("abcdef", "f"), "f");
1126 CHECK(RTStrStr("abcdef", "ef"), "ef");
1127 CHECK(RTStrStr("abcdef", "e"), "ef");
1128 CHECK_NULL(RTStrStr("abcdef", "z"));
1129 CHECK_NULL(RTStrStr("abcdef", "A"));
1130 CHECK_NULL(RTStrStr("abcdef", "F"));
1131
1132 RTTestSub(hTest, "RTStrIStr");
1133 CHECK(RTStrIStr("abcdef", ""), "abcdef");
1134 CHECK_NULL(RTStrIStr("abcdef", NULL));
1135 CHECK_NULL(RTStrIStr(NULL, ""));
1136 CHECK_NULL(RTStrIStr(NULL, NULL));
1137 CHECK(RTStrIStr("abcdef", "abcdef"), "abcdef");
1138 CHECK(RTStrIStr("abcdef", "Abcdef"), "abcdef");
1139 CHECK(RTStrIStr("abcdef", "ABcDeF"), "abcdef");
1140 CHECK(RTStrIStr("abcdef", "b"), "bcdef");
1141 CHECK(RTStrIStr("abcdef", "B"), "bcdef");
1142 CHECK(RTStrIStr("abcdef", "bcdef"), "bcdef");
1143 CHECK(RTStrIStr("abcdef", "BCdEf"), "bcdef");
1144 CHECK(RTStrIStr("abcdef", "bCdEf"), "bcdef");
1145 CHECK(RTStrIStr("abcdef", "bcdEf"), "bcdef");
1146 CHECK(RTStrIStr("abcdef", "BcdEf"), "bcdef");
1147 CHECK(RTStrIStr("abcdef", "cdef"), "cdef");
1148 CHECK(RTStrIStr("abcdef", "cde"), "cdef");
1149 CHECK(RTStrIStr("abcdef", "cd"), "cdef");
1150 CHECK(RTStrIStr("abcdef", "c"), "cdef");
1151 CHECK(RTStrIStr("abcdef", "f"), "f");
1152 CHECK(RTStrIStr("abcdeF", "F"), "F");
1153 CHECK(RTStrIStr("abcdef", "F"), "f");
1154 CHECK(RTStrIStr("abcdef", "ef"), "ef");
1155 CHECK(RTStrIStr("EeEef", "e"), "EeEef");
1156 CHECK(RTStrIStr("EeEef", "E"), "EeEef");
1157 CHECK(RTStrIStr("EeEef", "EE"), "EeEef");
1158 CHECK(RTStrIStr("EeEef", "EEE"), "EeEef");
1159 CHECK(RTStrIStr("EeEef", "EEEF"), "eEef");
1160 CHECK_NULL(RTStrIStr("EeEef", "z"));
1161
1162#undef CHECK
1163#undef CHECK_NULL
1164 RTTestSubDone(hTest);
1165}
1166
1167
1168void testUtf8Latin1(RTTEST hTest)
1169{
1170 RTTestSub(hTest, "Latin-1 <-> Utf-8 conversion functions");
1171
1172 /* Test Utf8 -> Latin1 */
1173 size_t cch_szAll = 0;
1174 size_t cbShort = RTStrCalcLatin1Len(g_szAll);
1175 RTTEST_CHECK(hTest, cbShort == 0);
1176 int rc = RTStrCalcLatin1LenEx(g_szAll, 383, &cch_szAll);
1177 RTTEST_CHECK(hTest, (cch_szAll == 255));
1178 rc = RTStrCalcLatin1LenEx(g_szAll, RTSTR_MAX, &cch_szAll);
1179 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1180 char *psz = NULL;
1181 char szShort[256] = { 0 };
1182 memcpy(szShort, g_szAll, 255);
1183 cbShort = RTStrCalcLatin1Len(szShort);
1184 RTTEST_CHECK(hTest, cbShort == 191);
1185 rc = RTStrToLatin1(szShort, &psz);
1186 RTTEST_CHECK_RC_OK(hTest, rc);
1187 if (RT_SUCCESS(rc))
1188 {
1189 RTTEST_CHECK(hTest, (strlen(psz) == 191));
1190 for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1191 if (psz[i] != (char) j)
1192 {
1193 RTTestFailed(hTest, "conversion of g_szAll to Latin1 failed at position %u\n", i);
1194 break;
1195 }
1196 }
1197 RTStrFree(psz);
1198 rc = RTStrToLatin1(g_szAll, &psz);
1199 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1200 char sz[512];
1201 char *psz2 = &sz[0];
1202 size_t cchActual = 0;
1203 rc = RTStrToLatin1Ex(g_szAll, sizeof(sz) - 1, &psz2, sizeof(sz),
1204 &cchActual);
1205 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1206 RTTEST_CHECK_MSG(hTest, cchActual == 0,
1207 (hTest, "cchActual=%lu\n", cchActual));
1208 rc = RTStrToLatin1Ex(g_szAll, 383, &psz2, sizeof(sz),
1209 &cchActual);
1210 RTTEST_CHECK_RC_OK(hTest, rc);
1211 if (RT_SUCCESS(rc))
1212 {
1213 RTTEST_CHECK(hTest, (cchActual == 255));
1214 RTTEST_CHECK(hTest, (cchActual == strlen(sz)));
1215 for (unsigned i = 0, j = 1; psz2[i] != '\0'; ++i, ++j)
1216 if (psz2[i] != (char) j)
1217 {
1218 RTTestFailed(hTest, "second conversion of g_szAll to Latin1 failed at position %u\n", i);
1219 break;
1220 }
1221 }
1222 rc = RTStrToLatin1Ex(g_szAll, 129, &psz2, 128, &cchActual);
1223 RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1224 RTTEST_CHECK_MSG(hTest, cchActual == 128,
1225 (hTest, "cchActual=%lu\n", cchActual));
1226 rc = RTStrToLatin1Ex(g_szAll, 383, &psz, 0, &cchActual);
1227 RTTEST_CHECK_RC_OK(hTest, rc);
1228 if (RT_SUCCESS(rc))
1229 {
1230 RTTEST_CHECK(hTest, (cchActual == 255));
1231 RTTEST_CHECK(hTest, (cchActual == strlen(psz)));
1232 for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1233 if ( ((j < 0x100) && (psz[i] != (char) j))
1234 || ((j > 0xff) && psz[i] != '?'))
1235 {
1236 RTTestFailed(hTest, "third conversion of g_szAll to Latin1 failed at position %u\n", i);
1237 break;
1238 }
1239 }
1240 const char *pszBad = "Hello\xDC\xD8";
1241 rc = RTStrToLatin1Ex(pszBad, RTSTR_MAX, &psz2, sizeof(sz),
1242 &cchActual);
1243 RTTEST_CHECK_RC(hTest, rc, VERR_INVALID_UTF8_ENCODING);
1244 RTStrFree(psz);
1245
1246 /* Test Latin1 -> Utf8 */
1247 const char *pszLat1 = "\x01\x20\x40\x80\x81";
1248 RTTEST_CHECK(hTest, RTLatin1CalcUtf8Len(pszLat1) == 7);
1249 rc = RTLatin1CalcUtf8LenEx(pszLat1, 3, &cchActual);
1250 RTTEST_CHECK_RC_OK(hTest, rc);
1251 if (RT_SUCCESS(rc))
1252 RTTEST_CHECK(hTest, cchActual == 3);
1253 rc = RTLatin1CalcUtf8LenEx(pszLat1, RTSTR_MAX, &cchActual);
1254 RTTEST_CHECK_RC_OK(hTest, rc);
1255 if (RT_SUCCESS(rc))
1256 RTTEST_CHECK(hTest, cchActual == 7);
1257 char *pch = NULL;
1258 char ch[8];
1259 char *pch2 = &ch[0];
1260 cchActual = 0;
1261 rc = RTLatin1ToUtf8(pszLat1, &pch);
1262 RTTEST_CHECK_RC_OK(hTest, rc);
1263 if (RT_SUCCESS(rc))
1264 RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1265 RTStrFree(pch);
1266 rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch, 0, &cchActual);
1267 RTTEST_CHECK_RC_OK(hTest, rc);
1268 if (RT_SUCCESS(rc))
1269 {
1270 RTTEST_CHECK(hTest, (cchActual == 7));
1271 RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1272 }
1273 RTStrFree(pch);
1274 rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch, 0, NULL);
1275 RTTEST_CHECK_RC_OK(hTest, rc);
1276 if (RT_SUCCESS(rc))
1277 RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1278 RTStrFree(pch);
1279 rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch2, RT_ELEMENTS(ch),
1280 &cchActual);
1281 RTTEST_CHECK_RC_OK(hTest, rc);
1282 if (RT_SUCCESS(rc))
1283 {
1284 RTTEST_CHECK(hTest, (cchActual == 7));
1285 RTTEST_CHECK(hTest, !strcmp(pch2, "\x01\x20\x40\xC2\x80\xC2\x81"));
1286 }
1287 rc = RTLatin1ToUtf8Ex(pszLat1, 3, &pch2, RT_ELEMENTS(ch),
1288 &cchActual);
1289 RTTEST_CHECK_RC_OK(hTest, rc);
1290 if (RT_SUCCESS(rc))
1291 {
1292 RTTEST_CHECK(hTest, (cchActual == 3));
1293 RTTEST_CHECK(hTest, !strcmp(pch2, "\x01\x20\x40"));
1294 }
1295 rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch2, RT_ELEMENTS(ch) - 1,
1296 &cchActual);
1297 RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1298 RTTEST_CHECK(hTest, (cchActual == 7));
1299 RTTestSubDone(hTest);
1300}
1301
1302
1303void testUtf16Latin1(RTTEST hTest)
1304{
1305 RTTestSub(hTest, "Latin-1 <-> Utf-16 conversion functions");
1306
1307 /* Test Utf16 -> Latin1 */
1308 size_t cch_szAll = 0;
1309 size_t cbShort = RTUtf16CalcLatin1Len(g_wszAll);
1310 RTTEST_CHECK(hTest, cbShort == 0);
1311 int rc = RTUtf16CalcLatin1LenEx(g_wszAll, 255, &cch_szAll);
1312 RTTEST_CHECK(hTest, (cch_szAll == 255));
1313 rc = RTUtf16CalcLatin1LenEx(g_wszAll, RTSTR_MAX, &cch_szAll);
1314 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1315 char *psz = NULL;
1316 RTUTF16 wszShort[256] = { 0 };
1317 for (unsigned i = 0; i < 255; ++i)
1318 wszShort[i] = i + 1;
1319 cbShort = RTUtf16CalcLatin1Len(wszShort);
1320 RTTEST_CHECK(hTest, cbShort == 255);
1321 rc = RTUtf16ToLatin1(wszShort, &psz);
1322 RTTEST_CHECK_RC_OK(hTest, rc);
1323 if (RT_SUCCESS(rc))
1324 {
1325 RTTEST_CHECK(hTest, (strlen(psz) == 255));
1326 for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1327 if (psz[i] != (char) j)
1328 {
1329 RTTestFailed(hTest, "conversion of g_wszAll to Latin1 failed at position %u\n", i);
1330 break;
1331 }
1332 }
1333 RTStrFree(psz);
1334 rc = RTUtf16ToLatin1(g_wszAll, &psz);
1335 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1336 char sz[512];
1337 char *psz2 = &sz[0];
1338 size_t cchActual = 0;
1339 rc = RTUtf16ToLatin1Ex(g_wszAll, sizeof(sz) - 1, &psz2, sizeof(sz),
1340 &cchActual);
1341 RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1342 RTTEST_CHECK_MSG(hTest, cchActual == 0,
1343 (hTest, "cchActual=%lu\n", cchActual));
1344 rc = RTUtf16ToLatin1Ex(g_wszAll, 255, &psz2, sizeof(sz),
1345 &cchActual);
1346 RTTEST_CHECK_RC_OK(hTest, rc);
1347 if (RT_SUCCESS(rc))
1348 {
1349 RTTEST_CHECK(hTest, (cchActual == 255));
1350 RTTEST_CHECK(hTest, (cchActual == strlen(sz)));
1351 for (unsigned i = 0, j = 1; psz2[i] != '\0'; ++i, ++j)
1352 if (psz2[i] != (char) j)
1353 {
1354 RTTestFailed(hTest, "second conversion of g_wszAll to Latin1 failed at position %u\n", i);
1355 break;
1356 }
1357 }
1358 rc = RTUtf16ToLatin1Ex(g_wszAll, 128, &psz2, 128, &cchActual);
1359 RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1360 RTTEST_CHECK_MSG(hTest, cchActual == 128,
1361 (hTest, "cchActual=%lu\n", cchActual));
1362 rc = RTUtf16ToLatin1Ex(g_wszAll, 255, &psz, 0, &cchActual);
1363 RTTEST_CHECK_RC_OK(hTest, rc);
1364 if (RT_SUCCESS(rc))
1365 {
1366 RTTEST_CHECK(hTest, (cchActual == 255));
1367 RTTEST_CHECK(hTest, (cchActual == strlen(psz)));
1368 for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1369 if ( ((j < 0x100) && (psz[i] != (char) j))
1370 || ((j > 0xff) && psz[i] != '?'))
1371 {
1372 RTTestFailed(hTest, "third conversion of g_wszAll to Latin1 failed at position %u\n", i);
1373 break;
1374 }
1375 }
1376 const char *pszBad = "H\0e\0l\0l\0o\0\0\xDC\0\xD8\0";
1377 rc = RTUtf16ToLatin1Ex((RTUTF16 *) pszBad, RTSTR_MAX, &psz2, sizeof(sz),
1378 &cchActual);
1379 RTTEST_CHECK_RC(hTest, rc, VERR_INVALID_UTF16_ENCODING);
1380 RTStrFree(psz);
1381
1382 /* Test Latin1 -> Utf16 */
1383 const char *pszLat1 = "\x01\x20\x40\x80\x81";
1384 RTTEST_CHECK(hTest, RTLatin1CalcUtf16Len(pszLat1) == 5);
1385 rc = RTLatin1CalcUtf16LenEx(pszLat1, 3, &cchActual);
1386 RTTEST_CHECK_RC_OK(hTest, rc);
1387 if (RT_SUCCESS(rc))
1388 RTTEST_CHECK(hTest, cchActual == 3);
1389 rc = RTLatin1CalcUtf16LenEx(pszLat1, RTSTR_MAX, &cchActual);
1390 RTTEST_CHECK_RC_OK(hTest, rc);
1391 if (RT_SUCCESS(rc))
1392 RTTEST_CHECK(hTest, cchActual == 5);
1393 RTUTF16 *pwc = NULL;
1394 RTUTF16 wc[6];
1395 RTUTF16 *pwc2 = &wc[0];
1396 size_t cwActual = 0;
1397 rc = RTLatin1ToUtf16(pszLat1, &pwc);
1398 RTTEST_CHECK_RC_OK(hTest, rc);
1399 if (RT_SUCCESS(rc))
1400 RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1401 && (pwc[2] == 0x40) && (pwc[3] == 0x80)
1402 && (pwc[4] == 0x81) && (pwc[5] == '\0'));
1403 RTUtf16Free(pwc);
1404 rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc, 0, &cwActual);
1405 RTTEST_CHECK_RC_OK(hTest, rc);
1406 if (RT_SUCCESS(rc))
1407 {
1408 RTTEST_CHECK(hTest, (cwActual == 5));
1409 RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1410 && (pwc[2] == 0x40) && (pwc[3] == 0x80)
1411 && (pwc[4] == 0x81) && (pwc[5] == '\0'));
1412 }
1413 RTUtf16Free(pwc);
1414 rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc, 0, NULL);
1415 RTTEST_CHECK_RC_OK(hTest, rc);
1416 if (RT_SUCCESS(rc))
1417 RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1418 && (pwc[2] == 0x40) && (pwc[3] == 0x80)
1419 && (pwc[4] == 0x81) && (pwc[5] == '\0'));
1420 RTUtf16Free(pwc);
1421 rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc2, RT_ELEMENTS(wc),
1422 &cwActual);
1423 RTTEST_CHECK_RC_OK(hTest, rc);
1424 if (RT_SUCCESS(rc))
1425 {
1426 RTTEST_CHECK(hTest, (cwActual == 5));
1427 RTTEST_CHECK(hTest, (wc[0] == 1) && (wc[1] == 0x20)
1428 && (wc[2] == 0x40) && (wc[3] == 0x80)
1429 && (wc[4] == 0x81) && (wc[5] == '\0'));
1430 }
1431 rc = RTLatin1ToUtf16Ex(pszLat1, 3, &pwc2, RT_ELEMENTS(wc),
1432 &cwActual);
1433 RTTEST_CHECK_RC_OK(hTest, rc);
1434 if (RT_SUCCESS(rc))
1435 {
1436 RTTEST_CHECK(hTest, (cwActual == 3));
1437 RTTEST_CHECK(hTest, (wc[0] == 1) && (wc[1] == 0x20)
1438 && (wc[2] == 0x40) && (wc[3] == '\0'));
1439 }
1440 rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc2, RT_ELEMENTS(wc) - 1,
1441 &cwActual);
1442 RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1443 RTTEST_CHECK(hTest, (cwActual == 5));
1444 RTTestSubDone(hTest);
1445}
1446
1447
1448static void testNoTransation(RTTEST hTest)
1449{
1450 /*
1451 * Try trigger a VERR_NO_TRANSLATION error in convert to
1452 * current CP to latin-1.
1453 */
1454 const RTUTF16 s_swzTest1[] = { 0x2358, 0x2242, 0x2357, 0x2359, 0x22f9, 0x2c4e, 0x0030, 0x0060,
1455 0x0092, 0x00c1, 0x00f2, 0x1f80, 0x0088, 0x2c38, 0x2c30, 0x0000 };
1456 char *pszTest1;
1457 int rc = RTUtf16ToUtf8(s_swzTest1, &pszTest1);
1458 RTTESTI_CHECK_RC_RETV(rc, VINF_SUCCESS);
1459
1460 RTTestSub(hTest, "VERR_NO_TRANSLATION/RTStrUtf8ToCurrentCP");
1461 char *pszOut;
1462 rc = RTStrUtf8ToCurrentCP(&pszOut, pszTest1);
1463 if (rc == VINF_SUCCESS)
1464 {
1465 RTTestIPrintf(RTTESTLVL_ALWAYS, "CurrentCP is UTF-8 or similar (LC_ALL=%s LANG=%s LC_CTYPE=%s)\n",
1466 RTEnvGet("LC_ALL"), RTEnvGet("LANG"), RTEnvGet("LC_CTYPE"));
1467 if (strcmp(pszOut, pszTest1))
1468 RTTestFailed(hTest, "mismatch\nutf8: %.*Rhxs\n got: %.*Rhxs\n", strlen(pszTest1), pszTest1, strlen(pszOut), pszOut);
1469 RTStrFree(pszOut);
1470 }
1471 else
1472 RTTESTI_CHECK_MSG(rc == VWRN_NO_TRANSLATION || rc == VERR_NO_TRANSLATION, ("rc=%Rrc\n", rc));
1473
1474 RTTestSub(hTest, "VERR_NO_TRANSLATION/RTUtf16ToLatin1");
1475 rc = RTUtf16ToLatin1(s_swzTest1, &pszOut);
1476 RTTESTI_CHECK_RC(rc, VERR_NO_TRANSLATION);
1477 if (RT_SUCCESS(rc))
1478 RTStrFree(pszOut);
1479
1480 RTStrFree(pszTest1);
1481 RTTestSubDone(hTest);
1482}
1483
1484static void testGetPut(RTTEST hTest)
1485{
1486 /*
1487 * Test RTStrPutCp, RTStrGetCp and RTStrGetCpEx.
1488 */
1489 RTTestSub(hTest, "RTStrPutCp, RTStrGetCp and RTStrGetCpEx");
1490
1491 RTUNICP uc = 0;
1492 while (uc <= 0x10fffd)
1493 {
1494 /* Figure the range - skip illegal ranges. */
1495 RTUNICP ucFirst = uc;
1496 if (ucFirst - UINT32_C(0xd800) <= 0x7ff)
1497 ucFirst = 0xe000;
1498 else if (ucFirst == UINT32_C(0xfffe) || ucFirst == UINT32_C(0xffff))
1499 ucFirst = 0x10000;
1500
1501 RTUNICP ucLast = ucFirst + 1023;
1502 if (ucLast - UINT32_C(0xd800) <= 0x7ff)
1503 ucLast = 0xd7ff;
1504 else if (ucLast == UINT32_C(0xfffe) || ucLast == UINT32_C(0xffff))
1505 ucLast = 0xfffd;
1506
1507 /* Encode the range into a string, decode each code point as we go along. */
1508 char sz1[8192];
1509 char *pszDst = sz1;
1510 for (uc = ucFirst; uc <= ucLast; uc++)
1511 {
1512 char *pszBefore = pszDst;
1513 pszDst = RTStrPutCp(pszDst, uc);
1514 RTTESTI_CHECK(pszBefore - pszDst < 6);
1515
1516 RTUNICP uc2 = RTStrGetCp(pszBefore);
1517 RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1518
1519 const char *pszSrc = pszBefore;
1520 RTUNICP uc3 = 42;
1521 RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1522 RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1523 RTTESTI_CHECK_MSG(pszSrc == pszDst, ("pszSrc=%p pszDst=%p\n", pszSrc, pszDst));
1524 }
1525
1526 /* Decode and re-encode it. */
1527 const char *pszSrc = pszDst = sz1;
1528 for (uc = ucFirst; uc <= ucLast; uc++)
1529 {
1530 RTUNICP uc2 = RTStrGetCp(pszSrc);
1531 RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1532
1533 RTUNICP uc3 = 42;
1534 RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1535 RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1536
1537 pszDst = RTStrPutCp(pszDst, uc);
1538 RTTESTI_CHECK_MSG(pszSrc == pszDst, ("pszSrc=%p pszDst=%p\n", pszSrc, pszDst));
1539 pszSrc = pszDst;
1540 }
1541
1542 /* Decode and wipe it (checking compiler optimizations). */
1543 pszSrc = pszDst = sz1;
1544 for (uc = ucFirst; uc <= ucLast; uc++)
1545 {
1546 RTUNICP uc2 = RTStrGetCp(pszSrc);
1547 RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1548
1549 RTUNICP uc3 = 42;
1550 RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1551 RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1552
1553 pszDst = RTStrPutCp(pszDst, 0);
1554 }
1555
1556 /* advance */
1557 uc = ucLast + 1;
1558 }
1559
1560}
1561
1562
1563int main()
1564{
1565 /*
1566 * Init the runtime, test and say hello.
1567 */
1568 RTTEST hTest;
1569 RTEXITCODE rcExit = RTTestInitAndCreate("tstUtf8", &hTest);
1570 if (rcExit != RTEXITCODE_SUCCESS)
1571 return rcExit;
1572 RTTestBanner(hTest);
1573
1574 /*
1575 * Run the tests.
1576 */
1577 InitStrings();
1578 test1(hTest);
1579 test2(hTest);
1580 test3(hTest);
1581 TstRTStrXCmp(hTest);
1582 TstRTStrPurgeEncoding(hTest);
1583 /* TstRT*PurgeComplementSet test conditions which assert. */
1584 TstRTStrPurgeComplementSet(hTest);
1585 TstRTUtf16PurgeComplementSet(hTest);
1586 testStrEnd(hTest);
1587 testStrStr(hTest);
1588 testUtf8Latin1(hTest);
1589 testUtf16Latin1(hTest);
1590 testNoTransation(hTest);
1591 testGetPut(hTest);
1592
1593 Benchmarks(hTest);
1594
1595 /*
1596 * Summary
1597 */
1598 return RTTestSummaryAndDestroy(hTest);
1599}
1600
Note: See TracBrowser for help on using the repository browser.

© 2024 Oracle Support Privacy / Do Not Sell My Info Terms of Use Trademark Policy Automated Access Etiquette