tstUtf8.cpp@ 93174

Last change on this file since 93174 was 93115, checked in by vboxsync, 3 years ago
scm --update-copyright-year
Property svn:eol-style set to `native` Property svn:keywords set to `Id Revision`
File size: 57.6 KB

Line
1	/* $Id: tstUtf8.cpp 93115 2022-01-01 11:31:46Z vboxsync $ */
2	/** @file
3	* IPRT Testcase - UTF-8 and UTF-16 string conversions.
4	*/
5
6	/*
7	* Copyright (C) 2006-2022 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	*/
51	static 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	*/
65	static 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	RTStrFree(pszUtf8);
102	rc = RTStrCurrentCPToUtf8(&pszUtf8, pszCurrent);
103	if (rc == VINF_SUCCESS)
104	RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> Current -> UTF-8 successful.\n");
105	else
106	RTTestFailed(hTest, "%d: The third part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
107	__LINE__, rc);
108	if (RT_SUCCESS(rc))
109	RTStrFree(pszUtf8);
110	RTStrFree(pszCurrent);
111	}
112	else
113	{
114	if (rc == VERR_NO_TRANSLATION)
115	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");
116	else if (rc == VWRN_NO_TRANSLATION)
117	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");
118	else
119	RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
120	__LINE__, rc);
121	if (RT_SUCCESS(rc))
122	RTStrFree(pszCurrent);
123	RTStrFree(pszUtf8);
124	}
125	}
126	else
127	RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> Current -> UTF-8 failed with return value %Rrc.",
128	__LINE__, rc);
129	RTMemFree(pwszRand);
130
131	/*
132	* Generate a new random string.
133	*/
134	RTTestSub(hTest, "Random UTF-16 -> UTF-8 -> UTF-16");
135	pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
136	for (int i = 0; i < 30; i++)
137	pwszRand[i] = GetRandUtf16();
138	pwszRand[30] = 0;
139	rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
140	if (rc == VINF_SUCCESS)
141	{
142	rc = RTStrToUtf16(pszUtf8, &pwsz);
143	if (rc == VINF_SUCCESS)
144	{
145	int i;
146	for (i = 0; pwszRand[i] == pwsz[i] && pwsz[i] != 0; i++)
147	/* nothing */;
148	if (pwszRand[i] == pwsz[i] && pwsz[i] == 0)
149	RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> UTF-16 successful.\n");
150	else
151	{
152	RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> UTF-16 failed.", __LINE__);
153	RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz[i]);
154	}
155	RTUtf16Free(pwsz);
156	}
157	else
158	RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> UTF-16 failed with return value %Rrc.",
159	__LINE__, rc);
160	RTStrFree(pszUtf8);
161	}
162	else
163	RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> UTF-16 failed with return value %Rrc.",
164	__LINE__, rc);
165	RTMemFree(pwszRand);
166
167	/*
168	* Generate yet another random string and convert it to a buffer.
169	*/
170	RTTestSub(hTest, "Random RTUtf16ToUtf8Ex + RTStrToUtf16");
171	pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
172	for (int i = 0; i < 30; i++)
173	pwszRand[i] = GetRandUtf16();
174	pwszRand[30] = 0;
175
176	char szUtf8Array[120];
177	char *pszUtf8Array = szUtf8Array;
178	rc = RTUtf16ToUtf8Ex(pwszRand, RTSTR_MAX, &pszUtf8Array, 120, NULL);
179	if (rc == 0)
180	{
181	rc = RTStrToUtf16(pszUtf8Array, &pwsz);
182	if (rc == 0)
183	{
184	int i;
185	for (i = 0; pwszRand[i] == pwsz[i] && pwsz[i] != 0; i++)
186	;
187	if (pwsz[i] == 0 && i >= 8)
188	RTTestPassed(hTest, "Random UTF-16 -> fixed length UTF-8 -> UTF-16 successful.\n");
189	else
190	{
191	RTTestFailed(hTest, "%d: Incorrect conversion of UTF-16 -> fixed length UTF-8 -> UTF-16.\n", __LINE__);
192	RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz[i]);
193	}
194	RTUtf16Free(pwsz);
195	}
196	else
197	RTTestFailed(hTest, "%d: The second part of random UTF-16 -> fixed length UTF-8 -> UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
198	}
199	else
200	RTTestFailed(hTest, "%d: The first part of random UTF-16 -> fixed length UTF-8 -> UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
201	RTMemFree(pwszRand);
202
203	/*
204	* And again.
205	*/
206	RTTestSub(hTest, "Random RTUtf16ToUtf8 + RTStrToUtf16Ex");
207	pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
208	for (int i = 0; i < 30; i++)
209	pwszRand[i] = GetRandUtf16();
210	pwszRand[30] = 0;
211
212	RTUTF16 wszBuf[70];
213	PRTUTF16 pwsz2Buf = wszBuf;
214	rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
215	if (rc == 0)
216	{
217	rc = RTStrToUtf16Ex(pszUtf8, RTSTR_MAX, &pwsz2Buf, 70, NULL);
218	if (rc == 0)
219	{
220	int i;
221	for (i = 0; pwszRand[i] == pwsz2Buf[i] && pwsz2Buf[i] != 0; i++)
222	;
223	if (pwszRand[i] == 0 && pwsz2Buf[i] == 0)
224	RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> fixed length UTF-16 successful.\n");
225	else
226	{
227	RTTestFailed(hTest, "%d: Incorrect conversion of random UTF-16 -> UTF-8 -> fixed length UTF-16.\n", __LINE__);
228	RTTestPrintf(hTest, RTTESTLVL_FAILURE, "First differing character is at position %d and has the value %x.\n", i, pwsz2Buf[i]);
229	}
230	}
231	else
232	RTTestFailed(hTest, "%d: The second part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n", __LINE__, rc);
233	RTStrFree(pszUtf8);
234	}
235	else
236	RTTestFailed(hTest, "%d: The first part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n",
237	__LINE__, rc);
238	RTMemFree(pwszRand);
239
240	pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
241	for (int i = 0; i < 30; i++)
242	pwszRand[i] = GetRandUtf16();
243	pwszRand[30] = 0;
244
245	rc = RTUtf16ToUtf8Ex(pwszRand, RTSTR_MAX, &pszUtf8Array, 20, NULL);
246	if (rc == VERR_BUFFER_OVERFLOW)
247	RTTestPassed(hTest, "Random UTF-16 -> fixed length UTF-8 with too short buffer successfully rejected.\n");
248	else
249	RTTestFailed(hTest, "%d: Random UTF-16 -> fixed length UTF-8 with too small buffer returned value %d instead of VERR_BUFFER_OVERFLOW.\n",
250	__LINE__, rc);
251	RTMemFree(pwszRand);
252
253	/*
254	* last time...
255	*/
256	RTTestSub(hTest, "Random RTUtf16ToUtf8 + RTStrToUtf16Ex");
257	pwszRand = (PRTUTF16)RTMemAlloc(31 * sizeof(*pwsz));
258	for (int i = 0; i < 30; i++)
259	pwszRand[i] = GetRandUtf16();
260	pwszRand[30] = 0;
261
262	rc = RTUtf16ToUtf8(pwszRand, &pszUtf8);
263	if (rc == VINF_SUCCESS)
264	{
265	rc = RTStrToUtf16Ex(pszUtf8, RTSTR_MAX, &pwsz2Buf, 20, NULL);
266	if (rc == VERR_BUFFER_OVERFLOW)
267	RTTestPassed(hTest, "Random UTF-16 -> UTF-8 -> fixed length UTF-16 with too short buffer successfully rejected.\n");
268	else
269	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",
270	__LINE__, rc);
271	RTStrFree(pszUtf8);
272	}
273	else
274	RTTestFailed(hTest, "%d:The first part of random UTF-16 -> UTF-8 -> fixed length UTF-16 failed with return value %Rrc.\n",
275	__LINE__, rc);
276	RTMemFree(pwszRand);
277
278	RTTestSubDone(hTest);
279	}
280
281
282	static RTUNICP g_uszAll[0x110000 - 1 - 0x800 - 2 + 1];
283	static RTUTF16 g_wszAll[0xfffe - (0xe000 - 0xd800) + (0x110000 - 0x10000) * 2];
284	static char g_szAll[0x7f + (0x800 - 0x80) * 2 + (0xfffe - 0x800 - (0xe000 - 0xd800))* 3 + (0x110000 - 0x10000) * 4 + 1];
285
286	static void whereami(int cBits, size_t off)
287	{
288	if (cBits == 8)
289	{
290	if (off < 0x7f)
291	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", off + 1);
292	else if (off < 0xf7f)
293	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x7f) / 2 + 0x80);
294	else if (off < 0x27f7f)
295	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0xf7f) / 3 + 0x800);
296	else if (off < 0x2df79)
297	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x27f7f) / 3 + 0xe000);
298	else if (off < 0x42df79)
299	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 U+%#x\n", (off - 0x2df79) / 4 + 0x10000);
300	else
301	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-8 ???\n");
302	}
303	else if (cBits == 16)
304	{
305	if (off < 0xd7ff*2)
306	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", off / 2 + 1);
307	else if (off < 0xf7fd*2)
308	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", (off - 0xd7ff*2) / 2 + 0xe000);
309	else if (off < 0x20f7fd)
310	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 U+%#x\n", (off - 0xf7fd*2) / 4 + 0x10000);
311	else
312	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "UTF-16 ???\n");
313	}
314	else
315	{
316	if (off < (0xd800 - 1) * sizeof(RTUNICP))
317	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 1);
318	else if (off < (0xfffe - 0x800 - 1) * sizeof(RTUNICP))
319	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 0x800 + 1);
320	else
321	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "RTUNICP U+%#x\n", off / sizeof(RTUNICP) + 0x800 + 1 + 2);
322	}
323	}
324
325	int mymemcmp(const void pv1, const void pv2, size_t cb, int cBits)
326	{
327	const uint8_t pb1 = (const uint8_t )pv1;
328	const uint8_t pb2 = (const uint8_t )pv2;
329	for (size_t off = 0; off < cb; off++)
330	{
331	if (pb1[off] != pb2[off])
332	{
333	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "mismatch at %#x: ", off);
334	whereami(cBits, off);
335	if (off > 0)
336	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, " %#x: %02x != %02x!\n", off-1, pb1[off-1], pb2[off-1]);
337	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, "*%#x: %02x != %02x!\n", off, pb1[off], pb2[off]);
338	for (size_t i = 1; i < 10; i++)
339	if (off + i < cb)
340	RTTestPrintf(NIL_RTTEST, RTTESTLVL_FAILURE, " %#x: %02x != %02x!\n", off+i, pb1[off+i], pb2[off+i]);
341	return 1;
342	}
343	}
344	return 0;
345	}
346
347
348	void InitStrings()
349	{
350	/*
351	* Generate unicode string containing all the legal UTF-16 codepoints, both UTF-16 and UTF-8 version.
352	*/
353	/* the simple code point array first */
354	unsigned i = 0;
355	RTUNICP uc = 1;
356	while (uc < 0xd800)
357	g_uszAll[i++] = uc++;
358	uc = 0xe000;
359	while (uc < 0xfffe)
360	g_uszAll[i++] = uc++;
361	uc = 0x10000;
362	while (uc < 0x110000)
363	g_uszAll[i++] = uc++;
364	g_uszAll[i++] = 0;
365	Assert(RT_ELEMENTS(g_uszAll) == i);
366
367	/* the utf-16 one */
368	i = 0;
369	uc = 1;
370	//RTPrintf("tstUtf8: %#x=%#x", i, uc);
371	while (uc < 0xd800)
372	g_wszAll[i++] = uc++;
373	uc = 0xe000;
374	//RTPrintf(" %#x=%#x", i, uc);
375	while (uc < 0xfffe)
376	g_wszAll[i++] = uc++;
377	uc = 0x10000;
378	//RTPrintf(" %#x=%#x", i, uc);
379	while (uc < 0x110000)
380	{
381	g_wszAll[i++] = 0xd800 \| ((uc - 0x10000) >> 10);
382	g_wszAll[i++] = 0xdc00 \| ((uc - 0x10000) & 0x3ff);
383	uc++;
384	}
385	//RTPrintf(" %#x=%#x\n", i, uc);
386	g_wszAll[i++] = '\0';
387	Assert(RT_ELEMENTS(g_wszAll) == i);
388
389	/*
390	* The utf-8 one
391	*/
392	i = 0;
393	uc = 1;
394	//RTPrintf("tstUtf8: %#x=%#x", i, uc);
395	while (uc < 0x80)
396	g_szAll[i++] = uc++;
397	//RTPrintf(" %#x=%#x", i, uc);
398	while (uc < 0x800)
399	{
400	g_szAll[i++] = 0xc0 \| (uc >> 6);
401	g_szAll[i++] = 0x80 \| (uc & 0x3f);
402	Assert(!((uc >> 6) & ~0x1f));
403	uc++;
404	}
405	//RTPrintf(" %#x=%#x", i, uc);
406	while (uc < 0xd800)
407	{
408	g_szAll[i++] = 0xe0 \| (uc >> 12);
409	g_szAll[i++] = 0x80 \| ((uc >> 6) & 0x3f);
410	g_szAll[i++] = 0x80 \| (uc & 0x3f);
411	Assert(!((uc >> 12) & ~0xf));
412	uc++;
413	}
414	uc = 0xe000;
415	//RTPrintf(" %#x=%#x", i, uc);
416	while (uc < 0xfffe)
417	{
418	g_szAll[i++] = 0xe0 \| (uc >> 12);
419	g_szAll[i++] = 0x80 \| ((uc >> 6) & 0x3f);
420	g_szAll[i++] = 0x80 \| (uc & 0x3f);
421	Assert(!((uc >> 12) & ~0xf));
422	uc++;
423	}
424	uc = 0x10000;
425	//RTPrintf(" %#x=%#x", i, uc);
426	while (uc < 0x110000)
427	{
428	g_szAll[i++] = 0xf0 \| (uc >> 18);
429	g_szAll[i++] = 0x80 \| ((uc >> 12) & 0x3f);
430	g_szAll[i++] = 0x80 \| ((uc >> 6) & 0x3f);
431	g_szAll[i++] = 0x80 \| (uc & 0x3f);
432	Assert(!((uc >> 18) & ~0x7));
433	uc++;
434	}
435	//RTPrintf(" %#x=%#x\n", i, uc);
436	g_szAll[i++] = '\0';
437	Assert(RT_ELEMENTS(g_szAll) == i);
438	}
439
440
441	void test2(RTTEST hTest)
442	{
443	/*
444	* Convert to UTF-8 and back.
445	*/
446	RTTestSub(hTest, "UTF-16 -> UTF-8 -> UTF-16");
447	char *pszUtf8;
448	int rc = RTUtf16ToUtf8(&g_wszAll[0], &pszUtf8);
449	if (rc == VINF_SUCCESS)
450	{
451	pszUtf8[0] = 1;
452	if (mymemcmp(pszUtf8, g_szAll, sizeof(g_szAll), 8))
453	RTTestFailed(hTest, "UTF-16 -> UTF-8 mismatch!");
454
455	PRTUTF16 pwszUtf16;
456	rc = RTStrToUtf16(pszUtf8, &pwszUtf16);
457	if (rc == VINF_SUCCESS)
458	{
459	if (mymemcmp(pwszUtf16, g_wszAll, sizeof(g_wszAll), 16))
460	RTTestFailed(hTest, "UTF-8 -> UTF-16 failed compare!");
461	RTUtf16Free(pwszUtf16);
462	}
463	else
464	RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
465	RTStrFree(pszUtf8);
466	}
467	else
468	RTTestFailed(hTest, "UTF-16 -> UTF-8 failed, rc=%Rrc.", rc);
469
470
471	/*
472	* Convert to UTF-16 and back. (just in case the above test fails)
473	*/
474	RTTestSub(hTest, "UTF-8 -> UTF-16 -> UTF-8");
475	PRTUTF16 pwszUtf16;
476	rc = RTStrToUtf16(&g_szAll[0], &pwszUtf16);
477	if (rc == VINF_SUCCESS)
478	{
479	if (mymemcmp(pwszUtf16, g_wszAll, sizeof(g_wszAll), 16))
480	RTTestFailed(hTest, "UTF-8 -> UTF-16 failed compare!");
481
482	rc = RTUtf16ToUtf8(pwszUtf16, &pszUtf8);
483	if (rc == VINF_SUCCESS)
484	{
485	if (mymemcmp(pszUtf8, g_szAll, sizeof(g_szAll), 8))
486	RTTestFailed(hTest, "UTF-16 -> UTF-8 failed compare!");
487	RTStrFree(pszUtf8);
488	}
489	else
490	RTTestFailed(hTest, "UTF-16 -> UTF-8 failed, rc=%Rrc.", rc);
491	RTUtf16Free(pwszUtf16);
492	}
493	else
494	RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
495
496	/*
497	* Convert UTF-8 to CPs.
498	*/
499	RTTestSub(hTest, "UTF-8 -> UNI -> UTF-8");
500	PRTUNICP paCps;
501	rc = RTStrToUni(g_szAll, &paCps);
502	if (rc == VINF_SUCCESS)
503	{
504	if (mymemcmp(paCps, g_uszAll, sizeof(g_uszAll), 32))
505	RTTestFailed(hTest, "UTF-8 -> UTF-16 failed, rc=%Rrc.", rc);
506
507	size_t cCps;
508	rc = RTStrToUniEx(g_szAll, RTSTR_MAX, &paCps, RT_ELEMENTS(g_uszAll), &cCps);
509	if (rc == VINF_SUCCESS)
510	{
511	if (cCps != RT_ELEMENTS(g_uszAll) - 1)
512	RTTestFailed(hTest, "wrong Code Point count %zu, expected %zu\n", cCps, RT_ELEMENTS(g_uszAll) - 1);
513	}
514	else
515	RTTestFailed(hTest, "UTF-8 -> Code Points failed, rc=%Rrc.\n", rc);
516
517	/** @todo RTCpsToUtf8 or something. */
518	RTUniFree(paCps);
519	}
520	else
521	RTTestFailed(hTest, "UTF-8 -> Code Points failed, rc=%Rrc.\n", rc);
522
523	/*
524	* Check the various string lengths.
525	*/
526	RTTestSub(hTest, "Lengths");
527	size_t cuc1 = RTStrCalcUtf16Len(g_szAll);
528	size_t cuc2 = RTUtf16Len(g_wszAll);
529	if (cuc1 != cuc2)
530	RTTestFailed(hTest, "cuc1=%zu != cuc2=%zu\n", cuc1, cuc2);
531	//size_t cuc3 = RTUniLen(g_uszAll);
532
533
534	/*
535	* Enumerate the strings.
536	*/
537	RTTestSub(hTest, "Code Point Getters and Putters");
538	char pszPut1Base = (char )RTMemAlloc(sizeof(g_szAll));
539	AssertRelease(pszPut1Base);
540	char *pszPut1 = pszPut1Base;
541	PRTUTF16 pwszPut2Base = (PRTUTF16)RTMemAlloc(sizeof(g_wszAll));
542	AssertRelease(pwszPut2Base);
543	PRTUTF16 pwszPut2 = pwszPut2Base;
544	const char *psz1 = g_szAll;
545	const char *psz2 = g_szAll;
546	PCRTUTF16 pwsz3 = g_wszAll;
547	PCRTUTF16 pwsz4 = g_wszAll;
548	for (;;)
549	{
550	/*
551	* getters
552	*/
553	RTUNICP uc1;
554	rc = RTStrGetCpEx(&psz1, &uc1);
555	if (RT_FAILURE(rc))
556	{
557	RTTestFailed(hTest, "RTStrGetCpEx failed with rc=%Rrc at %.10Rhxs", rc, psz2);
558	whereami(8, psz2 - &g_szAll[0]);
559	break;
560	}
561	char *pszPrev1 = RTStrPrevCp(g_szAll, psz1);
562	if (pszPrev1 != psz2)
563	{
564	RTTestFailed(hTest, "RTStrPrevCp returned %p expected %p!", pszPrev1, psz2);
565	whereami(8, psz2 - &g_szAll[0]);
566	break;
567	}
568	RTUNICP uc2 = RTStrGetCp(psz2);
569	if (uc2 != uc1)
570	{
571	RTTestFailed(hTest, "RTStrGetCpEx and RTStrGetCp returned different CPs: %RTunicp != %RTunicp", uc2, uc1);
572	whereami(8, psz2 - &g_szAll[0]);
573	break;
574	}
575	psz2 = RTStrNextCp(psz2);
576	if (psz2 != psz1)
577	{
578	RTTestFailed(hTest, "RTStrGetCpEx and RTStrGetNext returned different next pointer!");
579	whereami(8, psz2 - &g_szAll[0]);
580	break;
581	}
582
583	RTUNICP uc3;
584	rc = RTUtf16GetCpEx(&pwsz3, &uc3);
585	if (RT_FAILURE(rc))
586	{
587	RTTestFailed(hTest, "RTUtf16GetCpEx failed with rc=%Rrc at %.10Rhxs", rc, pwsz4);
588	whereami(16, pwsz4 - &g_wszAll[0]);
589	break;
590	}
591	if (uc3 != uc2)
592	{
593	RTTestFailed(hTest, "RTUtf16GetCpEx and RTStrGetCp returned different CPs: %RTunicp != %RTunicp", uc3, uc2);
594	whereami(16, pwsz4 - &g_wszAll[0]);
595	break;
596	}
597	RTUNICP uc4 = RTUtf16GetCp(pwsz4);
598	if (uc3 != uc4)
599	{
600	RTTestFailed(hTest, "RTUtf16GetCpEx and RTUtf16GetCp returned different CPs: %RTunicp != %RTunicp", uc3, uc4);
601	whereami(16, pwsz4 - &g_wszAll[0]);
602	break;
603	}
604	pwsz4 = RTUtf16NextCp(pwsz4);
605	if (pwsz4 != pwsz3)
606	{
607	RTTestFailed(hTest, "RTUtf16GetCpEx and RTUtf16GetNext returned different next pointer!");
608	whereami(8, pwsz4 - &g_wszAll[0]);
609	break;
610	}
611
612
613	/*
614	* putters
615	*/
616	pszPut1 = RTStrPutCp(pszPut1, uc1);
617	if (pszPut1 - pszPut1Base != psz1 - &g_szAll[0])
618	{
619	RTTestFailed(hTest, "RTStrPutCp is not at the same offset! %p != %p",
620	pszPut1 - pszPut1Base, psz1 - &g_szAll[0]);
621	whereami(8, psz2 - &g_szAll[0]);
622	break;
623	}
624
625	pwszPut2 = RTUtf16PutCp(pwszPut2, uc3);
626	if (pwszPut2 - pwszPut2Base != pwsz3 - &g_wszAll[0])
627	{
628	RTTestFailed(hTest, "RTStrPutCp is not at the same offset! %p != %p",
629	pwszPut2 - pwszPut2Base, pwsz3 - &g_wszAll[0]);
630	whereami(8, pwsz4 - &g_wszAll[0]);
631	break;
632	}
633
634
635	/* the end? */
636	if (!uc1)
637	break;
638	}
639
640	/* check output if we seems to have made it thru it all. */
641	if (psz2 == &g_szAll[sizeof(g_szAll)])
642	{
643	if (mymemcmp(pszPut1Base, g_szAll, sizeof(g_szAll), 8))
644	RTTestFailed(hTest, "RTStrPutCp encoded the string incorrectly.");
645	if (mymemcmp(pwszPut2Base, g_wszAll, sizeof(g_wszAll), 16))
646	RTTestFailed(hTest, "RTUtf16PutCp encoded the string incorrectly.");
647	}
648
649	RTMemFree(pszPut1Base);
650	RTMemFree(pwszPut2Base);
651
652	RTTestSubDone(hTest);
653	}
654
655
656	/**
657	* Check case insensitivity.
658	*/
659	void test3(RTTEST hTest)
660	{
661	RTTestSub(hTest, "Case Sensitivity");
662
663	if ( RTUniCpToLower('a') != 'a'
664	\|\| RTUniCpToLower('A') != 'a'
665	\|\| RTUniCpToLower('b') != 'b'
666	\|\| RTUniCpToLower('B') != 'b'
667	\|\| RTUniCpToLower('Z') != 'z'
668	\|\| RTUniCpToLower('z') != 'z'
669	\|\| RTUniCpToUpper('c') != 'C'
670	\|\| RTUniCpToUpper('C') != 'C'
671	\|\| RTUniCpToUpper('z') != 'Z'
672	\|\| RTUniCpToUpper('Z') != 'Z')
673	RTTestFailed(hTest, "RTUniToUpper/Lower failed basic tests.\n");
674
675	if (RTUtf16ICmp(g_wszAll, g_wszAll))
676	RTTestFailed(hTest, "RTUtf16ICmp failed the basic test.\n");
677
678	if (RTUtf16Cmp(g_wszAll, g_wszAll))
679	RTTestFailed(hTest, "RTUtf16Cmp failed the basic test.\n");
680
681	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 };
682	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 };
683	if ( RTUtf16ICmp(s_wszTst1b, s_wszTst1b)
684	\|\| RTUtf16ICmp(s_wszTst1a, s_wszTst1a)
685	\|\| RTUtf16ICmp(s_wszTst1a, s_wszTst1b)
686	\|\| RTUtf16ICmp(s_wszTst1b, s_wszTst1a)
687	)
688	RTTestFailed(hTest, "RTUtf16ICmp failed the alphabet test.\n");
689
690	if ( RTUtf16Cmp(s_wszTst1b, s_wszTst1b)
691	\|\| RTUtf16Cmp(s_wszTst1a, s_wszTst1a)
692	\|\| !RTUtf16Cmp(s_wszTst1a, s_wszTst1b)
693	\|\| !RTUtf16Cmp(s_wszTst1b, s_wszTst1a)
694	)
695	RTTestFailed(hTest, "RTUtf16Cmp failed the alphabet test.\n");
696
697	RTTestSubDone(hTest);
698	}
699
700
701	/**
702	* Test the RTStr*Cmp functions.
703	*/
704	void TstRTStrXCmp(RTTEST hTest)
705	{
706	#define CHECK_DIFF(expr, op) \
707	do \
708	{ \
709	int iDiff = expr; \
710	if (!(iDiff op 0)) \
711	RTTestFailed(hTest, "%d: %d " #op " 0: %s\n", __LINE__, iDiff, #expr); \
712	} while (0)
713
714	/** @todo test the non-ascii bits. */
715
716	RTTestSub(hTest, "RTStrCmp");
717	CHECK_DIFF(RTStrCmp(NULL, NULL), == );
718	CHECK_DIFF(RTStrCmp(NULL, ""), < );
719	CHECK_DIFF(RTStrCmp("", NULL), > );
720	CHECK_DIFF(RTStrCmp("", ""), == );
721	CHECK_DIFF(RTStrCmp("abcdef", "abcdef"), == );
722	CHECK_DIFF(RTStrCmp("abcdef", "abcde"), > );
723	CHECK_DIFF(RTStrCmp("abcde", "abcdef"), < );
724	CHECK_DIFF(RTStrCmp("abcdeg", "abcdef"), > );
725	CHECK_DIFF(RTStrCmp("abcdef", "abcdeg"), < );
726	CHECK_DIFF(RTStrCmp("abcdeF", "abcdef"), < );
727	CHECK_DIFF(RTStrCmp("abcdef", "abcdeF"), > );
728
729
730	RTTestSub(hTest, "RTStrNCmp");
731	CHECK_DIFF(RTStrNCmp(NULL, NULL, RTSTR_MAX), == );
732	CHECK_DIFF(RTStrNCmp(NULL, "", RTSTR_MAX), < );
733	CHECK_DIFF(RTStrNCmp("", NULL, RTSTR_MAX), > );
734	CHECK_DIFF(RTStrNCmp("", "", RTSTR_MAX), == );
735	CHECK_DIFF(RTStrNCmp("abcdef", "abcdef", RTSTR_MAX), == );
736	CHECK_DIFF(RTStrNCmp("abcdef", "abcde", RTSTR_MAX), > );
737	CHECK_DIFF(RTStrNCmp("abcde", "abcdef", RTSTR_MAX), < );
738	CHECK_DIFF(RTStrNCmp("abcdeg", "abcdef", RTSTR_MAX), > );
739	CHECK_DIFF(RTStrNCmp("abcdef", "abcdeg", RTSTR_MAX), < );
740	CHECK_DIFF(RTStrNCmp("abcdeF", "abcdef", RTSTR_MAX), < );
741	CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", RTSTR_MAX), > );
742
743	CHECK_DIFF(RTStrNCmp("abcdef", "fedcba", 0), ==);
744	CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", 5), ==);
745	CHECK_DIFF(RTStrNCmp("abcdef", "abcdeF", 6), > );
746
747
748	RTTestSub(hTest, "RTStrICmp");
749	CHECK_DIFF(RTStrICmp(NULL, NULL), == );
750	CHECK_DIFF(RTStrICmp(NULL, ""), < );
751	CHECK_DIFF(RTStrICmp("", NULL), > );
752	CHECK_DIFF(RTStrICmp("", ""), == );
753	CHECK_DIFF(RTStrICmp("abcdef", "abcdef"), == );
754	CHECK_DIFF(RTStrICmp("abcdef", "abcde"), > );
755	CHECK_DIFF(RTStrICmp("abcde", "abcdef"), < );
756	CHECK_DIFF(RTStrICmp("abcdeg", "abcdef"), > );
757	CHECK_DIFF(RTStrICmp("abcdef", "abcdeg"), < );
758
759	CHECK_DIFF(RTStrICmp("abcdeF", "abcdef"), ==);
760	CHECK_DIFF(RTStrICmp("abcdef", "abcdeF"), ==);
761	CHECK_DIFF(RTStrICmp("ABCDEF", "abcdef"), ==);
762	CHECK_DIFF(RTStrICmp("abcdef", "ABCDEF"), ==);
763	CHECK_DIFF(RTStrICmp("AbCdEf", "aBcDeF"), ==);
764	CHECK_DIFF(RTStrICmp("AbCdEg", "aBcDeF"), > );
765	CHECK_DIFF(RTStrICmp("AbCdEG", "aBcDef"), > ); /* diff performed on the lower case cp. */
766
767
768	RTTestSub(hTest, "RTStrICmpAscii");
769	CHECK_DIFF(RTStrICmpAscii(NULL, NULL), == );
770	CHECK_DIFF(RTStrICmpAscii(NULL, ""), < );
771	CHECK_DIFF(RTStrICmpAscii("", NULL), > );
772	CHECK_DIFF(RTStrICmpAscii("", ""), == );
773	CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdef"), == );
774	CHECK_DIFF(RTStrICmpAscii("abcdef", "abcde"), > );
775	CHECK_DIFF(RTStrICmpAscii("abcde", "abcdef"), < );
776	CHECK_DIFF(RTStrICmpAscii("abcdeg", "abcdef"), > );
777	CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdeg"), < );
778
779	CHECK_DIFF(RTStrICmpAscii("abcdeF", "abcdef"), ==);
780	CHECK_DIFF(RTStrICmpAscii("abcdef", "abcdeF"), ==);
781	CHECK_DIFF(RTStrICmpAscii("ABCDEF", "abcdef"), ==);
782	CHECK_DIFF(RTStrICmpAscii("abcdef", "ABCDEF"), ==);
783	CHECK_DIFF(RTStrICmpAscii("AbCdEf", "aBcDeF"), ==);
784	CHECK_DIFF(RTStrICmpAscii("AbCdEg", "aBcDeF"), > );
785	CHECK_DIFF(RTStrICmpAscii("AbCdEG", "aBcDef"), > ); /* diff performed on the lower case cp. */
786
787
788	RTTestSub(hTest, "RTStrNICmp");
789	CHECK_DIFF(RTStrNICmp(NULL, NULL, RTSTR_MAX), == );
790	CHECK_DIFF(RTStrNICmp(NULL, "", RTSTR_MAX), < );
791	CHECK_DIFF(RTStrNICmp("", NULL, RTSTR_MAX), > );
792	CHECK_DIFF(RTStrNICmp("", "", RTSTR_MAX), == );
793	CHECK_DIFF(RTStrNICmp(NULL, NULL, 0), == );
794	CHECK_DIFF(RTStrNICmp(NULL, "", 0), == );
795	CHECK_DIFF(RTStrNICmp("", NULL, 0), == );
796	CHECK_DIFF(RTStrNICmp("", "", 0), == );
797	CHECK_DIFF(RTStrNICmp("abcdef", "abcdef", RTSTR_MAX), == );
798	CHECK_DIFF(RTStrNICmp("abcdef", "abcde", RTSTR_MAX), > );
799	CHECK_DIFF(RTStrNICmp("abcde", "abcdef", RTSTR_MAX), < );
800	CHECK_DIFF(RTStrNICmp("abcdeg", "abcdef", RTSTR_MAX), > );
801	CHECK_DIFF(RTStrNICmp("abcdef", "abcdeg", RTSTR_MAX), < );
802
803	CHECK_DIFF(RTStrNICmp("abcdeF", "abcdef", RTSTR_MAX), ==);
804	CHECK_DIFF(RTStrNICmp("abcdef", "abcdeF", RTSTR_MAX), ==);
805	CHECK_DIFF(RTStrNICmp("ABCDEF", "abcdef", RTSTR_MAX), ==);
806	CHECK_DIFF(RTStrNICmp("abcdef", "ABCDEF", RTSTR_MAX), ==);
807	CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDeF", RTSTR_MAX), ==);
808	CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", RTSTR_MAX), > );
809	CHECK_DIFF(RTStrNICmp("AbCdEG", "aBcDef", RTSTR_MAX), > ); /* diff performed on the lower case cp. */
810
811	CHECK_DIFF(RTStrNICmp("ABCDEF", "fedcba", 0), ==);
812	CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", 5), ==);
813	CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDeF", 5), ==);
814	CHECK_DIFF(RTStrNICmp("AbCdE", "aBcDe", 5), ==);
815	CHECK_DIFF(RTStrNICmp("AbCdE", "aBcDeF", 5), ==);
816	CHECK_DIFF(RTStrNICmp("AbCdEf", "aBcDe", 5), ==);
817	CHECK_DIFF(RTStrNICmp("AbCdEg", "aBcDeF", 6), > );
818	CHECK_DIFF(RTStrNICmp("AbCdEG", "aBcDef", 6), > ); /* diff performed on the lower case cp. */
819	/* We should continue using byte comparison when we hit the invalid CP. Will assert in debug builds. */
820	// CHECK_DIFF(RTStrNICmp("AbCd\xff""eg", "aBcD\xff""eF", 6), ==);
821
822	RTTestSubDone(hTest);
823	}
824
825
826
827	/**
828	* Check UTF-8 encoding purging.
829	*/
830	void TstRTStrPurgeEncoding(RTTEST hTest)
831	{
832	RTTestSub(hTest, "RTStrPurgeEncoding");
833
834	/*
835	* Test some good strings.
836	*/
837	char sz1[] = "1234567890wertyuiopsdfghjklzxcvbnm";
838	char sz1Copy[sizeof(sz1)];
839	memcpy(sz1Copy, sz1, sizeof(sz1));
840
841	RTTESTI_CHECK_RETV(RTStrPurgeEncoding(sz1) == 0);
842	RTTESTI_CHECK_RETV(!memcmp(sz1, sz1Copy, sizeof(sz1)));
843
844	char *pszAll = RTStrDup(g_szAll);
845	if (pszAll)
846	{
847	RTTESTI_CHECK(RTStrPurgeEncoding(pszAll) == 0);
848	RTTESTI_CHECK(!memcmp(pszAll, g_szAll, sizeof(g_szAll)));
849	RTStrFree(pszAll);
850	}
851
852	/*
853	* Test some bad stuff.
854	*/
855	struct
856	{
857	size_t cErrors;
858	unsigned char szIn[5];
859	const char *pszExpect;
860	} aTests[] =
861	{
862	{ 0, { '1', '2', '3', '4', '\0' }, "1234" },
863	{ 1, { 0x80, '2', '3', '4', '\0' }, "?234" },
864	{ 1, { '1', 0x80, '3', '4', '\0' }, "1?34" },
865	{ 1, { '1', '2', 0x80, '4', '\0' }, "12?4" },
866	{ 1, { '1', '2', '3', 0x80, '\0' }, "123?" },
867	{ 2, { 0x80, 0x81, '3', '4', '\0' }, "??34" },
868	{ 2, { '1', 0x80, 0x81, '4', '\0' }, "1??4" },
869	{ 2, { '1', '2', 0x80, 0x81, '\0' }, "12??" },
870	};
871	for (size_t i = 0; i < RT_ELEMENTS(aTests); i++)
872	{
873	size_t cErrors = RTStrPurgeEncoding((char *)aTests[i].szIn);
874	if (cErrors != aTests[i].cErrors)
875	RTTestFailed(hTest, "#%u: cErrors=%u expected %u\n", i, cErrors, aTests[i].cErrors);
876	else if (strcmp((char *)aTests[i].szIn, aTests[i].pszExpect))
877	RTTestFailed(hTest, "#%u: %.5Rhxs expected %.5Rhxs (%s)\n", i, aTests[i].szIn, aTests[i].pszExpect, aTests[i].pszExpect);
878	}
879
880	RTTestSubDone(hTest);
881	}
882
883
884	/**
885	* Check string sanitising.
886	*/
887	void TstRTStrPurgeComplementSet(RTTEST hTest)
888	{
889	RTTestSub(hTest, "RTStrPurgeComplementSet");
890	RTUNICP aCpSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
891	'\0' };
892	RTUNICP aCpBadSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
893	'7', '\0' }; /* Contains an incomplete pair. */
894	struct
895	{
896	const char *pcszIn;
897	const char *pcszOut;
898	PCRTUNICP pcCpSet;
899	char chReplacement;
900	ssize_t cExpected;
901	}
902	aTests[] =
903	{
904	{ "1234werttrew4321", "1234werttrew4321", aCpSet, '_', 0 },
905	{ "123654wert\xc2\xa2trew\xe2\x82\xac""4321",
906	"123_54wert__trew___4321", aCpSet, '_', 3 },
907	{ "hjhj8766", "????????", aCpSet, '?', 8 },
908	{ "123\xf0\xa4\xad\xa2""4", "123____4", aCpSet, '_', 1 },
909	{ "\xff", "\xff", aCpSet, '_', -1 },
910	{ "____", "____", aCpBadSet, '_', -1 }
911	};
912	enum { MAX_IN_STRING = 256 };
913
914	for (unsigned i = 0; i < RT_ELEMENTS(aTests); ++i)
915	{
916	char szCopy[MAX_IN_STRING];
917	ssize_t cReplacements;
918	AssertRC(RTStrCopy(szCopy, RT_ELEMENTS(szCopy), aTests[i].pcszIn));
919	RTTestDisableAssertions(hTest);
920	cReplacements = RTStrPurgeComplementSet(szCopy, aTests[i].pcCpSet, aTests[i].chReplacement);
921	RTTestRestoreAssertions(hTest);
922	if (cReplacements != aTests[i].cExpected)
923	RTTestFailed(hTest, "#%u: expected %lld, actual %lld\n", i,
924	(long long) aTests[i].cExpected,
925	(long long) cReplacements);
926	if (strcmp(aTests[i].pcszOut, szCopy))
927	RTTestFailed(hTest, "#%u: expected %s, actual %s\n", i,
928	aTests[i].pcszOut, szCopy);
929	}
930	}
931
932
933	/**
934	* Check string sanitising.
935	*/
936	void TstRTUtf16PurgeComplementSet(RTTEST hTest)
937	{
938	RTTestSub(hTest, "RTUtf16PurgeComplementSet");
939	RTUNICP aCpSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
940	'\0' };
941	RTUNICP aCpBadSet[] = { '1', '5', 'w', 'w', 'r', 'r', 'e', 'f', 't', 't',
942	'7', '\0' }; /* Contains an incomplete pair. */
943	struct
944	{
945	const char *pcszIn;
946	const char *pcszOut;
947	size_t cwc; /* Zero means the strings are Utf-8. */
948	PCRTUNICP pcCpSet;
949	char chReplacement;
950	ssize_t cExpected;
951	}
952	aTests[] =
953	{
954	{ "1234werttrew4321", "1234werttrew4321", 0, aCpSet, '_', 0 },
955	{ "123654wert\xc2\xa2trew\xe2\x82\xac""4321",
956	"123_54wert_trew_4321", 0, aCpSet, '_', 3 },
957	{ "hjhj8766", "????????", 0, aCpSet, '?', 8 },
958	{ "123\xf0\xa4\xad\xa2""4", "123__4", 0, aCpSet, '_', 1 },
959	{ "\xff\xff\0", "\xff\xff\0", 2, aCpSet, '_', -1 },
960	{ "\xff\xff\0", "\xff\xff\0", 2, aCpSet, '_', -1 },
961	{ "____", "____", 0, aCpBadSet, '_', -1 }
962	};
963	enum { MAX_IN_STRING = 256 };
964
965	for (unsigned i = 0; i < RT_ELEMENTS(aTests); ++i)
966	{
967	RTUTF16 wszInCopy[MAX_IN_STRING], *pwszInCopy = wszInCopy;
968	RTUTF16 wszOutCopy[MAX_IN_STRING], *pwszOutCopy = wszOutCopy;
969	ssize_t cReplacements;
970	if (!aTests[i].cwc)
971	{
972	AssertRC(RTStrToUtf16Ex(aTests[i].pcszIn, RTSTR_MAX, &pwszInCopy,
973	RT_ELEMENTS(wszInCopy), NULL));
974	AssertRC(RTStrToUtf16Ex(aTests[i].pcszOut, RTSTR_MAX, &pwszOutCopy,
975	RT_ELEMENTS(wszOutCopy), NULL));
976	}
977	else
978	{
979	Assert(aTests[i].cwc <= RT_ELEMENTS(wszInCopy));
980	memcpy(wszInCopy, aTests[i].pcszIn, aTests[i].cwc * 2);
981	memcpy(wszOutCopy, aTests[i].pcszOut, aTests[i].cwc * 2);
982	}
983
984	RTTestDisableAssertions(hTest);
985	cReplacements = RTUtf16PurgeComplementSet(wszInCopy, aTests[i].pcCpSet, aTests[i].chReplacement);
986	RTTestRestoreAssertions(hTest);
987
988	if (cReplacements != aTests[i].cExpected)
989	RTTestFailed(hTest, "#%u: expected %lld, actual %lld\n", i,
990	(long long) aTests[i].cExpected,
991	(long long) cReplacements);
992	if (RTUtf16Cmp(wszInCopy, wszOutCopy))
993	RTTestFailed(hTest, "#%u: expected %ls, actual %ls\n", i,
994	wszOutCopy, wszInCopy);
995	}
996	}
997
998
999	/**
1000	* Benchmark stuff.
1001	*/
1002	void Benchmarks(RTTEST hTest)
1003	{
1004	static union
1005	{
1006	RTUTF16 wszBuf[sizeof(g_wszAll)];
1007	char szBuf[sizeof(g_szAll)];
1008	} s_Buf;
1009
1010	RTTestSub(hTest, "Benchmarks");
1011	/** @todo add RTTest* methods for reporting benchmark results. */
1012	RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "Benchmarking RTStrToUtf16Ex: "); /** @todo figure this stuff into the test framework. */
1013	PRTUTF16 pwsz = &s_Buf.wszBuf[0];
1014	int rc = RTStrToUtf16Ex(&g_szAll[0], RTSTR_MAX, &pwsz, RT_ELEMENTS(s_Buf.wszBuf), NULL);
1015	if (RT_SUCCESS(rc))
1016	{
1017	int i;
1018	uint64_t u64Start = RTTimeNanoTS();
1019	for (i = 0; i < 100; i++)
1020	{
1021	rc = RTStrToUtf16Ex(&g_szAll[0], RTSTR_MAX, &pwsz, RT_ELEMENTS(s_Buf.wszBuf), NULL);
1022	if (RT_FAILURE(rc))
1023	{
1024	RTTestFailed(hTest, "UTF-8 -> UTF-16 benchmark failed at i=%d, rc=%Rrc\n", i, rc);
1025	break;
1026	}
1027	}
1028	uint64_t u64Elapsed = RTTimeNanoTS() - u64Start;
1029	RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "%d in %'RI64 ns\n", i, u64Elapsed);
1030	}
1031
1032	RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "Benchmarking RTUtf16ToUtf8Ex: ");
1033	char *psz = &s_Buf.szBuf[0];
1034	rc = RTUtf16ToUtf8Ex(&g_wszAll[0], RTSTR_MAX, &psz, RT_ELEMENTS(s_Buf.szBuf), NULL);
1035	if (RT_SUCCESS(rc))
1036	{
1037	int i;
1038	uint64_t u64Start = RTTimeNanoTS();
1039	for (i = 0; i < 100; i++)
1040	{
1041	rc = RTUtf16ToUtf8Ex(&g_wszAll[0], RTSTR_MAX, &psz, RT_ELEMENTS(s_Buf.szBuf), NULL);
1042	if (RT_FAILURE(rc))
1043	{
1044	RTTestFailed(hTest, "UTF-16 -> UTF-8 benchmark failed at i=%d, rc=%Rrc\n", i, rc);
1045	break;
1046	}
1047	}
1048	uint64_t u64Elapsed = RTTimeNanoTS() - u64Start;
1049	RTTestPrintf(hTest, RTTESTLVL_ALWAYS, "%d in %'RI64 ns\n", i, u64Elapsed);
1050	}
1051
1052	RTTestSubDone(hTest);
1053	}
1054
1055
1056	/**
1057	* Tests RTStrEnd
1058	*/
1059	static void testStrEnd(RTTEST hTest)
1060	{
1061	RTTestSub(hTest, "RTStrEnd");
1062
1063	static char const s_szEmpty[1] = "";
1064	RTTESTI_CHECK(RTStrEnd(s_szEmpty, 0) == NULL);
1065	RTTESTI_CHECK(RTStrEnd(s_szEmpty, 1) == &s_szEmpty[0]);
1066	for (size_t i = 0; i < _1M; i++)
1067	RTTESTI_CHECK(RTStrEnd(s_szEmpty, ~i) == &s_szEmpty[0]);
1068
1069	/* Check the implementation won't ever overshoot the '\0' in the input in
1070	anyway that may lead to a SIGSEV. (VC++ 14.1 does this) */
1071	size_t const cchStr = 1023;
1072	char pszStr = (char )RTTestGuardedAllocTail(hTest, cchStr + 1);
1073	memset(pszStr, ' ', cchStr);
1074	char * const pszStrEnd = &pszStr[cchStr];
1075	*pszStrEnd = '\0';
1076	RTTEST_CHECK_RETV(hTest, strlen(pszStr) == cchStr);
1077
1078	for (size_t off = 0; off <= cchStr; off++)
1079	{
1080	RTTEST_CHECK(hTest, RTStrEnd(&pszStr[off], cchStr + 1 - off) == pszStrEnd);
1081	RTTEST_CHECK(hTest, RTStrEnd(&pszStr[off], RTSTR_MAX) == pszStrEnd);
1082
1083	RTTEST_CHECK(hTest, memchr(&pszStr[off], '\0', cchStr + 1 - off) == pszStrEnd);
1084	RTTEST_CHECK(hTest, strchr(&pszStr[off], '\0') == pszStrEnd);
1085	RTTEST_CHECK(hTest, strchr(&pszStr[off], '?') == NULL);
1086
1087	size_t cchMax = 0;
1088	for (; cchMax <= cchStr - off; cchMax++)
1089	{
1090	const char *pszRet = RTStrEnd(&pszStr[off], cchMax);
1091	if (pszRet != NULL)
1092	{
1093	RTTestFailed(hTest, "off=%zu cchMax=%zu: %p, expected NULL\n", off, cchMax, pszRet);
1094	break;
1095	}
1096	}
1097	for (; cchMax <= _8K; cchMax++)
1098	{
1099	const char *pszRet = RTStrEnd(&pszStr[off], cchMax);
1100	if (pszRet != pszStrEnd)
1101	{
1102	RTTestFailed(hTest, "off=%zu cchMax=%zu: off by %p\n", off, cchMax, pszRet);
1103	break;
1104	}
1105	}
1106	}
1107	RTTestGuardedFree(hTest, pszStr);
1108	}
1109
1110
1111	/**
1112	* Tests RTStrStr and RTStrIStr.
1113	*/
1114	static void testStrStr(RTTEST hTest)
1115	{
1116	#define CHECK_NULL(expr) \
1117	do { \
1118	const char *pszRet = expr; \
1119	if (pszRet != NULL) \
1120	RTTestFailed(hTest, "%d: %#x -> %s expected NULL", __LINE__, #expr, pszRet); \
1121	} while (0)
1122
1123	#define CHECK(expr, expect) \
1124	do { \
1125	const char *pszRet = expr; \
1126	const char *pszExpect = (expect); \
1127	if ( (pszRet != NULL && pszExpect == NULL) \
1128	\|\| (pszRet == NULL && pszExpect != NULL) \
1129	\|\| strcmp(pszRet, pszExpect) \
1130	) \
1131	RTTestFailed(hTest, "%d: %#x -> %s expected %s", __LINE__, #expr, pszRet, pszExpect); \
1132	} while (0)
1133
1134
1135	RTTestSub(hTest, "RTStrStr");
1136	CHECK(RTStrStr("abcdef", ""), "abcdef");
1137	CHECK_NULL(RTStrStr("abcdef", NULL));
1138	CHECK_NULL(RTStrStr(NULL, ""));
1139	CHECK_NULL(RTStrStr(NULL, NULL));
1140	CHECK(RTStrStr("abcdef", "abcdef"), "abcdef");
1141	CHECK(RTStrStr("abcdef", "b"), "bcdef");
1142	CHECK(RTStrStr("abcdef", "bcdef"), "bcdef");
1143	CHECK(RTStrStr("abcdef", "cdef"), "cdef");
1144	CHECK(RTStrStr("abcdef", "cde"), "cdef");
1145	CHECK(RTStrStr("abcdef", "cd"), "cdef");
1146	CHECK(RTStrStr("abcdef", "c"), "cdef");
1147	CHECK(RTStrStr("abcdef", "f"), "f");
1148	CHECK(RTStrStr("abcdef", "ef"), "ef");
1149	CHECK(RTStrStr("abcdef", "e"), "ef");
1150	CHECK_NULL(RTStrStr("abcdef", "z"));
1151	CHECK_NULL(RTStrStr("abcdef", "A"));
1152	CHECK_NULL(RTStrStr("abcdef", "F"));
1153
1154	RTTestSub(hTest, "RTStrIStr");
1155	CHECK(RTStrIStr("abcdef", ""), "abcdef");
1156	CHECK_NULL(RTStrIStr("abcdef", NULL));
1157	CHECK_NULL(RTStrIStr(NULL, ""));
1158	CHECK_NULL(RTStrIStr(NULL, NULL));
1159	CHECK(RTStrIStr("abcdef", "abcdef"), "abcdef");
1160	CHECK(RTStrIStr("abcdef", "Abcdef"), "abcdef");
1161	CHECK(RTStrIStr("abcdef", "ABcDeF"), "abcdef");
1162	CHECK(RTStrIStr("abcdef", "b"), "bcdef");
1163	CHECK(RTStrIStr("abcdef", "B"), "bcdef");
1164	CHECK(RTStrIStr("abcdef", "bcdef"), "bcdef");
1165	CHECK(RTStrIStr("abcdef", "BCdEf"), "bcdef");
1166	CHECK(RTStrIStr("abcdef", "bCdEf"), "bcdef");
1167	CHECK(RTStrIStr("abcdef", "bcdEf"), "bcdef");
1168	CHECK(RTStrIStr("abcdef", "BcdEf"), "bcdef");
1169	CHECK(RTStrIStr("abcdef", "cdef"), "cdef");
1170	CHECK(RTStrIStr("abcdef", "cde"), "cdef");
1171	CHECK(RTStrIStr("abcdef", "cd"), "cdef");
1172	CHECK(RTStrIStr("abcdef", "c"), "cdef");
1173	CHECK(RTStrIStr("abcdef", "f"), "f");
1174	CHECK(RTStrIStr("abcdeF", "F"), "F");
1175	CHECK(RTStrIStr("abcdef", "F"), "f");
1176	CHECK(RTStrIStr("abcdef", "ef"), "ef");
1177	CHECK(RTStrIStr("EeEef", "e"), "EeEef");
1178	CHECK(RTStrIStr("EeEef", "E"), "EeEef");
1179	CHECK(RTStrIStr("EeEef", "EE"), "EeEef");
1180	CHECK(RTStrIStr("EeEef", "EEE"), "EeEef");
1181	CHECK(RTStrIStr("EeEef", "EEEF"), "eEef");
1182	CHECK_NULL(RTStrIStr("EeEef", "z"));
1183
1184	#undef CHECK
1185	#undef CHECK_NULL
1186	RTTestSubDone(hTest);
1187	}
1188
1189
1190	void testUtf8Latin1(RTTEST hTest)
1191	{
1192	RTTestSub(hTest, "Latin-1 <-> Utf-8 conversion functions");
1193
1194	/* Test Utf8 -> Latin1 */
1195	size_t cch_szAll = 0;
1196	size_t cbShort = RTStrCalcLatin1Len(g_szAll);
1197	RTTEST_CHECK(hTest, cbShort == 0);
1198	int rc = RTStrCalcLatin1LenEx(g_szAll, 383, &cch_szAll);
1199	RTTEST_CHECK(hTest, (cch_szAll == 255));
1200	rc = RTStrCalcLatin1LenEx(g_szAll, RTSTR_MAX, &cch_szAll);
1201	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1202	char *psz = NULL;
1203	char szShort[256] = { 0 };
1204	memcpy(szShort, g_szAll, 255);
1205	cbShort = RTStrCalcLatin1Len(szShort);
1206	RTTEST_CHECK(hTest, cbShort == 191);
1207	rc = RTStrToLatin1(szShort, &psz);
1208	RTTEST_CHECK_RC_OK(hTest, rc);
1209	if (RT_SUCCESS(rc))
1210	{
1211	RTTEST_CHECK(hTest, (strlen(psz) == 191));
1212	for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1213	if (psz[i] != (char) j)
1214	{
1215	RTTestFailed(hTest, "conversion of g_szAll to Latin1 failed at position %u\n", i);
1216	break;
1217	}
1218	}
1219	RTStrFree(psz);
1220	rc = RTStrToLatin1(g_szAll, &psz);
1221	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1222	char sz[512];
1223	char *psz2 = &sz[0];
1224	size_t cchActual = 0;
1225	rc = RTStrToLatin1Ex(g_szAll, sizeof(sz) - 1, &psz2, sizeof(sz),
1226	&cchActual);
1227	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1228	RTTEST_CHECK_MSG(hTest, cchActual == 0,
1229	(hTest, "cchActual=%lu\n", cchActual));
1230	rc = RTStrToLatin1Ex(g_szAll, 383, &psz2, sizeof(sz),
1231	&cchActual);
1232	RTTEST_CHECK_RC_OK(hTest, rc);
1233	if (RT_SUCCESS(rc))
1234	{
1235	RTTEST_CHECK(hTest, (cchActual == 255));
1236	RTTEST_CHECK(hTest, (cchActual == strlen(sz)));
1237	for (unsigned i = 0, j = 1; psz2[i] != '\0'; ++i, ++j)
1238	if (psz2[i] != (char) j)
1239	{
1240	RTTestFailed(hTest, "second conversion of g_szAll to Latin1 failed at position %u\n", i);
1241	break;
1242	}
1243	}
1244	rc = RTStrToLatin1Ex(g_szAll, 129, &psz2, 128, &cchActual);
1245	RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1246	RTTEST_CHECK_MSG(hTest, cchActual == 128,
1247	(hTest, "cchActual=%lu\n", cchActual));
1248	rc = RTStrToLatin1Ex(g_szAll, 383, &psz, 0, &cchActual);
1249	RTTEST_CHECK_RC_OK(hTest, rc);
1250	if (RT_SUCCESS(rc))
1251	{
1252	RTTEST_CHECK(hTest, (cchActual == 255));
1253	RTTEST_CHECK(hTest, (cchActual == strlen(psz)));
1254	for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1255	if ( ((j < 0x100) && (psz[i] != (char) j))
1256	\|\| ((j > 0xff) && psz[i] != '?'))
1257	{
1258	RTTestFailed(hTest, "third conversion of g_szAll to Latin1 failed at position %u\n", i);
1259	break;
1260	}
1261	}
1262	const char *pszBad = "Hello\xDC\xD8";
1263	rc = RTStrToLatin1Ex(pszBad, RTSTR_MAX, &psz2, sizeof(sz),
1264	&cchActual);
1265	RTTEST_CHECK_RC(hTest, rc, VERR_INVALID_UTF8_ENCODING);
1266	RTStrFree(psz);
1267
1268	/* Test Latin1 -> Utf8 */
1269	const char *pszLat1 = "\x01\x20\x40\x80\x81";
1270	RTTEST_CHECK(hTest, RTLatin1CalcUtf8Len(pszLat1) == 7);
1271	rc = RTLatin1CalcUtf8LenEx(pszLat1, 3, &cchActual);
1272	RTTEST_CHECK_RC_OK(hTest, rc);
1273	if (RT_SUCCESS(rc))
1274	RTTEST_CHECK(hTest, cchActual == 3);
1275	rc = RTLatin1CalcUtf8LenEx(pszLat1, RTSTR_MAX, &cchActual);
1276	RTTEST_CHECK_RC_OK(hTest, rc);
1277	if (RT_SUCCESS(rc))
1278	RTTEST_CHECK(hTest, cchActual == 7);
1279	char *pch = NULL;
1280	char ch[8];
1281	char *pch2 = &ch[0];
1282	cchActual = 0;
1283	rc = RTLatin1ToUtf8(pszLat1, &pch);
1284	RTTEST_CHECK_RC_OK(hTest, rc);
1285	if (RT_SUCCESS(rc))
1286	RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1287	RTStrFree(pch);
1288	rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch, 0, &cchActual);
1289	RTTEST_CHECK_RC_OK(hTest, rc);
1290	if (RT_SUCCESS(rc))
1291	{
1292	RTTEST_CHECK(hTest, (cchActual == 7));
1293	RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1294	}
1295	RTStrFree(pch);
1296	rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch, 0, NULL);
1297	RTTEST_CHECK_RC_OK(hTest, rc);
1298	if (RT_SUCCESS(rc))
1299	RTTEST_CHECK(hTest, !strcmp(pch, "\x01\x20\x40\xC2\x80\xC2\x81"));
1300	RTStrFree(pch);
1301	rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch2, RT_ELEMENTS(ch),
1302	&cchActual);
1303	RTTEST_CHECK_RC_OK(hTest, rc);
1304	if (RT_SUCCESS(rc))
1305	{
1306	RTTEST_CHECK(hTest, (cchActual == 7));
1307	RTTEST_CHECK(hTest, !strcmp(pch2, "\x01\x20\x40\xC2\x80\xC2\x81"));
1308	}
1309	rc = RTLatin1ToUtf8Ex(pszLat1, 3, &pch2, RT_ELEMENTS(ch),
1310	&cchActual);
1311	RTTEST_CHECK_RC_OK(hTest, rc);
1312	if (RT_SUCCESS(rc))
1313	{
1314	RTTEST_CHECK(hTest, (cchActual == 3));
1315	RTTEST_CHECK(hTest, !strcmp(pch2, "\x01\x20\x40"));
1316	}
1317	rc = RTLatin1ToUtf8Ex(pszLat1, RTSTR_MAX, &pch2, RT_ELEMENTS(ch) - 1,
1318	&cchActual);
1319	RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1320	RTTEST_CHECK(hTest, (cchActual == 7));
1321	RTTestSubDone(hTest);
1322	}
1323
1324
1325	void testUtf16Latin1(RTTEST hTest)
1326	{
1327	RTTestSub(hTest, "Latin-1 <-> Utf-16 conversion functions");
1328
1329	/* Test Utf16 -> Latin1 */
1330	size_t cch_szAll = 0;
1331	size_t cbShort = RTUtf16CalcLatin1Len(g_wszAll);
1332	RTTEST_CHECK(hTest, cbShort == 0);
1333	int rc = RTUtf16CalcLatin1LenEx(g_wszAll, 255, &cch_szAll);
1334	RTTEST_CHECK(hTest, (cch_szAll == 255));
1335	rc = RTUtf16CalcLatin1LenEx(g_wszAll, RTSTR_MAX, &cch_szAll);
1336	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1337	char *psz = NULL;
1338	RTUTF16 wszShort[256] = { 0 };
1339	for (unsigned i = 0; i < 255; ++i)
1340	wszShort[i] = i + 1;
1341	cbShort = RTUtf16CalcLatin1Len(wszShort);
1342	RTTEST_CHECK(hTest, cbShort == 255);
1343	rc = RTUtf16ToLatin1(wszShort, &psz);
1344	RTTEST_CHECK_RC_OK(hTest, rc);
1345	if (RT_SUCCESS(rc))
1346	{
1347	RTTEST_CHECK(hTest, (strlen(psz) == 255));
1348	for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1349	if (psz[i] != (char) j)
1350	{
1351	RTTestFailed(hTest, "conversion of g_wszAll to Latin1 failed at position %u\n", i);
1352	break;
1353	}
1354	}
1355	RTStrFree(psz);
1356	rc = RTUtf16ToLatin1(g_wszAll, &psz);
1357	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1358	char sz[512];
1359	char *psz2 = &sz[0];
1360	size_t cchActual = 0;
1361	rc = RTUtf16ToLatin1Ex(g_wszAll, sizeof(sz) - 1, &psz2, sizeof(sz),
1362	&cchActual);
1363	RTTEST_CHECK_RC(hTest, rc, VERR_NO_TRANSLATION);
1364	RTTEST_CHECK_MSG(hTest, cchActual == 0,
1365	(hTest, "cchActual=%lu\n", cchActual));
1366	rc = RTUtf16ToLatin1Ex(g_wszAll, 255, &psz2, sizeof(sz),
1367	&cchActual);
1368	RTTEST_CHECK_RC_OK(hTest, rc);
1369	if (RT_SUCCESS(rc))
1370	{
1371	RTTEST_CHECK(hTest, (cchActual == 255));
1372	RTTEST_CHECK(hTest, (cchActual == strlen(sz)));
1373	for (unsigned i = 0, j = 1; psz2[i] != '\0'; ++i, ++j)
1374	if (psz2[i] != (char) j)
1375	{
1376	RTTestFailed(hTest, "second conversion of g_wszAll to Latin1 failed at position %u\n", i);
1377	break;
1378	}
1379	}
1380	rc = RTUtf16ToLatin1Ex(g_wszAll, 128, &psz2, 128, &cchActual);
1381	RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1382	RTTEST_CHECK_MSG(hTest, cchActual == 128,
1383	(hTest, "cchActual=%lu\n", cchActual));
1384	rc = RTUtf16ToLatin1Ex(g_wszAll, 255, &psz, 0, &cchActual);
1385	RTTEST_CHECK_RC_OK(hTest, rc);
1386	if (RT_SUCCESS(rc))
1387	{
1388	RTTEST_CHECK(hTest, (cchActual == 255));
1389	RTTEST_CHECK(hTest, (cchActual == strlen(psz)));
1390	for (unsigned i = 0, j = 1; psz[i] != '\0'; ++i, ++j)
1391	if ( ((j < 0x100) && (psz[i] != (char) j))
1392	\|\| ((j > 0xff) && psz[i] != '?'))
1393	{
1394	RTTestFailed(hTest, "third conversion of g_wszAll to Latin1 failed at position %u\n", i);
1395	break;
1396	}
1397	}
1398	const char *pszBad = "H\0e\0l\0l\0o\0\0\xDC\0\xD8\0";
1399	rc = RTUtf16ToLatin1Ex((RTUTF16 *) pszBad, RTSTR_MAX, &psz2, sizeof(sz),
1400	&cchActual);
1401	RTTEST_CHECK_RC(hTest, rc, VERR_INVALID_UTF16_ENCODING);
1402	RTStrFree(psz);
1403
1404	/* Test Latin1 -> Utf16 */
1405	const char *pszLat1 = "\x01\x20\x40\x80\x81";
1406	RTTEST_CHECK(hTest, RTLatin1CalcUtf16Len(pszLat1) == 5);
1407	rc = RTLatin1CalcUtf16LenEx(pszLat1, 3, &cchActual);
1408	RTTEST_CHECK_RC_OK(hTest, rc);
1409	if (RT_SUCCESS(rc))
1410	RTTEST_CHECK(hTest, cchActual == 3);
1411	rc = RTLatin1CalcUtf16LenEx(pszLat1, RTSTR_MAX, &cchActual);
1412	RTTEST_CHECK_RC_OK(hTest, rc);
1413	if (RT_SUCCESS(rc))
1414	RTTEST_CHECK(hTest, cchActual == 5);
1415	RTUTF16 *pwc = NULL;
1416	RTUTF16 wc[6];
1417	RTUTF16 *pwc2 = &wc[0];
1418	size_t cwActual = 0;
1419	rc = RTLatin1ToUtf16(pszLat1, &pwc);
1420	RTTEST_CHECK_RC_OK(hTest, rc);
1421	if (RT_SUCCESS(rc))
1422	RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1423	&& (pwc[2] == 0x40) && (pwc[3] == 0x80)
1424	&& (pwc[4] == 0x81) && (pwc[5] == '\0'));
1425	RTUtf16Free(pwc);
1426	rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc, 0, &cwActual);
1427	RTTEST_CHECK_RC_OK(hTest, rc);
1428	if (RT_SUCCESS(rc))
1429	{
1430	RTTEST_CHECK(hTest, (cwActual == 5));
1431	RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1432	&& (pwc[2] == 0x40) && (pwc[3] == 0x80)
1433	&& (pwc[4] == 0x81) && (pwc[5] == '\0'));
1434	}
1435	RTUtf16Free(pwc);
1436	rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc, 0, NULL);
1437	RTTEST_CHECK_RC_OK(hTest, rc);
1438	if (RT_SUCCESS(rc))
1439	RTTEST_CHECK(hTest, (pwc[0] == 1) && (pwc[1] == 0x20)
1440	&& (pwc[2] == 0x40) && (pwc[3] == 0x80)
1441	&& (pwc[4] == 0x81) && (pwc[5] == '\0'));
1442	RTUtf16Free(pwc);
1443	rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc2, RT_ELEMENTS(wc),
1444	&cwActual);
1445	RTTEST_CHECK_RC_OK(hTest, rc);
1446	if (RT_SUCCESS(rc))
1447	{
1448	RTTEST_CHECK(hTest, (cwActual == 5));
1449	RTTEST_CHECK(hTest, (wc[0] == 1) && (wc[1] == 0x20)
1450	&& (wc[2] == 0x40) && (wc[3] == 0x80)
1451	&& (wc[4] == 0x81) && (wc[5] == '\0'));
1452	}
1453	rc = RTLatin1ToUtf16Ex(pszLat1, 3, &pwc2, RT_ELEMENTS(wc),
1454	&cwActual);
1455	RTTEST_CHECK_RC_OK(hTest, rc);
1456	if (RT_SUCCESS(rc))
1457	{
1458	RTTEST_CHECK(hTest, (cwActual == 3));
1459	RTTEST_CHECK(hTest, (wc[0] == 1) && (wc[1] == 0x20)
1460	&& (wc[2] == 0x40) && (wc[3] == '\0'));
1461	}
1462	rc = RTLatin1ToUtf16Ex(pszLat1, RTSTR_MAX, &pwc2, RT_ELEMENTS(wc) - 1,
1463	&cwActual);
1464	RTTEST_CHECK_RC(hTest, rc, VERR_BUFFER_OVERFLOW);
1465	RTTEST_CHECK(hTest, (cwActual == 5));
1466	RTTestSubDone(hTest);
1467	}
1468
1469
1470	static void testNoTransation(RTTEST hTest)
1471	{
1472	/*
1473	* Try trigger a VERR_NO_TRANSLATION error in convert to
1474	* current CP to latin-1.
1475	*/
1476	const RTUTF16 s_swzTest1[] = { 0x2358, 0x2242, 0x2357, 0x2359, 0x22f9, 0x2c4e, 0x0030, 0x0060,
1477	0x0092, 0x00c1, 0x00f2, 0x1f80, 0x0088, 0x2c38, 0x2c30, 0x0000 };
1478	char *pszTest1;
1479	int rc = RTUtf16ToUtf8(s_swzTest1, &pszTest1);
1480	RTTESTI_CHECK_RC_RETV(rc, VINF_SUCCESS);
1481
1482	RTTestSub(hTest, "VERR_NO_TRANSLATION/RTStrUtf8ToCurrentCP");
1483	char *pszOut;
1484	rc = RTStrUtf8ToCurrentCP(&pszOut, pszTest1);
1485	if (rc == VINF_SUCCESS)
1486	{
1487	RTTestIPrintf(RTTESTLVL_ALWAYS, "CurrentCP is UTF-8 or similar (LC_ALL=%s LANG=%s LC_CTYPE=%s)\n",
1488	RTEnvGet("LC_ALL"), RTEnvGet("LANG"), RTEnvGet("LC_CTYPE"));
1489	if (strcmp(pszOut, pszTest1))
1490	RTTestFailed(hTest, "mismatch\nutf8: %.Rhxs\n got: %.Rhxs\n", strlen(pszTest1), pszTest1, strlen(pszOut), pszOut);
1491	RTStrFree(pszOut);
1492	}
1493	else
1494	RTTESTI_CHECK_MSG(rc == VWRN_NO_TRANSLATION \|\| rc == VERR_NO_TRANSLATION, ("rc=%Rrc\n", rc));
1495
1496	RTTestSub(hTest, "VERR_NO_TRANSLATION/RTUtf16ToLatin1");
1497	rc = RTUtf16ToLatin1(s_swzTest1, &pszOut);
1498	RTTESTI_CHECK_RC(rc, VERR_NO_TRANSLATION);
1499	if (RT_SUCCESS(rc))
1500	RTStrFree(pszOut);
1501
1502	RTStrFree(pszTest1);
1503	RTTestSubDone(hTest);
1504	}
1505
1506	static void testGetPut(RTTEST hTest)
1507	{
1508	/*
1509	* Test RTStrPutCp, RTStrGetCp and RTStrGetCpEx.
1510	*/
1511	RTTestSub(hTest, "RTStrPutCp, RTStrGetCp and RTStrGetCpEx");
1512
1513	RTUNICP uc = 0;
1514	while (uc <= 0x10fffd)
1515	{
1516	/* Figure the range - skip illegal ranges. */
1517	RTUNICP ucFirst = uc;
1518	if (ucFirst - UINT32_C(0xd800) <= 0x7ff)
1519	ucFirst = 0xe000;
1520	else if (ucFirst == UINT32_C(0xfffe) \|\| ucFirst == UINT32_C(0xffff))
1521	ucFirst = 0x10000;
1522
1523	RTUNICP ucLast = ucFirst + 1023;
1524	if (ucLast - UINT32_C(0xd800) <= 0x7ff)
1525	ucLast = 0xd7ff;
1526	else if (ucLast == UINT32_C(0xfffe) \|\| ucLast == UINT32_C(0xffff))
1527	ucLast = 0xfffd;
1528
1529	/* Encode the range into a string, decode each code point as we go along. */
1530	char sz1[8192];
1531	char *pszDst = sz1;
1532	for (uc = ucFirst; uc <= ucLast; uc++)
1533	{
1534	char *pszBefore = pszDst;
1535	pszDst = RTStrPutCp(pszDst, uc);
1536	RTTESTI_CHECK(pszBefore - pszDst < 6);
1537
1538	RTUNICP uc2 = RTStrGetCp(pszBefore);
1539	RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1540
1541	const char *pszSrc = pszBefore;
1542	RTUNICP uc3 = 42;
1543	RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1544	RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1545	RTTESTI_CHECK_MSG(pszSrc == pszDst, ("pszSrc=%p pszDst=%p\n", pszSrc, pszDst));
1546	}
1547
1548	/* Decode and re-encode it. */
1549	const char *pszSrc = pszDst = sz1;
1550	for (uc = ucFirst; uc <= ucLast; uc++)
1551	{
1552	RTUNICP uc2 = RTStrGetCp(pszSrc);
1553	RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1554
1555	RTUNICP uc3 = 42;
1556	RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1557	RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1558
1559	pszDst = RTStrPutCp(pszDst, uc);
1560	RTTESTI_CHECK_MSG(pszSrc == pszDst, ("pszSrc=%p pszDst=%p\n", pszSrc, pszDst));
1561	pszSrc = pszDst;
1562	}
1563
1564	/* Decode and wipe it (checking compiler optimizations). */
1565	pszSrc = pszDst = sz1;
1566	for (uc = ucFirst; uc <= ucLast; uc++)
1567	{
1568	RTUNICP uc2 = RTStrGetCp(pszSrc);
1569	RTTESTI_CHECK_MSG(uc2 == uc, ("uc2=%#x uc=%#x\n", uc2, uc));
1570
1571	RTUNICP uc3 = 42;
1572	RTTESTI_CHECK_RC(RTStrGetCpEx(&pszSrc, &uc3), VINF_SUCCESS);
1573	RTTESTI_CHECK_MSG(uc3 == uc, ("uc3=%#x uc=%#x\n", uc3, uc));
1574
1575	pszDst = RTStrPutCp(pszDst, 0);
1576	}
1577
1578	/* advance */
1579	uc = ucLast + 1;
1580	}
1581
1582	}
1583
1584
1585	int main()
1586	{
1587	/*
1588	* Init the runtime, test and say hello.
1589	*/
1590	RTTEST hTest;
1591	RTEXITCODE rcExit = RTTestInitAndCreate("tstUtf8", &hTest);
1592	if (rcExit != RTEXITCODE_SUCCESS)
1593	return rcExit;
1594	RTTestBanner(hTest);
1595
1596	/*
1597	* Run the tests.
1598	*/
1599	InitStrings();
1600	test1(hTest);
1601	test2(hTest);
1602	test3(hTest);
1603	TstRTStrXCmp(hTest);
1604	TstRTStrPurgeEncoding(hTest);
1605	/* TstRTPurgeComplementSet test conditions which assert. /
1606	TstRTStrPurgeComplementSet(hTest);
1607	TstRTUtf16PurgeComplementSet(hTest);
1608	testStrEnd(hTest);
1609	testStrStr(hTest);
1610	testUtf8Latin1(hTest);
1611	testUtf16Latin1(hTest);
1612	testNoTransation(hTest);
1613	testGetPut(hTest);
1614
1615	Benchmarks(hTest);
1616
1617	/*
1618	* Summary
1619	*/
1620	return RTTestSummaryAndDestroy(hTest);
1621	}
1622

Note: See TracBrowser for help on using the repository browser.

source: vbox/trunk/src/VBox/Runtime/testcase/tstUtf8.cpp@ 93174

Download in other formats: