1 /*-
2 * Copyright (c) 2002 Tim J. Robbins
3 * All rights reserved.
4 *
5 * Copyright (c) 2013 Ed Schouten <ed@FreeBSD.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27 * SUCH DAMAGE.
28 */
29 /*
30 * Test program for mbrtoc16() as specified by ISO/IEC 9899:2011.
31 */
32
33 #include <sys/cdefs.h>
34 #include <errno.h>
35 #include <limits.h>
36 #include <locale.h>
37 #include <stdio.h>
38 #include <string.h>
39 #include <uchar.h>
40
41 #include <atf-c.h>
42
43 static void
require_lc_ctype(const char * locale_name)44 require_lc_ctype(const char *locale_name)
45 {
46 char *lc_ctype_set;
47
48 lc_ctype_set = setlocale(LC_CTYPE, locale_name);
49 if (lc_ctype_set == NULL)
50 atf_tc_fail("setlocale(LC_CTYPE, \"%s\") failed; errno=%d",
51 locale_name, errno);
52
53 ATF_REQUIRE(strcmp(lc_ctype_set, locale_name) == 0);
54 }
55
56 static mbstate_t s;
57 static char16_t c16;
58
59 ATF_TC_WITHOUT_HEAD(mbrtoc16_c_locale_test);
ATF_TC_BODY(mbrtoc16_c_locale_test,tc)60 ATF_TC_BODY(mbrtoc16_c_locale_test, tc)
61 {
62
63 require_lc_ctype("C");
64
65 /* Null wide character, internal state. */
66 ATF_REQUIRE(mbrtoc16(&c16, "", 1, NULL) == 0);
67 ATF_REQUIRE(c16 == 0);
68
69 /* Null wide character. */
70 memset(&s, 0, sizeof(s));
71 ATF_REQUIRE(mbrtoc16(&c16, "", 1, &s) == 0);
72 ATF_REQUIRE(c16 == 0);
73
74 /* Latin letter A, internal state. */
75 ATF_REQUIRE(mbrtoc16(NULL, 0, 0, NULL) == 0);
76 ATF_REQUIRE(mbrtoc16(&c16, "A", 1, NULL) == 1);
77 ATF_REQUIRE(c16 == L'A');
78
79 /* Latin letter A. */
80 memset(&s, 0, sizeof(s));
81 ATF_REQUIRE(mbrtoc16(&c16, "A", 1, &s) == 1);
82 ATF_REQUIRE(c16 == L'A');
83
84 /* Incomplete character sequence. */
85 c16 = L'z';
86 memset(&s, 0, sizeof(s));
87 ATF_REQUIRE(mbrtoc16(&c16, "", 0, &s) == (size_t)-2);
88 ATF_REQUIRE(c16 == L'z');
89
90 /* Check that mbrtoc16() doesn't access the buffer when n == 0. */
91 c16 = L'z';
92 memset(&s, 0, sizeof(s));
93 ATF_REQUIRE(mbrtoc16(&c16, "", 0, &s) == (size_t)-2);
94 ATF_REQUIRE(c16 == L'z');
95
96 /* Check that mbrtoc16() doesn't read ahead too aggressively. */
97 memset(&s, 0, sizeof(s));
98 ATF_REQUIRE(mbrtoc16(&c16, "AB", 2, &s) == 1);
99 ATF_REQUIRE(c16 == L'A');
100 ATF_REQUIRE(mbrtoc16(&c16, "C", 1, &s) == 1);
101 ATF_REQUIRE(c16 == L'C');
102
103 }
104
105 ATF_TC_WITHOUT_HEAD(mbrtoc16_iso_8859_1_test);
ATF_TC_BODY(mbrtoc16_iso_8859_1_test,tc)106 ATF_TC_BODY(mbrtoc16_iso_8859_1_test, tc)
107 {
108
109 require_lc_ctype("en_US.ISO8859-1");
110
111 /* Currency sign. */
112 memset(&s, 0, sizeof(s));
113 ATF_REQUIRE(mbrtoc16(&c16, "\xa4", 1, &s) == 1);
114 ATF_REQUIRE(c16 == 0xa4);
115 }
116
117 ATF_TC_WITHOUT_HEAD(mbrtoc16_iso_8859_15_test);
ATF_TC_BODY(mbrtoc16_iso_8859_15_test,tc)118 ATF_TC_BODY(mbrtoc16_iso_8859_15_test, tc)
119 {
120
121 require_lc_ctype("en_US.ISO8859-15");
122
123 /* Euro sign. */
124 memset(&s, 0, sizeof(s));
125 ATF_REQUIRE(mbrtoc16(&c16, "\xa4", 1, &s) == 1);
126 ATF_REQUIRE(c16 == 0x20ac);
127 }
128
129 ATF_TC_WITHOUT_HEAD(mbrtoc16_utf_8_test);
ATF_TC_BODY(mbrtoc16_utf_8_test,tc)130 ATF_TC_BODY(mbrtoc16_utf_8_test, tc)
131 {
132
133 require_lc_ctype("en_US.UTF-8");
134
135 /* Null wide character, internal state. */
136 ATF_REQUIRE(mbrtoc16(NULL, 0, 0, NULL) == 0);
137 ATF_REQUIRE(mbrtoc16(&c16, "", 1, NULL) == 0);
138 ATF_REQUIRE(c16 == 0);
139
140 /* Null wide character. */
141 memset(&s, 0, sizeof(s));
142 ATF_REQUIRE(mbrtoc16(&c16, "", 1, &s) == 0);
143 ATF_REQUIRE(c16 == 0);
144
145 /* Latin letter A, internal state. */
146 ATF_REQUIRE(mbrtoc16(NULL, 0, 0, NULL) == 0);
147 ATF_REQUIRE(mbrtoc16(&c16, "A", 1, NULL) == 1);
148 ATF_REQUIRE(c16 == L'A');
149
150 /* Latin letter A. */
151 memset(&s, 0, sizeof(s));
152 ATF_REQUIRE(mbrtoc16(&c16, "A", 1, &s) == 1);
153 ATF_REQUIRE(c16 == L'A');
154
155 /* Incomplete character sequence (zero length). */
156 c16 = L'z';
157 memset(&s, 0, sizeof(s));
158 ATF_REQUIRE(mbrtoc16(&c16, "", 0, &s) == (size_t)-2);
159 ATF_REQUIRE(c16 == L'z');
160
161 /* Incomplete character sequence (truncated double-byte). */
162 memset(&s, 0, sizeof(s));
163 c16 = 0;
164 ATF_REQUIRE(mbrtoc16(&c16, "\xc3", 1, &s) == (size_t)-2);
165
166 /* Same as above, but complete. */
167 memset(&s, 0, sizeof(s));
168 c16 = 0;
169 ATF_REQUIRE(mbrtoc16(&c16, "\xc3\x84", 2, &s) == 2);
170 ATF_REQUIRE(c16 == 0xc4);
171
172 /* Test restarting behaviour. */
173 memset(&s, 0, sizeof(s));
174 c16 = 0;
175 ATF_REQUIRE(mbrtoc16(&c16, "\xc3", 1, &s) == (size_t)-2);
176 ATF_REQUIRE(c16 == 0);
177 ATF_REQUIRE(mbrtoc16(&c16, "\xb7", 1, &s) == 1);
178 ATF_REQUIRE(c16 == 0xf7);
179
180 /* Surrogate pair. */
181 memset(&s, 0, sizeof(s));
182 c16 = 0;
183 ATF_REQUIRE(mbrtoc16(&c16, "\xf0\x9f\x92\xa9", 4, &s) == 4);
184 ATF_REQUIRE(c16 == 0xd83d);
185 ATF_REQUIRE(mbrtoc16(&c16, "", 0, &s) == (size_t)-3);
186 ATF_REQUIRE(c16 == 0xdca9);
187
188 /* Letter e with acute, precomposed. */
189 memset(&s, 0, sizeof(s));
190 c16 = 0;
191 ATF_REQUIRE(mbrtoc16(&c16, "\xc3\xa9", 2, &s) == 2);
192 ATF_REQUIRE(c16 == 0xe9);
193
194 /* Letter e with acute, combined. */
195 memset(&s, 0, sizeof(s));
196 c16 = 0;
197 ATF_REQUIRE(mbrtoc16(&c16, "\x65\xcc\x81", 3, &s) == 1);
198 ATF_REQUIRE(c16 == 0x65);
199 ATF_REQUIRE(mbrtoc16(&c16, "\xcc\x81", 2, &s) == 2);
200 ATF_REQUIRE(c16 == 0x301);
201 }
202
ATF_TP_ADD_TCS(tp)203 ATF_TP_ADD_TCS(tp)
204 {
205
206 ATF_TP_ADD_TC(tp, mbrtoc16_c_locale_test);
207 ATF_TP_ADD_TC(tp, mbrtoc16_iso_8859_1_test);
208 ATF_TP_ADD_TC(tp, mbrtoc16_iso_8859_15_test);
209 ATF_TP_ADD_TC(tp, mbrtoc16_utf_8_test);
210
211 return (atf_no_error());
212 }
213