1 /* Test of word breaks in UTF-32 strings.
2 Copyright (C) 2009-2025 Free Software Foundation, Inc.
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <https://www.gnu.org/licenses/>. */
17 /* Written by Bruno Haible <bruno@clisp.org>, 2009. */
30 /* Test case n = 0. */
31 u32_wordbreaks (NULL
, 0, NULL
);
34 static const uint32_t input
[61] =
35 /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a) 日本語,中文,한글" */
36 { 'G', 'r', 0x00FC, 0x00DF, ' ', 'G', 'o', 't', 't', '.', ' ',
37 0x0417, 0x0434, 0x0440, 0x0430, 0x0432, 0x0441, 0x0442, 0x0432, 0x0443,
38 0x0439, 0x0442, 0x0435, '!', ' ',
39 'x', '=', '(', '-', 'b', 0x00B1, 's', 'q', 'r', 't', '(', 'b', 0x00B2,
40 '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')', ' ', ' ',
41 0x65E5, 0x672C, 0x8A9E, ',', 0x4E2D, 0x6587, ',', 0xD55C, 0xAE00, '\n'
43 char *p
= (char *) malloc (SIZEOF (input
));
46 u32_wordbreaks (input
, SIZEOF (input
), p
);
48 for (i
= 0; i
< 61; i
++)
50 ASSERT (p
[i
] == ((i
>= 4 && i
<= 5)
51 || (i
>= 9 && i
<= 11)
52 || (i
>= 23 && i
<= 31)
53 || (i
>= 35 && i
<= 39)
54 || (i
>= 42 && i
<= 46)
55 || (i
>= 48 && i
<= 49)
56 || (i
>= 51 && i
<= 58)
64 /* Same input string, decomposed. */
65 static const uint32_t input
[67] =
66 /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a) 日本語,中文,한글" */
67 { 'G', 'r', 0x0075, 0x0308, 0x00DF, ' ', 'G', 'o', 't', 't', '.', ' ',
68 0x0417, 0x0434, 0x0440, 0x0430, 0x0432, 0x0441, 0x0442, 0x0432, 0x0443,
69 0x0438, 0x0306, 0x0442, 0x0435, '!', ' ',
70 'x', '=', '(', '-', 'b', 0x00B1, 's', 'q', 'r', 't', '(', 'b', 0x00B2,
71 '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')', ' ', ' ',
72 0x65E5, 0x672C, 0x8A9E, ',', 0x4E2D, 0x6587, ',',
73 0x1112, 0x1161, 0x11AB, 0x1100, 0x1173, 0x11AF, '\n'
75 char *p
= (char *) malloc (SIZEOF (input
));
78 u32_wordbreaks (input
, SIZEOF (input
), p
);
80 for (i
= 0; i
< 67; i
++)
82 ASSERT (p
[i
] == ((i
>= 5 && i
<= 6)
83 || (i
>= 10 && i
<= 12)
84 || (i
>= 25 && i
<= 33)
85 || (i
>= 37 && i
<= 41)
86 || (i
>= 44 && i
<= 48)
87 || (i
>= 50 && i
<= 51)
88 || (i
>= 53 && i
<= 60)
97 static const uint32_t input
[8] =
98 { 'a', '\n', 'b', '\r', 'c', '\r', '\n', 'd' };
99 char *p
= (char *) malloc (SIZEOF (input
));
102 u32_wordbreaks (input
, SIZEOF (input
), p
);
103 for (i
= 0; i
< 8; i
++)
105 ASSERT (p
[i
] == (i
== 1 || i
== 2 || i
== 3 || i
== 4 || i
== 5
112 /* Test regional indicators. */
114 static const uint32_t input
[6] =
115 { '.', 0x1F1E9, 0x1F1EA, 0x1F1EB, 0x1F1F7, '.' };
116 char *p
= (char *) malloc (SIZEOF (input
));
119 u32_wordbreaks (input
, SIZEOF (input
), p
);
120 for (i
= 0; i
< 6; i
++)
122 ASSERT (p
[i
] == (i
== 1 || i
== 3 || i
== 5 ? 1 : 0));
127 return test_exit_status
;