1 | /* Test collation function via transformation using real data. |
2 | Copyright (C) 1997-2022 Free Software Foundation, Inc. |
3 | This file is part of the GNU C Library. |
4 | |
5 | The GNU C Library is free software; you can redistribute it and/or |
6 | modify it under the terms of the GNU Lesser General Public |
7 | License as published by the Free Software Foundation; either |
8 | version 2.1 of the License, or (at your option) any later version. |
9 | |
10 | The GNU C Library is distributed in the hope that it will be useful, |
11 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
12 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
13 | Lesser General Public License for more details. |
14 | |
15 | You should have received a copy of the GNU Lesser General Public |
16 | License along with the GNU C Library; if not, see |
17 | <https://www.gnu.org/licenses/>. */ |
18 | |
19 | #include <ctype.h> |
20 | #include <error.h> |
21 | #include <locale.h> |
22 | #include <stdio.h> |
23 | #include <stdlib.h> |
24 | #include <string.h> |
25 | #include <stdbool.h> |
26 | |
27 | /* Keep in sync with string/strxfrm_l.c. */ |
28 | #define SMALL_STR_SIZE 4095 |
29 | |
30 | struct lines |
31 | { |
32 | char *xfrm; |
33 | char *line; |
34 | }; |
35 | |
36 | static int xstrcmp (const void *, const void *); |
37 | |
38 | int |
39 | main (int argc, char *argv[]) |
40 | { |
41 | int result = 0; |
42 | bool nocache = false; |
43 | size_t nstrings, nstrings_max; |
44 | struct lines *strings; |
45 | char *line = NULL; |
46 | size_t len = 0; |
47 | size_t n; |
48 | |
49 | if (argc < 2) |
50 | error (status: 1, errnum: 0, format: "usage: %s <random seed> [-nocache]" , argv[0]); |
51 | |
52 | if (argc == 3) |
53 | { |
54 | if (strcmp (argv[2], "-nocache" ) == 0) |
55 | nocache = true; |
56 | else |
57 | { |
58 | printf (format: "Unknown option %s!\n" , argv[2]); |
59 | exit (1); |
60 | } |
61 | } |
62 | |
63 | setlocale (LC_ALL, "" ); |
64 | |
65 | nstrings_max = 100; |
66 | nstrings = 0; |
67 | strings = (struct lines *) malloc (size: nstrings_max * sizeof (struct lines)); |
68 | if (strings == NULL) |
69 | { |
70 | perror (argv[0]); |
71 | exit (1); |
72 | } |
73 | |
74 | while (1) |
75 | { |
76 | char saved, *word, *newp; |
77 | size_t l, line_len, needed; |
78 | |
79 | if (getline (lineptr: &line, n: &len, stdin) < 0) |
80 | break; |
81 | |
82 | if (nstrings == nstrings_max) |
83 | { |
84 | strings = (struct lines *) realloc (ptr: strings, |
85 | size: (nstrings_max *= 2) |
86 | * sizeof (*strings)); |
87 | if (strings == NULL) |
88 | { |
89 | perror (argv[0]); |
90 | exit (1); |
91 | } |
92 | } |
93 | strings[nstrings].line = strdup (s: line); |
94 | l = strcspn (line, ":(;" ); |
95 | while (l > 0 && isspace (line[l - 1])) |
96 | --l; |
97 | |
98 | saved = line[l]; |
99 | line[l] = '\0'; |
100 | |
101 | if (nocache) |
102 | { |
103 | line_len = strlen (line); |
104 | word = malloc (size: line_len + SMALL_STR_SIZE + 1); |
105 | if (word == NULL) |
106 | { |
107 | printf (format: "malloc failed: %m\n" ); |
108 | exit (1); |
109 | } |
110 | memset (word, ' ', SMALL_STR_SIZE); |
111 | memcpy (word + SMALL_STR_SIZE, line, line_len); |
112 | word[line_len + SMALL_STR_SIZE] = '\0'; |
113 | } |
114 | else |
115 | word = line; |
116 | |
117 | needed = strxfrm (NULL, src: word, n: 0); |
118 | newp = malloc (size: needed + 1); |
119 | if (newp == NULL) |
120 | { |
121 | printf (format: "malloc failed: %m\n" ); |
122 | exit (1); |
123 | } |
124 | strxfrm (dest: newp, src: word, n: needed + 1); |
125 | strings[nstrings].xfrm = newp; |
126 | |
127 | if (nocache) |
128 | free (ptr: word); |
129 | line[l] = saved; |
130 | ++nstrings; |
131 | } |
132 | free (ptr: line); |
133 | |
134 | /* First shuffle. */ |
135 | srandom (seed: atoi (argv[1])); |
136 | for (n = 0; n < 10 * nstrings; ++n) |
137 | { |
138 | int r1, r2, r; |
139 | size_t idx1 = random () % nstrings; |
140 | size_t idx2 = random () % nstrings; |
141 | struct lines tmp = strings[idx1]; |
142 | strings[idx1] = strings[idx2]; |
143 | strings[idx2] = tmp; |
144 | |
145 | /* While we are at it a first little test. */ |
146 | r1 = strcmp (strings[idx1].xfrm, strings[idx2].xfrm); |
147 | r2 = strcmp (strings[idx2].xfrm, strings[idx1].xfrm); |
148 | r = -(r1 ^ r2); |
149 | if (r) |
150 | r /= abs (x: r1 ^ r2); |
151 | |
152 | if (r < 0 || (r == 0 && (r1 != 0 || r2 != 0)) |
153 | || (r > 0 && (r1 ^ r2) >= 0)) |
154 | printf (format: "collate wrong: %d vs. %d\n" , r1, r2); |
155 | } |
156 | |
157 | /* Now sort. */ |
158 | qsort (strings, nstrings, sizeof (struct lines), xstrcmp); |
159 | |
160 | /* Print the result. */ |
161 | for (n = 0; n < nstrings; ++n) |
162 | { |
163 | fputs (strings[n].line, stdout); |
164 | free (ptr: strings[n].line); |
165 | free (ptr: strings[n].xfrm); |
166 | } |
167 | free (ptr: strings); |
168 | |
169 | return result; |
170 | } |
171 | |
172 | |
173 | static int |
174 | xstrcmp (const void *ptr1, const void *ptr2) |
175 | { |
176 | const struct lines *l1 = (const struct lines *) ptr1; |
177 | const struct lines *l2 = (const struct lines *) ptr2; |
178 | |
179 | return strcmp (l1->xfrm, l2->xfrm); |
180 | } |
181 | |