1/* Test of compatibility normalization of UTF-32 strings.
2   Copyright (C) 2009, 2010 Free Software Foundation, Inc.
3
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.
8
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.
13
14   You should have received a copy of the GNU General Public License
15   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
16
17/* Written by Bruno Haible <bruno@clisp.org>, 2009.  */
18
19#include <config.h>
20
21#if GNULIB_TEST_UNINORM_U32_NORMALIZE
22
23#include "uninorm.h"
24
25#include <signal.h>
26#include <stdlib.h>
27#include <unistd.h>
28
29#include "unistr.h"
30#include "macros.h"
31
32static int
33check (const uint32_t *input, size_t input_length,
34       const uint32_t *expected, size_t expected_length)
35{
36  size_t length;
37  uint32_t *result;
38
39  /* Test return conventions with resultbuf == NULL.  */
40  result = u32_normalize (UNINORM_NFKC, input, input_length, NULL, &length);
41  if (!(result != NULL))
42    return 1;
43  if (!(length == expected_length))
44    return 2;
45  if (!(u32_cmp (result, expected, expected_length) == 0))
46    return 3;
47  free (result);
48
49  /* Test return conventions with resultbuf too small.  */
50  if (expected_length > 0)
51    {
52      uint32_t *preallocated;
53
54      length = expected_length - 1;
55      preallocated = (uint32_t *) malloc (length * sizeof (uint32_t));
56      result = u32_normalize (UNINORM_NFKC, input, input_length, preallocated, &length);
57      if (!(result != NULL))
58        return 4;
59      if (!(result != preallocated))
60        return 5;
61      if (!(length == expected_length))
62        return 6;
63      if (!(u32_cmp (result, expected, expected_length) == 0))
64        return 7;
65      free (result);
66      free (preallocated);
67    }
68
69  /* Test return conventions with resultbuf large enough.  */
70  {
71    uint32_t *preallocated;
72
73    length = expected_length;
74    preallocated = (uint32_t *) malloc (length * sizeof (uint32_t));
75    result = u32_normalize (UNINORM_NFKC, input, input_length, preallocated, &length);
76    if (!(result != NULL))
77      return 8;
78    if (!(preallocated == NULL || result == preallocated))
79      return 9;
80    if (!(length == expected_length))
81      return 10;
82    if (!(u32_cmp (result, expected, expected_length) == 0))
83      return 11;
84    free (preallocated);
85  }
86
87  return 0;
88}
89
90void
91test_u32_nfkc (void)
92{
93  { /* Empty string.  */
94    ASSERT (check (NULL, 0, NULL, 0) == 0);
95  }
96  { /* SPACE */
97    static const uint32_t input[]    = { 0x0020 };
98    ASSERT (check (input, SIZEOF (input), input, SIZEOF (input)) == 0);
99  }
100
101  { /* LATIN CAPITAL LETTER A WITH DIAERESIS */
102    static const uint32_t input[]      = { 0x00C4 };
103    static const uint32_t decomposed[] = { 0x0041, 0x0308 };
104    ASSERT (check (input, SIZEOF (input),           input, SIZEOF (input)) == 0);
105    ASSERT (check (decomposed, SIZEOF (decomposed), input, SIZEOF (input)) == 0);
106  }
107
108  { /* LATIN CAPITAL LETTER A WITH DIAERESIS AND MACRON */
109    static const uint32_t input[]      = { 0x01DE };
110    static const uint32_t decomposed[] = { 0x0041, 0x0308, 0x0304 };
111    ASSERT (check (input, SIZEOF (input),           input, SIZEOF (input)) == 0);
112    ASSERT (check (decomposed, SIZEOF (decomposed), input, SIZEOF (input)) == 0);
113  }
114
115  { /* ANGSTROM SIGN */
116    static const uint32_t input[]      = { 0x212B };
117    static const uint32_t decomposed[] = { 0x0041, 0x030A };
118    static const uint32_t expected[]   = { 0x00C5 };
119    ASSERT (check (input, SIZEOF (input),           expected, SIZEOF (expected)) == 0);
120    ASSERT (check (decomposed, SIZEOF (decomposed), expected, SIZEOF (expected)) == 0);
121    ASSERT (check (expected, SIZEOF (expected),     expected, SIZEOF (expected)) == 0);
122  }
123
124  { /* GREEK DIALYTIKA AND PERISPOMENI */
125    static const uint32_t input[]      = { 0x1FC1 };
126    static const uint32_t decomposed[] = { 0x0020, 0x0308, 0x0342 };
127    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
128    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
129  }
130
131  { /* SCRIPT SMALL L */
132    static const uint32_t input[]      = { 0x2113 };
133    static const uint32_t decomposed[] = { 0x006C };
134    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
135    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
136  }
137
138  { /* NO-BREAK SPACE */
139    static const uint32_t input[]      = { 0x00A0 };
140    static const uint32_t decomposed[] = { 0x0020 };
141    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
142    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
143  }
144
145  { /* ARABIC LETTER VEH INITIAL FORM */
146    static const uint32_t input[]      = { 0xFB6C };
147    static const uint32_t decomposed[] = { 0x06A4 };
148    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
149    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
150  }
151
152  { /* ARABIC LETTER VEH MEDIAL FORM */
153    static const uint32_t input[]      = { 0xFB6D };
154    static const uint32_t decomposed[] = { 0x06A4 };
155    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
156    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
157  }
158
159  { /* ARABIC LETTER VEH FINAL FORM */
160    static const uint32_t input[]      = { 0xFB6B };
161    static const uint32_t decomposed[] = { 0x06A4 };
162    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
163    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
164  }
165
166  { /* ARABIC LETTER VEH ISOLATED FORM */
167    static const uint32_t input[]      = { 0xFB6A };
168    static const uint32_t decomposed[] = { 0x06A4 };
169    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
170    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
171  }
172
173  { /* CIRCLED NUMBER FIFTEEN */
174    static const uint32_t input[]      = { 0x246E };
175    static const uint32_t decomposed[] = { 0x0031, 0x0035 };
176    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
177    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
178  }
179
180  { /* TRADE MARK SIGN */
181    static const uint32_t input[]      = { 0x2122 };
182    static const uint32_t decomposed[] = { 0x0054, 0x004D };
183    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
184    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
185  }
186
187  { /* LATIN SUBSCRIPT SMALL LETTER I */
188    static const uint32_t input[]      = { 0x1D62 };
189    static const uint32_t decomposed[] = { 0x0069 };
190    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
191    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
192  }
193
194  { /* PRESENTATION FORM FOR VERTICAL LEFT PARENTHESIS */
195    static const uint32_t input[]      = { 0xFE35 };
196    static const uint32_t decomposed[] = { 0x0028 };
197    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
198    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
199  }
200
201  { /* FULLWIDTH LATIN CAPITAL LETTER A */
202    static const uint32_t input[]      = { 0xFF21 };
203    static const uint32_t decomposed[] = { 0x0041 };
204    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
205    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
206  }
207
208  { /* HALFWIDTH IDEOGRAPHIC COMMA */
209    static const uint32_t input[]      = { 0xFF64 };
210    static const uint32_t decomposed[] = { 0x3001 };
211    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
212    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
213  }
214
215  { /* SMALL IDEOGRAPHIC COMMA */
216    static const uint32_t input[]      = { 0xFE51 };
217    static const uint32_t decomposed[] = { 0x3001 };
218    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
219    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
220  }
221
222  { /* SQUARE MHZ */
223    static const uint32_t input[]      = { 0x3392 };
224    static const uint32_t decomposed[] = { 0x004D, 0x0048, 0x007A };
225    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
226    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
227  }
228
229  { /* VULGAR FRACTION THREE EIGHTHS */
230    static const uint32_t input[]      = { 0x215C };
231    static const uint32_t decomposed[] = { 0x0033, 0x2044, 0x0038 };
232    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
233    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
234  }
235
236  { /* MICRO SIGN */
237    static const uint32_t input[]      = { 0x00B5 };
238    static const uint32_t decomposed[] = { 0x03BC };
239    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
240    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
241  }
242
243  { /* ARABIC LIGATURE SALLALLAHOU ALAYHE WASALLAM */
244    static const uint32_t input[]      = { 0xFDFA };
245    static const uint32_t decomposed[] =
246      { 0x0635, 0x0644, 0x0649, 0x0020, 0x0627, 0x0644, 0x0644, 0x0647, 0x0020,
247        0x0639, 0x0644, 0x064A, 0x0647, 0x0020, 0x0648, 0x0633, 0x0644, 0x0645
248      };
249    ASSERT (check (input, SIZEOF (input),           decomposed, SIZEOF (decomposed)) == 0);
250    ASSERT (check (decomposed, SIZEOF (decomposed), decomposed, SIZEOF (decomposed)) == 0);
251  }
252
253  { /* HANGUL SYLLABLE GEUL */
254    static const uint32_t input[]      = { 0xAE00 };
255    static const uint32_t decomposed[] = { 0x1100, 0x1173, 0x11AF };
256    ASSERT (check (input, SIZEOF (input),           input, SIZEOF (input)) == 0);
257    ASSERT (check (decomposed, SIZEOF (decomposed), input, SIZEOF (input)) == 0);
258  }
259
260  { /* HANGUL SYLLABLE GEU */
261    static const uint32_t input[]      = { 0xADF8 };
262    static const uint32_t decomposed[] = { 0x1100, 0x1173 };
263    ASSERT (check (input, SIZEOF (input),           input, SIZEOF (input)) == 0);
264    ASSERT (check (decomposed, SIZEOF (decomposed), input, SIZEOF (input)) == 0);
265  }
266
267  { /* "Gr���� Gott. ������������������������! x=(-b��sqrt(b��-4ac))/(2a)  ���������,������,������" */
268    static const uint32_t input[] =
269      { 'G', 'r', 0x00FC, 0x00DF, ' ', 'G', 'o', 't', 't', '.', ' ',
270        0x0417, 0x0434, 0x0440, 0x0430, 0x0432, 0x0441, 0x0442, 0x0432, 0x0443,
271        0x0439, 0x0442, 0x0435, '!', ' ',
272        'x', '=', '(', '-', 'b', 0x00B1, 's', 'q', 'r', 't', '(', 'b', 0x00B2,
273        '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')', ' ', ' ',
274        0x65E5, 0x672C, 0x8A9E, ',', 0x4E2D, 0x6587, ',', 0xD55C, 0xAE00, '\n'
275      };
276    static const uint32_t decomposed[] =
277      { 'G', 'r', 0x0075, 0x0308, 0x00DF, ' ', 'G', 'o', 't', 't', '.', ' ',
278        0x0417, 0x0434, 0x0440, 0x0430, 0x0432, 0x0441, 0x0442, 0x0432, 0x0443,
279        0x0438, 0x0306, 0x0442, 0x0435, '!', ' ',
280        'x', '=', '(', '-', 'b', 0x00B1, 's', 'q', 'r', 't', '(', 'b', 0x0032,
281        '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')', ' ', ' ',
282        0x65E5, 0x672C, 0x8A9E, ',', 0x4E2D, 0x6587, ',',
283        0x1112, 0x1161, 0x11AB, 0x1100, 0x1173, 0x11AF, '\n'
284      };
285    static const uint32_t expected[] =
286      { 'G', 'r', 0x00FC, 0x00DF, ' ', 'G', 'o', 't', 't', '.', ' ',
287        0x0417, 0x0434, 0x0440, 0x0430, 0x0432, 0x0441, 0x0442, 0x0432, 0x0443,
288        0x0439, 0x0442, 0x0435, '!', ' ',
289        'x', '=', '(', '-', 'b', 0x00B1, 's', 'q', 'r', 't', '(', 'b', 0x0032,
290        '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')', ' ', ' ',
291        0x65E5, 0x672C, 0x8A9E, ',', 0x4E2D, 0x6587, ',', 0xD55C, 0xAE00, '\n'
292      };
293    ASSERT (check (input, SIZEOF (input),           expected, SIZEOF (expected)) == 0);
294    ASSERT (check (decomposed, SIZEOF (decomposed), expected, SIZEOF (expected)) == 0);
295    ASSERT (check (expected, SIZEOF (expected),     expected, SIZEOF (expected)) == 0);
296  }
297
298#if HAVE_DECL_ALARM
299  /* Declare failure if test takes too long, by using default abort
300     caused by SIGALRM.  */
301  signal (SIGALRM, SIG_DFL);
302  alarm (50);
303#endif
304
305  /* Check that the sorting is not O(n��) but O(n log n).  */
306  {
307    int pass;
308    for (pass = 0; pass < 3; pass++)
309      {
310        size_t repeat = 1;
311        size_t m = 100000;
312        uint32_t *input = (uint32_t *) malloc (2 * m * sizeof (uint32_t));
313        if (input != NULL)
314          {
315            uint32_t *expected = input + m;
316            size_t m1 = m / 2;
317            size_t m2 = (m - 1) / 2;
318            /* NB: m1 + m2 == m - 1.  */
319            uint32_t *p;
320            size_t i;
321
322            input[0] = 0x0041;
323            p = input + 1;
324            switch (pass)
325              {
326              case 0:
327                for (i = 0; i < m1; i++)
328                  *p++ = 0x0319;
329                for (i = 0; i < m2; i++)
330                  *p++ = 0x0300;
331                break;
332
333              case 1:
334                for (i = 0; i < m2; i++)
335                  *p++ = 0x0300;
336                for (i = 0; i < m1; i++)
337                  *p++ = 0x0319;
338                break;
339
340              case 2:
341                for (i = 0; i < m2; i++)
342                  {
343                    *p++ = 0x0319;
344                    *p++ = 0x0300;
345                  }
346                for (; i < m1; i++)
347                  *p++ = 0x0319;
348                break;
349
350              default:
351                abort ();
352              }
353
354            expected[0] = 0x00C0;
355            p = expected + 1;
356            for (i = 0; i < m1; i++)
357              *p++ = 0x0319;
358            for (i = 0; i < m2 - 1; i++)
359              *p++ = 0x0300;
360
361            for (; repeat > 0; repeat--)
362              {
363                ASSERT (check (input, m,        expected, m - 1) == 0);
364                ASSERT (check (expected, m - 1, expected, m - 1) == 0);
365              }
366
367            free (input);
368          }
369      }
370  }
371}
372
373#else
374
375void
376test_u32_nfkc (void)
377{
378}
379
380#endif
381