glib/tests/utf8-performance.c

   1 /* GLIB - Library of useful routines for C programming
   2  *
   3  * Copyright (C) 2010 Mikhail Zabaluev <mikhail.zabaluev@gmail.com>
   4  *
   5  * This 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 of the License, or (at your option) any later version.
   9  *
  10  * This 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 this library; if not, see <http://www.gnu.org/licenses/>.
  17  */
  18
  19 #include <string.h>
  20
  21 #include <glib.h>
  22
  23 #define NUM_ITERATIONS 500000
  24
  25 static const char str_ascii[] =
  26     "The quick brown fox jumps over the lazy dog";
  27
  28 static const gchar str_latin1[] =
  29     "Zwölf Boxkämpfer jagen Viktor quer über den großen Sylter Deich";
  30
  31 /* Energizing GOELRO-talk in Russian, used by KDE */
  32 static const char str_cyrillic[] =
  33     "Широкая электрификация южных губерний даст мощный толчок подъёму "
  34     "сельского хозяйства.";
  35
  36 /* First sentence from the Wikipedia article:
  37  * http://zh.wikipedia.org/w/index.php?title=%E6%B1%89%E5%AD%97&oldid=13053137 */
  38 static const char str_chinese[] =
  39     "漢字，亦稱中文字、中国字，在台灣又被稱為國字，是漢字文化圈廣泛使用的一種文字，屬於表意文字的詞素音節文字";
  40
  41 typedef int (* GrindFunc) (const char *, gsize);
  42
  43 static int
  44 grind_get_char (const char *str, gsize len)
  45 {
  46   gunichar acc = 0;
  47   int i;
  48   for (i = 0; i < NUM_ITERATIONS; i++)
  49     {
  50       const char *p = str;
  51       while (*p) {
  52         acc += g_utf8_get_char (p);
  53         p = g_utf8_next_char (p);
  54       }
  55     }
  56   return acc;
  57 }
  58
  59 static int
  60 grind_get_char_validated (const char *str, gsize len)
  61 {
  62   gunichar acc = 0;
  63   int i;
  64   for (i = 0; i < NUM_ITERATIONS; i++)
  65     {
  66       const char *p = str;
  67       while (*p) {
  68         acc += g_utf8_get_char_validated (p, -1);
  69         p = g_utf8_next_char (p);
  70       }
  71     }
  72   return acc;
  73 }
  74
  75 static int
  76 grind_utf8_to_ucs4 (const char *str, gsize len)
  77 {
  78   int i;
  79   for (i = 0; i < NUM_ITERATIONS; i++)
  80     {
  81       gunichar *ustr;
  82       ustr = g_utf8_to_ucs4 (str, -1, NULL, NULL, NULL);
  83       g_free (ustr);
  84     }
  85   return 0;
  86 }
  87
  88 static int
  89 grind_get_char_backwards (const char *str, gsize len)
  90 {
  91   gunichar acc = 0;
  92   int i;
  93   for (i = 0; i < NUM_ITERATIONS; i++)
  94     {
  95       const char *p = str + len;
  96       do
  97         {
  98           p = g_utf8_prev_char (p);
  99           acc += g_utf8_get_char (p);
 100         }
 101       while (p != str);
 102     }
 103   return acc;
 104 }
 105
 106 static int
 107 grind_utf8_to_ucs4_sized (const char *str, gsize len)
 108 {
 109   int i;
 110   for (i = 0; i < NUM_ITERATIONS; i++)
 111     {
 112       gunichar *ustr;
 113       ustr = g_utf8_to_ucs4 (str, len, NULL, NULL, NULL);
 114       g_free (ustr);
 115     }
 116   return 0;
 117 }
 118
 119 static int
 120 grind_utf8_to_ucs4_fast (const char *str, gsize len)
 121 {
 122   int i;
 123   for (i = 0; i < NUM_ITERATIONS; i++)
 124     {
 125       gunichar *ustr;
 126       ustr = g_utf8_to_ucs4_fast (str, -1, NULL);
 127       g_free (ustr);
 128     }
 129   return 0;
 130 }
 131
 132 static int
 133 grind_utf8_to_ucs4_fast_sized (const char *str, gsize len)
 134 {
 135   int i;
 136   for (i = 0; i < NUM_ITERATIONS; i++)
 137     {
 138       gunichar *ustr;
 139       ustr = g_utf8_to_ucs4_fast (str, len, NULL);
 140       g_free (ustr);
 141     }
 142   return 0;
 143 }
 144
 145 static void
 146 perform_for (GrindFunc grind_func, const char *str, const char *label)
 147 {
 148   gsize len;
 149   gulong bytes_ground;
 150   gdouble time_elapsed;
 151   gdouble result;
 152
 153   len = strlen (str);
 154   bytes_ground = (gulong) len * NUM_ITERATIONS;
 155
 156   g_test_timer_start ();
 157
 158   grind_func (str, len);
 159
 160   time_elapsed = g_test_timer_elapsed ();
 161
 162   result = ((gdouble) bytes_ground / time_elapsed) * 1.0e-6;
 163
 164   g_test_maximized_result (result, "%-9s %6.1f MB/s", label, result);
 165 }
 166
 167 static void
 168 perform (gconstpointer data)
 169 {
 170   GrindFunc grind_func = (GrindFunc) data;
 171
 172   if (!g_test_perf ())
 173     return;
 174
 175   perform_for (grind_func, str_ascii, "ASCII:");
 176   perform_for (grind_func, str_latin1, "Latin-1:");
 177   perform_for (grind_func, str_cyrillic, "Cyrillic:");
 178   perform_for (grind_func, str_chinese, "Chinese:");
 179 }
 180
 181 int
 182 main (int argc, char **argv)
 183 {
 184   g_test_init (&argc, &argv, NULL);
 185
 186   if (g_test_perf ())
 187     {
 188       g_test_add_data_func ("/utf8/perf/get_char", grind_get_char, perform);
 189       g_test_add_data_func ("/utf8/perf/get_char-backwards", grind_get_char_backwards, perform);
 190       g_test_add_data_func ("/utf8/perf/get_char_validated", grind_get_char_validated, perform);
 191       g_test_add_data_func ("/utf8/perf/utf8_to_ucs4", grind_utf8_to_ucs4, perform);
 192       g_test_add_data_func ("/utf8/perf/utf8_to_ucs4-sized", grind_utf8_to_ucs4_sized, perform);
 193       g_test_add_data_func ("/utf8/perf/utf8_to_ucs4_fast", grind_utf8_to_ucs4_fast, perform);
 194       g_test_add_data_func ("/utf8/perf/utf8_to_ucs4_fast-sized", grind_utf8_to_ucs4_fast_sized, perform);
 195     }
 196
 197   return g_test_run ();
 198 }