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_han[] =
  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 #define GRIND_LOOP_BEGIN                 \
  44   {                                      \
  45     int i;                               \
  46     for (i = 0; i < NUM_ITERATIONS; i++)
  47
  48 #define GRIND_LOOP_END \
  49   }
  50
  51 static int
  52 grind_get_char (const char *str, gsize len)
  53 {
  54   gunichar acc = 0;
  55   GRIND_LOOP_BEGIN
  56     {
  57       const char *p = str;
  58       while (*p)
  59         {
  60           acc += g_utf8_get_char (p);
  61           p = g_utf8_next_char (p);
  62         }
  63     }
  64   GRIND_LOOP_END;
  65   return acc;
  66 }
  67
  68 static int
  69 grind_get_char_validated (const char *str, gsize len)
  70 {
  71   gunichar acc = 0;
  72   GRIND_LOOP_BEGIN
  73     {
  74       const char *p = str;
  75       while (*p)
  76         {
  77           acc += g_utf8_get_char_validated (p, -1);
  78           p = g_utf8_next_char (p);
  79         }
  80     }
  81   GRIND_LOOP_END;
  82   return acc;
  83 }
  84
  85 static int
  86 grind_utf8_to_ucs4 (const char *str, gsize len)
  87 {
  88   GRIND_LOOP_BEGIN
  89     {
  90       gunichar *ustr;
  91       ustr = g_utf8_to_ucs4 (str, -1, NULL, NULL, NULL);
  92       g_free (ustr);
  93     }
  94   GRIND_LOOP_END;
  95   return 0;
  96 }
  97
  98 static int
  99 grind_get_char_backwards (const char *str, gsize len)
 100 {
 101   gunichar acc = 0;
 102   GRIND_LOOP_BEGIN
 103     {
 104       const char *p = str + len;
 105       do
 106         {
 107           p = g_utf8_prev_char (p);
 108           acc += g_utf8_get_char (p);
 109         }
 110       while (p != str);
 111     }
 112   GRIND_LOOP_END;
 113   return acc;
 114 }
 115
 116 static int
 117 grind_utf8_to_ucs4_sized (const char *str, gsize len)
 118 {
 119   GRIND_LOOP_BEGIN
 120     {
 121       gunichar *ustr;
 122       ustr = g_utf8_to_ucs4 (str, len, NULL, NULL, NULL);
 123       g_free (ustr);
 124     }
 125   GRIND_LOOP_END;
 126   return 0;
 127 }
 128
 129 static int
 130 grind_utf8_to_ucs4_fast (const char *str, gsize len)
 131 {
 132   GRIND_LOOP_BEGIN
 133     {
 134       gunichar *ustr;
 135       ustr = g_utf8_to_ucs4_fast (str, -1, NULL);
 136       g_free (ustr);
 137     }
 138   GRIND_LOOP_END;
 139   return 0;
 140 }
 141
 142 static int
 143 grind_utf8_to_ucs4_fast_sized (const char *str, gsize len)
 144 {
 145   GRIND_LOOP_BEGIN
 146     {
 147       gunichar *ustr;
 148       ustr = g_utf8_to_ucs4_fast (str, len, NULL);
 149       g_free (ustr);
 150     }
 151   GRIND_LOOP_END;
 152   return 0;
 153 }
 154
 155 static int
 156 grind_utf8_validate (const char *str, gsize len)
 157 {
 158   GRIND_LOOP_BEGIN
 159     g_utf8_validate (str, -1, NULL);
 160   GRIND_LOOP_END;
 161   return 0;
 162 }
 163
 164 static int
 165 grind_utf8_validate_sized (const char *str, gsize len)
 166 {
 167   GRIND_LOOP_BEGIN
 168     g_utf8_validate (str, len, NULL);
 169   GRIND_LOOP_END;
 170   return 0;
 171 }
 172
 173 typedef struct _GrindData {
 174   GrindFunc func;
 175   const char *str;
 176 } GrindData;
 177
 178 static void
 179 perform (gconstpointer data)
 180 {
 181   GrindData *gd = (GrindData *) data;
 182   GrindFunc grind_func = gd->func;
 183   const char *str = gd->str;
 184   gsize len;
 185   gulong bytes_ground;
 186   gdouble time_elapsed;
 187   gdouble result;
 188
 189   len = strlen (str);
 190   bytes_ground = (gulong) len * NUM_ITERATIONS;
 191
 192   g_test_timer_start ();
 193
 194   grind_func (str, len);
 195
 196   time_elapsed = g_test_timer_elapsed ();
 197
 198   result = ((gdouble) bytes_ground / time_elapsed) * 1.0e-6;
 199
 200   g_test_maximized_result (result, "%7.1f MB/s", result);
 201
 202   g_slice_free (GrindData, gd);
 203 }
 204
 205 static void
 206 add_cases(const char *path, GrindFunc func)
 207 {
 208 #define ADD_CASE(script)                              \
 209   G_STMT_START {                                      \
 210     GrindData *gd;                                    \
 211     gchar *full_path;                                 \
 212     gd = g_slice_new0(GrindData);                     \
 213     gd->func = func;                                  \
 214     gd->str = str_##script;                           \
 215     full_path = g_strdup_printf("%s/" #script, path); \
 216     g_test_add_data_func (full_path, gd, perform);    \
 217     g_free (full_path);                               \
 218   } G_STMT_END
 219
 220   ADD_CASE(ascii);
 221   ADD_CASE(latin1);
 222   ADD_CASE(cyrillic);
 223   ADD_CASE(han);
 224
 225 #undef ADD_CASE
 226 }
 227
 228 int
 229 main (int argc, char **argv)
 230 {
 231   g_test_init (&argc, &argv, NULL);
 232
 233   if (g_test_perf ())
 234     {
 235       add_cases ("/utf8/perf/get_char", grind_get_char);
 236       add_cases ("/utf8/perf/get_char-backwards", grind_get_char_backwards);
 237       add_cases ("/utf8/perf/get_char_validated", grind_get_char_validated);
 238       add_cases ("/utf8/perf/utf8_to_ucs4", grind_utf8_to_ucs4);
 239       add_cases ("/utf8/perf/utf8_to_ucs4-sized", grind_utf8_to_ucs4_sized);
 240       add_cases ("/utf8/perf/utf8_to_ucs4_fast", grind_utf8_to_ucs4_fast);
 241       add_cases ("/utf8/perf/utf8_to_ucs4_fast-sized", grind_utf8_to_ucs4_fast_sized);
 242       add_cases ("/utf8/perf/utf8_validate", grind_utf8_validate);
 243       add_cases ("/utf8/perf/utf8_validate-sized", grind_utf8_validate_sized);
 244     }
 245
 246   return g_test_run ();
 247 }