gqsort.c

   1 /* GLIB - Library of useful routines for C programming
   2  * Copyright (C) 1991, 1992, 1996, 1997 Free Software Foundation, Inc.
   3  * Copyright (C) 2000 Eazel, Inc.
   4  * Copyright (C) 1995-1997  Peter Mattis, Spencer Kimball and Josh MacDonald
   5  *
   6  * This library is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Lesser General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2 of the License, or (at your option) any later version.
  10  *
  11  * This library is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14  * Lesser General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Lesser General Public
  17  * License along with this library; if not, write to the
  18  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  19  * Boston, MA 02111-1307, USA.
  20  */
  21
  22 /*
  23  * This file was originally part of the GNU C Library, and was modified to allow
  24  * user data to be passed in to the sorting function.
  25  *
  26  * Written by Douglas C. Schmidt (schmidt@ics.uci.edu).
  27  * Modified by Maciej Stachowiak (mjs@eazel.com)
  28  *
  29  * Modified by the GLib Team and others 1997-2000.  See the AUTHORS
  30  * file for a list of people on the GLib Team.  See the ChangeLog
  31  * files for a list of changes.  These files are distributed with GLib
  32  * at ftp://ftp.gtk.org/pub/gtk/.
  33  */
  34
  35 #include <string.h>
  36
  37 #include "glib.h"
  38
  39 /* Byte-wise swap two items of size SIZE. */
  40 #define SWAP(a, b, size)                                                      \
  41   do                                                                          \
  42     {                                                                         \
  43       register size_t __size = (size);                                        \
  44       register char *__a = (a), *__b = (b);                                   \
  45       do                                                                      \
  46         {                                                                     \
  47           char __tmp = *__a;                                                  \
  48           *__a++ = *__b;                                                      \
  49           *__b++ = __tmp;                                                     \
  50         } while (--__size > 0);                                               \
  51     } while (0)
  52
  53 /* Discontinue quicksort algorithm when partition gets below this size.
  54    This particular magic number was chosen to work best on a Sun 4/260. */
  55 #define MAX_THRESH 4
  56
  57 /* Stack node declarations used to store unfulfilled partition obligations. */
  58 typedef struct
  59 {
  60   char *lo;
  61   char *hi;
  62 }
  63 stack_node;
  64
  65 /* The next 4 #defines implement a very fast in-line stack abstraction. */
  66 #define STACK_SIZE      (8 * sizeof(unsigned long int))
  67 #define PUSH(low, high) ((void) ((top->lo = (low)), (top->hi = (high)), ++top))
  68 #define POP(low, high)  ((void) (--top, (low = top->lo), (high = top->hi)))
  69 #define STACK_NOT_EMPTY (stack < top)
  70
  71
  72 /* Order size using quicksort.  This implementation incorporates
  73  * four optimizations discussed in Sedgewick:
  74  *
  75  * 1. Non-recursive, using an explicit stack of pointer that store the next
  76  *    array partition to sort.  To save time, this maximum amount of space
  77  *    required to store an array of MAX_INT is allocated on the stack.  Assuming
  78  *    a 32-bit integer, this needs only 32 * sizeof(stack_node) == 136 bits.
  79  *    Pretty cheap, actually.
  80  *
  81  * 2. Chose the pivot element using a median-of-three decision tree.  This
  82  *    reduces the probability of selecting a bad pivot value and eliminates
  83  *    certain * extraneous comparisons.
  84  *
  85  * 3. Only quicksorts TOTAL_ELEMS / MAX_THRESH partitions, leaving insertion
  86  *    sort to order the MAX_THRESH items within each partition.  This is a big
  87  *    win, since insertion sort is faster for small, mostly sorted array
  88  *    segments.
  89  *
  90  * 4. The larger of the two sub-partitions is always pushed onto the stack
  91  *    first, with the algorithm then concentrating on the smaller partition.
  92  *    This *guarantees* no more than log (n) stack size is needed (actually O(1)
  93  *    in this case)!
  94  */
  95
  96 /**
  97  * g_qsort_with_data:
  98  * @pbase: start of array to sort
  99  * @total_elems: elements in the array
 100  * @size: size of each element
 101  * @compare_func: function to compare elements
 102  * @user_data: data to pass to @compare_func
 103  *
 104  * This is just like the standard C qsort() function, but
 105  * the comparison routine accepts a user data argument.
 106  *
 107  **/
 108 void
 109 g_qsort_with_data (gconstpointer    pbase,
 110                    gint             total_elems,
 111                    size_t           size,
 112                    GCompareDataFunc compare_func,
 113                    gpointer         user_data)
 114 {
 115   register char *base_ptr = (char *) pbase;
 116
 117   /* Allocating SIZE bytes for a pivot buffer facilitates a better
 118    * algorithm below since we can do comparisons directly on the pivot.
 119    */
 120   char *pivot_buffer = (char *) alloca (size);
 121   const size_t max_thresh = MAX_THRESH * size;
 122
 123   g_return_if_fail (total_elems > 0);
 124   g_return_if_fail (pbase != NULL);
 125   g_return_if_fail (compare_func != NULL);
 126
 127   if (total_elems > MAX_THRESH)
 128     {
 129       char *lo = base_ptr;
 130       char *hi = &lo[size * (total_elems - 1)];
 131       /* Largest size needed for 32-bit int!!! */
 132       stack_node stack[STACK_SIZE];
 133       stack_node *top = stack + 1;
 134
 135       while (STACK_NOT_EMPTY)
 136         {
 137           char *left_ptr;
 138           char *right_ptr;
 139
 140           char *pivot = pivot_buffer;
 141
 142           /* Select median value from among LO, MID, and HI. Rearrange
 143            * LO and HI so the three values are sorted. This lowers the
 144            * probability of picking a pathological pivot value and
 145            * skips a comparison for both the LEFT_PTR and RIGHT_PTR. */
 146
 147           char *mid = lo + size * ((hi - lo) / size >> 1);
 148
 149           if ((*compare_func) ((void *) mid, (void *) lo, user_data) < 0)
 150             SWAP (mid, lo, size);
 151           if ((*compare_func) ((void *) hi, (void *) mid, user_data) < 0)
 152             SWAP (mid, hi, size);
 153           else
 154             goto jump_over;
 155           if ((*compare_func) ((void *) mid, (void *) lo, user_data) < 0)
 156             SWAP (mid, lo, size);
 157         jump_over:;
 158           memcpy (pivot, mid, size);
 159           pivot = pivot_buffer;
 160
 161           left_ptr = lo + size;
 162           right_ptr = hi - size;
 163
 164           /* Here's the famous ``collapse the walls'' section of quicksort.
 165            * Gotta like those tight inner loops!  They are the main reason
 166            * that this algorithm runs much faster than others. */
 167           do
 168             {
 169               while ((*compare_func)
 170                      ((void *) left_ptr, (void *) pivot,
 171                       user_data) < 0)
 172                 left_ptr += size;
 173
 174               while ((*compare_func)
 175                      ((void *) pivot, (void *) right_ptr,
 176                       user_data) < 0)
 177                 right_ptr -= size;
 178
 179               if (left_ptr < right_ptr)
 180                 {
 181                   SWAP (left_ptr, right_ptr, size);
 182                   left_ptr += size;
 183                   right_ptr -= size;
 184                 }
 185               else if (left_ptr == right_ptr)
 186                 {
 187                   left_ptr += size;
 188                   right_ptr -= size;
 189                   break;
 190                 }
 191             }
 192           while (left_ptr <= right_ptr);
 193
 194           /* Set up pointers for next iteration.  First determine whether
 195            * left and right partitions are below the threshold size.  If so,
 196            * ignore one or both.  Otherwise, push the larger partition's
 197            * bounds on the stack and continue sorting the smaller one. */
 198
 199           if ((size_t) (right_ptr - lo) <= max_thresh)
 200             {
 201               if ((size_t) (hi - left_ptr) <= max_thresh)
 202                 /* Ignore both small partitions. */
 203                 POP (lo, hi);
 204               else
 205                 /* Ignore small left partition. */
 206                 lo = left_ptr;
 207             }
 208           else if ((size_t) (hi - left_ptr) <= max_thresh)
 209                                 /* Ignore small right partition. */
 210             hi = right_ptr;
 211           else if ((right_ptr - lo) > (hi - left_ptr))
 212             {
 213                                 /* Push larger left partition indices. */
 214               PUSH (lo, right_ptr);
 215               lo = left_ptr;
 216
 217             }
 218           else
 219             {
 220                                 /* Push larger right partition indices. */
 221               PUSH (left_ptr, hi);
 222               hi = right_ptr;
 223             }
 224         }
 225     }
 226
 227   /* Once the BASE_PTR array is partially sorted by quicksort the rest
 228    * is completely sorted using insertion sort, since this is efficient
 229    * for partitions below MAX_THRESH size. BASE_PTR points to the beginning
 230    * of the array to sort, and END_PTR points at the very last element in
 231    * the array (*not* one beyond it!). */
 232
 233   {
 234     char *const end_ptr = &base_ptr[size * (total_elems - 1)];
 235     char *tmp_ptr = base_ptr;
 236     char *thresh = MIN (end_ptr, base_ptr + max_thresh);
 237     register char *run_ptr;
 238
 239     /* Find smallest element in first threshold and place it at the
 240      * array's beginning.  This is the smallest array element,
 241      * and the operation speeds up insertion sort's inner loop. */
 242
 243     for (run_ptr = tmp_ptr + size; run_ptr <= thresh;
 244          run_ptr +=
 245            size) if ((*compare_func) ((void *) run_ptr, (void *) tmp_ptr,
 246                                       user_data) < 0)
 247              tmp_ptr = run_ptr;
 248
 249     if (tmp_ptr != base_ptr)
 250       SWAP (tmp_ptr, base_ptr, size);
 251
 252     /* Insertion sort, running from left-hand-side up to right-hand-side.  */
 253
 254     run_ptr = base_ptr + size;
 255     while ((run_ptr += size) <= end_ptr)
 256       {
 257         tmp_ptr = run_ptr - size;
 258         while ((*compare_func)
 259                ((void *) run_ptr, (void *) tmp_ptr,
 260                 user_data) < 0)
 261           tmp_ptr -= size;
 262
 263         tmp_ptr += size;
 264         if (tmp_ptr != run_ptr)
 265           {
 266             char *trav;
 267
 268             trav = run_ptr + size;
 269             while (--trav >= run_ptr)
 270               {
 271                 char c = *trav;
 272                 char *hi, *lo;
 273
 274                 for (hi = lo = trav;
 275                      (lo -= size) >= tmp_ptr; hi = lo)
 276                   *hi = *lo;
 277                 *hi = c;
 278               }
 279           }
 280       }
 281   }
 282 }