stdlib/FreeBSD/radixsort.c

   1 /*-
   2  * Copyright (c) 1990, 1993
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * This code is derived from software contributed to Berkeley by
   6  * Peter McIlroy and by Dan Bernstein at New York University,
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 4. Neither the name of the University nor the names of its contributors
  17  *    may be used to endorse or promote products derived from this software
  18  *    without specific prior written permission.
  19  *
  20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  30  * SUCH DAMAGE.
  31  */
  32
  33 #if defined(LIBC_SCCS) && !defined(lint)
  34 static char sccsid[] = "@(#)radixsort.c 8.2 (Berkeley) 4/28/95";
  35 #endif /* LIBC_SCCS and not lint */
  36 #include <sys/cdefs.h>
  37 __FBSDID("$FreeBSD: src/lib/libc/stdlib/radixsort.c,v 1.8 2007/01/09 00:28:10 imp Exp $");
  38
  39 /*
  40  * Radixsort routines.
  41  *
  42  * Program r_sort_a() is unstable but uses O(logN) extra memory for a stack.
  43  * Use radixsort(a, n, trace, endchar) for this case.
  44  *
  45  * For stable sorting (using N extra pointers) use sradixsort(), which calls
  46  * r_sort_b().
  47  *
  48  * For a description of this code, see D. McIlroy, P. McIlroy, K. Bostic,
  49  * "Engineering Radix Sort".
  50  */
  51
  52 #include <sys/types.h>
  53 #include <stdlib.h>
  54 #include <stddef.h>
  55 #include <errno.h>
  56
  57 typedef struct {
  58         const u_char **sa;
  59         int sn, si;
  60 } stack;
  61
  62 static inline void simplesort
  63 (const u_char **, int, int, const u_char *, u_int);
  64 static void r_sort_a(const u_char **, int, int, const u_char *, u_int);
  65 static void r_sort_b(const u_char **, const u_char **, int, int,
  66     const u_char *, u_int);
  67
  68 #define THRESHOLD       20              /* Divert to simplesort(). */
  69 #define SIZE            512             /* Default stack size. */
  70
  71 #define SETUP {                                                         \
  72         if (tab == NULL) {                                              \
  73                 tr = tr0;                                               \
  74                 for (c = 0; c < endch; c++)                             \
  75                         tr0[c] = c + 1;                                 \
  76                 tr0[c] = 0;                                             \
  77                 for (c++; c < 256; c++)                                 \
  78                         tr0[c] = c;                                     \
  79                 endch = 0;                                              \
  80         } else {                                                        \
  81                 endch = tab[endch];                                     \
  82                 tr = tab;                                               \
  83                 if (endch != 0 && endch != 255) {                       \
  84                         errno = EINVAL;                                 \
  85                         return (-1);                                    \
  86                 }                                                       \
  87         }                                                               \
  88 }
  89
  90 int
  91 radixsort(a, n, tab, endch)
  92         const u_char **a, *tab;
  93         int n;
  94         u_int endch;
  95 {
  96         const u_char *tr;
  97         int c;
  98         u_char tr0[256];
  99
 100         SETUP;
 101         r_sort_a(a, n, 0, tr, endch);
 102         return (0);
 103 }
 104
 105 int
 106 sradixsort(a, n, tab, endch)
 107         const u_char **a, *tab;
 108         int n;
 109         u_int endch;
 110 {
 111         const u_char *tr, **ta;
 112         int c;
 113         u_char tr0[256];
 114
 115         SETUP;
 116         if (n < THRESHOLD)
 117                 simplesort(a, n, 0, tr, endch);
 118         else {
 119                 if ((ta = malloc(n * sizeof(a))) == NULL)
 120                         return (-1);
 121                 r_sort_b(a, ta, n, 0, tr, endch);
 122                 free(ta);
 123         }
 124         return (0);
 125 }
 126
 127 #define empty(s)        (s >= sp)
 128 #define pop(a, n, i)    a = (--sp)->sa, n = sp->sn, i = sp->si
 129 #define push(a, n, i)   sp->sa = a, sp->sn = n, (sp++)->si = i
 130 #define swap(a, b, t)   t = a, a = b, b = t
 131
 132 /* Unstable, in-place sort. */
 133 static void
 134 r_sort_a(a, n, i, tr, endch)
 135         const u_char **a;
 136         int n, i;
 137         const u_char *tr;
 138         u_int endch;
 139 {
 140         static int count[256], nc, bmin;
 141         int c;
 142         const u_char **ak, *r;
 143         stack s[SIZE], *sp, *sp0, *sp1, temp;
 144         int *cp, bigc;
 145         const u_char **an, *t, **aj, **top[256];
 146
 147         /* Set up stack. */
 148         sp = s;
 149         push(a, n, i);
 150         while (!empty(s)) {
 151                 pop(a, n, i);
 152                 if (n < THRESHOLD) {
 153                         simplesort(a, n, i, tr, endch);
 154                         continue;
 155                 }
 156                 an = a + n;
 157
 158                 /* Make character histogram. */
 159                 if (nc == 0) {
 160                         bmin = 255;     /* First occupied bin, excluding eos. */
 161                         for (ak = a; ak < an;) {
 162                                 c = tr[(*ak++)[i]];
 163                                 if (++count[c] == 1 && c != endch) {
 164                                         if (c < bmin)
 165                                                 bmin = c;
 166                                         nc++;
 167                                 }
 168                         }
 169                         if (sp + nc > s + SIZE) {       /* Get more stack. */
 170                                 r_sort_a(a, n, i, tr, endch);
 171                                 continue;
 172                         }
 173                 }
 174
 175                 /*
 176                  * Special case: if all strings have the same
 177                  * character at position i, move on to the next
 178                  * character.
 179                  */
 180                 if (nc == 1 && count[bmin] == n) {
 181                         push(a, n, i+1);
 182                         nc = count[bmin] = 0;
 183                         continue;
 184                 }
 185
 186                 /*
 187                  * Set top[]; push incompletely sorted bins onto stack.
 188                  * top[] = pointers to last out-of-place element in bins.
 189                  * count[] = counts of elements in bins.
 190                  * Before permuting: top[c-1] + count[c] = top[c];
 191                  * during deal: top[c] counts down to top[c-1].
 192                  */
 193                 sp0 = sp1 = sp;         /* Stack position of biggest bin. */
 194                 bigc = 2;               /* Size of biggest bin. */
 195                 if (endch == 0)         /* Special case: set top[eos]. */
 196                         top[0] = ak = a + count[0];
 197                 else {
 198                         ak = a;
 199                         top[255] = an;
 200                 }
 201                 for (cp = count + bmin; nc > 0; cp++) {
 202                         while (*cp == 0)        /* Find next non-empty pile. */
 203                                 cp++;
 204                         if (*cp > 1) {
 205                                 if (*cp > bigc) {
 206                                         bigc = *cp;
 207                                         sp1 = sp;
 208                                 }
 209                                 push(ak, *cp, i+1);
 210                         }
 211                         top[cp-count] = ak += *cp;
 212                         nc--;
 213                 }
 214                 swap(*sp0, *sp1, temp); /* Play it safe -- biggest bin last. */
 215
 216                 /*
 217                  * Permute misplacements home.  Already home: everything
 218                  * before aj, and in bin[c], items from top[c] on.
 219                  * Inner loop:
 220                  *      r = next element to put in place;
 221                  *      ak = top[r[i]] = location to put the next element.
 222                  *      aj = bottom of 1st disordered bin.
 223                  * Outer loop:
 224                  *      Once the 1st disordered bin is done, ie. aj >= ak,
 225                  *      aj<-aj + count[c] connects the bins in a linked list;
 226                  *      reset count[c].
 227                  */
 228                 for (aj = a; aj < an;  *aj = r, aj += count[c], count[c] = 0)
 229                         for (r = *aj;  aj < (ak = --top[c = tr[r[i]]]);)
 230                                 swap(*ak, r, t);
 231         }
 232 }
 233
 234 /* Stable sort, requiring additional memory. */
 235 static void
 236 r_sort_b(a, ta, n, i, tr, endch)
 237         const u_char **a, **ta;
 238         int n, i;
 239         const u_char *tr;
 240         u_int endch;
 241 {
 242         static int count[256], nc, bmin;
 243         int c;
 244         const u_char **ak, **ai;
 245         stack s[512], *sp, *sp0, *sp1, temp;
 246         const u_char **top[256];
 247         int *cp, bigc;
 248
 249         sp = s;
 250         push(a, n, i);
 251         while (!empty(s)) {
 252                 pop(a, n, i);
 253                 if (n < THRESHOLD) {
 254                         simplesort(a, n, i, tr, endch);
 255                         continue;
 256                 }
 257
 258                 if (nc == 0) {
 259                         bmin = 255;
 260                         for (ak = a + n; --ak >= a;) {
 261                                 c = tr[(*ak)[i]];
 262                                 if (++count[c] == 1 && c != endch) {
 263                                         if (c < bmin)
 264                                                 bmin = c;
 265                                         nc++;
 266                                 }
 267                         }
 268                         if (sp + nc > s + SIZE) {
 269                                 r_sort_b(a, ta, n, i, tr, endch);
 270                                 continue;
 271                         }
 272                 }
 273
 274                 sp0 = sp1 = sp;
 275                 bigc = 2;
 276                 if (endch == 0) {
 277                         top[0] = ak = a + count[0];
 278                         count[0] = 0;
 279                 } else {
 280                         ak = a;
 281                         top[255] = a + n;
 282                         count[255] = 0;
 283                 }
 284                 for (cp = count + bmin; nc > 0; cp++) {
 285                         while (*cp == 0)
 286                                 cp++;
 287                         if ((c = *cp) > 1) {
 288                                 if (c > bigc) {
 289                                         bigc = c;
 290                                         sp1 = sp;
 291                                 }
 292                                 push(ak, c, i+1);
 293                         }
 294                         top[cp-count] = ak += c;
 295                         *cp = 0;                        /* Reset count[]. */
 296                         nc--;
 297                 }
 298                 swap(*sp0, *sp1, temp);
 299
 300                 for (ak = ta + n, ai = a+n; ak > ta;)   /* Copy to temp. */
 301                         *--ak = *--ai;
 302                 for (ak = ta+n; --ak >= ta;)            /* Deal to piles. */
 303                         *--top[tr[(*ak)[i]]] = *ak;
 304         }
 305 }
 306
 307 static inline void
 308 simplesort(a, n, b, tr, endch)  /* insertion sort */
 309         const u_char **a;
 310         int n, b;
 311         const u_char *tr;
 312         u_int endch;
 313 {
 314         u_char ch;
 315         const u_char  **ak, **ai, *s, *t;
 316
 317         for (ak = a+1; --n >= 1; ak++)
 318                 for (ai = ak; ai > a; ai--) {
 319                         for (s = ai[0] + b, t = ai[-1] + b;
 320                             (ch = tr[*s]) != endch; s++, t++)
 321                                 if (ch != tr[*t])
 322                                         break;
 323                         if (ch >= tr[*t])
 324                                 break;
 325                         swap(ai[0], ai[-1], s);
 326                 }
 327 }