fsort.c revision 1.19 1 /* $NetBSD: fsort.c,v 1.19 2001/05/15 11:19:45 jdolecek Exp $ */
2
3 /*-
4 * Copyright (c) 1993
5 * The Regents of the University of California. All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * Peter McIlroy.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 * 1. Redistributions of source code must retain the above copyright
14 * notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in the
17 * documentation and/or other materials provided with the distribution.
18 * 3. All advertising materials mentioning features or use of this software
19 * must display the following acknowledgement:
20 * This product includes software developed by the University of
21 * California, Berkeley and its contributors.
22 * 4. Neither the name of the University nor the names of its contributors
23 * may be used to endorse or promote products derived from this software
24 * without specific prior written permission.
25 *
26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
36 * SUCH DAMAGE.
37 */
38
39 /*
40 * Read in the next bin. If it fits in one segment sort it;
41 * otherwise refine it by segment deeper by one character,
42 * and try again on smaller bins. Sort the final bin at this level
43 * of recursion to keep the head of fstack at 0.
44 * After PANIC passes, abort to merge sort.
45 */
46 #include "sort.h"
47 #include "fsort.h"
48
49 #ifndef lint
50 __RCSID("$NetBSD: fsort.c,v 1.19 2001/05/15 11:19:45 jdolecek Exp $");
51 __SCCSID("@(#)fsort.c 8.1 (Berkeley) 6/6/93");
52 #endif /* not lint */
53
54 #include <stdlib.h>
55 #include <string.h>
56
57 static const u_char **keylist = 0;
58 u_char *buffer = 0, *linebuf = 0;
59 size_t bufsize = DEFBUFSIZE;
60 size_t linebuf_size;
61 struct tempfile fstack[MAXFCT];
62 extern char *toutpath;
63 #define FSORTMAX 4
64 int PANIC = FSORTMAX;
65
66 #define MSTART (MAXFCT - MERGE_FNUM)
67 #define SALIGN(n) ((n+sizeof(length_t)-1) & ~(sizeof(length_t)-1))
68
69 void
70 fsort(binno, depth, top, filelist, nfiles, outfp, ftbl)
71 int binno, depth, top;
72 struct filelist *filelist;
73 int nfiles;
74 FILE *outfp;
75 struct field *ftbl;
76 {
77 const u_char **keypos;
78 u_char *bufend, *tmpbuf;
79 u_char *weights;
80 int ntfiles, mfct = 0, total, i, maxb, lastb, panic = 0;
81 int c, nelem, base;
82 long sizes [NBINS+1];
83 get_func_t get;
84 struct recheader *crec;
85 struct field tfield[2];
86 FILE *prevfp, *tailfp[FSORTMAX+1];
87
88 memset(tailfp, 0, sizeof(tailfp));
89 prevfp = outfp;
90 memset(tfield, 0, sizeof(tfield));
91 if (ftbl[0].flags & R)
92 tfield[0].weights = Rascii;
93 else
94 tfield[0].weights = ascii;
95 tfield[0].icol.num = 1;
96 weights = ftbl[0].weights;
97 if (!buffer) {
98 buffer = malloc(bufsize);
99 keylist = malloc(MAXNUM * sizeof(u_char *));
100 memset(keylist, 0, MAXNUM * sizeof(u_char *));
101 if (!SINGL_FLD) {
102 linebuf_size = DEFLLEN;
103 if ((linebuf = malloc(linebuf_size)) == NULL)
104 errx(2, "cannot allocate memory");
105 }
106 }
107 bufend = buffer + bufsize;
108 if (binno >= 0) {
109 base = top + nfiles;
110 get = getnext;
111 } else {
112 base = 0;
113 if (SINGL_FLD)
114 get = makeline;
115 else
116 get = makekey;
117 }
118 for (;;) {
119 memset(sizes, 0, sizeof(sizes));
120 c = ntfiles = 0;
121 if (binno == weights[REC_D] &&
122 !(SINGL_FLD && ftbl[0].flags & F)) { /* pop */
123 rd_append(weights[REC_D], top,
124 nfiles, prevfp, buffer, bufend);
125 break;
126 } else if (binno == weights[REC_D]) {
127 depth = 0; /* start over on flat weights */
128 ftbl = tfield;
129 weights = ftbl[0].weights;
130 }
131 while (c != EOF) {
132 keypos = keylist;
133 nelem = 0;
134 crec = (RECHEADER *) buffer;
135
136 do_read:
137 while((c = get(binno, top, filelist, nfiles, crec,
138 bufend, ftbl)) == 0) {
139 *keypos++ = crec->data + depth;
140 if (++nelem == MAXNUM) {
141 c = BUFFEND;
142 break;
143 }
144 crec =(RECHEADER *) ((char *) crec +
145 SALIGN(crec->length) + sizeof(TRECHEADER));
146 }
147
148 if (c == BUFFEND && nelem < MAXNUM
149 && bufsize < MAXBUFSIZE) {
150 const u_char **keyp;
151 u_char *oldb = buffer;
152
153 /* buffer was too small for data, allocate
154 * bigger buffer */
155 bufsize *= 2;
156 buffer = realloc(buffer, bufsize);
157 if (!buffer) {
158 err(2, "failed to realloc buffer to %ld bytes",
159 (unsigned long) bufsize);
160 }
161 bufend = buffer + bufsize;
162
163 /* patch up keylist[] */
164 for(keyp = &keypos[-1]; keyp >= keylist; keyp--)
165 *keyp = buffer + (*keyp - oldb);
166
167 crec = (RECHEADER *) (buffer + ((u_char *)crec - oldb));
168 goto do_read;
169 }
170
171 if (c != BUFFEND && !ntfiles && !mfct) {
172 /* do not push */
173 continue;
174 }
175
176 /* push */
177 if (panic >= PANIC) {
178 fstack[MSTART + mfct].fp = ftmp();
179 if ((stable_sort)
180 ? sradixsort(keylist, nelem,
181 weights, REC_D)
182 : radixsort(keylist, nelem,
183 weights, REC_D) )
184 err(2, NULL);
185 append(keylist, nelem, depth,
186 fstack[MSTART + mfct].fp, putrec,
187 ftbl);
188 mfct++;
189 /* reduce number of open files */
190 if (mfct == MERGE_FNUM ||(c == EOF && ntfiles)) {
191 tmpbuf = malloc(bufend -
192 crec->data);
193 memmove(tmpbuf, crec->data,
194 bufend - crec->data);
195 fstack[base + ntfiles].fp
196 = ftmp();
197 fmerge(0, MSTART, filelist,
198 mfct, geteasy,
199 fstack[base].fp,
200 putrec, ftbl);
201 ntfiles++;
202 mfct = 0;
203 memmove(crec->data, tmpbuf,
204 bufend - crec->data);
205 free(tmpbuf);
206 }
207 } else {
208 fstack[base + ntfiles].fp= ftmp();
209 onepass(keylist, depth, nelem, sizes,
210 weights, fstack[base + ntfiles].fp);
211 ntfiles++;
212 }
213 }
214 if (!ntfiles && !mfct) { /* everything in memory--pop */
215 if (nelem > 1
216 && ((stable_sort)
217 ? sradixsort(keylist, nelem, weights, REC_D)
218 : radixsort(keylist, nelem, weights, REC_D) ))
219 err(2, NULL);
220 if (nelem > 0)
221 append(keylist, nelem, depth, outfp, putline, ftbl);
222 break; /* pop */
223 }
224 if (panic >= PANIC) {
225 if (!ntfiles)
226 fmerge(0, MSTART, filelist, mfct, geteasy,
227 outfp, putline, ftbl);
228 else
229 fmerge(0, base, filelist, ntfiles, geteasy,
230 outfp, putline, ftbl);
231 break;
232
233 }
234 total = maxb = lastb = 0; /* find if one bin dominates */
235 for (i = 0; i < NBINS; i++)
236 if (sizes[i]) {
237 if (sizes[i] > sizes[maxb])
238 maxb = i;
239 lastb = i;
240 total += sizes[i];
241 }
242 if (sizes[maxb] < max((total / 2) , BUFSIZE))
243 maxb = lastb; /* otherwise pop after last bin */
244 fstack[base].lastb = lastb;
245 fstack[base].maxb = maxb;
246
247 /* start refining next level. */
248 getnext(-1, base, NULL, ntfiles, crec, bufend, 0); /* rewind */
249 for (i = 0; i < maxb; i++) {
250 if (!sizes[i]) /* bin empty; step ahead file offset */
251 getnext(i, base, NULL,ntfiles, crec, bufend, 0);
252 else
253 fsort(i, depth+1, base, filelist, ntfiles,
254 outfp, ftbl);
255 }
256
257 get = getnext;
258
259 if (lastb != maxb) {
260 if (prevfp != outfp)
261 tailfp[panic] = prevfp;
262 prevfp = ftmp();
263 for (i = maxb+1; i <= lastb; i++)
264 if (!sizes[i])
265 getnext(i, base, NULL, ntfiles, crec,
266 bufend,0);
267 else
268 fsort(i, depth+1, base, filelist,
269 ntfiles, prevfp, ftbl);
270 }
271
272 /* sort biggest (or last) bin at this level */
273 depth++;
274 panic++;
275 binno = maxb;
276 top = base;
277 nfiles = ntfiles; /* so overwrite them */
278 }
279 if (prevfp != outfp) {
280 concat(outfp, prevfp);
281 fclose(prevfp);
282 }
283 for (i = panic; i >= 0; --i)
284 if (tailfp[i]) {
285 concat(outfp, tailfp[i]);
286 fclose(tailfp[i]);
287 }
288
289 /* If on top level, free our structures */
290 if (depth == 0) {
291 free(keylist), keylist = NULL;
292 free(buffer), buffer = NULL;
293 }
294 }
295
296 /*
297 * This is one pass of radix exchange, dumping the bins to disk.
298 */
299 #define swap(a, b, t) t = a, a = b, b = t
300 void
301 onepass(a, depth, n, sizes, tr, fp)
302 const u_char **a;
303 int depth;
304 long n, sizes[];
305 u_char *tr;
306 FILE *fp;
307 {
308 size_t tsizes[NBINS+1];
309 const u_char **bin[257], ***bp, ***bpmax, **top[256], ***tp;
310 static int histo[256];
311 int *hp;
312 int c;
313 const u_char **an, *t, **aj;
314 const u_char **ak, *r;
315
316 memset(tsizes, 0, sizeof(tsizes));
317 depth += sizeof(TRECHEADER);
318 an = &a[n];
319 for (ak = a; ak < an; ak++) {
320 histo[c = tr[**ak]]++;
321 tsizes[c] += ((const RECHEADER *) (*ak -= depth))->length;
322 }
323
324 bin[0] = a;
325 bpmax = bin + 256;
326 tp = top, hp = histo;
327 for (bp = bin; bp < bpmax; bp++) {
328 *tp++ = *(bp+1) = *bp + (c = *hp);
329 *hp++ = 0;
330 if (c <= 1)
331 continue;
332 }
333 for (aj = a; aj < an; *aj = r, aj = bin[c+1])
334 for (r = *aj; aj < (ak = --top[c = tr[r[depth]]]) ;)
335 swap(*ak, r, t);
336
337 for (ak = a, c = 0; c < 256; c++) {
338 an = bin[c+1];
339 n = an - ak;
340 tsizes[c] += n * sizeof(TRECHEADER);
341 /* tell getnext how many elements in this bin, this segment. */
342 EWRITE(&tsizes[c], sizeof(size_t), 1, fp);
343 sizes[c] += tsizes[c];
344 for (; ak < an; ++ak)
345 putrec((const RECHEADER *) *ak, fp);
346 }
347 }
348