1 | /**************************************************************************
|
---|
2 | *
|
---|
3 | * mg_text_estimate.c -- Program to estimate the compressed text size
|
---|
4 | * Copyright (C) 1994 Neil Sharman
|
---|
5 | *
|
---|
6 | * This program is free software; you can redistribute it and/or modify
|
---|
7 | * it under the terms of the GNU General Public License as published by
|
---|
8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
9 | * (at your option) any later version.
|
---|
10 | *
|
---|
11 | * This program 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
|
---|
14 | * GNU General Public License for more details.
|
---|
15 | *
|
---|
16 | * You should have received a copy of the GNU General Public License
|
---|
17 | * along with this program; if not, write to the Free Software
|
---|
18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
19 | *
|
---|
20 | * $Id: mg_text_estimate.c 25147 2012-02-28 00:59:00Z kjdon $
|
---|
21 | *
|
---|
22 | **************************************************************************/
|
---|
23 |
|
---|
24 | #include "sysfuncs.h"
|
---|
25 |
|
---|
26 | #include "memlib.h"
|
---|
27 | #include "messages.h"
|
---|
28 | #include "local_strings.h"
|
---|
29 | #include "bitio_gen.h"
|
---|
30 | #include "bitio_m.h"
|
---|
31 | #include "bitio_m_stdio.h"
|
---|
32 | #include "timing.h"
|
---|
33 | #include "mgheap.h"
|
---|
34 | #include "netorder.h" /* [RPAP - Jan 97: Endian Ordering] */
|
---|
35 |
|
---|
36 | #include "mg_files.h"
|
---|
37 | #include "locallib.h"
|
---|
38 | #include "invf.h"
|
---|
39 | #include "text.h"
|
---|
40 | #include "words.h"
|
---|
41 | #include "mg.h"
|
---|
42 | #include "comp_dict.h"
|
---|
43 | #include "hash.h"
|
---|
44 |
|
---|
45 |
|
---|
46 |
|
---|
47 |
|
---|
48 |
|
---|
49 | /*
|
---|
50 | $Log$
|
---|
51 | Revision 1.1 2003/02/20 21:18:24 mdewsnip
|
---|
52 | Addition of MG package for search and retrieval
|
---|
53 |
|
---|
54 | Revision 1.1 1999/08/10 21:18:14 sjboddie
|
---|
55 | renamed mg-1.3d directory mg
|
---|
56 |
|
---|
57 | Revision 1.4 1999/01/15 03:05:51 rjmcnab
|
---|
58 |
|
---|
59 | Renamed lib/heap to be lib/mgheap (it was causing some problems with
|
---|
60 | some versions of the STL libraries which contained a heap.h)
|
---|
61 |
|
---|
62 | Revision 1.3 1998/12/17 09:12:53 rjmcnab
|
---|
63 |
|
---|
64 | Altered mg to process utf-8 encoded Unicode. The main changes
|
---|
65 | are in the parsing of the input, the casefolding, and the stemming.
|
---|
66 |
|
---|
67 | Revision 1.2 1998/11/25 07:55:48 rjmcnab
|
---|
68 |
|
---|
69 | Modified mg to that you can specify the stemmer you want
|
---|
70 | to use via a command line option. You specify it to
|
---|
71 | mg_passes during the build process. The number of the
|
---|
72 | stemmer that you used is stored within the inverted
|
---|
73 | dictionary header and the stemmed dictionary header so
|
---|
74 | the correct stemmer is used in later stages of building
|
---|
75 | and querying.
|
---|
76 |
|
---|
77 | Revision 1.1 1998/11/17 09:35:19 rjmcnab
|
---|
78 | *** empty log message ***
|
---|
79 |
|
---|
80 | * Revision 1.3 1994/10/20 03:56:59 tes
|
---|
81 | * I have rewritten the boolean query optimiser and abstracted out the
|
---|
82 | * components of the boolean query.
|
---|
83 | *
|
---|
84 | * Revision 1.2 1994/09/20 04:41:54 tes
|
---|
85 | * For version 1.1
|
---|
86 | *
|
---|
87 | */
|
---|
88 |
|
---|
89 | static char *RCSID = "$Id: mg_text_estimate.c 25147 2012-02-28 00:59:00Z kjdon $";
|
---|
90 |
|
---|
91 | typedef struct stats_t
|
---|
92 | {
|
---|
93 | mg_u_long freq, occur;
|
---|
94 | mg_u_long len;
|
---|
95 | }
|
---|
96 | stats_t;
|
---|
97 |
|
---|
98 | static mg_u_long huff_count = 0;
|
---|
99 | static mg_u_long escape_count = 0;
|
---|
100 |
|
---|
101 |
|
---|
102 | char mode = 'H';
|
---|
103 |
|
---|
104 | /* [RJM 07/97: 4G limit] */
|
---|
105 | static double ReadInWordsStandard (char *name, double *num_bytes);
|
---|
106 | static double ReadInWordsSpecial (char *name, double *num_bytes,
|
---|
107 | double *num_extra_bytes,
|
---|
108 | mg_u_long *aux_compressed);
|
---|
109 |
|
---|
110 |
|
---|
111 |
|
---|
112 |
|
---|
113 | int main (int argc, char **argv)
|
---|
114 | {
|
---|
115 | char *stats_dict = NULL, *comp_dict = NULL;
|
---|
116 | ProgTime StartTime;
|
---|
117 | double num_bytes, num_extra_bytes = 0; /* [RJM 07/97: 4G limit] */
|
---|
118 | mg_u_long aux_compressed;
|
---|
119 | int ch;
|
---|
120 | double bytes;
|
---|
121 | opterr = 0;
|
---|
122 | msg_prefix = argv[0];
|
---|
123 | while ((ch = getopt (argc, argv, "BMDYHh")) != -1)
|
---|
124 | switch (ch)
|
---|
125 | {
|
---|
126 | case 'B':
|
---|
127 | Message ("The compressed text size for this method cannot be\n"
|
---|
128 | "estimated exactly, so an upper bound will be calculated.");
|
---|
129 | mode = ch;
|
---|
130 | break;
|
---|
131 | case 'M':
|
---|
132 | Message ("The compressed text size for this method cannot be\n"
|
---|
133 | "estimated exactly, so an upper bound will be calculated.");
|
---|
134 | mode = 'Y';
|
---|
135 | break;
|
---|
136 | case 'D':
|
---|
137 | case 'Y':
|
---|
138 | case 'H':
|
---|
139 | mode = ch;
|
---|
140 | break;
|
---|
141 | case 'h':
|
---|
142 | case '?':
|
---|
143 | fprintf (stderr, "usage: %s [-h] [B|M|D|Y|H] stats_dict comp_dict\n",
|
---|
144 | argv[0]);
|
---|
145 | exit (1);
|
---|
146 | }
|
---|
147 |
|
---|
148 | if (optind < argc)
|
---|
149 | stats_dict = argv[optind++];
|
---|
150 | else
|
---|
151 | stats_dict = NULL;
|
---|
152 | if (optind < argc)
|
---|
153 | comp_dict = argv[optind++];
|
---|
154 | else
|
---|
155 | {
|
---|
156 | fprintf (stderr, "usage: %s [-h] [B|M|D|Y|H] stats_dict comp_dict\n",
|
---|
157 | argv[0]);
|
---|
158 | exit (1);
|
---|
159 | }
|
---|
160 |
|
---|
161 |
|
---|
162 |
|
---|
163 | GetTime (&StartTime);
|
---|
164 |
|
---|
165 | if (LoadCompressionDictionary (comp_dict) == COMPERROR)
|
---|
166 | exit (1);
|
---|
167 |
|
---|
168 | if (mode == 'H')
|
---|
169 | bytes = ReadInWordsStandard (stats_dict, &num_bytes);
|
---|
170 | else
|
---|
171 | bytes = ReadInWordsSpecial (stats_dict, &num_bytes, &num_extra_bytes,
|
---|
172 | &aux_compressed);
|
---|
173 |
|
---|
174 |
|
---|
175 | Message ("Compressed Text %.2f Mb ( %.0f bytes %0.3f %%)",
|
---|
176 | bytes / 1024.0 / 1024, bytes, bytes * 100 / num_bytes);
|
---|
177 | Message ("Words portion of the dictionary %.2f Mb ( %.0f bytes %0.3f %%)",
|
---|
178 | Words_disk / 1024.0 / 1024, Words_disk * 1.0,
|
---|
179 | (Words_disk * 100.0) / num_bytes);
|
---|
180 | if (mode == 'H')
|
---|
181 | Message ("Huffman info for chars in dictionary %.2f Mb ( %.0f bytes %0.3f %%)",
|
---|
182 | Chars_disk / 1024.0 / 1024, Chars_disk * 1.0,
|
---|
183 | (Chars_disk * 100.0) / num_bytes);
|
---|
184 | else
|
---|
185 | {
|
---|
186 | Message ("Aux dictionary (Uncompressed) %.2f Mb ( %.0f bytes %0.3f %%)",
|
---|
187 | num_extra_bytes / 1024.0 / 1024, num_extra_bytes * 1.0,
|
---|
188 | (num_extra_bytes * 100.0) / num_bytes);
|
---|
189 | Message ("Aux dictionary (Compressed) %.2f Mb ( %.0f bytes %0.3f %%)",
|
---|
190 | aux_compressed / 1024.0 / 1024, aux_compressed * 1.0,
|
---|
191 | (aux_compressed * 100.0) / num_bytes);
|
---|
192 | }
|
---|
193 | Message ("Words and non-words in the text %u", huff_count + escape_count);
|
---|
194 | Message ("Words and non-words coded by escapes %u ( %0.3f %%)", escape_count,
|
---|
195 | escape_count * 100.0 / huff_count);
|
---|
196 | Message ("%s", ElapsedTime (&StartTime, NULL));
|
---|
197 | return 0;
|
---|
198 | }
|
---|
199 |
|
---|
200 |
|
---|
201 |
|
---|
202 |
|
---|
203 |
|
---|
204 | static double
|
---|
205 | ReadInWordsStandard (char *name, double *num_bytes)
|
---|
206 | {
|
---|
207 | compression_stats_header csh;
|
---|
208 | FILE *f;
|
---|
209 | mg_u_long which;
|
---|
210 | double bytes, bits = 0;
|
---|
211 |
|
---|
212 | f = open_named_file (name, "rb", MAGIC_STATS_DICT, MG_ABORT); /* [RPAP - Feb 97: WIN32 Port] */
|
---|
213 |
|
---|
214 | fread (&csh, sizeof (csh), 1, f);
|
---|
215 |
|
---|
216 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
217 | NTOHUL(csh.num_docs);
|
---|
218 | NTOHD(csh.num_bytes); /* [RJM 07/97: 4G limit] */
|
---|
219 |
|
---|
220 | for (which = 0; which < 2; which++)
|
---|
221 | {
|
---|
222 | frags_stats_header fsh;
|
---|
223 | int j;
|
---|
224 | fread (&fsh, sizeof (fsh), 1, f);
|
---|
225 |
|
---|
226 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
227 | NTOHUL(fsh.num_frags);
|
---|
228 | NTOHUL(fsh.mem_for_frags);
|
---|
229 |
|
---|
230 | for (j = 0; j < fsh.num_frags; j++)
|
---|
231 | {
|
---|
232 | int len, freq, res, occur_num;
|
---|
233 | u_char Word[16];
|
---|
234 | fread (&freq, sizeof (freq), 1, f);
|
---|
235 | fread (&occur_num, sizeof (occur_num), 1, f);
|
---|
236 | len = fgetc (f);
|
---|
237 |
|
---|
238 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
239 | NTOHSI(freq);
|
---|
240 | NTOHSI(occur_num);
|
---|
241 |
|
---|
242 | Word[0] = len;
|
---|
243 | fread (Word + 1, len, 1, f);
|
---|
244 |
|
---|
245 | /* Search the hash table for Word */
|
---|
246 | if (ht[which])
|
---|
247 | {
|
---|
248 | register mg_u_long hashval, step;
|
---|
249 | register int tsize = ht[which]->size;
|
---|
250 | register u_char **wptr;
|
---|
251 | HASH (hashval, step, Word, tsize);
|
---|
252 | for (;;)
|
---|
253 | {
|
---|
254 | register u_char *s1;
|
---|
255 | register u_char *s2;
|
---|
256 | register int len;
|
---|
257 | wptr = ht[which]->table[hashval];
|
---|
258 | if (wptr == NULL)
|
---|
259 | {
|
---|
260 | res = COMPERROR;
|
---|
261 | break;
|
---|
262 | }
|
---|
263 |
|
---|
264 | /* Compare the words */
|
---|
265 | s1 = Word;
|
---|
266 | s2 = *wptr;
|
---|
267 | len = *s1 + 1;
|
---|
268 | for (; len; len--)
|
---|
269 | if (*s1++ != *s2++)
|
---|
270 | break;
|
---|
271 |
|
---|
272 | if (len)
|
---|
273 | {
|
---|
274 | hashval += step;
|
---|
275 | if (hashval >= tsize)
|
---|
276 | hashval -= tsize;
|
---|
277 | }
|
---|
278 | else
|
---|
279 | {
|
---|
280 | res = ht[which]->table[hashval] - ht[which]->words;
|
---|
281 | break;
|
---|
282 | }
|
---|
283 | }
|
---|
284 | }
|
---|
285 | else
|
---|
286 | res = COMPERROR;
|
---|
287 |
|
---|
288 | /* Check that the word was found in the dictionary */
|
---|
289 | if (res == COMPERROR)
|
---|
290 | {
|
---|
291 | escape_count += freq;
|
---|
292 | if (cdh.dict_type == MG_COMPLETE_DICTIONARY)
|
---|
293 | Message ("Unknown word \"%.*s\"\n", *Word, Word + 1);
|
---|
294 | if (cdh.dict_type == MG_PARTIAL_DICTIONARY)
|
---|
295 | {
|
---|
296 | unsigned i;
|
---|
297 | if (ht[which])
|
---|
298 | {
|
---|
299 | res = ht[which]->hd->num_codes - 1;
|
---|
300 | bits += ht[which]->hd->clens[res] * freq;
|
---|
301 | }
|
---|
302 | bits += lens_huff[which].clens[Word[0]] * freq;
|
---|
303 | for (i = 0; i < Word[0]; i++)
|
---|
304 | bits += char_huff[which].clens[Word[i + 1]] * freq;
|
---|
305 | }
|
---|
306 | if (cdh.dict_type == MG_SEED_DICTIONARY)
|
---|
307 | {
|
---|
308 | unsigned i;
|
---|
309 | if (ht[which])
|
---|
310 | {
|
---|
311 | res = ht[which]->hd->num_codes - 1;
|
---|
312 | bits += ht[which]->hd->clens[res] * freq;
|
---|
313 | }
|
---|
314 | bits += lens_huff[which].clens[Word[0]] * freq;
|
---|
315 | for (i = 0; i < Word[0]; i++)
|
---|
316 | bits += char_huff[which].clens[Word[i + 1]] * freq;
|
---|
317 | }
|
---|
318 | }
|
---|
319 | else
|
---|
320 | {
|
---|
321 | bits += ht[which]->hd->clens[res] * freq;
|
---|
322 | huff_count += freq;
|
---|
323 | }
|
---|
324 | }
|
---|
325 | }
|
---|
326 | fclose (f);
|
---|
327 |
|
---|
328 |
|
---|
329 | /*
|
---|
330 | * The 5.5 bits added on per document consist of
|
---|
331 | * 1 bit for the word/non-word flag at the start of the document.
|
---|
332 | * 1 bit on average for document termination.
|
---|
333 | * 3.5 bits on average for padding to a byte boundary.
|
---|
334 | */
|
---|
335 | bits += 5.5 * csh.num_docs;
|
---|
336 |
|
---|
337 | bytes = bits / 8 + sizeof (compressed_text_header) + sizeof (mg_u_long);
|
---|
338 |
|
---|
339 | *num_bytes = csh.num_bytes;
|
---|
340 |
|
---|
341 | return bytes;
|
---|
342 | }
|
---|
343 |
|
---|
344 |
|
---|
345 | static int
|
---|
346 | stats_comp (const void *a, const void *b)
|
---|
347 | {
|
---|
348 | return ((stats_t *) a)->occur - ((stats_t *) b)->occur;
|
---|
349 | }
|
---|
350 |
|
---|
351 |
|
---|
352 |
|
---|
353 |
|
---|
354 |
|
---|
355 | static double
|
---|
356 | ReadInWordsSpecial (char *name, double *num_bytes,
|
---|
357 | double *num_extra_bytes,
|
---|
358 | mg_u_long *aux_compressed)
|
---|
359 | {
|
---|
360 | compression_stats_header csh;
|
---|
361 | FILE *f;
|
---|
362 | mg_u_long magic, which;
|
---|
363 | double bytes, bits = 0;
|
---|
364 | double size[2];
|
---|
365 |
|
---|
366 | if (!(f = fopen (name, "rb"))) /* [RPAP - Feb 97: WIN32 Port] */
|
---|
367 | FatalError (1, "Unable to open file \"%s\"\n", name);
|
---|
368 |
|
---|
369 | if (fread (&magic, sizeof (magic), 1, f) != 1)
|
---|
370 | FatalError (1, "Unable to read magic number");
|
---|
371 |
|
---|
372 | NTOHUL(magic); /* [RPAP - Jan 97: Endian Ordering] */
|
---|
373 |
|
---|
374 | if (magic != MAGIC_STATS_DICT)
|
---|
375 | FatalError (1, "Bad magic number");
|
---|
376 |
|
---|
377 | fread (&csh, sizeof (csh), 1, f);
|
---|
378 |
|
---|
379 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
380 | NTOHUL(csh.num_docs);
|
---|
381 | NTOHD(csh.num_bytes); /* [RJM 07/97: 4G limit] */
|
---|
382 |
|
---|
383 | for (which = 0; which < 2; which++)
|
---|
384 | {
|
---|
385 | stats_t *stats;
|
---|
386 | frags_stats_header fsh;
|
---|
387 | int j, num = 0;
|
---|
388 | mg_s_long lens[16], chars[256];
|
---|
389 |
|
---|
390 | bzero ((char *) lens, sizeof (lens));
|
---|
391 | bzero ((char *) chars, sizeof (chars));
|
---|
392 |
|
---|
393 | fread (&fsh, sizeof (fsh), 1, f);
|
---|
394 |
|
---|
395 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
396 | NTOHUL(fsh.num_frags);
|
---|
397 | NTOHUL(fsh.mem_for_frags);
|
---|
398 |
|
---|
399 | stats = Xmalloc (fsh.num_frags * sizeof (*stats));
|
---|
400 | if (!stats)
|
---|
401 | FatalError (1, "Unable to allocate memory for stats");
|
---|
402 | for (j = 0; j < fsh.num_frags; j++)
|
---|
403 | {
|
---|
404 | int len, res;
|
---|
405 | u_char Word[16];
|
---|
406 | fread (&stats[num].freq, sizeof (stats[num].freq), 1, f);
|
---|
407 | fread (&stats[num].occur, sizeof (stats[num].occur), 1, f);
|
---|
408 | len = fgetc (f);
|
---|
409 |
|
---|
410 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
411 | NTOHSI(stats[num].freq);
|
---|
412 | NTOHSI(stats[num].occur);
|
---|
413 |
|
---|
414 | stats[num].len = len + 1;
|
---|
415 |
|
---|
416 | Word[0] = len;
|
---|
417 | fread (Word + 1, len, 1, f);
|
---|
418 |
|
---|
419 | /* Search the hash table for Word */
|
---|
420 | if (ht[which])
|
---|
421 | {
|
---|
422 | register mg_u_long hashval, step;
|
---|
423 | register int tsize = ht[which]->size;
|
---|
424 | register u_char **wptr;
|
---|
425 | HASH (hashval, step, Word, tsize);
|
---|
426 | for (;;)
|
---|
427 | {
|
---|
428 | register u_char *s1;
|
---|
429 | register u_char *s2;
|
---|
430 | register int len;
|
---|
431 | wptr = ht[which]->table[hashval];
|
---|
432 | if (wptr == NULL)
|
---|
433 | {
|
---|
434 | res = COMPERROR;
|
---|
435 | break;
|
---|
436 | }
|
---|
437 |
|
---|
438 | /* Compare the words */
|
---|
439 | s1 = Word;
|
---|
440 | s2 = *wptr;
|
---|
441 | len = *s1 + 1;
|
---|
442 | for (; len; len--)
|
---|
443 | if (*s1++ != *s2++)
|
---|
444 | break;
|
---|
445 |
|
---|
446 | if (len)
|
---|
447 | {
|
---|
448 | hashval += step;
|
---|
449 | if (hashval >= tsize)
|
---|
450 | hashval -= tsize;
|
---|
451 | }
|
---|
452 | else
|
---|
453 | {
|
---|
454 | res = ht[which]->table[hashval] - ht[which]->words;
|
---|
455 | break;
|
---|
456 | }
|
---|
457 | }
|
---|
458 | }
|
---|
459 | else
|
---|
460 | res = COMPERROR;
|
---|
461 |
|
---|
462 | assert (stats[num].freq > 0);
|
---|
463 |
|
---|
464 | /* Check that the word was found in the dictionary */
|
---|
465 | if (res == COMPERROR)
|
---|
466 | {
|
---|
467 | if (cdh.dict_type == MG_COMPLETE_DICTIONARY)
|
---|
468 | Message ("Unknown word \"%.*s\"\n", *Word, Word + 1);
|
---|
469 | if ((cdh.dict_type == MG_PARTIAL_DICTIONARY ||
|
---|
470 | cdh.dict_type == MG_SEED_DICTIONARY) &&
|
---|
471 | ht[which])
|
---|
472 | {
|
---|
473 | int k;
|
---|
474 | res = ht[which]->hd->num_codes - 1;
|
---|
475 | bits += ht[which]->hd->clens[res] * stats[num].freq;
|
---|
476 | lens[Word[0]]++;
|
---|
477 | for (k = 0; k < (int) Word[0]; k++)
|
---|
478 | chars[Word[k + 1]]++;
|
---|
479 | }
|
---|
480 | num++;
|
---|
481 | }
|
---|
482 | else
|
---|
483 | {
|
---|
484 | bits += ht[which]->hd->clens[res] * stats[num].freq;
|
---|
485 | }
|
---|
486 | }
|
---|
487 |
|
---|
488 | {
|
---|
489 | int j, num_buckets = 0;
|
---|
490 | int start[10], end[10];
|
---|
491 | mg_u_long flens[16], fchars[256];
|
---|
492 |
|
---|
493 | for (j = 0; j < 256; j++)
|
---|
494 | if (!chars[j] && PESINAWORD (j) == which)
|
---|
495 | fchars[j] = 1;
|
---|
496 | else
|
---|
497 | fchars[j] = chars[j];
|
---|
498 | for (j = 0; j < 16; j++)
|
---|
499 | if (!lens[j])
|
---|
500 | flens[j] = 1;
|
---|
501 | else
|
---|
502 | flens[j] = lens[j];
|
---|
503 |
|
---|
504 | size[which] = Calculate_Huffman_Size (16, flens, lens) +
|
---|
505 | Calculate_Huffman_Size (256, fchars, chars);
|
---|
506 |
|
---|
507 |
|
---|
508 | qsort (stats, num, sizeof (*stats), stats_comp);
|
---|
509 |
|
---|
510 |
|
---|
511 | if (mode == 'Y')
|
---|
512 | {
|
---|
513 | int k;
|
---|
514 | num_buckets = 1;
|
---|
515 | start[0] = 0;
|
---|
516 | end[0] = cdh.num_words[which] - 1;
|
---|
517 | Message ("There are %d novel %swords", num, which ? "" : "non-");
|
---|
518 | while (end[num_buckets - 1] < num)
|
---|
519 | {
|
---|
520 | start[num_buckets] = end[num_buckets - 1] + 1;
|
---|
521 | end[num_buckets] = start[num_buckets] +
|
---|
522 | (end[num_buckets - 1] - start[num_buckets - 1]) * 2;
|
---|
523 | num_buckets++;
|
---|
524 | }
|
---|
525 | for (k = 0; k < num_buckets; k++)
|
---|
526 | Message ("Bucket %d %d <= x <= %d\n", k + 1, start[k], end[k]);
|
---|
527 | }
|
---|
528 |
|
---|
529 | for (j = 0; j < num; j++)
|
---|
530 | {
|
---|
531 | (*num_extra_bytes) += stats[j].len;
|
---|
532 | switch (mode)
|
---|
533 | {
|
---|
534 | case 'D':
|
---|
535 | bits += BIO_Delta_Length (j + 1) * stats[j].freq;
|
---|
536 | break;
|
---|
537 | case 'B':
|
---|
538 | bits += BIO_Binary_Length (j + 1, num) * stats[j].freq;
|
---|
539 | break;
|
---|
540 | case 'Y':
|
---|
541 | {
|
---|
542 | int k = 0;
|
---|
543 | while (j > end[k])
|
---|
544 | k++;
|
---|
545 | if (!(j - start[k] + 1 >= 1 &&
|
---|
546 | j - start[k] + 1 <= end[k] - start[k] + 1))
|
---|
547 | {
|
---|
548 | Message ("k = %d, j = %d", k, j);
|
---|
549 | assert (j - start[k] + 1 >= 1 &&
|
---|
550 | j - start[k] + 1 <= end[k] - start[k] + 1);
|
---|
551 | }
|
---|
552 | else
|
---|
553 | {
|
---|
554 | bits += (BIO_Gamma_Length (k + 1) +
|
---|
555 | BIO_Binary_Length (j - start[k] + 1,
|
---|
556 | end[k] - start[k] + 1)) *
|
---|
557 | stats[j].freq;
|
---|
558 | }
|
---|
559 | }
|
---|
560 | }
|
---|
561 | }
|
---|
562 | if (mode == 'B' && num)
|
---|
563 | bits += BIO_Delta_Length (num) * csh.num_docs;
|
---|
564 | }
|
---|
565 | Xfree (stats);
|
---|
566 | }
|
---|
567 | fclose (f);
|
---|
568 |
|
---|
569 | *aux_compressed = (size[0] + size[1]) / 8;
|
---|
570 |
|
---|
571 | /*
|
---|
572 | * The 5.5 bits added on per document consist of
|
---|
573 | * 1 bit for the word/non-word flag at the start of the document.
|
---|
574 | * 1 bit on average for document termination.
|
---|
575 | * 3.5 bits on average for padding to a byte boundary.
|
---|
576 | */
|
---|
577 | bits += 5.5 * csh.num_docs;
|
---|
578 |
|
---|
579 | bytes = bits / 8 + sizeof (compressed_text_header) + sizeof (mg_u_long);
|
---|
580 |
|
---|
581 | *num_bytes = csh.num_bytes;
|
---|
582 |
|
---|
583 | return bytes;
|
---|
584 | }
|
---|