1 | /**************************************************************************
|
---|
2 | *
|
---|
3 | * mgpp_perf_hash_build.cpp -- Program to build a perfect hash function
|
---|
4 | * Copyright (C) 1999 Rodger McNab
|
---|
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 | **************************************************************************/
|
---|
21 |
|
---|
22 | #define _XOPEN_SOURCE 1
|
---|
23 | #define _XOPEN_SOURCE_EXTENDED 1
|
---|
24 |
|
---|
25 | // need this to avoid bizarre compiler problems under VC++ 6.0
|
---|
26 | #if defined (__WIN32__) && !defined (GSDL_USE_IOS_H)
|
---|
27 | # include <iostream>
|
---|
28 | #endif
|
---|
29 |
|
---|
30 | /* getopt is in posix.2, so cygwin should have it in unistd, but doesn't */
|
---|
31 | #if defined (__WIN32__) || defined (__CYGWIN__)
|
---|
32 | # include "getopt_old.h"
|
---|
33 | #else
|
---|
34 | # include <unistd.h>
|
---|
35 | #endif
|
---|
36 |
|
---|
37 | #include "sysfuncs.h"
|
---|
38 | #include "memlib.h"
|
---|
39 | #include "messages.h"
|
---|
40 | #include "local_strings.h"
|
---|
41 | #include "perf_hash.h"
|
---|
42 | #include "netorder.h"
|
---|
43 |
|
---|
44 | #include "mg_files.h"
|
---|
45 | #include "invf.h"
|
---|
46 | #include "locallib.h"
|
---|
47 | #include "words.h"
|
---|
48 | #include "mg.h"
|
---|
49 |
|
---|
50 | #define POOL_SIZE 1024*1024
|
---|
51 |
|
---|
52 | static void ProcessFiles (char *filename, int r) {
|
---|
53 | FILE *dictFile, *hashFile;
|
---|
54 | unsigned long i;
|
---|
55 | invf_dict_header idh;
|
---|
56 | perf_hash_data *phd;
|
---|
57 | u_char *pool;
|
---|
58 | unsigned long pool_left;
|
---|
59 | u_char **starts;
|
---|
60 |
|
---|
61 | // read in the dictionary
|
---|
62 | dictFile = open_file (filename, INVF_DICT_SUFFIX, "rb",
|
---|
63 | MAGIC_STEM_BUILD, MG_ABORT);
|
---|
64 | if (dictFile==NULL) {
|
---|
65 | FatalError(1, "unable to open file");
|
---|
66 | }
|
---|
67 | idh.Read (dictFile);
|
---|
68 |
|
---|
69 | //cerr << idh.lookback<<" "<<idh.word_dict_start<<endl;
|
---|
70 | // go to the start of the word dictionary
|
---|
71 | fseek (dictFile, idh.word_dict_start, SEEK_SET);
|
---|
72 |
|
---|
73 | if (!(pool = (u_char *) Xmalloc (POOL_SIZE)))
|
---|
74 | FatalError (1, "Out of memory");
|
---|
75 | pool_left = POOL_SIZE;
|
---|
76 |
|
---|
77 | if (!(starts = (u_char **) Xmalloc (sizeof (u_char *) * idh.word_dict_size)))
|
---|
78 | FatalError (1, "Out of memory");
|
---|
79 | //cerr << "size= "<< idh.word_dict_size<<endl;
|
---|
80 | word_dict_el wordEl;
|
---|
81 | wordEl.SetNumLevels (idh.num_levels);
|
---|
82 | for (i = 0; i < idh.word_dict_size; ++i) {
|
---|
83 | // read the next word and associated information
|
---|
84 | wordEl.Read (dictFile, idh.num_levels);
|
---|
85 |
|
---|
86 | // push string onto pool data
|
---|
87 | register unsigned long l;
|
---|
88 | l = wordEl.el.size() + 1;
|
---|
89 | if (pool_left < l) {
|
---|
90 | pool = (u_char *) Xmalloc (POOL_SIZE);
|
---|
91 | pool_left = POOL_SIZE;
|
---|
92 | }
|
---|
93 | starts[i] = pool;
|
---|
94 |
|
---|
95 | *pool++ = wordEl.el.size();
|
---|
96 | memcpy ((char *) pool, &(wordEl.el[0]), wordEl.el.size());
|
---|
97 | //cerr << pool<<" " <<starts[i]<<endl;
|
---|
98 | pool += wordEl.el.size();
|
---|
99 | pool_left -= l;
|
---|
100 |
|
---|
101 | }
|
---|
102 | fclose (dictFile);
|
---|
103 | //cerr << pool<<" " <<starts[i-1]<<endl;
|
---|
104 | //cerr<<"starts "<<starts[113529]<<endl;
|
---|
105 | //cerr << starts[17][1] << " "<<starts[25][4]<<endl;
|
---|
106 | // create perfect hash file
|
---|
107 | hashFile = create_file (filename, INVF_DICT_HASH_SUFFIX, "wb",
|
---|
108 | MAGIC_HASH, MG_ABORT);
|
---|
109 | if (!(phd = gen_hash_func (idh.word_dict_size, starts, r)))
|
---|
110 | FatalError (1, "Unable to generate hash function");
|
---|
111 | if (write_perf_hash_data (hashFile, phd) == -1)
|
---|
112 | FatalError (1, "Unable to write hash function");
|
---|
113 | fclose (hashFile);
|
---|
114 | }
|
---|
115 |
|
---|
116 |
|
---|
117 |
|
---|
118 | int main (int argc, char **argv) {
|
---|
119 | int r = -1;
|
---|
120 | char *filename = "";
|
---|
121 | int ch;
|
---|
122 | msg_prefix = argv[0];
|
---|
123 | opterr = 0;
|
---|
124 |
|
---|
125 | while ((ch = getopt (argc, argv, "f:d:r:h")) != -1) {
|
---|
126 | switch (ch) {
|
---|
127 | case 'f': // input file
|
---|
128 | filename = optarg;
|
---|
129 | break;
|
---|
130 | case 'd':
|
---|
131 | set_basepath (optarg);
|
---|
132 | break;
|
---|
133 | case 'r':
|
---|
134 | r = atoi (optarg);
|
---|
135 | break;
|
---|
136 | case 'h':
|
---|
137 | case '?':
|
---|
138 | fprintf (stderr, "usage: %s [-f input_file]"
|
---|
139 | "[-d data directory] [-r random seed] [-h]\n", argv[0]);
|
---|
140 | exit (1);
|
---|
141 | }
|
---|
142 | }
|
---|
143 |
|
---|
144 | ProcessFiles (filename, r);
|
---|
145 | return 0;
|
---|
146 | }
|
---|