[3745] | 1 | /**************************************************************************
|
---|
| 2 | *
|
---|
| 3 | * mg_perf_hash_build.c -- Program to build a perfect hash function
|
---|
| 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_perf_hash_build.c 16583 2008-07-29 10:20:36Z davidb $
|
---|
| 21 | *
|
---|
| 22 | **************************************************************************/
|
---|
| 23 |
|
---|
| 24 | #include "sysfuncs.h"
|
---|
| 25 | #include "memlib.h"
|
---|
| 26 | #include "messages.h"
|
---|
| 27 | #include "timing.h"
|
---|
| 28 | #include "local_strings.h"
|
---|
| 29 | #include "perf_hash.h"
|
---|
| 30 | #include "netorder.h" /* [RPAP - Jan 97: Endian Ordering] */
|
---|
| 31 |
|
---|
| 32 | #include "mg_files.h"
|
---|
| 33 | #include "invf.h"
|
---|
| 34 | #include "locallib.h"
|
---|
| 35 | #include "words.h"
|
---|
| 36 | #include "mg.h"
|
---|
| 37 |
|
---|
| 38 | /*
|
---|
| 39 | $Log$
|
---|
| 40 | Revision 1.1 2003/02/20 21:18:24 mdewsnip
|
---|
| 41 | Addition of MG package for search and retrieval
|
---|
| 42 |
|
---|
| 43 | Revision 1.1 1999/08/10 21:18:13 sjboddie
|
---|
| 44 | renamed mg-1.3d directory mg
|
---|
| 45 |
|
---|
| 46 | Revision 1.2 1998/11/25 07:55:47 rjmcnab
|
---|
| 47 |
|
---|
| 48 | Modified mg to that you can specify the stemmer you want
|
---|
| 49 | to use via a command line option. You specify it to
|
---|
| 50 | mg_passes during the build process. The number of the
|
---|
| 51 | stemmer that you used is stored within the inverted
|
---|
| 52 | dictionary header and the stemmed dictionary header so
|
---|
| 53 | the correct stemmer is used in later stages of building
|
---|
| 54 | and querying.
|
---|
| 55 |
|
---|
| 56 | Revision 1.1 1998/11/17 09:35:15 rjmcnab
|
---|
| 57 | *** empty log message ***
|
---|
| 58 |
|
---|
| 59 | * Revision 1.3 1994/10/20 03:56:58 tes
|
---|
| 60 | * I have rewritten the boolean query optimiser and abstracted out the
|
---|
| 61 | * components of the boolean query.
|
---|
| 62 | *
|
---|
| 63 | * Revision 1.2 1994/09/20 04:41:53 tes
|
---|
| 64 | * For version 1.1
|
---|
| 65 | *
|
---|
| 66 | */
|
---|
| 67 |
|
---|
| 68 | static char *RCSID = "$Id: mg_perf_hash_build.c 16583 2008-07-29 10:20:36Z davidb $";
|
---|
| 69 |
|
---|
| 70 |
|
---|
| 71 |
|
---|
| 72 | #define POOL_SIZE 1024*1024
|
---|
| 73 |
|
---|
| 74 | static void process_files (char *filename);
|
---|
| 75 |
|
---|
| 76 | int r = -1;
|
---|
| 77 |
|
---|
| 78 | int main (int argc, char **argv)
|
---|
| 79 | {
|
---|
| 80 | ProgTime start;
|
---|
| 81 | char *file_name = "";
|
---|
| 82 | int ch;
|
---|
| 83 | msg_prefix = argv[0];
|
---|
| 84 | opterr = 0;
|
---|
| 85 | while ((ch = getopt (argc, argv, "f:d:r:h")) != -1)
|
---|
| 86 | switch (ch)
|
---|
| 87 | {
|
---|
| 88 | case 'f': /* input file */
|
---|
| 89 | file_name = optarg;
|
---|
| 90 | break;
|
---|
| 91 | case 'd':
|
---|
| 92 | set_basepath (optarg);
|
---|
| 93 | break;
|
---|
| 94 | case 'r':
|
---|
| 95 | r = atoi (optarg);
|
---|
| 96 | break;
|
---|
| 97 | case 'h':
|
---|
| 98 | case '?':
|
---|
| 99 | fprintf (stderr, "usage: %s [-f input_file]"
|
---|
| 100 | "[-d data directory] [-r random seed] [-h]\n", argv[0]);
|
---|
| 101 | exit (1);
|
---|
| 102 | }
|
---|
| 103 |
|
---|
| 104 | GetTime (&start);
|
---|
| 105 | process_files (file_name);
|
---|
| 106 | Message ("%s\n", ElapsedTime (&start, NULL));
|
---|
| 107 | return 0;
|
---|
| 108 | }
|
---|
| 109 |
|
---|
| 110 |
|
---|
| 111 |
|
---|
| 112 |
|
---|
| 113 |
|
---|
| 114 | static void
|
---|
| 115 | process_files (char *filename)
|
---|
| 116 | {
|
---|
| 117 | FILE *dict, *hash;
|
---|
| 118 | unsigned long i;
|
---|
| 119 | u_char prev[MAXSTEMLEN + 1];
|
---|
| 120 | struct invf_dict_header idh;
|
---|
| 121 | perf_hash_data *phd;
|
---|
| 122 | u_char *pool;
|
---|
| 123 | int pool_left;
|
---|
| 124 | u_char **starts;
|
---|
| 125 |
|
---|
| 126 |
|
---|
| 127 | dict = open_file (filename, INVF_DICT_SUFFIX, "rb",
|
---|
| 128 | MAGIC_STEM_BUILD, MG_ABORT); /* [RPAP - Feb 97: WIN32 Port] */
|
---|
| 129 |
|
---|
| 130 | fread ((char *) &idh, sizeof (idh), 1, dict);
|
---|
| 131 |
|
---|
| 132 | /* [RPAP - Jan 97: Endian Ordering] */
|
---|
| 133 | NTOHUL(idh.lookback);
|
---|
| 134 | NTOHUL(idh.dict_size);
|
---|
| 135 | NTOHUL(idh.total_bytes);
|
---|
| 136 | NTOHUL(idh.index_string_bytes);
|
---|
| 137 | NTOHD(idh.input_bytes); /* [RJM 07/97: 4G limit] */
|
---|
| 138 | NTOHUL(idh.num_of_docs);
|
---|
| 139 | NTOHUL(idh.static_num_of_docs);
|
---|
| 140 | NTOHUL(idh.num_of_words);
|
---|
| 141 | NTOHUL(idh.stemmer_num);
|
---|
| 142 | NTOHUL(idh.stem_method);
|
---|
| 143 |
|
---|
| 144 | hash = create_file (filename, INVF_DICT_HASH_SUFFIX, "wb",
|
---|
| 145 | MAGIC_HASH, MG_ABORT); /* [RPAP - Feb 97: WIN32 Port] */
|
---|
| 146 |
|
---|
| 147 | if (!(pool = Xmalloc (POOL_SIZE)))
|
---|
| 148 | FatalError (1, "Out of memory");
|
---|
| 149 | pool_left = POOL_SIZE;
|
---|
| 150 |
|
---|
| 151 | if (!(starts = Xmalloc (sizeof (u_char *) * idh.dict_size)))
|
---|
| 152 | FatalError (1, "Out of memory");
|
---|
| 153 |
|
---|
| 154 | for (i = 0; i < idh.dict_size; i++)
|
---|
| 155 | {
|
---|
| 156 | register unsigned long copy, suff, l;
|
---|
| 157 | unsigned long wcnt, fcnt;
|
---|
| 158 |
|
---|
| 159 | /* build a new word on top of prev */
|
---|
| 160 | copy = getc (dict);
|
---|
| 161 | suff = getc (dict);
|
---|
| 162 | *prev = copy + suff;
|
---|
| 163 | fread (prev + copy + 1, sizeof (u_char), suff, dict);
|
---|
| 164 |
|
---|
| 165 | /* read other data, but no need to store it */
|
---|
| 166 | fread (&fcnt, sizeof (fcnt), 1, dict);
|
---|
| 167 | fread (&wcnt, sizeof (wcnt), 1, dict);
|
---|
| 168 |
|
---|
| 169 | l = *prev + 1;
|
---|
| 170 | if (pool_left < l)
|
---|
| 171 | {
|
---|
| 172 | pool = Xmalloc (POOL_SIZE);
|
---|
| 173 | pool_left = POOL_SIZE;
|
---|
| 174 | }
|
---|
| 175 | starts[i] = pool;
|
---|
| 176 | bcopy ((char *) prev, (char *) pool, l);
|
---|
| 177 | pool += l;
|
---|
| 178 | pool_left -= l;
|
---|
| 179 | }
|
---|
| 180 | if (!(phd = gen_hash_func (idh.dict_size, starts, r)))
|
---|
| 181 | FatalError (1, "Unable to generate hash function");
|
---|
| 182 | if (write_perf_hash_data (hash, phd) == -1)
|
---|
| 183 | FatalError (1, "Unable to write hash function");
|
---|
| 184 | fclose (dict);
|
---|
| 185 | fclose (hash);
|
---|
| 186 | }
|
---|