| 1 |
/* |
| 2 |
** Copyright (C) 1995, 1996, 1997, 1998 Hewlett-Packard Company |
| 3 |
** Originally by Kevin Hughes, kev@kevcom.com, 3/11/94 |
| 4 |
** |
| 5 |
** This program and library is free software; you can redistribute it and/or |
| 6 |
** modify it under the terms of the GNU (Library) General Public License |
| 7 |
** as published by the Free Software Foundation; either version 2 |
| 8 |
** of the License, or any later version. |
| 9 |
** |
| 10 |
** This program is distributed in the hope that it will be useful, |
| 11 |
** but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 |
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 13 |
** GNU (Library) General Public License for more details. |
| 14 |
** |
| 15 |
** You should have received a copy of the GNU (Library) General Public License |
| 16 |
** along with this program; if not, write to the Free Software |
| 17 |
** Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. |
| 18 |
**--------------------------------------------------------- |
| 19 |
** Added addStopList to support printing of common words |
| 20 |
** G. Hill 4/7/97 ghill@library.berkeley.edu |
| 21 |
** |
| 22 |
** change sprintf to snprintf to avoid corruption |
| 23 |
** SRE 11/17/99 |
| 24 |
** |
| 25 |
** 04/00 - Jose Ruiz |
| 26 |
** change hash for bighash in mergeresultlists for better performance |
| 27 |
** when big searchs (a* or b* or c*) |
| 28 |
** |
| 29 |
*/ |
| 30 |
|
| 31 |
#include "swish.h" |
| 32 |
#include "string.h" |
| 33 |
#include "hash.h" |
| 34 |
#include "mem.h" |
| 35 |
#include "search.h" |
| 36 |
|
| 37 |
/* Hashes a string. |
| 38 |
*/ |
| 39 |
|
| 40 |
unsigned hash(s) |
| 41 |
char *s; |
| 42 |
{ |
| 43 |
unsigned hashval; |
| 44 |
|
| 45 |
for (hashval = 0; *s != '\0'; s++) |
| 46 |
hashval = (int) ((unsigned char) *s) + 31 * hashval; |
| 47 |
return hashval % HASHSIZE; |
| 48 |
} |
| 49 |
|
| 50 |
/* Hashes a string for a larger hash table. |
| 51 |
*/ |
| 52 |
|
| 53 |
unsigned bighash(s) |
| 54 |
char *s; |
| 55 |
{ |
| 56 |
unsigned hashval; |
| 57 |
|
| 58 |
for (hashval = 0; *s != '\0'; s++) |
| 59 |
hashval = (int) ((unsigned char) *s) + 31 * hashval; |
| 60 |
return hashval % BIGHASHSIZE; |
| 61 |
} |
| 62 |
|
| 63 |
/* Hashes a int. |
| 64 |
*/ |
| 65 |
|
| 66 |
unsigned numhash(i) |
| 67 |
int i; |
| 68 |
{ |
| 69 |
return i % HASHSIZE; |
| 70 |
} |
| 71 |
|
| 72 |
/* Hashes a int for a larger hash table. |
| 73 |
*/ |
| 74 |
|
| 75 |
unsigned bignumhash(i) |
| 76 |
int i; |
| 77 |
{ |
| 78 |
return i % BIGHASHSIZE; |
| 79 |
} |
| 80 |
|
| 81 |
/* Hashes a string for a larger hash table (for search). |
| 82 |
*/ |
| 83 |
|
| 84 |
unsigned verybighash(s) |
| 85 |
char *s; |
| 86 |
{ |
| 87 |
unsigned hashval; |
| 88 |
|
| 89 |
for (hashval = 0; *s != '\0'; s++) |
| 90 |
hashval = (int) ((unsigned char) *s) + 31 * hashval; |
| 91 |
return hashval % VERYBIGHASHSIZE; |
| 92 |
} |
| 93 |
|
| 94 |
|
| 95 |
/* Adds a stop word to the list of removed common words */ |
| 96 |
void addStopList(INDEXDATAHEADER *header, char *word) |
| 97 |
{ |
| 98 |
char *arrayWord; |
| 99 |
|
| 100 |
if (isstopword(header, word)) |
| 101 |
return; |
| 102 |
|
| 103 |
/* Another BUG!! Jose Ruiz 04/00 |
| 104 |
The dimension of the array was not checked |
| 105 |
Fixed */ |
| 106 |
if (header->stopPos == header->stopMaxSize) |
| 107 |
{ |
| 108 |
header->stopMaxSize += 100; |
| 109 |
if (!header->stopList) |
| 110 |
header->stopList = (char **) emalloc(header->stopMaxSize * sizeof(char *)); |
| 111 |
|
| 112 |
else |
| 113 |
header->stopList = (char **) erealloc(header->stopList, header->stopMaxSize * sizeof(char *)); |
| 114 |
} |
| 115 |
arrayWord = (char *) estrdup(word); |
| 116 |
header->stopList[header->stopPos++] = arrayWord; |
| 117 |
} |
| 118 |
|
| 119 |
|
| 120 |
/* Adds a stop word to a hash table. |
| 121 |
*/ |
| 122 |
|
| 123 |
void addstophash(INDEXDATAHEADER *header, char *word) |
| 124 |
{ |
| 125 |
unsigned hashval; |
| 126 |
struct swline *sp; |
| 127 |
|
| 128 |
if (isstopword(header, word)) |
| 129 |
return; |
| 130 |
|
| 131 |
sp = (struct swline *) emalloc(sizeof(struct swline)); |
| 132 |
|
| 133 |
sp->line = (char *) estrdup(word); |
| 134 |
|
| 135 |
hashval = hash(word); |
| 136 |
sp->next = header->hashstoplist[hashval]; |
| 137 |
header->hashstoplist[hashval] = sp; |
| 138 |
} |
| 139 |
|
| 140 |
/* Sees if a word is a stop word by looking it up in the hash table. |
| 141 |
*/ |
| 142 |
|
| 143 |
int isstopword(INDEXDATAHEADER *header, char *word) |
| 144 |
{ |
| 145 |
unsigned hashval; |
| 146 |
struct swline *sp; |
| 147 |
|
| 148 |
hashval = hash(word); |
| 149 |
sp = header->hashstoplist[hashval]; |
| 150 |
|
| 151 |
while (sp != NULL) |
| 152 |
{ |
| 153 |
if (!strcmp(sp->line, word)) |
| 154 |
return 1; |
| 155 |
sp = sp->next; |
| 156 |
} |
| 157 |
return 0; |
| 158 |
} |
| 159 |
|
| 160 |
|
| 161 |
|
| 162 |
/* Adds a buzzword to a hash table.*/ |
| 163 |
|
| 164 |
void addbuzzwordhash(INDEXDATAHEADER *header, char *word) |
| 165 |
{ |
| 166 |
unsigned hashval; |
| 167 |
struct swline *sp; |
| 168 |
|
| 169 |
if (isbuzzword(header, word)) |
| 170 |
return; |
| 171 |
|
| 172 |
header->buzzwords_used_flag++; |
| 173 |
|
| 174 |
sp = (struct swline *) emalloc(sizeof(struct swline)); |
| 175 |
|
| 176 |
sp->line = (char *) estrdup(word); |
| 177 |
|
| 178 |
|
| 179 |
/* should buzzwords be case sensitive? */ |
| 180 |
strtolower( sp->line ); |
| 181 |
|
| 182 |
hashval = hash( sp->line ); |
| 183 |
|
| 184 |
|
| 185 |
sp->next = header->hashbuzzwordlist[hashval]; |
| 186 |
header->hashbuzzwordlist[hashval] = sp; |
| 187 |
} |
| 188 |
|
| 189 |
void freebuzzwordhash(INDEXDATAHEADER *header) |
| 190 |
{ |
| 191 |
int i; |
| 192 |
struct swline *sp, |
| 193 |
*tmp; |
| 194 |
|
| 195 |
for (i = 0; i < HASHSIZE; i++) |
| 196 |
if (header->hashbuzzwordlist[i]) |
| 197 |
{ |
| 198 |
sp = (struct swline *) header->hashbuzzwordlist[i]; |
| 199 |
while (sp) |
| 200 |
{ |
| 201 |
tmp = sp->next; |
| 202 |
efree(sp->line); |
| 203 |
efree(sp); |
| 204 |
sp = tmp; |
| 205 |
} |
| 206 |
header->hashbuzzwordlist[i] = NULL; |
| 207 |
} |
| 208 |
} |
| 209 |
|
| 210 |
|
| 211 |
/* Sees if a word is a buzzword by looking it up in the hash table. */ |
| 212 |
|
| 213 |
int isbuzzword(INDEXDATAHEADER *header, char *word) |
| 214 |
{ |
| 215 |
unsigned hashval; |
| 216 |
struct swline *sp; |
| 217 |
|
| 218 |
hashval = hash(word); |
| 219 |
sp = header->hashbuzzwordlist[hashval]; |
| 220 |
|
| 221 |
while (sp != NULL) |
| 222 |
{ |
| 223 |
if (!strcmp(sp->line, word)) |
| 224 |
return 1; |
| 225 |
sp = sp->next; |
| 226 |
} |
| 227 |
return 0; |
| 228 |
} |
| 229 |
|
| 230 |
|
| 231 |
|
| 232 |
void freestophash(INDEXDATAHEADER *header) |
| 233 |
{ |
| 234 |
int i; |
| 235 |
struct swline *sp, |
| 236 |
*tmp; |
| 237 |
|
| 238 |
for (i = 0; i < HASHSIZE; i++) |
| 239 |
if (header->hashstoplist[i]) |
| 240 |
{ |
| 241 |
sp = (struct swline *) header->hashstoplist[i]; |
| 242 |
while (sp) |
| 243 |
{ |
| 244 |
tmp = sp->next; |
| 245 |
efree(sp->line); |
| 246 |
efree(sp); |
| 247 |
sp = tmp; |
| 248 |
} |
| 249 |
header->hashstoplist[i] = NULL; |
| 250 |
} |
| 251 |
} |
| 252 |
|
| 253 |
void freeStopList(INDEXDATAHEADER *header) |
| 254 |
{ |
| 255 |
int i; |
| 256 |
|
| 257 |
for (i = 0; i < header->stopPos; i++) |
| 258 |
efree(header->stopList[i]); |
| 259 |
if (header->stopList) |
| 260 |
efree(header->stopList); |
| 261 |
header->stopList = NULL; |
| 262 |
header->stopPos = header->stopMaxSize = 0; |
| 263 |
} |
| 264 |
|
| 265 |
/* Adds a "use" word to a hash table. |
| 266 |
*/ |
| 267 |
|
| 268 |
void addusehash(INDEXDATAHEADER *header, char *word) |
| 269 |
{ |
| 270 |
unsigned hashval; |
| 271 |
struct swline *sp; |
| 272 |
|
| 273 |
if (isuseword(header, word)) |
| 274 |
return; |
| 275 |
|
| 276 |
sp = (struct swline *) emalloc(sizeof(struct swline)); |
| 277 |
|
| 278 |
sp->line = (char *) estrdup(word); |
| 279 |
|
| 280 |
hashval = hash(word); |
| 281 |
sp->next = header->hashuselist[hashval]; |
| 282 |
header->hashuselist[hashval] = sp; |
| 283 |
} |
| 284 |
|
| 285 |
/* Sees if a word is a "use" word by looking it up in the hash table. |
| 286 |
*/ |
| 287 |
|
| 288 |
int isuseword(INDEXDATAHEADER *header, char *word) |
| 289 |
{ |
| 290 |
unsigned hashval; |
| 291 |
struct swline *sp; |
| 292 |
|
| 293 |
hashval = hash(word); |
| 294 |
sp = header->hashuselist[hashval]; |
| 295 |
|
| 296 |
while (sp != NULL) |
| 297 |
{ |
| 298 |
if (!strcmp(sp->line, word)) |
| 299 |
return 1; |
| 300 |
sp = sp->next; |
| 301 |
} |
| 302 |
return 0; |
| 303 |
} |
Parent Directory
| 
Revision Log
| 
Revision Graph