| 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 |
** fixed non-int subscripting pointed out by "gcc -Wall" |
| 20 |
** SRE 2/22/00 |
| 21 |
** |
| 22 |
** 2001-03-08 rasc rewritten and enhanced suffix routines |
| 23 |
** |
| 24 |
*/ |
| 25 |
|
| 26 |
#include "swish.h" |
| 27 |
#include "check.h" |
| 28 |
#include "hash.h" |
| 29 |
#include "string.h" |
| 30 |
#include "mem.h" |
| 31 |
|
| 32 |
/* Check if a file with a particular suffix should be indexed |
| 33 |
** according to the settings in the configuration file. |
| 34 |
*/ |
| 35 |
|
| 36 |
/* Should a word be indexed? Consults the stopword hash list |
| 37 |
** and checks if the word is of a reasonable length... |
| 38 |
** If you have any good rules that can work with most languages, |
| 39 |
** please let me know... |
| 40 |
*/ |
| 41 |
|
| 42 |
int isokword(sw, word, indexf) |
| 43 |
SWISH *sw; |
| 44 |
char *word; |
| 45 |
IndexFILE *indexf; |
| 46 |
{ |
| 47 |
int i, |
| 48 |
same, |
| 49 |
hasnumber, |
| 50 |
hasvowel, |
| 51 |
hascons, |
| 52 |
numberrow, |
| 53 |
vowelrow, |
| 54 |
consrow, |
| 55 |
wordlen; |
| 56 |
char lastchar; |
| 57 |
|
| 58 |
if (word[0] == '\0') |
| 59 |
return 0; |
| 60 |
|
| 61 |
if (isstopword(&indexf->header, word)) |
| 62 |
return 0; |
| 63 |
|
| 64 |
wordlen = strlen(word); |
| 65 |
if ((wordlen < indexf->header.minwordlimit) || (wordlen > indexf->header.maxwordlimit)) |
| 66 |
return 0; |
| 67 |
|
| 68 |
lastchar = '\0'; |
| 69 |
same = 0; |
| 70 |
hasnumber = hasvowel = hascons = 0; |
| 71 |
numberrow = vowelrow = consrow = 0; |
| 72 |
|
| 73 |
for (i = 0; word[i] != '\0'; i++) |
| 74 |
{ |
| 75 |
/* Max number of times a char can repeat in a word */ |
| 76 |
if (word[i] == lastchar) |
| 77 |
{ |
| 78 |
same++; |
| 79 |
if (same > IGNORESAME) |
| 80 |
return 0; |
| 81 |
} |
| 82 |
else |
| 83 |
same = 0; |
| 84 |
|
| 85 |
/* Max number of consecutive digits */ |
| 86 |
if (isdigit((int) ( (unsigned char) word[i]))) |
| 87 |
{ |
| 88 |
hasnumber = 1; |
| 89 |
numberrow++; |
| 90 |
if (numberrow > IGNOREROWN) |
| 91 |
return 0; |
| 92 |
vowelrow = 0; |
| 93 |
consrow = 0; |
| 94 |
} |
| 95 |
|
| 96 |
/* maximum number of consecutive vowels a word can have */ |
| 97 |
else if (isvowel(sw, word[i])) |
| 98 |
{ |
| 99 |
hasvowel = 1; |
| 100 |
vowelrow++; |
| 101 |
if (vowelrow > IGNOREROWV) |
| 102 |
return 0; |
| 103 |
numberrow = 0; |
| 104 |
consrow = 0; |
| 105 |
} |
| 106 |
|
| 107 |
/* maximum number of consecutive consonants a word can have */ |
| 108 |
else if (!ispunct((int) ( (unsigned char) word[i]))) |
| 109 |
{ |
| 110 |
hascons = 1; |
| 111 |
consrow++; |
| 112 |
if (consrow > IGNOREROWC) |
| 113 |
return 0; |
| 114 |
numberrow = 0; |
| 115 |
vowelrow = 0; |
| 116 |
} |
| 117 |
lastchar = word[i]; |
| 118 |
} |
| 119 |
|
| 120 |
/* If IGNOREALLV is 1, words containing all vowels won't be indexed. */ |
| 121 |
if (IGNOREALLV) |
| 122 |
if (hasvowel && !hascons) |
| 123 |
return 0; |
| 124 |
|
| 125 |
/* If IGNOREALLC is 1, words containing all consonants won't be indexed */ |
| 126 |
if (IGNOREALLC) |
| 127 |
if (hascons && !hasvowel) |
| 128 |
return 0; |
| 129 |
|
| 130 |
/* If IGNOREALLN is 1, words containing all digits won't be indexed */ |
| 131 |
if (IGNOREALLN) |
| 132 |
if (hasnumber && !hasvowel && !hascons) |
| 133 |
return 0; |
| 134 |
|
| 135 |
return 1; |
| 136 |
} |
| 137 |
|
| 138 |
|
| 139 |
/* |
| 140 |
-- Determine document type by checking the file extension |
| 141 |
-- of the filename |
| 142 |
-- Return: doctype |
| 143 |
-- 2001-03-08 rasc rewritten (optimize and match also |
| 144 |
-- e.g. ".htm", ".htm.de" or ".html.gz") |
| 145 |
*/ |
| 146 |
|
| 147 |
int getdoctype(char *filename, struct IndexContents *indexcontents) |
| 148 |
{ |
| 149 |
struct swline *swl; |
| 150 |
char *s, |
| 151 |
*fe; |
| 152 |
|
| 153 |
|
| 154 |
if (!indexcontents) |
| 155 |
return NODOCTYPE; |
| 156 |
|
| 157 |
/* basically do a right to left compare */ |
| 158 |
fe = (filename + strlen(filename)); |
| 159 |
while (indexcontents) |
| 160 |
{ |
| 161 |
swl = indexcontents->patt; |
| 162 |
|
| 163 |
while (swl) |
| 164 |
{ |
| 165 |
s = fe - strlen(swl->line); |
| 166 |
if (s >= filename) |
| 167 |
{ /* no negative overflow! */ |
| 168 |
if (!strcasecmp(swl->line, s)) |
| 169 |
{ |
| 170 |
return indexcontents->DocType;; |
| 171 |
} |
| 172 |
} |
| 173 |
swl = swl->next; |
| 174 |
} |
| 175 |
|
| 176 |
indexcontents = indexcontents->next; |
| 177 |
} |
| 178 |
|
| 179 |
return NODOCTYPE; |
| 180 |
} |
| 181 |
|
| 182 |
|
| 183 |
|
| 184 |
|
| 185 |
|
| 186 |
struct StoreDescription *hasdescription(int doctype, struct StoreDescription *sd) |
| 187 |
{ |
| 188 |
while (sd) |
| 189 |
{ |
| 190 |
if (sd->DocType == doctype) |
| 191 |
return sd; |
| 192 |
sd = sd->next; |
| 193 |
} |
| 194 |
return NULL; |
| 195 |
} |
Parent Directory
| 
Revision Log
| 
Revision Graph