Rev 11124 | Rev 18181 | Go to most recent revision | Details | Compare with Previous | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
10 | reyssat | 1 | /* Copyright (C) 1998-2003 XIAO, Gang of Universite de Nice - Sophia Antipolis |
2 | * |
||
3 | * This program is free software; you can redistribute it and/or modify |
||
4 | * it under the terms of the GNU General Public License as published by |
||
5 | * the Free Software Foundation; either version 2 of the License, or |
||
6 | * (at your option) any later version. |
||
7 | * |
||
8 | * This program is distributed in the hope that it will be useful, |
||
9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
||
10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
||
11 | * GNU General Public License for more details. |
||
12 | * |
||
13 | * You should have received a copy of the GNU General Public License |
||
14 | * along with this program; if not, write to the Free Software |
||
15 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
||
16 | */ |
||
17 | |||
8123 | bpr | 18 | #include "../Lib/libwims.h" |
19 | #include "suffix.h" |
||
20 | |||
6895 | bpr | 21 | #define suflim 256 |
10 | reyssat | 22 | #define sufbuflim 102400 |
23 | |||
24 | int suffixcnt; |
||
25 | struct { |
||
12248 | bpr | 26 | unsigned char *original; |
27 | int olen; |
||
28 | unsigned char *replace; |
||
6895 | bpr | 29 | } |
10 | reyssat | 30 | suf[suflim]; |
31 | char *sufbuf; |
||
32 | int sufwordlen, sufminlen; |
||
33 | |||
6895 | bpr | 34 | /* Suffix translation, to be used within translator. */ |
10 | reyssat | 35 | |
36 | int sufcomp(int t, const unsigned char *s2) |
||
37 | { |
||
12248 | bpr | 38 | int k; |
6895 | bpr | 39 | |
12248 | bpr | 40 | for(k=0;k<suf[t].olen && k<sufwordlen |
41 | && suf[t].original[k]==s2[sufwordlen-k-1];k++); |
||
42 | if(k>=suf[t].olen) { |
||
43 | if(sufwordlen>k) return -1; else return 0; |
||
44 | } |
||
45 | else return suf[t].original[k]-s2[sufwordlen-k-1]; |
||
10 | reyssat | 46 | } |
47 | |||
6895 | bpr | 48 | /* searches a list. Returns index if found, -1 if nomatch. |
49 | * This routine is faster than naive one by one comparisons, |
||
8100 | bpr | 50 | * and is especially suited for large lists. |
51 | */ |
||
3247 | bpr | 52 | int suffix_list(void *list, int items, size_t item_size, const unsigned char *str) |
10 | reyssat | 53 | { |
12248 | bpr | 54 | int i1,i2,j,k,t,v; |
55 | unsigned char c,d; |
||
6895 | bpr | 56 | |
12248 | bpr | 57 | if(items<=0) return -1; |
58 | k=sufcomp(0,str); |
||
59 | if(k==0) return 0; |
||
60 | if(k>0) return -1; |
||
61 | j=items-1; k=sufcomp(j,str); |
||
62 | if(k==0) return j; |
||
63 | if(k>0) for(i1=0,i2=j;i2>i1+1;) { |
||
64 | j=i1+(i2-i1)/2; k=sufcomp(j,str); |
||
10 | reyssat | 65 | if(k==0) return j; |
12248 | bpr | 66 | if(k>0) {i2=j; continue;} |
67 | if(k<0) {i1=j; continue;} |
||
68 | } |
||
69 | if(k>0 && j>0) j--; |
||
70 | backcheck: |
||
71 | v=j;for(t=0;t<suf[j].olen && t<sufwordlen |
||
72 | && suf[j].original[t]==str[sufwordlen-t-1];t++); |
||
73 | if(t<sufminlen) return -1; |
||
74 | if(t>=suf[j].olen) return j; |
||
75 | for(j--,c=str[sufwordlen-1],d=str[sufwordlen-t]; |
||
76 | j>=0 && suf[j].original[0]==c && suf[j].olen>t |
||
77 | && suf[j].original[t-1]==d;j--); |
||
78 | if(j>=0 && suf[j].original[0]==c && |
||
3247 | bpr | 79 | strncmp((char*)suf[j].original,(char*)suf[v].original,suf[j].olen)==0) |
12248 | bpr | 80 | return j; |
81 | else goto backcheck; |
||
10 | reyssat | 82 | } |
83 | |||
6895 | bpr | 84 | /* Prepare dictionary. */ |
10 | reyssat | 85 | void suffix_dic(char *sdicname) |
86 | { |
||
12248 | bpr | 87 | int i,l; |
88 | FILE *suff; |
||
89 | char *p1, *p2, *pp; |
||
90 | long int flen; |
||
10 | reyssat | 91 | |
12248 | bpr | 92 | suffixcnt=0; sufminlen=100000; |
93 | suff=fopen(sdicname,"r"); if(suff==NULL) return; |
||
94 | fseek(suff,0,SEEK_END);flen=ftell(suff); fseek(suff,0,SEEK_SET); |
||
95 | if(flen>sufbuflim) return; |
||
96 | sufbuf=xmalloc(flen+16);flen=fread(sufbuf,1,flen,suff); |
||
97 | fclose(suff); |
||
98 | if(flen>0 && flen<sufbuflim) sufbuf[flen]=0; |
||
99 | else return; |
||
100 | for(i=0,p1=sufbuf;p1!=NULL && *p1!=0 && i<suflim;p1=p2) { |
||
101 | p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0; |
||
102 | pp=strchr(p1,':'); if(pp==NULL) continue; |
||
103 | *pp++=0; |
||
104 | strip_trailing_spaces2(p1); strip_trailing_spaces2(pp); |
||
105 | singlespace2(p1); |
||
106 | p1=find_word_start(p1); pp=find_word_start(pp); |
||
107 | if(*p1==0) continue; |
||
108 | suf[i].original=(unsigned char*)p1; suf[i].olen=l=strlen(p1); |
||
109 | if(l<sufminlen) sufminlen=l; |
||
110 | suf[i].replace=(unsigned char*)pp; i++; |
||
111 | } |
||
112 | suffixcnt=i; |
||
10 | reyssat | 113 | } |
114 | |||
6895 | bpr | 115 | /* Suffix translation. */ |
116 | /* FIXME : ne rien faire si le résultat est de longueur inferieur à 2 |
||
117 | * car ensuite cela sera neglige. |
||
118 | */ |
||
119 | |||
10 | reyssat | 120 | void suffix_translate(char *p) |
121 | { |
||
12248 | bpr | 122 | char *p1, *p2; |
123 | int t; |
||
10 | reyssat | 124 | |
12248 | bpr | 125 | for(p1=find_word_start(p); |
126 | p1!=NULL && p1-p<MAX_LINELEN && *p1!=0; |
||
127 | p1=p2) { |
||
128 | if(!isalpha(*p1)) {p2=p1+1; continue;} |
||
129 | for(p2=p1;isalpha(*p2);p2++); |
||
130 | if(*p2!=0 && strchr(" ,.?!'\"\n`:;()[]{}<>",*p2)==NULL) continue; |
||
131 | sufwordlen=p2-p1; |
||
132 | t=suffix_list(suf,suffixcnt,sizeof(suf[0]),(unsigned char*)p1); |
||
133 | if(t<0) continue; |
||
134 | string_modify3(p,p2-suf[t].olen,p2,(char*)suf[t].replace); |
||
135 | p2=p2-suf[t].olen+strlen((char*)suf[t].replace); |
||
136 | } |
||
137 | p[MAX_LINELEN]=0; |
||
10 | reyssat | 138 | } |
139 | |||
140 | void suffix(char *p, char *sdicname) |
||
141 | { |
||
12248 | bpr | 142 | suffix_dic(sdicname); if(suffixcnt>0) suffix_translate(p); |
10 | reyssat | 143 | } |
144 |