Rev 8161 | 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 | */ |
||
8161 | bpr | 17 | #include "symtext.h" |
10 | reyssat | 18 | |
8113 | bpr | 19 | #define suflim 256 |
10 | reyssat | 20 | #define sufbuflim 102400 |
21 | |||
22 | int suffixcnt; |
||
23 | struct { |
||
24 | unsigned char *original; |
||
25 | int olen; |
||
26 | unsigned char *replace; |
||
8113 | bpr | 27 | } |
10 | reyssat | 28 | suf[suflim]; |
29 | char *sufbuf; |
||
30 | int sufwordlen, sufminlen; |
||
31 | |||
8113 | bpr | 32 | /* Suffix translation, to be used within translator. */ |
10 | reyssat | 33 | |
34 | int sufcomp(int t, const unsigned char *s2) |
||
35 | { |
||
36 | int k; |
||
8113 | bpr | 37 | |
10 | reyssat | 38 | for(k=0;k<suf[t].olen && k<sufwordlen |
8113 | bpr | 39 | && suf[t].original[k]==s2[sufwordlen-k-1];k++); |
10 | reyssat | 40 | if(k>=suf[t].olen) { |
8113 | bpr | 41 | if(sufwordlen>k) return -1; else return 0; |
10 | reyssat | 42 | } |
43 | else return suf[t].original[k]-s2[sufwordlen-k-1]; |
||
44 | } |
||
45 | |||
8113 | bpr | 46 | /* searches a list. Returns index if found, -1 if nomatch. |
47 | * This routine is faster than naive one by one comparisons, |
||
48 | * and is especially suited for large lists. |
||
49 | */ |
||
3808 | kbelabas | 50 | int suffix_list(void *list, int items, size_t item_size, const unsigned char *str) |
10 | reyssat | 51 | { |
52 | int i1,i2,j,k,t,v; |
||
3808 | kbelabas | 53 | unsigned char c,d; |
8113 | bpr | 54 | |
10 | reyssat | 55 | if(items<=0) return -1; |
56 | k=sufcomp(0,str); |
||
57 | if(k==0) return 0; if(k>0) return -1; |
||
58 | j=items-1; k=sufcomp(j,str); |
||
59 | if(k==0) return j; |
||
60 | if(k>0) for(i1=0,i2=j;i2>i1+1;) { |
||
8113 | bpr | 61 | j=i1+(i2-i1)/2; k=sufcomp(j,str); |
62 | if(k==0) return j; |
||
63 | if(k>0) {i2=j; continue;} |
||
64 | if(k<0) {i1=j; continue;} |
||
10 | reyssat | 65 | } |
66 | if(k>0 && j>0) j--; |
||
67 | backcheck: |
||
68 | v=j;for(t=0;t<suf[j].olen && t<sufwordlen |
||
8113 | bpr | 69 | && suf[j].original[t]==str[sufwordlen-t-1];t++); |
10 | reyssat | 70 | if(t<sufminlen) return -1; if(t>=suf[j].olen) return j; |
71 | for(j--,c=str[sufwordlen-1],d=str[sufwordlen-t]; |
||
8113 | bpr | 72 | j>=0 && suf[j].original[0]==c && suf[j].olen>t |
73 | && suf[j].original[t-1]==d;j--); |
||
74 | if(j>=0 && suf[j].original[0]==c && |
||
3808 | kbelabas | 75 | strncmp((char*)suf[j].original,(char*)suf[v].original,suf[j].olen)==0) |
10 | reyssat | 76 | return j; |
77 | else goto backcheck; |
||
78 | } |
||
79 | |||
8113 | bpr | 80 | /* Prepare dictionary. */ |
10 | reyssat | 81 | void suffix_dic(char *sdicname) |
82 | { |
||
83 | int i,k,l; |
||
84 | FILE *suff; |
||
85 | char *p1, *p2, *pp; |
||
86 | long int flen; |
||
87 | |||
88 | suffixcnt=0; sufminlen=100000; |
||
89 | suff=fopen(sdicname,"r"); if(suff==NULL) return; |
||
90 | fseek(suff,0,SEEK_END);flen=ftell(suff); fseek(suff,0,SEEK_SET); |
||
91 | if(flen>sufbuflim) return; |
||
92 | sufbuf=xmalloc(flen+16);flen=fread(sufbuf,1,flen,suff); |
||
93 | fclose(suff); |
||
94 | if(flen>0 && flen<sufbuflim) sufbuf[flen]=0; |
||
95 | else return; |
||
96 | for(i=0,p1=sufbuf;p1!=NULL && *p1!=0 && i<suflim;p1=p2) { |
||
8113 | bpr | 97 | p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0; |
98 | pp=strchr(p1,':'); if(pp==NULL) continue; |
||
99 | *pp++=0; |
||
100 | strip_trailing_spaces(p1); strip_trailing_spaces(pp); |
||
101 | p1=find_word_start(p1); pp=find_word_start(pp); |
||
102 | if(*p1==0) continue; |
||
103 | if(i>0) { |
||
104 | k=strcmp((char*)suf[i-1].original,p1); |
||
105 | if(k>0) { |
||
106 | pp=strrchr(sdicname,'/'); if(pp==NULL) pp=sdicname; else pp++; |
||
8195 | bpr | 107 | sym_error("unsorted_dictionary %s: %s > %s.\n", |
8113 | bpr | 108 | pp,suf[i-1].original,p1); |
109 | } |
||
110 | if(k==0) { |
||
111 | pp=strrchr(sdicname,'/'); if(pp==NULL) pp=sdicname; else pp++; |
||
8195 | bpr | 112 | sym_error("duplication_in_dictionary %s: %s.\n",pp,p1); |
8113 | bpr | 113 | } |
114 | } |
||
115 | suf[i].original=(unsigned char*)p1; suf[i].olen=l=strlen(p1); |
||
116 | if(l<sufminlen) sufminlen=l; |
||
117 | suf[i].replace=(unsigned char*)pp; i++; |
||
10 | reyssat | 118 | } |
119 | suffixcnt=i; |
||
120 | } |
||
121 | |||
8113 | bpr | 122 | /* Suffix translation. */ |
10 | reyssat | 123 | void suffix_translate(char *p) |
124 | { |
||
125 | char *p1, *p2; |
||
126 | int t; |
||
127 | |||
128 | for(p1=find_word_start(p); |
||
8113 | bpr | 129 | p1!=NULL && p1-p<MAX_LINELEN && *p1!=0; |
130 | p1=p2) { |
||
131 | if(!isalpha(*p1)) {p2=p1+1; continue;} |
||
132 | for(p2=p1;isalpha(*p2);p2++); |
||
133 | if(*p2!=0 && strchr(" ,.?!'\"\n`:;()[]{}<>",*p2)==NULL) continue; |
||
134 | sufwordlen=p2-p1; |
||
135 | t=suffix_list(suf,suffixcnt,sizeof(suf[0]),(unsigned char*)p1); |
||
136 | if(t<0) continue; |
||
137 | string_modify(p,p2-suf[t].olen,p2,(char*)suf[t].replace); |
||
138 | p2=p2-suf[t].olen+strlen((char*)suf[t].replace); |
||
139 | } |
||
140 | p[MAX_LINELEN]=0; |
||
10 | reyssat | 141 | } |
142 | |||
143 | void suffix(char *p, char *sdicname) |
||
144 | { |
||
145 | suffix_dic(sdicname); if(suffixcnt>0) suffix_translate(p); |
||
146 | } |
||
147 |