Rev 10 | Rev 6895 | Go to most recent revision | Show entire file | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed
| Rev 10 | Rev 3247 | ||
|---|---|---|---|
| Line 43... | Line 43... | ||
| 43 | } |
43 | } |
| 44 | 44 | ||
| 45 | /* searches a list. Returns index if found, -1 if nomatch. |
45 | /* searches a list. Returns index if found, -1 if nomatch. |
| 46 | * This routine is faster than naive one by one comparisons, |
46 | * This routine is faster than naive one by one comparisons, |
| 47 | * and is especially suited for large lists. */ |
47 | * and is especially suited for large lists. */ |
| 48 | int suffix_list(void *list, int items, size_t item_size, const char *str) |
48 | int suffix_list(void *list, int items, size_t item_size, const unsigned char *str) |
| 49 | { |
49 | { |
| 50 | int i1,i2,j,k,t,v; |
50 | int i1,i2,j,k,t,v; |
| 51 | char c,d; |
51 | unsigned char c,d; |
| 52 | 52 | ||
| 53 | if(items<=0) return -1; |
53 | if(items<=0) return -1; |
| 54 | k=sufcomp(0,str); |
54 | k=sufcomp(0,str); |
| 55 | if(k==0) return 0; if(k>0) return -1; |
55 | if(k==0) return 0; if(k>0) return -1; |
| 56 | j=items-1; k=sufcomp(j,str); |
56 | j=items-1; k=sufcomp(j,str); |
| Line 68... | Line 68... | ||
| 68 | if(t<sufminlen) return -1; if(t>=suf[j].olen) return j; |
68 | if(t<sufminlen) return -1; if(t>=suf[j].olen) return j; |
| 69 | for(j--,c=str[sufwordlen-1],d=str[sufwordlen-t]; |
69 | for(j--,c=str[sufwordlen-1],d=str[sufwordlen-t]; |
| 70 | j>=0 && suf[j].original[0]==c && suf[j].olen>t |
70 | j>=0 && suf[j].original[0]==c && suf[j].olen>t |
| 71 | && suf[j].original[t-1]==d;j--); |
71 | && suf[j].original[t-1]==d;j--); |
| 72 | if(j>=0 && suf[j].original[0]==c && |
72 | if(j>=0 && suf[j].original[0]==c && |
| 73 | strncmp( |
73 | strncmp((char*)suf[j].original,(char*)suf[v].original,suf[j].olen)==0) |
| 74 | return j; |
74 | return j; |
| 75 | else goto backcheck; |
75 | else goto backcheck; |
| 76 | } |
76 | } |
| 77 | 77 | ||
| 78 | /* Prepare dictionary. */ |
78 | /* Prepare dictionary. */ |
| Line 97... | Line 97... | ||
| 97 | *pp++=0; |
97 | *pp++=0; |
| 98 | strip_trailing_spaces(p1); strip_trailing_spaces(pp); |
98 | strip_trailing_spaces(p1); strip_trailing_spaces(pp); |
| 99 | singlespace(p1); |
99 | singlespace(p1); |
| 100 | p1=find_word_start(p1); pp=find_word_start(pp); |
100 | p1=find_word_start(p1); pp=find_word_start(pp); |
| 101 | if(*p1==0) continue; |
101 | if(*p1==0) continue; |
| 102 | suf[i].original=p1; suf[i].olen=l=strlen(p1); |
102 | suf[i].original=(unsigned char*)p1; suf[i].olen=l=strlen(p1); |
| 103 | if(l<sufminlen) sufminlen=l; |
103 | if(l<sufminlen) sufminlen=l; |
| 104 | suf[i].replace=pp; i++; |
104 | suf[i].replace=(unsigned char*)pp; i++; |
| 105 | } |
105 | } |
| 106 | suffixcnt=i; |
106 | suffixcnt=i; |
| 107 | } |
107 | } |
| 108 | 108 | ||
| 109 | /* Suffix translation. */ |
109 | /* Suffix translation. */ |
| Line 117... | Line 117... | ||
| 117 | p1=p2) { |
117 | p1=p2) { |
| 118 | if(!isalpha(*p1)) {p2=p1+1; continue;} |
118 | if(!isalpha(*p1)) {p2=p1+1; continue;} |
| 119 | for(p2=p1;isalpha(*p2);p2++); |
119 | for(p2=p1;isalpha(*p2);p2++); |
| 120 | if(*p2!=0 && strchr(" ,.?!'\"\n`:;()[]{}<>",*p2)==NULL) continue; |
120 | if(*p2!=0 && strchr(" ,.?!'\"\n`:;()[]{}<>",*p2)==NULL) continue; |
| 121 | sufwordlen=p2-p1; |
121 | sufwordlen=p2-p1; |
| 122 | t=suffix_list(suf,suffixcnt,sizeof(suf[0]),p1); |
122 | t=suffix_list(suf,suffixcnt,sizeof(suf[0]),(unsigned char*)p1); |
| 123 | if(t<0) continue; |
123 | if(t<0) continue; |
| 124 | string_modify(p,p2-suf[t].olen,p2,suf[t].replace); |
124 | string_modify(p,p2-suf[t].olen,p2,(char*)suf[t].replace); |
| 125 | p2=p2-suf[t].olen+strlen(suf[t].replace); |
125 | p2=p2-suf[t].olen+strlen((char*)suf[t].replace); |
| 126 | } |
126 | } |
| 127 | p[MAX_LINELEN]=0; |
127 | p[MAX_LINELEN]=0; |
| 128 | } |
128 | } |
| 129 | 129 | ||
| 130 | void suffix(char *p, char *sdicname) |
130 | void suffix(char *p, char *sdicname) |