Rev 8149 | Rev 8863 | 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 | |||
7676 | bpr | 18 | /* Versatile translation according to a dictionary */ |
10 | reyssat | 19 | |
20 | /*************** Customization: change values hereafter ****************/ |
||
21 | |||
7676 | bpr | 22 | /* limit of dictionary entries */ |
10 | reyssat | 23 | #define entrylim 32768 |
7676 | bpr | 24 | /* limit of dictionary length */ |
25 | #define diclim 2*1024*1024 |
||
26 | /* limit of source length */ |
||
10 | reyssat | 27 | #define sourcelim 16*1024*1024 |
28 | |||
29 | /***************** Nothing should need change hereafter *****************/ |
||
8185 | bpr | 30 | #include "../Lib/libwims.h" |
8123 | bpr | 31 | #include "suffix.h" |
10 | reyssat | 32 | |
33 | char *inpbuf, outbuf[2*(MAX_LINELEN+1)]; |
||
34 | char *dicbuf; |
||
35 | struct entry { |
||
36 | unsigned char *original, *replace; |
||
37 | int olen,earlier; |
||
38 | } entry[entrylim]; |
||
39 | int entrycount; |
||
40 | |||
41 | enum { |
||
42 | unk_delete, unk_leave, unk_replace |
||
43 | }; |
||
44 | |||
45 | int has_digits=0; |
||
46 | int unknown_type=unk_delete; |
||
47 | int nocase=0,leaveline=0,fromfile=0; |
||
48 | char *unknown, unkbuf[1024]; |
||
49 | |||
7676 | bpr | 50 | /* Exit without translating anything */ |
10 | reyssat | 51 | void escape(void) |
52 | { |
||
53 | printf("%s",inpbuf); exit(0); |
||
54 | } |
||
55 | |||
7676 | bpr | 56 | /* strip trailing spaces; return string end. */ |
8100 | bpr | 57 | char *strip_trailing_spaces2(char *p) |
10 | reyssat | 58 | { |
59 | char *pp; |
||
60 | if(*p==0) return p; |
||
61 | for(pp=p+strlen(p)-1; pp>=p && isspace(*pp); *(pp--)=0); |
||
62 | return pp; |
||
63 | } |
||
64 | |||
65 | int compare(int i1, const char *s2) |
||
66 | { |
||
67 | int k; |
||
3247 | bpr | 68 | if(nocase) k=strncasecmp((char*)entry[i1].original,s2,entry[i1].olen); |
69 | else k=strncmp((char*)entry[i1].original,s2,entry[i1].olen); |
||
10 | reyssat | 70 | if(k==0 && (isalnum(*(s2+entry[i1].olen)) || (*(s2+entry[i1].olen)&128)!=0)) return -1; |
71 | else return k; |
||
72 | } |
||
73 | |||
7676 | bpr | 74 | /* searches a list. Returns index if found, -1 if nomatch. |
8149 | bpr | 75 | * Uses binary search, list must be sorted. |
76 | */ |
||
8100 | bpr | 77 | int search_list2(struct entry *list, int items, size_t item_size, const char *str) |
10 | reyssat | 78 | { |
79 | int i1,i2,j,k,t,t1; |
||
80 | unsigned char c; |
||
81 | |||
82 | if(items<=0) return -1; |
||
83 | j=0; c=str[0]; |
||
84 | k=list[0].original[0]-c; if(k==0) k=compare(0,str); |
||
85 | if(k==0) goto more; if(k>0) return -1; |
||
86 | j=items-1; k=list[j].original[0]-c; if(k==0) k=compare(j,str); |
||
87 | if(k==0) return j; |
||
88 | if(k>0) for(i1=0,i2=j;i2>i1+1;) { |
||
7676 | bpr | 89 | j=i1+(i2-i1)/2; |
90 | k=list[j].original[0]-c; if(k==0) k=compare(j,str); |
||
91 | if(k==0) goto more; |
||
92 | if(k>0) {i2=j; continue;} |
||
93 | if(k<0) {i1=j; continue;} |
||
10 | reyssat | 94 | } |
95 | if(k>0) {j--;k=compare(j,str);} |
||
96 | more: |
||
97 | if((t=list[j].earlier)<0) { |
||
7676 | bpr | 98 | if(k==0) return j; else return -1; |
10 | reyssat | 99 | } |
100 | if(compare(t,str)!=0) return -1; |
||
101 | for(j=t1=t,k=0;j<items && list[j].earlier==t1 && (k=compare(j,str))<=0; j++) { |
||
7676 | bpr | 102 | if(k==0) t=j; |
10 | reyssat | 103 | } |
104 | return t; |
||
105 | } |
||
106 | |||
8123 | bpr | 107 | /* modify a string. Bufferlen must be at least MAX_LINELEN */ |
8100 | bpr | 108 | void string_modify3(char *start, char *bad_beg, char *bad_end, char *good,...) |
10 | reyssat | 109 | { |
110 | char buf[MAX_LINELEN+1]; |
||
111 | va_list vp; |
||
7676 | bpr | 112 | |
10 | reyssat | 113 | va_start(vp,good); |
114 | vsnprintf(buf,sizeof(buf),good,vp); va_end(vp); |
||
115 | if(strlen(start)-(bad_end-bad_beg)+strlen(buf)>=2*MAX_LINELEN) |
||
7676 | bpr | 116 | return; /* this is an error situation. */ |
10 | reyssat | 117 | strcat(buf,bad_end); |
3718 | reyssat | 118 | ovlstrcpy(bad_beg,buf); |
10 | reyssat | 119 | } |
120 | |||
7676 | bpr | 121 | /* change all spaces into ' ', and collapse multiple occurences */ |
8100 | bpr | 122 | void singlespace2(char *p) |
10 | reyssat | 123 | { |
124 | char *pp, *p2; |
||
125 | for(pp=p;*pp;pp++) { |
||
7676 | bpr | 126 | if(!isspace(*pp)) continue; |
127 | if(leaveline) { |
||
128 | if(*pp==13) ovlstrcpy(pp,pp+1); |
||
129 | if(*pp=='\n') { |
||
130 | pp++; |
||
131 | gopt: for(p2=pp; isspace(*p2) && *p2!='\n'; p2++); |
||
132 | if(p2>pp) ovlstrcpy(pp,p2); pp--; |
||
133 | } |
||
134 | else { |
||
135 | pp++; if(!isspace(*pp) || *pp=='\n') continue; |
||
136 | goto gopt; |
||
137 | } |
||
138 | } |
||
139 | else { |
||
140 | if(*pp!=' ') *pp=' '; |
||
141 | pp++; if(!isspace(*pp)) continue; |
||
142 | for(p2=pp;isspace(*p2);p2++); |
||
143 | ovlstrcpy(pp,p2); pp--; |
||
144 | } |
||
10 | reyssat | 145 | } |
146 | } |
||
147 | |||
7676 | bpr | 148 | /* Prepare dictionary */ |
10 | reyssat | 149 | void prepare_dic(void) |
150 | { |
||
151 | int i,l; |
||
152 | FILE *dicf; |
||
153 | char *fname, *p1, *p2, *pp, buf[1024]; |
||
154 | long int flen; |
||
155 | fname=getenv("w_dictionary"); |
||
156 | if(fname==NULL || *fname==0 || *fname=='/' || strstr(fname,"..")) { |
||
7676 | bpr | 157 | p1=getenv("w_module"); if(p1 && strncmp(p1,"classes/",strlen("classes/"))==0) { |
158 | p1=getenv("w_wims_class"); p2=getenv("w_wims_home"); |
||
159 | if(p1 && p2) { |
||
160 | snprintf(buf,sizeof(buf),"%s/log/classes/%s/",p2,p1); |
||
161 | if(strncmp(fname,buf,strlen(buf))!=0) escape(); |
||
162 | } |
||
163 | else escape(); |
||
164 | } |
||
165 | else { |
||
166 | p1=getenv("untrust"); if(p1 && strstr(p1,"yes")) escape(); |
||
167 | } |
||
10 | reyssat | 168 | } |
6876 | bpr | 169 | /* replace escape() by return if there is some suffix dictionary, */ |
170 | |||
171 | dicf=fopen(fname,"r"); if(dicf==NULL) return; |
||
10 | reyssat | 172 | fseek(dicf,0,SEEK_END);flen=ftell(dicf); fseek(dicf,0,SEEK_SET); |
173 | if(flen>diclim) escape(); |
||
174 | dicbuf=xmalloc(flen+16);flen=fread(dicbuf,1,flen,dicf); |
||
175 | fclose(dicf); |
||
176 | if(flen>0 && flen<diclim) dicbuf[flen]=0; |
||
177 | else escape(); |
||
178 | for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) { |
||
7676 | bpr | 179 | p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0; |
180 | pp=strchr(p1,':'); if(pp==NULL) continue; |
||
181 | *pp++=0; |
||
8100 | bpr | 182 | strip_trailing_spaces2(p1); strip_trailing_spaces2(pp); |
183 | singlespace2(p1); |
||
7676 | bpr | 184 | p1=find_word_start(p1); pp=find_word_start(pp); |
185 | if(*p1==0) continue; |
||
186 | if(has_digits==0) { |
||
187 | char *p; |
||
188 | for(p=p1;*p!=0 && p<pp && !isdigit(*p);p++); |
||
189 | if(isdigit(*p)) has_digits=1; |
||
190 | } |
||
191 | entry[i].original=(unsigned char*)p1; |
||
8100 | bpr | 192 | entry[i].replace=(unsigned char*)pp; |
7676 | bpr | 193 | entry[i].olen=l=strlen(p1); entry[i].earlier=-1; |
194 | if(i>0) { |
||
195 | int l1,l2; |
||
196 | l1=entry[i-1].earlier; if(l1>=0) l2=entry[l1].olen; |
||
197 | else {l2=entry[i-1].olen;l1=i-1;} |
||
198 | if(l>l2 && isspace(p1[l2]) |
||
199 | && strncmp((char*)entry[l1].original,p1,l2)==0) |
||
200 | entry[i].earlier=entry[i-1].earlier=l1; |
||
201 | } |
||
202 | i++; |
||
10 | reyssat | 203 | } |
204 | entrycount=i; if(entrycount<=0) escape(); |
||
205 | } |
||
206 | |||
7676 | bpr | 207 | /* now make the translation. */ |
10 | reyssat | 208 | void translate(void) |
209 | { |
||
210 | char *p1, *p2, *pp; |
||
211 | int t; |
||
212 | |||
213 | for(p1=find_word_start(outbuf); |
||
7676 | bpr | 214 | p1!=NULL && p1-outbuf<MAX_LINELEN && *p1!=0; |
215 | p1=p2) { |
||
216 | p2=find_word_end(p1); |
||
217 | for(pp=p1;pp<p2 && |
||
218 | ((!has_digits && isalpha(*pp)) || |
||
219 | (has_digits && isalnum(*pp)) || (*pp&128)!=0 || |
||
220 | strchr("_",*pp)!=NULL);pp++); |
||
221 | p2=find_word_start(p2); |
||
222 | if(pp==p1 || |
||
223 | (has_digits==0 && isdigit(*pp)) || |
||
224 | (*pp!=0 && !isspace(*pp) && strchr(",.?!/;",*pp)==NULL)) continue; |
||
8100 | bpr | 225 | t=search_list2(entry,entrycount,sizeof(entry[0]),p1); |
7676 | bpr | 226 | if(t<0) { |
227 | switch(unknown_type) { |
||
228 | case unk_leave: break; |
||
229 | case unk_delete: { |
||
230 | ovlstrcpy(p1,find_word_start(pp)); p2=p1; |
||
231 | break; |
||
232 | } |
||
233 | case unk_replace: { |
||
8100 | bpr | 234 | string_modify3(outbuf,p1,pp,unkbuf); |
7676 | bpr | 235 | p2=find_word_start(p1+strlen(unkbuf)); |
236 | } |
||
237 | } |
||
238 | continue; |
||
239 | } |
||
8100 | bpr | 240 | string_modify3(outbuf,p1,p1+strlen((char*)entry[t].original), |
7676 | bpr | 241 | (char*)entry[t].replace); |
242 | p2=find_word_start(p1+strlen((char*)entry[t].replace)); |
||
10 | reyssat | 243 | } |
244 | outbuf[MAX_LINELEN]=0; printf("%s",outbuf); |
||
245 | } |
||
246 | |||
247 | void switches(void) |
||
248 | { |
||
249 | char *sw; |
||
250 | unknown=getenv("w_translator_unknown"); |
||
251 | if(unknown==NULL) unknown=""; |
||
252 | snprintf(unkbuf,sizeof(unkbuf),"%s",find_word_start(unknown)); |
||
253 | *find_word_end(unkbuf)=0; |
||
254 | if(strcasecmp(unkbuf,"leave")==0) unknown_type=unk_leave; |
||
255 | else if(unkbuf[0]!=0) unknown_type=unk_replace; |
||
256 | sw=getenv("w_translator_switch"); |
||
257 | if(sw==NULL) return; |
||
258 | if(strstr(sw,"leaveline")) leaveline=1; |
||
259 | if(strstr(sw,"nocase")) nocase=1; |
||
260 | if(strstr(sw,"file")) fromfile=1; |
||
261 | } |
||
262 | |||
263 | int main() |
||
264 | { |
||
265 | char c, *p1, *p2, *s; |
||
266 | unsigned int l; |
||
267 | |||
268 | switches(); |
||
269 | if(!fromfile) { |
||
7676 | bpr | 270 | s=getenv("wims_exec_parm"); |
271 | if(s==NULL || *s==0) return 0; /* Nothing to translate */ |
||
272 | l=strlen(s); if(l<=0 || l>sourcelim) return 0; /* too long */ |
||
273 | inpbuf=xmalloc(l+16); memmove(inpbuf,s,l+1); |
||
10 | reyssat | 274 | } |
275 | else { |
||
7676 | bpr | 276 | FILE *f; |
277 | s=getenv("translator_input"); if(s==NULL || *s==0) return 0; |
||
278 | f=fopen(s,"r"); if(f==NULL) return 0; /* no file */ |
||
279 | fseek(f,0,SEEK_END); l=ftell(f); fseek(f,0,SEEK_SET); |
||
280 | if(l<=0 || l>sourcelim) return 0; /* too long */ |
||
281 | inpbuf=xmalloc(l+16); (void)fread(inpbuf,1,l,f); fclose(f); inpbuf[l]=0; |
||
10 | reyssat | 282 | } |
283 | p1=inpbuf; prepare_dic(); |
||
284 | if(leaveline) c='\n'; else c=' '; |
||
285 | do { |
||
7676 | bpr | 286 | l=strlen(p1); |
287 | if(l>MAX_LINELEN-1024) l=MAX_LINELEN-1024; p2=p1+l; |
||
288 | if(*p2) { |
||
289 | while(p2>p1 && *p2!=c) p2--; |
||
290 | } |
||
291 | if(p2<=p1) return 0; |
||
292 | memmove(outbuf,p1,p2-p1); outbuf[p2-p1]=0; |
||
8100 | bpr | 293 | singlespace2(outbuf); |
7676 | bpr | 294 | s=getenv("w_suffix_dictionary"); |
295 | if(s!=NULL && *s!=0) suffix(outbuf,s); |
||
296 | translate(); |
||
297 | if(*p2==c) {printf("%c",c); p1=++p2;} |
||
10 | reyssat | 298 | } |
299 | while(*p2); |
||
300 | return 0; |
||
301 | } |