WebSVN – wimsdev – Diff – /trunk/wims/src/Misc/translator_.c



#define entrylim 32768
/* limit of dictionary length */
#define diclim 1024*1024
 
/***************** Nothing should need change hereafter *****************/
 
 
 
 
char inpbuf[MAX_LINELEN+1], outbuf[2*MAX_LINELEN+2];
char *dicbuf;
struct entry {

int unknown_type=unk_delete;
int nocase=0,leaveline=0;
char *unknown, unkbuf[1024];
 
int compare(int i1, const char *s2)
{
    int k;
    if(nocase) k=strncasecmp((char*)entry[i1].original,s2,entry[i1].olen);
    else k=strncmp((char*)entry[i1].original,s2,entry[i1].olen);
    if(k==0 && (isalnum(*(s2+entry[i1].olen)) || (*(s2+entry[i1].olen)&128)!=0)) return -1;
    else return k;
}
 
/* searches a list. Returns index if found, -1 if nomatch.
 * Uses binary search, list must be sorted. */
int search_list2(struct entry *list, int items, size_t item_size, const char *str)
{
    int i1,i2,j,k,t,t1;
    unsigned char c;
 
    if(items<=0) return -1;

    more:
    if((t=list[j].earlier)<0) {
      if(k==0) return j; else return -1;
    }
    if(compare(t,str)!=0) return -1;
    for(j=t1=t,k=0;j<items && list[j].earlier==t1 && (k=compare(j,str))<=0; j++) {
      if(k==0) t=j;
    }
    return t;
}
 
/* change all spaces into ' ', and collapse multiple occurences */
void singlespace2(char *p)
{
    char *pp, *p2;
    for(pp=p;*pp;pp++) {
      if(!isspace(*pp)) continue;
      if(leaveline) {

    else return;
    for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {
      p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;
      pp=strchr(p1,':'); if(pp==NULL) continue;
      *pp++=0;
      strip_trailing_spaces2(p1); strip_trailing_spaces2(pp);
      singlespace2(p1);
      p1=find_word_start(p1); pp=find_word_start(pp);
      if(*p1==0) continue;
      if(has_digits==0) {
          char *p;
          for(p=p1;*p!=0 && p<pp && !isdigit(*p);p++);

           strchr("_",*pp)!=NULL);pp++);
      p2=find_word_start(p2);
      if(pp==p1 ||
         (has_digits==0 && isdigit(*pp)) ||
         (*pp!=0 && !isspace(*pp) && strchr(",.?!/;",*pp)==NULL)) continue;
      t=search_list2(entry,entrycount,sizeof(entry[0]),p1);
      if(t<0) {
          switch(unknown_type) {
            case unk_leave: break;
            case unk_delete: {
                ovlstrcpy(p1,find_word_start(pp)); p2=p1;
                break;
            }
            case unk_replace: {
                string_modify3(outbuf,p1,pp,unkbuf);
                p2=find_word_start(p1+strlen(unkbuf));
            }
          }
          continue;
      }
      string_modify3(outbuf,p1,p1+strlen((char*)entry[t].original),
                  (char*)entry[t].replace);
      p2=find_word_start(p1+strlen((char*)entry[t].replace));
    }
    snprintf(p,MAX_LINELEN,"%s",outbuf);
}

Rev 7676	Rev 8100
Line 23...	Line 23...
23	#define entrylim 32768	23	#define entrylim 32768
24	/* limit of dictionary length */	24	/* limit of dictionary length */
25	#define diclim 1024*1024	25	#define diclim 1024*1024
26		26
27	/*************** Nothing should need change hereafter ***************/	27	/*************** Nothing should need change hereafter ***************/
28		-
29	#include "../Lib/basicstr.c"	-
30		28
31		29
32	char inpbuf[MAX_LINELEN+1], outbuf[2*MAX_LINELEN+2];	30	char inpbuf[MAX_LINELEN+1], outbuf[2*MAX_LINELEN+2];
33	char *dicbuf;	31	char *dicbuf;
34	struct entry {	32	struct entry {
Line 45...	Line 43...
45	int unknown_type=unk_delete;	43	int unknown_type=unk_delete;
46	int nocase=0,leaveline=0;	44	int nocase=0,leaveline=0;
47	char *unknown, unkbuf[1024];	45	char *unknown, unkbuf[1024];
48		46
49	int compare(int i1, const char *s2)	47	int compare(int i1, const char *s2)
50	{	48	{
51	int k;	49	int k;
52	if(nocase) k=strncasecmp((char*)entry[i1].original,s2,entry[i1].olen);	50	if(nocase) k=strncasecmp((char*)entry[i1].original,s2,entry[i1].olen);
53	else k=strncmp((char*)entry[i1].original,s2,entry[i1].olen);	51	else k=strncmp((char*)entry[i1].original,s2,entry[i1].olen);
54	if(k==0 && (isalnum((s2+entry[i1].olen)) \|\| ((s2+entry[i1].olen)&128)!=0)) return -1;	52	if(k==0 && (isalnum((s2+entry[i1].olen)) \|\| ((s2+entry[i1].olen)&128)!=0)) return -1;
55	else return k;	53	else return k;
56	}	54	}
57		55
58	/* searches a list. Returns index if found, -1 if nomatch.	56	/* searches a list. Returns index if found, -1 if nomatch.
59	* Uses binary search, list must be sorted. */	57	* Uses binary search, list must be sorted. */
60	int ~~search_list~~(struct entry list, int items, size_t item_size, const char str)	58	int search_list2(struct entry list, int items, size_t item_size, const char str)
61	{	59	{
62	int i1,i2,j,k,t,t1;	60	int i1,i2,j,k,t,t1;
63	unsigned char c;	61	unsigned char c;
64		62
65	if(items<=0) return -1;	63	if(items<=0) return -1;
Line 79...	Line 77...
79	more:	77	more:
80	if((t=list[j].earlier)<0) {	78	if((t=list[j].earlier)<0) {
81	if(k==0) return j; else return -1;	79	if(k==0) return j; else return -1;
82	}	80	}
83	if(compare(t,str)!=0) return -1;	81	if(compare(t,str)!=0) return -1;
84	for(j=t1=t,k=0;j<items && list[j].earlier==t1 && (k=compare(j,str))<=0; j++)	82	for(j=t1=t,k=0;j<items && list[j].earlier==t1 && (k=compare(j,str))<=0; j++) {
85	if(k==0) t=j;	83	if(k==0) t=j;
-		84	}
86	return t;	85	return t;
87	}	86	}
88		87
89	/* change all spaces into ' ', and collapse multiple occurences */	88	/* change all spaces into ' ', and collapse multiple occurences */
90	void ~~singlespace~~(char *p)	89	void singlespace2(char *p)
91	{	90	{
92	char pp, p2;	91	char pp, p2;
93	for(pp=p;*pp;pp++) {	92	for(pp=p;*pp;pp++) {
94	if(!isspace(*pp)) continue;	93	if(!isspace(*pp)) continue;
95	if(leaveline) {	94	if(leaveline) {
Line 133...	Line 132...
133	else return;	132	else return;
134	for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {	133	for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {
135	p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;	134	p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;
136	pp=strchr(p1,':'); if(pp==NULL) continue;	135	pp=strchr(p1,':'); if(pp==NULL) continue;
137	*pp++=0;	136	*pp++=0;
138	~~strip_trailing_spaces~~(p1); ~~strip_trailing_spaces~~(pp);	137	strip_trailing_spaces2(p1); strip_trailing_spaces2(pp);
139	~~singlespace~~(p1);	138	singlespace2(p1);
140	p1=find_word_start(p1); pp=find_word_start(pp);	139	p1=find_word_start(p1); pp=find_word_start(pp);
141	if(*p1==0) continue;	140	if(*p1==0) continue;
142	if(has_digits==0) {	141	if(has_digits==0) {
143	char *p;	142	char *p;
144	for(p=p1;p!=0 && p<pp && !isdigit(p);p++);	143	for(p=p1;p!=0 && p<pp && !isdigit(p);p++);
Line 178...	Line 177...
178	strchr("_",*pp)!=NULL);pp++);	177	strchr("_",*pp)!=NULL);pp++);
179	p2=find_word_start(p2);	178	p2=find_word_start(p2);
180	if(pp==p1 \|\|	179	if(pp==p1 \|\|
181	(has_digits==0 && isdigit(*pp)) \|\|	180	(has_digits==0 && isdigit(*pp)) \|\|
182	(pp!=0 && !isspace(pp) && strchr(",.?!/;",*pp)==NULL)) continue;	181	(pp!=0 && !isspace(pp) && strchr(",.?!/;",*pp)==NULL)) continue;
183	t=~~search_list~~(entry,entrycount,sizeof(entry[0]),p1);	182	t=search_list2(entry,entrycount,sizeof(entry[0]),p1);
184	if(t<0) {	183	if(t<0) {
185	switch(unknown_type) {	184	switch(unknown_type) {
186	case unk_leave: break;	185	case unk_leave: break;
187	case unk_delete: {	186	case unk_delete: {
188	ovlstrcpy(p1,find_word_start(pp)); p2=p1;	187	ovlstrcpy(p1,find_word_start(pp)); p2=p1;
189	break;	188	break;
190	}	189	}
191	case unk_replace: {	190	case unk_replace: {
192	~~string_modify~~(outbuf,p1,pp,unkbuf);	191	string_modify3(outbuf,p1,pp,unkbuf);
193	p2=find_word_start(p1+strlen(unkbuf));	192	p2=find_word_start(p1+strlen(unkbuf));
194	}	193	}
195	}	194	}
196	continue;	195	continue;
197	}	196	}
198	~~string_modify~~(outbuf,p1,p1+strlen((char*)entry[t].original),	197	string_modify3(outbuf,p1,p1+strlen((char*)entry[t].original),
199	(char*)entry[t].replace);	198	(char*)entry[t].replace);
200	p2=find_word_start(p1+strlen((char*)entry[t].replace));	199	p2=find_word_start(p1+strlen((char*)entry[t].replace));
201	}	200	}
202	snprintf(p,MAX_LINELEN,"%s",outbuf);	201	snprintf(p,MAX_LINELEN,"%s",outbuf);
203	}	202	}

Subversion Repositories wimsdev

(root)/trunk/wims/src/Misc/translator_.c – Rev 7676 → 8100