void regex::read_tokens(FILE *file){
unsigned i=0,j=0,k=0;
unsigned num_tokens=0;
num_short_tokens=0;
num_long_tokens=0;
char *token = allocate_char_array(1000);
int r = fscanf(file,"%s\n",token);
while(r!=EOF && r>0){
r=fscanf(file,"%s\n",token);
if (r!=EOF && r>0) num_tokens++;
}
free(token);
char **tokens = allocate_string_array(num_tokens);
rewind(file);
while (i<num_tokens){
char *token =(char *)malloc(1000);
fscanf(file,"%s\n",token);
tokens[i++]=token;
}
for (int i=0;i<num_tokens;i++){
if (strlen(tokens[i])==1 ||( strlen(tokens[i]) <=3 && tokens[i][0]=='\\') ||
( strlen(tokens[i]) <=4 && tokens[i][0]=='\\' && tokens[i][1]=='x'))
num_short_tokens++;
}
num_long_tokens=num_tokens-num_short_tokens;
short_tokens = allocate_string_array(num_short_tokens);
long_tokens = allocate_string_array(num_long_tokens);
for (int i=0;i<num_tokens;i++){
if (strlen(tokens[i])==1 ||( strlen(tokens[i]) <=3 && tokens[i][0]=='\\') ||
( strlen(tokens[i]) <=4 && tokens[i][0]=='\\' && tokens[i][1]=='x'))
short_tokens[j++]=tokens[i];
else
long_tokens[k++]=tokens[i];
}
free(tokens);
}
NFA *regex_parser::parse(FILE *file, int from, int to){
rewind(file);
char *re=allocate_char_array(4000);
char cmd[4000];
int i=0;
int j=0;
unsigned int c=fgetc(file);
// NFA
NFA *nfa=new NFA();
NFA *non_anchored = nfa->add_epsilon(); // for .* RegEx
NFA *anchored = nfa->add_epsilon(); // for anchored RegEx (^)
fprintf(stdout,"\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\n");
//parsing the RegEx and putting them in a NFA
while(c!=EOF){
if (c=='\n' || c=='\r'){
if(i!=0){
re[i]='\0';
if (re[0]!='#'){
j++;
if (j>=from && (to==-1 || j<=to)){
fprintf(stdout,"%dth regex: %s\n",j,re);
if (DEBUG) fprintf(stdout,"\n%d) processing regex:: <%s> ...\n",j,re);
parse_re(nfa, re);
}
}
i=0;
free(re);
re=allocate_char_array(4000);
}
}else{
re[i++]=c;
}
c=fgetc(file);
} //end while
if(i!=0){
re[i]='\0';
if (re[0]!= '#'){
j++;
if (j>=from && (to==-1 || j<=to)){
fprintf(stdout,"%dth regex: %s\n",j,re);
if (DEBUG) fprintf(stdout,"\n%d) processing regex:: <%s> ...\n",j,re);
parse_re(nfa,re);
}
}
free(re);
}
if (DEBUG) fprintf(stdout, "\nAll RegEx processed\n");
//if (re!=NULL) free(re);
//handle -m modifier
if (m_modifier && (!anchored->get_epsilon()->empty() || !anchored->get_transitions()->empty())){
non_anchored->add_transition('\n',anchored);
non_anchored->add_transition('\r',anchored);
}
//delete non_anchored, if necessary
if(non_anchored->get_epsilon()->empty() && non_anchored->get_transitions()->empty()){
nfa->get_epsilon()->remove(non_anchored);
delete non_anchored;
}else{
non_anchored->add_any(non_anchored);
}
return nfa->get_first();
}
NFA *regex_parser::group_regex(FILE *file, int group[]){
rewind(file);
char *re=allocate_char_array(1000);
int i=0, j=0, k=0;
int size=group[0];
unsigned int c=fgetc(file);
// NFA
NFA *nfa=new NFA();
NFA *non_anchored = nfa->add_epsilon(); // for .* RegEx
NFA *anchored = nfa->add_epsilon(); // for anchored RegEx (^)
// parsing the RegEx and putting them in a NFA
int min_j = group[1];
int max_j = group[1];
for (k=1; k<=size; k++){
if (group[k] > max_j)
max_j = group[k];
if (group[k] < min_j)
min_j = group[k];
}
if (DEBUG) fprintf(stdout, "@\n");
while (c!=EOF){
if (c=='\n' || c=='\r'){
if (i!=0){
re[i]='\0';
if (re[0]!='#'){
j++;
if (j>max_j)
break;
for (k=1; k<=size; k++){
if (j==group[k]){
if (DEBUG) fprintf(stdout,"%d) preprocessing regex:: <%s> ...\n",j,re);
//if (DEBUG) printf("%d ", j);
parse_re(nfa, re);
break;
}
if (j<min_j)
break;
}
}
i=0;
free(re);
re=NULL;
re=allocate_char_array(1000);
}
}
else{
re[i++]=c;
}
c=fgetc(file);
} //end while
if (i!=0){
re[i]='\0';
if (re[0]!= '#'){
j++;
if (j<=max_j)
for (k=1; k<=size; k++){
if (j==group[k]){
if (DEBUG) fprintf(stdout,"\n%d) preprocessing regex:: <%s> ...\n",j,re);
//if (DEBUG) printf("%d ", j);
parse_re(nfa, re);
break;
}
if (j<min_j)
break;
}
}
free(re);
re=NULL;
}
if (DEBUG) fprintf(stdout, "@\n");
if (DEBUG) fprintf(stdout, "All RegEx processed\n");
if (re!=NULL) free(re);
//handle -m modifier
if (m_modifier && (!anchored->get_epsilon()->empty() || !anchored->get_transitions()->empty())){
non_anchored->add_transition('\n',anchored);
non_anchored->add_transition('\r',anchored);
}
//delete non_anchored, if necessary
if (non_anchored->get_epsilon()->empty() && non_anchored->get_transitions()->empty()){
nfa->get_epsilon()->remove(non_anchored);
delete non_anchored;
}
else{
non_anchored->add_any(non_anchored);
}
return nfa->get_first();
}