Make inputparser more robust + add minimize
This commit is contained in:
parent
a4dbf3d195
commit
81f3c022c8
|
@ -12,24 +12,51 @@ namespace lxs {
|
||||||
auto enfavect = parseLines(is);
|
auto enfavect = parseLines(is);
|
||||||
auto enfas = linesToEnfa(enfavect);
|
auto enfas = linesToEnfa(enfavect);
|
||||||
auto enfa = merge(enfas);
|
auto enfa = merge(enfas);
|
||||||
auto dfa = mssc(enfa);
|
auto dfa = minimize(mssc(enfa));
|
||||||
return dfa;
|
return dfa;
|
||||||
}
|
}
|
||||||
|
|
||||||
std::vector<std::pair<std::string,std::string> > InputParser::parseLines(std::istream &is) {
|
std::vector<std::pair<std::string,std::string> > InputParser::parseLines(std::istream &is) {
|
||||||
std::string line;
|
std::string line;
|
||||||
std::vector<std::pair<std::string,std::string> > result;
|
std::vector<std::pair<std::string,std::string> > result;
|
||||||
|
unsigned int i=0;
|
||||||
while(std::getline(is,line)) {
|
while(std::getline(is,line)) {
|
||||||
int loc = line.find_first_of('=');
|
if(line.length() == 0) continue;
|
||||||
int start = line.find_first_not_of(" \t\v\f\r");
|
i++;
|
||||||
|
std::size_t loc = line.find_first_of('=');
|
||||||
|
if(loc == std::string::npos) {
|
||||||
|
std::cerr << "Invalid syntax on line " << i << ": no '=' found!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
|
size_t start = line.find_first_not_of(" \t\v\f\r");
|
||||||
|
if(start == loc) {
|
||||||
|
std::cerr << "Invalid syntax on line " << i << ": no valid tokenname specified!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
std::string tokenname = line.substr(start, loc);
|
std::string tokenname = line.substr(start, loc);
|
||||||
int end = tokenname.find_last_not_of(" \t\v\f\r");
|
if(tokenname.length() == 0) {
|
||||||
|
std::cerr << "Invalid syntax on line " << i << ": no valid tokenname specified!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
|
std::size_t end = tokenname.find_last_not_of(" \t\v\f\r");
|
||||||
tokenname = tokenname.substr(0,end + 1);
|
tokenname = tokenname.substr(0,end + 1);
|
||||||
std::string regex = line.substr(loc+1);
|
std::string regex = line.substr(loc+1);
|
||||||
start = regex.find_first_not_of(" ");
|
if(regex.length() == 0) {
|
||||||
|
std::cerr << "Invalid syntax on line " << i << ": no valid regex specified!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
|
start = regex.find_first_not_of(" \t\v\f\r");
|
||||||
|
if(start == std::string::npos) {
|
||||||
|
std::cerr << "Invalid syntax on line " << i << ": no valid regex specified!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
regex = regex.substr(start);
|
regex = regex.substr(start);
|
||||||
result.push_back(std::make_pair(tokenname,regex));
|
result.push_back(std::make_pair(tokenname,regex));
|
||||||
}
|
}
|
||||||
|
if(result.size() == 0) {
|
||||||
|
std::cerr << "No valid rules found in the input file!" << std::endl;
|
||||||
|
exit(1);
|
||||||
|
}
|
||||||
return result;
|
return result;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
Loading…
Reference in New Issue