yuwenxiong/js-in-cpp

a JavaScript interpreter implement by C

JS-in-C

A JS interpreter in C++, implemented via recursive decent parsing.

Lex

Usage

Lex l(const string& str)

or

Lex l;
l.getLex();

after that, directly use l.tokens:

vector<Token> tokens;

class Token{
public:
    Token() {};
    Token(TOKEN_TYPES _type, const string& _value): type(_type), value(_value) {};
    void setToken(TOKEN_TYPES _type, const string& _value){
        type = _type;
        value = _value;
    }
    TOKEN_TYPES type;
    string value;
};

SUPPORTS

• " " is not required, tokens can be adjacent to each other.
• tokens are defined in TOKEN_TYPES

Numbers: signed decimal;

positive octal(prefix '0'), hex(prefix '0x', '0X');

signed float:

"^[\+-]?((([1-9]\d*)?\.\d+|[1-9]\d*(\.\d*)?)[eE][\+-]?[1-9]\d*|([1-9]\d*)?\.\d+)"

• Comments removed. To remove the Comments correctly, you need to add '\n' to every line so that I can judge the range of '\'-type comments.

And if the script contains multi-line comments, you need to read the whole text, then send to Lex object.

ifstream input(filePath, std::ios::in);

while (input.getline(buffer, 1024)) {
    string line(buffer);
    wholeText += line + endline;
}

RESTRICTIONS

1. the tokens that can appear before numbers with '+/-' prefix:

   set<TOKEN_TYPES> tokenBeforePrefix{TK_L_BRACKET, TK_ASSIGN, TK_PLUS_EQUAL, TK_MINUS_EQUAL, TK_MULTI_EQUAL, TK_DIV_EQUAL, TK_MOD_EQUAL, TK_AND_EQUAL, TK_OR_EQUAL, TK_L_EQUAL, TK_G_EQUAL, };

2. do not support operators '>>>' and '>>>='