Simple C++ Lexer

lexer.cpp

// A simple Lexer meant to demonstrate a few theoretical concepts. It can
// support several parser concepts and is very fast (though speed is not its
// design goal).
//
// J. Arrieta, Nabla Zero Labs
//
// This code is released under the MIT License.
//
// Copyright 2018 Nabla Zero Labs
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
// of this software and associated documentation files(the "Software"), to deal
// in the Software without restriction, including without limitation the rights
// to use, copy, modify, merge, publish ,distribute, sublicense, and / or sell
// copies of the Software, and to permit persons to whom the Software is
// furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
// SOFTWARE.

#include <string>

class Token {
 public:
  enum class Kind {
    Number,
    Identifier,
    LeftParen,
    RightParen,
    LeftSquare,
    RightSquare,
    LeftCurly,
    RightCurly,
    LessThan,
    GreaterThan,
    Equal,
    Plus,
    Minus,
    Asterisk,
    Slash,
    Hash,
    Dot,
    Comma,
    Colon,
    Semicolon,
    SingleQuote,
    DoubleQuote,
    Comment,
    Pipe,
    End,
    Unexpected,
  };

  Token(Kind kind) noexcept : m_kind{kind} {}

  Token(Kind kind, const char* beg, std::size_t len) noexcept
      : m_kind{kind}, m_lexeme(beg, len) {}

  Token(Kind kind, const char* beg, const char* end) noexcept
      : m_kind{kind}, m_lexeme(beg, std::distance(beg, end)) {}

  Kind kind() const noexcept { return m_kind; }

  void kind(Kind kind) noexcept { m_kind = kind; }

  bool is(Kind kind) const noexcept { return m_kind == kind; }

  bool is_not(Kind kind) const noexcept { return m_kind != kind; }

  bool is_one_of(Kind k1, Kind k2) const noexcept { return is(k1) || is(k2); }

  template <typename... Ts>
  bool is_one_of(Kind k1, Kind k2, Ts... ks) const noexcept {
    return is(k1) || is_one_of(k2, ks...);
  }

  std::string_view lexeme() const noexcept { return m_lexeme; }

  void lexeme(std::string_view lexeme) noexcept {
    m_lexeme = std::move(lexeme);
  }

 private:
  Kind             m_kind{};
  std::string_view m_lexeme{};
};

class Lexer {
 public:
  Lexer(const char* beg) noexcept : m_beg{beg} {}

  Token next() noexcept;

 private:
  Token identifier() noexcept;
  Token number() noexcept;
  Token slash_or_comment() noexcept;
  Token atom(Token::Kind) noexcept;

  char peek() const noexcept { return *m_beg; }
  char get() noexcept { return *m_beg++; }

  const char* m_beg = nullptr;
};

bool is_space(char c) noexcept {
  switch (c) {
    case ' ':
    case '\t':
    case '\r':
    case '\n':
      return true;
    default:
      return false;
  }
}

bool is_digit(char c) noexcept {
  switch (c) {
    case '0':
    case '1':
    case '2':
    case '3':
    case '4':
    case '5':
    case '6':
    case '7':
    case '8':
    case '9':
      return true;
    default:
      return false;
  }
}

bool is_identifier_char(char c) noexcept {
  switch (c) {
    case 'a':
    case 'b':
    case 'c':
    case 'd':
    case 'e':
    case 'f':
    case 'g':
    case 'h':
    case 'i':
    case 'j':
    case 'k':
    case 'l':
    case 'm':
    case 'n':
    case 'o':
    case 'p':
    case 'q':
    case 'r':
    case 's':
    case 't':
    case 'u':
    case 'v':
    case 'w':
    case 'x':
    case 'y':
    case 'z':
    case 'A':
    case 'B':
    case 'C':
    case 'D':
    case 'E':
    case 'F':
    case 'G':
    case 'H':
    case 'I':
    case 'J':
    case 'K':
    case 'L':
    case 'M':
    case 'N':
    case 'O':
    case 'P':
    case 'Q':
    case 'R':
    case 'S':
    case 'T':
    case 'U':
    case 'V':
    case 'W':
    case 'X':
    case 'Y':
    case 'Z':
    case '0':
    case '1':
    case '2':
    case '3':
    case '4':
    case '5':
    case '6':
    case '7':
    case '8':
    case '9':
    case '_':
      return true;
    default:
      return false;
  }
}

Token Lexer::atom(Token::Kind kind) noexcept { return Token(kind, m_beg++, 1); }

Token Lexer::next() noexcept {
  while (is_space(peek())) get();

  switch (peek()) {
    case '\0':
      return Token(Token::Kind::End, m_beg, 1);
    default:
      return atom(Token::Kind::Unexpected);
    case 'a':
    case 'b':
    case 'c':
    case 'd':
    case 'e':
    case 'f':
    case 'g':
    case 'h':
    case 'i':
    case 'j':
    case 'k':
    case 'l':
    case 'm':
    case 'n':
    case 'o':
    case 'p':
    case 'q':
    case 'r':
    case 's':
    case 't':
    case 'u':
    case 'v':
    case 'w':
    case 'x':
    case 'y':
    case 'z':
    case 'A':
    case 'B':
    case 'C':
    case 'D':
    case 'E':
    case 'F':
    case 'G':
    case 'H':
    case 'I':
    case 'J':
    case 'K':
    case 'L':
    case 'M':
    case 'N':
    case 'O':
    case 'P':
    case 'Q':
    case 'R':
    case 'S':
    case 'T':
    case 'U':
    case 'V':
    case 'W':
    case 'X':
    case 'Y':
    case 'Z':
      return identifier();
    case '0':
    case '1':
    case '2':
    case '3':
    case '4':
    case '5':
    case '6':
    case '7':
    case '8':
    case '9':
      return number();
    case '(':
      return atom(Token::Kind::LeftParen);
    case ')':
      return atom(Token::Kind::RightParen);
    case '[':
      return atom(Token::Kind::LeftSquare);
    case ']':
      return atom(Token::Kind::RightSquare);
    case '{':
      return atom(Token::Kind::LeftCurly);
    case '}':
      return atom(Token::Kind::RightCurly);
    case '<':
      return atom(Token::Kind::LessThan);
    case '>':
      return atom(Token::Kind::GreaterThan);
    case '=':
      return atom(Token::Kind::Equal);
    case '+':
      return atom(Token::Kind::Plus);
    case '-':
      return atom(Token::Kind::Minus);
    case '*':
      return atom(Token::Kind::Asterisk);
    case '/':
      return slash_or_comment();
    case '#':
      return atom(Token::Kind::Hash);
    case '.':
      return atom(Token::Kind::Dot);
    case ',':
      return atom(Token::Kind::Comma);
    case ':':
      return atom(Token::Kind::Colon);
    case ';':
      return atom(Token::Kind::Semicolon);
    case '\'':
      return atom(Token::Kind::SingleQuote);
    case '"':
      return atom(Token::Kind::DoubleQuote);
    case '|':
      return atom(Token::Kind::Pipe);
  }
}

Token Lexer::identifier() noexcept {
  const char* start = m_beg;
  get();
  while (is_identifier_char(peek())) get();
  return Token(Token::Kind::Identifier, start, m_beg);
}

Token Lexer::number() noexcept {
  const char* start = m_beg;
  get();
  while (is_digit(peek())) get();
  return Token(Token::Kind::Number, start, m_beg);
}

Token Lexer::slash_or_comment() noexcept {
  const char* start = m_beg;
  get();
  if (peek() == '/') {
    get();
    start = m_beg;
    while (peek() != '\0') {
      if (get() == '\n') {
        return Token(Token::Kind::Comment, start,
                     std::distance(start, m_beg) - 1);
      }
    }
    return Token(Token::Kind::Unexpected, m_beg, 1);
  } else {
    return Token(Token::Kind::Slash, start, 1);
  }
}

#include <iomanip>
#include <iostream>

std::ostream& operator<<(std::ostream& os, const Token::Kind& kind) {
  static const char* const names[]{
      "Number",      "Identifier",  "LeftParen",  "RightParen", "LeftSquare",
      "RightSquare", "LeftCurly",   "RightCurly", "LessThan",   "GreaterThan",
      "Equal",       "Plus",        "Minus",      "Asterisk",   "Slash",
      "Hash",        "Dot",         "Comma",      "Colon",      "Semicolon",
      "SingleQuote", "DoubleQuote", "Comment",    "Pipe",       "End",
      "Unexpected",
  };
  return os << names[static_cast<int>(kind)];
}

int main() {
  auto code =
      "x = 2\n"
      "// This is a comment.\n"
      "var x\n"
      "var y\n"
      "var f = function(x, y) { sin(x) * sin(y) + x * y; }\n"
      "der(f, x)\n"
      "var g = function(x, y) { 2 * (x + der(f, y)); } // der(f, y) is a "
      "matrix\n"
      "var r{3}; // Vector of three elements\n"
      "var J{12, 12}; // Matrix of 12x12 elements\n"
      "var dot = function(u{:}, v{:}) -> scalar {\n"
      "          return u[i] * v[i]; // Einstein notation\n"
      "}\n"
      "var norm = function(u{:}) -> scalar { return sqrt(dot(u, u)); }\n"
      "<end>";

  Lexer lex(code);
  for (auto token = lex.next();
       not token.is_one_of(Token::Kind::End, Token::Kind::Unexpected);
       token = lex.next()) {
    std::cout << std::setw(12) << token.kind() << " |" << token.lexeme()
              << "|\n";
  }
}

Hello onurblt and asmwarrior. The code is released under the MIT License (I've updated the gist to reflect this fact). I hope you find it useful.

Hi, thanks for sharing your code. Is there any chance to add support for floats? If so, it would be better to use a finite state machine or just a regular expression?

Nice and easy example. Thanks! :)

@ice-bit for handling decimals as numbers just like those with integers is done
add this function:

bool is_decimal(char c) noexcept {
    switch(c) {
        case '.':
            return true;
        default:
            return false;
    }
}

and in the function Token Lexer::number() just make this very small change:
while (is_digit(peek()) || is_decimal(peek())) get();

With the following input

auto code =
      "x = 2.066721823991;\n"
      "var a = a.b()";

Output is:

Identifier |x|
Equal |=|
Number |2.066721823991|
Semicolon |;|
Identifier |var|
Identifier |a|
Equal |=|
Identifier |a|
Dot |.|
Identifier |b|
LeftParen |(|
RightParen |)|

@godoio the problem with that approach is that it would recognize, for example, 123.456.789 as a valid number. Parsing floats is not too difficult, but certainly much more involved. For that reason, I believe that floats belong in a parser, not in a lexer.

Nice code. Why not use the built in C functions instead of defining your own for is_digit, etc? You can just us the C function isdigit for numbers, isalpha for identifiers, etc.

@jakerieger No particular reason. Standard library functions would have worked.

As a minor detail, std::isdigit and std::isalpha expect an int. When one writes a lexer that lexes char, that's not a problem. However, if one were to write a lexer that lexes a stream of codepoints, then one would need to write their own is_xxx.

@arrieta Makes sense

MIT License

What would you recommend to anyone you want to make one form scratch with basic knowledge of C++ working on for my programing Language.