276 lines
7.6 KiB
C++
276 lines
7.6 KiB
C++
#include "elna/lexer.hpp"
|
|
|
|
#include <cstring>
|
|
|
|
namespace elna
|
|
{
|
|
namespace lex
|
|
{
|
|
using source_position = elna::source::position;
|
|
using source_error = elna::source::error;
|
|
using source_result = elna::source::result<std::vector<Token>>;
|
|
|
|
source::source(const std::string& buffer)
|
|
: m_buffer(buffer)
|
|
{
|
|
}
|
|
|
|
source::const_iterator source::begin() const
|
|
{
|
|
return source::const_iterator(std::cbegin(m_buffer));
|
|
}
|
|
|
|
source::const_iterator source::end() const
|
|
{
|
|
source_position end_position{ 0, 0 };
|
|
|
|
return source::const_iterator(std::cend(m_buffer), end_position);
|
|
}
|
|
|
|
source::const_iterator::const_iterator(std::string::const_iterator buffer,
|
|
const source_position start_position)
|
|
: m_buffer(buffer), m_position(start_position)
|
|
{
|
|
}
|
|
|
|
const source_position& source::const_iterator::position() const noexcept
|
|
{
|
|
return this->m_position;
|
|
}
|
|
|
|
source::const_iterator::reference source::const_iterator::operator*() const noexcept
|
|
{
|
|
return *m_buffer;
|
|
}
|
|
|
|
source::const_iterator::pointer source::const_iterator::operator->() const noexcept
|
|
{
|
|
return m_buffer.base();
|
|
}
|
|
|
|
source::const_iterator& source::const_iterator::operator++()
|
|
{
|
|
if (*this->m_buffer == '\n')
|
|
{
|
|
this->m_position.column = 1;
|
|
++this->m_position.line;
|
|
}
|
|
else
|
|
{
|
|
++this->m_position.column;
|
|
}
|
|
std::advance(this->m_buffer, 1);
|
|
|
|
return *this;
|
|
}
|
|
|
|
source::const_iterator& source::const_iterator::operator++(int)
|
|
{
|
|
auto tmp = *this;
|
|
++(*this);
|
|
return *this;
|
|
}
|
|
|
|
bool source::const_iterator::operator==(const source::const_iterator& that) const noexcept
|
|
{
|
|
return this->m_buffer == that.m_buffer;
|
|
}
|
|
|
|
bool source::const_iterator::operator!=(const source::const_iterator& that) const noexcept
|
|
{
|
|
return !(*this == that);
|
|
}
|
|
|
|
Token::Token(const type of, const char *value, source_position position)
|
|
: m_type(of), m_position(position)
|
|
{
|
|
std::size_t value_length = strlen(value);
|
|
char *buffer = reinterpret_cast<char *>(malloc(value_length + 1));
|
|
|
|
std::memcpy(buffer, value, value_length);
|
|
buffer[value_length] = 0;
|
|
|
|
m_value.identifier = buffer;
|
|
}
|
|
|
|
Token::Token(const type of, std::int32_t number, source_position position)
|
|
: m_type(of), m_position(position)
|
|
{
|
|
m_value.number = number;
|
|
}
|
|
|
|
Token::Token(const type of, source_position position)
|
|
: m_type(of), m_position(position)
|
|
{
|
|
}
|
|
|
|
Token::Token(const Token& that)
|
|
: m_type(that.of()), m_position(that.position())
|
|
{
|
|
*this = that;
|
|
}
|
|
|
|
Token::Token(Token&& that)
|
|
: m_type(that.of()), m_position(that.position())
|
|
{
|
|
*this = std::move(that);
|
|
}
|
|
|
|
Token::~Token()
|
|
{
|
|
if (m_type == type::identifier || m_type == type::term_operator || m_type == type::factor_operator)
|
|
{
|
|
std::free(const_cast<char*>(m_value.identifier));
|
|
}
|
|
}
|
|
|
|
Token& Token::operator=(const Token& that)
|
|
{
|
|
m_type = that.of();
|
|
m_position = that.position();
|
|
if (that.of() == type::identifier || that.of() == type::term_operator || m_type == type::factor_operator)
|
|
{
|
|
std::size_t value_length = strlen(that.identifier());
|
|
char *buffer = reinterpret_cast<char *>(malloc(value_length + 1));
|
|
|
|
std::memcpy(buffer, that.identifier(), value_length);
|
|
buffer[value_length] = 0;
|
|
|
|
m_value.identifier = buffer;
|
|
}
|
|
else if (that.of() == type::number)
|
|
{
|
|
m_value.number = that.number();
|
|
}
|
|
return *this;
|
|
}
|
|
|
|
Token& Token::operator=(Token&& that)
|
|
{
|
|
m_type = that.of();
|
|
m_position = that.position();
|
|
if (that.of() == type::identifier || that.of() == type::term_operator || that.of() == type::factor_operator)
|
|
{
|
|
m_value.identifier = that.identifier();
|
|
that.m_value.identifier = nullptr;
|
|
}
|
|
else if (that.of() == type::number)
|
|
{
|
|
m_value.number = that.number();
|
|
}
|
|
return *this;
|
|
}
|
|
|
|
Token::type Token::of() const noexcept
|
|
{
|
|
return m_type;
|
|
}
|
|
|
|
const char *Token::identifier() const noexcept
|
|
{
|
|
return m_value.identifier;
|
|
}
|
|
|
|
std::int32_t Token::number() const noexcept
|
|
{
|
|
return m_value.number;
|
|
}
|
|
|
|
const source_position& Token::position() const noexcept
|
|
{
|
|
return m_position;
|
|
}
|
|
|
|
source_result lex(const char *buffer)
|
|
{
|
|
std::vector<Token> tokens;
|
|
source input{ buffer };
|
|
|
|
for (auto iterator = input.begin(); iterator != input.end();)
|
|
{
|
|
if (*iterator == ' ' || *iterator == '\n')
|
|
{
|
|
}
|
|
else if (std::isdigit(*iterator))
|
|
{
|
|
tokens.emplace_back(
|
|
Token::type::number,
|
|
static_cast<std::int32_t>(*iterator - '0'),
|
|
iterator.position()
|
|
);
|
|
}
|
|
else if (*iterator == '=')
|
|
{
|
|
tokens.emplace_back(Token::type::equals, iterator.position());
|
|
}
|
|
else if (*iterator == '(')
|
|
{
|
|
tokens.emplace_back(Token::type::left_paren, iterator.position());
|
|
}
|
|
else if (*iterator == ')')
|
|
{
|
|
tokens.emplace_back(Token::type::right_paren, iterator.position());
|
|
}
|
|
else if (*iterator == ';')
|
|
{
|
|
tokens.emplace_back(Token::type::semicolon, iterator.position());
|
|
}
|
|
else if (*iterator == ',')
|
|
{
|
|
tokens.emplace_back(Token::type::comma, iterator.position());
|
|
}
|
|
else if (*iterator == '!')
|
|
{
|
|
tokens.emplace_back(Token::type::bang, iterator.position());
|
|
}
|
|
else if (*iterator == '.')
|
|
{
|
|
tokens.emplace_back(Token::type::dot, iterator.position());
|
|
}
|
|
else if (std::isalpha(*iterator))
|
|
{
|
|
std::string word;
|
|
auto i = iterator;
|
|
while (i != input.end() && std::isalpha(*i))
|
|
{
|
|
word.push_back(*i);
|
|
++i;
|
|
}
|
|
if (word == "const")
|
|
{
|
|
tokens.emplace_back(Token::type::let, iterator.position());
|
|
}
|
|
else if (word == "var")
|
|
{
|
|
tokens.emplace_back(Token::type::var, iterator.position());
|
|
}
|
|
else
|
|
{
|
|
tokens.emplace_back(Token::type::identifier, word.c_str(), iterator.position());
|
|
}
|
|
iterator = i;
|
|
continue;
|
|
}
|
|
else if (*iterator == '+' || *iterator == '-')
|
|
{
|
|
std::string _operator{ *iterator };
|
|
|
|
tokens.emplace_back(Token::type::term_operator, _operator.c_str(), iterator.position());
|
|
}
|
|
else if (*iterator == '*' || *iterator == '/')
|
|
{
|
|
std::string _operator{ *iterator };
|
|
|
|
tokens.emplace_back(Token::type::factor_operator, _operator.c_str(), iterator.position());
|
|
}
|
|
else
|
|
{
|
|
return source_result("Unexpected next character", iterator.position());
|
|
}
|
|
++iterator;
|
|
}
|
|
return source_result(tokens);
|
|
}
|
|
}
|
|
}
|