2020-03-11 18:27:43 +00:00
|
|
|
/*
|
|
|
|
* Copyright (c) 2020, Stephan Unverwerth <s.unverwerth@gmx.de>
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions are met:
|
|
|
|
*
|
|
|
|
* 1. Redistributions of source code must retain the above copyright notice, this
|
|
|
|
* list of conditions and the following disclaimer.
|
|
|
|
*
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright notice,
|
|
|
|
* this list of conditions and the following disclaimer in the documentation
|
|
|
|
* and/or other materials provided with the distribution.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
|
|
|
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
|
|
|
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
|
|
|
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
|
|
|
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
|
|
|
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
|
|
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "Lexer.h"
|
|
|
|
#include <AK/HashMap.h>
|
|
|
|
#include <AK/StringBuilder.h>
|
|
|
|
#include <ctype.h>
|
|
|
|
#include <stdio.h>
|
|
|
|
|
|
|
|
namespace JS {
|
|
|
|
|
|
|
|
HashMap<String, TokenType> Lexer::s_keywords;
|
2020-03-12 12:11:33 +00:00
|
|
|
HashMap<String, TokenType> Lexer::s_three_char_tokens;
|
2020-03-11 18:27:43 +00:00
|
|
|
HashMap<String, TokenType> Lexer::s_two_char_tokens;
|
|
|
|
HashMap<char, TokenType> Lexer::s_single_char_tokens;
|
|
|
|
|
|
|
|
Lexer::Lexer(StringView source)
|
|
|
|
: m_source(source)
|
2020-04-05 09:34:03 +00:00
|
|
|
, m_current_token(TokenType::Eof, StringView(nullptr), StringView(nullptr), 0, 0)
|
2020-03-11 18:27:43 +00:00
|
|
|
{
|
|
|
|
if (s_keywords.is_empty()) {
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("await", TokenType::Await);
|
2020-03-29 11:09:54 +00:00
|
|
|
s_keywords.set("break", TokenType::Break);
|
|
|
|
s_keywords.set("case", TokenType::Case);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("catch", TokenType::Catch);
|
|
|
|
s_keywords.set("class", TokenType::Class);
|
|
|
|
s_keywords.set("const", TokenType::Const);
|
2020-04-04 22:22:42 +00:00
|
|
|
s_keywords.set("continue", TokenType::Continue);
|
2020-04-30 16:26:27 +00:00
|
|
|
s_keywords.set("debugger", TokenType::Debugger);
|
2020-03-29 11:09:54 +00:00
|
|
|
s_keywords.set("default", TokenType::Default);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("delete", TokenType::Delete);
|
|
|
|
s_keywords.set("do", TokenType::Do);
|
|
|
|
s_keywords.set("else", TokenType::Else);
|
2020-05-12 16:43:35 +00:00
|
|
|
s_keywords.set("enum", TokenType::Enum);
|
|
|
|
s_keywords.set("export", TokenType::Export);
|
|
|
|
s_keywords.set("extends", TokenType::Extends);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("false", TokenType::BoolLiteral);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("finally", TokenType::Finally);
|
2020-03-12 12:12:12 +00:00
|
|
|
s_keywords.set("for", TokenType::For);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("function", TokenType::Function);
|
|
|
|
s_keywords.set("if", TokenType::If);
|
2020-05-12 16:43:35 +00:00
|
|
|
s_keywords.set("import", TokenType::Import);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("in", TokenType::In);
|
|
|
|
s_keywords.set("instanceof", TokenType::Instanceof);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("let", TokenType::Let);
|
|
|
|
s_keywords.set("new", TokenType::New);
|
|
|
|
s_keywords.set("null", TokenType::NullLiteral);
|
|
|
|
s_keywords.set("return", TokenType::Return);
|
2020-05-12 16:43:35 +00:00
|
|
|
s_keywords.set("super", TokenType::Super);
|
2020-03-29 11:09:54 +00:00
|
|
|
s_keywords.set("switch", TokenType::Switch);
|
2020-04-12 22:42:14 +00:00
|
|
|
s_keywords.set("this", TokenType::This);
|
2020-03-24 21:03:50 +00:00
|
|
|
s_keywords.set("throw", TokenType::Throw);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("true", TokenType::BoolLiteral);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("try", TokenType::Try);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("typeof", TokenType::Typeof);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("var", TokenType::Var);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("void", TokenType::Void);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_keywords.set("while", TokenType::While);
|
2020-05-12 16:43:35 +00:00
|
|
|
s_keywords.set("with", TokenType::With);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_keywords.set("yield", TokenType::Yield);
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
2020-03-12 12:11:33 +00:00
|
|
|
if (s_three_char_tokens.is_empty()) {
|
|
|
|
s_three_char_tokens.set("===", TokenType::EqualsEqualsEquals);
|
|
|
|
s_three_char_tokens.set("!==", TokenType::ExclamationMarkEqualsEquals);
|
2020-05-04 22:03:35 +00:00
|
|
|
s_three_char_tokens.set("**=", TokenType::DoubleAsteriskEquals);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_three_char_tokens.set("<<=", TokenType::ShiftLeftEquals);
|
|
|
|
s_three_char_tokens.set(">>=", TokenType::ShiftRightEquals);
|
|
|
|
s_three_char_tokens.set(">>>", TokenType::UnsignedShiftRight);
|
2020-04-27 06:05:37 +00:00
|
|
|
s_three_char_tokens.set("...", TokenType::TripleDot);
|
2020-03-12 12:11:33 +00:00
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
if (s_two_char_tokens.is_empty()) {
|
2020-03-30 13:26:09 +00:00
|
|
|
s_two_char_tokens.set("=>", TokenType::Arrow);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_two_char_tokens.set("+=", TokenType::PlusEquals);
|
|
|
|
s_two_char_tokens.set("-=", TokenType::MinusEquals);
|
|
|
|
s_two_char_tokens.set("*=", TokenType::AsteriskEquals);
|
|
|
|
s_two_char_tokens.set("/=", TokenType::SlashEquals);
|
|
|
|
s_two_char_tokens.set("%=", TokenType::PercentEquals);
|
|
|
|
s_two_char_tokens.set("&=", TokenType::AmpersandEquals);
|
|
|
|
s_two_char_tokens.set("|=", TokenType::PipeEquals);
|
2020-05-04 21:34:45 +00:00
|
|
|
s_two_char_tokens.set("^=", TokenType::CaretEquals);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_two_char_tokens.set("&&", TokenType::DoubleAmpersand);
|
|
|
|
s_two_char_tokens.set("||", TokenType::DoublePipe);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_two_char_tokens.set("??", TokenType::DoubleQuestionMark);
|
|
|
|
s_two_char_tokens.set("**", TokenType::DoubleAsterisk);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_two_char_tokens.set("==", TokenType::EqualsEquals);
|
2020-03-23 13:10:23 +00:00
|
|
|
s_two_char_tokens.set("<=", TokenType::LessThanEquals);
|
|
|
|
s_two_char_tokens.set(">=", TokenType::GreaterThanEquals);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_two_char_tokens.set("!=", TokenType::ExclamationMarkEquals);
|
|
|
|
s_two_char_tokens.set("--", TokenType::MinusMinus);
|
|
|
|
s_two_char_tokens.set("++", TokenType::PlusPlus);
|
|
|
|
s_two_char_tokens.set("<<", TokenType::ShiftLeft);
|
|
|
|
s_two_char_tokens.set(">>", TokenType::ShiftRight);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_two_char_tokens.set("?.", TokenType::QuestionMarkPeriod);
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (s_single_char_tokens.is_empty()) {
|
|
|
|
s_single_char_tokens.set('&', TokenType::Ampersand);
|
|
|
|
s_single_char_tokens.set('*', TokenType::Asterisk);
|
|
|
|
s_single_char_tokens.set('[', TokenType::BracketOpen);
|
|
|
|
s_single_char_tokens.set(']', TokenType::BracketClose);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_single_char_tokens.set('^', TokenType::Caret);
|
2020-03-21 00:29:00 +00:00
|
|
|
s_single_char_tokens.set(':', TokenType::Colon);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_single_char_tokens.set(',', TokenType::Comma);
|
|
|
|
s_single_char_tokens.set('{', TokenType::CurlyOpen);
|
|
|
|
s_single_char_tokens.set('}', TokenType::CurlyClose);
|
|
|
|
s_single_char_tokens.set('=', TokenType::Equals);
|
|
|
|
s_single_char_tokens.set('!', TokenType::ExclamationMark);
|
|
|
|
s_single_char_tokens.set('-', TokenType::Minus);
|
|
|
|
s_single_char_tokens.set('(', TokenType::ParenOpen);
|
|
|
|
s_single_char_tokens.set(')', TokenType::ParenClose);
|
|
|
|
s_single_char_tokens.set('%', TokenType::Percent);
|
|
|
|
s_single_char_tokens.set('.', TokenType::Period);
|
|
|
|
s_single_char_tokens.set('|', TokenType::Pipe);
|
|
|
|
s_single_char_tokens.set('+', TokenType::Plus);
|
|
|
|
s_single_char_tokens.set('?', TokenType::QuestionMark);
|
|
|
|
s_single_char_tokens.set(';', TokenType::Semicolon);
|
|
|
|
s_single_char_tokens.set('/', TokenType::Slash);
|
2020-03-12 22:02:41 +00:00
|
|
|
s_single_char_tokens.set('~', TokenType::Tilde);
|
2020-03-11 18:27:43 +00:00
|
|
|
s_single_char_tokens.set('<', TokenType::LessThan);
|
|
|
|
s_single_char_tokens.set('>', TokenType::GreaterThan);
|
|
|
|
}
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
|
|
|
|
void Lexer::consume()
|
|
|
|
{
|
2020-03-13 20:29:32 +00:00
|
|
|
if (m_position >= m_source.length()) {
|
|
|
|
m_position = m_source.length() + 1;
|
2020-03-11 18:27:43 +00:00
|
|
|
m_current_char = EOF;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2020-04-05 09:34:03 +00:00
|
|
|
if (m_current_char == '\n') {
|
|
|
|
m_line_number++;
|
|
|
|
m_line_column = 1;
|
|
|
|
} else {
|
|
|
|
m_line_column++;
|
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
m_current_char = m_source[m_position++];
|
|
|
|
}
|
|
|
|
|
2020-04-05 12:20:58 +00:00
|
|
|
void Lexer::consume_exponent()
|
|
|
|
{
|
|
|
|
consume();
|
|
|
|
if (m_current_char == '-' || m_current_char == '+')
|
|
|
|
consume();
|
|
|
|
while (isdigit(m_current_char)) {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-04-13 17:50:58 +00:00
|
|
|
bool Lexer::match(char a, char b) const
|
|
|
|
{
|
|
|
|
if (m_position >= m_source.length())
|
|
|
|
return false;
|
|
|
|
|
|
|
|
return m_current_char == a
|
|
|
|
&& m_source[m_position] == b;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::match(char a, char b, char c) const
|
|
|
|
{
|
|
|
|
if (m_position + 1 >= m_source.length())
|
|
|
|
return false;
|
|
|
|
|
|
|
|
return m_current_char == a
|
|
|
|
&& m_source[m_position] == b
|
|
|
|
&& m_source[m_position + 1] == c;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::match(char a, char b, char c, char d) const
|
|
|
|
{
|
|
|
|
if (m_position + 2 >= m_source.length())
|
|
|
|
return false;
|
|
|
|
|
|
|
|
return m_current_char == a
|
|
|
|
&& m_source[m_position] == b
|
|
|
|
&& m_source[m_position + 1] == c
|
|
|
|
&& m_source[m_position + 2] == d;
|
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
bool Lexer::is_eof() const
|
|
|
|
{
|
2020-03-13 20:29:32 +00:00
|
|
|
return m_current_char == EOF;
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::is_identifier_start() const
|
|
|
|
{
|
|
|
|
return isalpha(m_current_char) || m_current_char == '_' || m_current_char == '$';
|
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::is_identifier_middle() const
|
|
|
|
{
|
|
|
|
return is_identifier_start() || isdigit(m_current_char);
|
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::is_line_comment_start() const
|
|
|
|
{
|
2020-04-13 17:50:58 +00:00
|
|
|
return match('/', '/') || match('<', '!', '-', '-') || match('-', '-', '>');
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::is_block_comment_start() const
|
|
|
|
{
|
2020-04-13 17:50:58 +00:00
|
|
|
return match('/', '*');
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
bool Lexer::is_block_comment_end() const
|
|
|
|
{
|
2020-04-13 17:50:58 +00:00
|
|
|
return match('*', '/');
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
2020-04-05 12:20:58 +00:00
|
|
|
bool Lexer::is_numeric_literal_start() const
|
|
|
|
{
|
|
|
|
return isdigit(m_current_char) || (m_current_char == '.' && m_position < m_source.length() && isdigit(m_source[m_position]));
|
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
Token Lexer::next()
|
|
|
|
{
|
|
|
|
size_t trivia_start = m_position;
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
auto in_template = !m_template_states.is_empty();
|
2020-03-11 18:27:43 +00:00
|
|
|
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
if (!in_template || m_template_states.last().in_expr) {
|
|
|
|
// consume whitespace and comments
|
|
|
|
while (true) {
|
|
|
|
if (isspace(m_current_char)) {
|
|
|
|
do {
|
|
|
|
consume();
|
|
|
|
} while (isspace(m_current_char));
|
|
|
|
} else if (is_line_comment_start()) {
|
2020-03-11 18:27:43 +00:00
|
|
|
consume();
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
do {
|
|
|
|
consume();
|
|
|
|
} while (!is_eof() && m_current_char != '\n');
|
|
|
|
} else if (is_block_comment_start()) {
|
2020-03-11 18:27:43 +00:00
|
|
|
consume();
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
do {
|
|
|
|
consume();
|
|
|
|
} while (!is_eof() && !is_block_comment_end());
|
|
|
|
consume(); // consume *
|
|
|
|
consume(); // consume /
|
|
|
|
} else {
|
|
|
|
break;
|
|
|
|
}
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
size_t value_start = m_position;
|
2020-03-12 12:11:33 +00:00
|
|
|
auto token_type = TokenType::Invalid;
|
2020-03-11 18:27:43 +00:00
|
|
|
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
if (m_current_char == '`') {
|
|
|
|
consume();
|
|
|
|
|
|
|
|
if (!in_template) {
|
|
|
|
token_type = TokenType::TemplateLiteralStart;
|
|
|
|
m_template_states.append({ false, 0 });
|
|
|
|
} else {
|
|
|
|
if (m_template_states.last().in_expr) {
|
|
|
|
m_template_states.append({ false, 0 });
|
|
|
|
token_type = TokenType::TemplateLiteralStart;
|
|
|
|
} else {
|
|
|
|
m_template_states.take_last();
|
|
|
|
token_type = TokenType::TemplateLiteralEnd;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else if (in_template && m_template_states.last().in_expr && m_template_states.last().open_bracket_count == 0 && m_current_char == '}') {
|
|
|
|
consume();
|
|
|
|
token_type = TokenType::TemplateLiteralExprEnd;
|
|
|
|
m_template_states.last().in_expr = false;
|
|
|
|
} else if (in_template && !m_template_states.last().in_expr) {
|
|
|
|
if (is_eof()) {
|
|
|
|
token_type = TokenType::UnterminatedTemplateLiteral;
|
|
|
|
m_template_states.take_last();
|
|
|
|
} else if (match('$', '{')) {
|
|
|
|
token_type = TokenType::TemplateLiteralExprStart;
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
m_template_states.last().in_expr = true;
|
|
|
|
} else {
|
|
|
|
while (!match('$', '{') && m_current_char != '`' && !is_eof()) {
|
|
|
|
if (match('\\', '$') || match('\\', '`'))
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
|
|
|
|
token_type = TokenType::TemplateLiteralString;
|
|
|
|
}
|
|
|
|
} else if (is_identifier_start()) {
|
2020-03-11 18:27:43 +00:00
|
|
|
// identifier or keyword
|
|
|
|
do {
|
|
|
|
consume();
|
|
|
|
} while (is_identifier_middle());
|
|
|
|
|
|
|
|
StringView value = m_source.substring_view(value_start - 1, m_position - value_start);
|
|
|
|
auto it = s_keywords.find(value);
|
|
|
|
if (it == s_keywords.end()) {
|
|
|
|
token_type = TokenType::Identifier;
|
|
|
|
} else {
|
|
|
|
token_type = it->value;
|
|
|
|
}
|
2020-04-05 12:20:58 +00:00
|
|
|
} else if (is_numeric_literal_start()) {
|
|
|
|
if (m_current_char == '0') {
|
2020-03-11 18:27:43 +00:00
|
|
|
consume();
|
2020-04-05 12:20:58 +00:00
|
|
|
if (m_current_char == '.') {
|
|
|
|
// decimal
|
|
|
|
consume();
|
|
|
|
while (isdigit(m_current_char)) {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
if (m_current_char == 'e' || m_current_char == 'E') {
|
|
|
|
consume_exponent();
|
|
|
|
}
|
|
|
|
} else if (m_current_char == 'e' || m_current_char == 'E') {
|
|
|
|
consume_exponent();
|
|
|
|
} else if (m_current_char == 'o' || m_current_char == 'O') {
|
|
|
|
// octal
|
|
|
|
consume();
|
|
|
|
while (m_current_char >= '0' && m_current_char <= '7') {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
} else if (m_current_char == 'b' || m_current_char == 'B') {
|
|
|
|
// binary
|
|
|
|
consume();
|
|
|
|
while (m_current_char == '0' || m_current_char == '1') {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
} else if (m_current_char == 'x' || m_current_char == 'X') {
|
|
|
|
// hexadecimal
|
|
|
|
consume();
|
|
|
|
while (isxdigit(m_current_char)) {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
} else if (isdigit(m_current_char)) {
|
|
|
|
// octal without 'O' prefix. Forbidden in 'strict mode'
|
|
|
|
// FIXME: We need to make sure this produces a syntax error when in strict mode
|
|
|
|
do {
|
|
|
|
consume();
|
|
|
|
} while (isdigit(m_current_char));
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// 1...9 or period
|
|
|
|
while (isdigit(m_current_char)) {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
if (m_current_char == '.') {
|
|
|
|
consume();
|
|
|
|
while (isdigit(m_current_char)) {
|
|
|
|
consume();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (m_current_char == 'e' || m_current_char == 'E') {
|
|
|
|
consume_exponent();
|
|
|
|
}
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
token_type = TokenType::NumericLiteral;
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
} else if (m_current_char == '"' || m_current_char == '\'') {
|
2020-03-14 10:45:25 +00:00
|
|
|
char stop_char = m_current_char;
|
2020-03-12 12:05:06 +00:00
|
|
|
consume();
|
2020-03-14 10:45:25 +00:00
|
|
|
while (m_current_char != stop_char && m_current_char != '\n' && !is_eof()) {
|
|
|
|
if (m_current_char == '\\') {
|
|
|
|
consume();
|
|
|
|
}
|
2020-03-12 12:05:06 +00:00
|
|
|
consume();
|
|
|
|
}
|
2020-03-14 10:45:25 +00:00
|
|
|
if (m_current_char != stop_char) {
|
|
|
|
token_type = TokenType::UnterminatedStringLiteral;
|
|
|
|
} else {
|
|
|
|
consume();
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
token_type = TokenType::StringLiteral;
|
2020-03-14 10:45:25 +00:00
|
|
|
}
|
2020-03-11 18:27:43 +00:00
|
|
|
} else if (m_current_char == EOF) {
|
|
|
|
token_type = TokenType::Eof;
|
|
|
|
} else {
|
2020-03-12 22:02:41 +00:00
|
|
|
// There is only one four-char operator: >>>=
|
|
|
|
bool found_four_char_token = false;
|
2020-04-13 17:50:58 +00:00
|
|
|
if (match('>', '>', '>', '=')) {
|
|
|
|
found_four_char_token = true;
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
token_type = TokenType::UnsignedShiftRightEquals;
|
2020-03-12 22:02:41 +00:00
|
|
|
}
|
|
|
|
|
2020-03-12 12:11:33 +00:00
|
|
|
bool found_three_char_token = false;
|
2020-03-12 22:02:41 +00:00
|
|
|
if (!found_four_char_token && m_position + 1 < m_source.length()) {
|
2020-03-12 12:52:54 +00:00
|
|
|
char second_char = m_source[m_position];
|
|
|
|
char third_char = m_source[m_position + 1];
|
|
|
|
char three_chars[] { (char)m_current_char, second_char, third_char, 0 };
|
|
|
|
auto it = s_three_char_tokens.find(three_chars);
|
2020-03-12 12:11:33 +00:00
|
|
|
if (it != s_three_char_tokens.end()) {
|
|
|
|
found_three_char_token = true;
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
token_type = it->value;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
bool found_two_char_token = false;
|
2020-03-12 22:02:41 +00:00
|
|
|
if (!found_four_char_token && !found_three_char_token && m_position < m_source.length()) {
|
2020-03-12 12:52:54 +00:00
|
|
|
char second_char = m_source[m_position];
|
|
|
|
char two_chars[] { (char)m_current_char, second_char, 0 };
|
|
|
|
auto it = s_two_char_tokens.find(two_chars);
|
2020-03-11 18:27:43 +00:00
|
|
|
if (it != s_two_char_tokens.end()) {
|
|
|
|
found_two_char_token = true;
|
|
|
|
consume();
|
|
|
|
consume();
|
|
|
|
token_type = it->value;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-03-12 12:11:33 +00:00
|
|
|
bool found_one_char_token = false;
|
2020-03-12 22:02:41 +00:00
|
|
|
if (!found_four_char_token && !found_three_char_token && !found_two_char_token) {
|
2020-03-11 18:27:43 +00:00
|
|
|
auto it = s_single_char_tokens.find(m_current_char);
|
|
|
|
if (it != s_single_char_tokens.end()) {
|
2020-03-12 12:11:33 +00:00
|
|
|
found_one_char_token = true;
|
2020-03-11 18:27:43 +00:00
|
|
|
consume();
|
|
|
|
token_type = it->value;
|
|
|
|
}
|
|
|
|
}
|
2020-03-12 12:11:33 +00:00
|
|
|
|
2020-03-12 22:02:41 +00:00
|
|
|
if (!found_four_char_token && !found_three_char_token && !found_two_char_token && !found_one_char_token) {
|
2020-03-12 12:11:33 +00:00
|
|
|
consume();
|
|
|
|
token_type = TokenType::Invalid;
|
|
|
|
}
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|
|
|
|
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
if (!m_template_states.is_empty() && m_template_states.last().in_expr) {
|
|
|
|
if (token_type == TokenType::CurlyOpen) {
|
|
|
|
m_template_states.last().open_bracket_count++;
|
|
|
|
} else if (token_type == TokenType::CurlyClose) {
|
|
|
|
m_template_states.last().open_bracket_count--;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
m_current_token = Token(
|
|
|
|
token_type,
|
|
|
|
m_source.substring_view(trivia_start - 1, value_start - trivia_start),
|
2020-04-05 09:34:03 +00:00
|
|
|
m_source.substring_view(value_start - 1, m_position - value_start),
|
|
|
|
m_line_number,
|
2020-04-05 12:20:49 +00:00
|
|
|
m_line_column - m_position + value_start);
|
2020-03-11 18:27:43 +00:00
|
|
|
|
|
|
|
return m_current_token;
|
|
|
|
}
|
LibJS: Add template literals
Adds fully functioning template literals. Because template literals
contain expressions, most of the work has to be done in the Lexer rather
than the Parser. And because of the complexity of template literals
(expressions, nesting, escapes, etc), the Lexer needs to have some
template-related state.
When entering a new template literal, a TemplateLiteralStart token is
emitted. When inside a literal, all text will be parsed up until a '${'
or '`' (or EOF, but that's a syntax error) is seen, and then a
TemplateLiteralExprStart token is emitted. At this point, the Lexer
proceeds as normal, however it keeps track of the number of opening
and closing curly braces it has seen in order to determine the close
of the expression. Once it finds a matching curly brace for the '${',
a TemplateLiteralExprEnd token is emitted and the state is updated
accordingly.
When the Lexer is inside of a template literal, but not an expression,
and sees a '`', this must be the closing grave: a TemplateLiteralEnd
token is emitted.
The state required to correctly parse template strings consists of a
vector (for nesting) of two pieces of information: whether or not we
are in a template expression (as opposed to a template string); and
the count of the number of unmatched open curly braces we have seen
(only applicable if the Lexer is currently in a template expression).
TODO: Add support for template literal newlines in the JS REPL (this will
cause a syntax error currently):
> `foo
> bar`
'foo
bar'
2020-05-03 22:41:14 +00:00
|
|
|
|
2020-03-11 18:27:43 +00:00
|
|
|
}
|