RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 22:17:45 +00:00
Code Issues Activity Actions

LibCpp: Fix lexing & parsing of non-terminated strings

Browse source
This commit is contained in:
Itamar 2021-02-10 22:15:38 +02:00 • committed by Andreas Kling
parent 2da5ecba41
commit 8ace2cfa18
3 changed files with 25 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

29
Userland/Libraries/LibCpp/Parser.cpp
View file

@ -53,7 +53,7 @@ Parser::Parser(const StringView& program)
dbgln("{}", m_program);
dbgln("Tokens:");
for (auto& token : m_tokens) {
dbgln("{}", token.to_string());
dbgln("{} ({}:{}-{}:{})", token.to_string(), token.start().line, token.start().column, token.end().line, token.end().column);
}
#endif
}
@ -673,9 +673,14 @@ StringView Parser::text_of_token(const Cpp::Token& token) const
StringView Parser::text_of_node(const ASTNode& node) const
{
if (node.start().line == node.end().line) {
ASSERT(node.start().column <= node.end().column);
return m_lines[node.start().line].substring_view(node.start().column, node.end().column - node.start().column + 1);
return text_of_range(node.start(), node.end());
}
StringView Parser::text_of_range(Position start, Position end) const
{
if (start.line == end.line) {
ASSERT(start.column <= end.column);
return m_lines[start.line].substring_view(start.column, end.column - start.column + 1);
}
auto index_of_position([this](auto position) {
@ -686,8 +691,8 @@ StringView Parser::text_of_node(const ASTNode& node) const
start_index += position.column;
return start_index;
});
auto start_index = index_of_position(node.start());
auto end_index = index_of_position(node.end());
auto start_index = index_of_position(start);
auto end_index = index_of_position(end);
ASSERT(end_index >= start_index);
return m_program.substring_view(start_index, end_index - start_index);
}
@ -717,6 +722,7 @@ bool Parser::match_expression()
return token_type == Token::Type::Integer
|| token_type == Token::Type::Float
|| token_type == Token::Type::Identifier
|| token_type == Token::Type::DoubleQuotedString
|| match_unary_expression();
}
@ -821,7 +827,6 @@ NonnullRefPtr<StringLiteral> Parser::parse_string_literal(ASTNode& parent)
auto token = peek();
if (token.type() != Token::Type::DoubleQuotedString && token.type() != Token::Type::EscapeSequence) {
ASSERT(start_token_index.has_value());
// TODO: don't consume
end_token_index = m_state.token_index - 1;
break;
}
@ -829,13 +834,19 @@ NonnullRefPtr<StringLiteral> Parser::parse_string_literal(ASTNode& parent)
start_token_index = m_state.token_index;
consume();
}
// String was not terminated
if (!end_token_index.has_value()) {
end_token_index = m_tokens.size() - 1;
}
ASSERT(start_token_index.has_value());
ASSERT(end_token_index.has_value());
Token start_token = m_tokens[start_token_index.value()];
Token end_token = m_tokens[end_token_index.value()];
ASSERT(start_token.start().line == end_token.start().line);
auto text = m_lines[start_token.start().line].substring_view(start_token.start().column, end_token.end().column - start_token.start().column + 1);
auto text = text_of_range(start_token.start(), end_token.end());
auto string_literal = create_ast_node<StringLiteral>(parent, start_token.start(), end_token.end());
string_literal->m_value = text;
return string_literal;