mirror of
				https://github.com/ethereum/solidity
				synced 2023-10-03 13:03:40 +00:00 
			
		
		
		
	
		
			
				
	
	
		
			212 lines
		
	
	
		
			6.7 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
			
		
		
	
	
			212 lines
		
	
	
		
			6.7 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
| // Copyright 2006-2012, the V8 project authors. All rights reserved.
 | |
| // Redistribution and use in source and binary forms, with or without
 | |
| // modification, are permitted provided that the following conditions are
 | |
| // met:
 | |
| //
 | |
| //    * Redistributions of source code must retain the above copyright
 | |
| //      notice, this list of conditions and the following disclaimer.
 | |
| //    * 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.
 | |
| //    * Neither the name of Google Inc. nor the names of its
 | |
| //      contributors may be used to endorse or promote products derived
 | |
| //      from this software without specific prior written permission.
 | |
| //
 | |
| // 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
 | |
| // OWNER 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.
 | |
| //
 | |
| // Modifications as part of solidity under the following license:
 | |
| //
 | |
| // solidity is free software: you can redistribute it and/or modify
 | |
| // it under the terms of the GNU General Public License as published by
 | |
| // the Free Software Foundation, either version 3 of the License, or
 | |
| // (at your option) any later version.
 | |
| //
 | |
| // solidity is distributed in the hope that it will be useful,
 | |
| // but WITHOUT ANY WARRANTY; without even the implied warranty of
 | |
| // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 | |
| // GNU General Public License for more details.
 | |
| //
 | |
| // You should have received a copy of the GNU General Public License
 | |
| // along with solidity.  If not, see <http://www.gnu.org/licenses/>.
 | |
| 
 | |
| #include <liblangutil/Token.h>
 | |
| #include <boost/range/iterator_range.hpp>
 | |
| #include <map>
 | |
| 
 | |
| using namespace std;
 | |
| 
 | |
| namespace solidity::langutil
 | |
| {
 | |
| 
 | |
| void ElementaryTypeNameToken::assertDetails(Token _baseType, unsigned const& _first, unsigned const& _second)
 | |
| {
 | |
| 	solAssert(TokenTraits::isElementaryTypeName(_baseType), "Expected elementary type name: " + string(TokenTraits::toString(_baseType)));
 | |
| 	if (_baseType == Token::BytesM)
 | |
| 	{
 | |
| 		solAssert(_second == 0, "There should not be a second size argument to type bytesM.");
 | |
| 		solAssert(_first <= 32, "No elementary type bytes" + to_string(_first) + ".");
 | |
| 	}
 | |
| 	else if (_baseType == Token::UIntM || _baseType == Token::IntM)
 | |
| 	{
 | |
| 		solAssert(_second == 0, "There should not be a second size argument to type " + string(TokenTraits::toString(_baseType)) + ".");
 | |
| 		solAssert(
 | |
| 			_first <= 256 && _first % 8 == 0,
 | |
| 			"No elementary type " + string(TokenTraits::toString(_baseType)) + to_string(_first) + "."
 | |
| 		);
 | |
| 	}
 | |
| 	else if (_baseType == Token::UFixedMxN || _baseType == Token::FixedMxN)
 | |
| 	{
 | |
| 		solAssert(
 | |
| 			_first >= 8 && _first <= 256 && _first % 8 == 0 && _second <= 80,
 | |
| 			"No elementary type " + string(TokenTraits::toString(_baseType)) + to_string(_first) + "x" + to_string(_second) + "."
 | |
| 		);
 | |
| 	}
 | |
| 	else
 | |
| 		solAssert(_first == 0 && _second == 0, "Unexpected size arguments");
 | |
| 
 | |
| 	m_token = _baseType;
 | |
| 	m_firstNumber = _first;
 | |
| 	m_secondNumber = _second;
 | |
| }
 | |
| 
 | |
| namespace TokenTraits
 | |
| {
 | |
| char const* toString(Token tok)
 | |
| {
 | |
| 	switch (tok)
 | |
| 	{
 | |
| #define T(name, string, precedence) case Token::name: return string;
 | |
| 		TOKEN_LIST(T, T)
 | |
| #undef T
 | |
| 		default: // Token::NUM_TOKENS:
 | |
| 			return "";
 | |
| 	}
 | |
| }
 | |
| 
 | |
| char const* name(Token tok)
 | |
| {
 | |
| #define T(name, string, precedence) #name,
 | |
| 	static char const* const names[TokenTraits::count()] = { TOKEN_LIST(T, T) };
 | |
| #undef T
 | |
| 
 | |
| 	solAssert(static_cast<size_t>(tok) < TokenTraits::count(), "");
 | |
| 	return names[static_cast<size_t>(tok)];
 | |
| }
 | |
| 
 | |
| std::string friendlyName(Token tok)
 | |
| {
 | |
| 	char const* ret = toString(tok);
 | |
| 	if (ret)
 | |
| 		return std::string(ret);
 | |
| 
 | |
| 	ret = name(tok);
 | |
| 	solAssert(ret != nullptr, "");
 | |
| 	return std::string(ret);
 | |
| }
 | |
| 
 | |
| #define T(name, string, precedence) precedence,
 | |
| int precedence(Token tok)
 | |
| {
 | |
| 	int8_t const static precs[TokenTraits::count()] =
 | |
| 	{
 | |
| 		TOKEN_LIST(T, T)
 | |
| 	};
 | |
| 	return precs[static_cast<size_t>(tok)];
 | |
| }
 | |
| #undef T
 | |
| 
 | |
| int parseSize(string::const_iterator _begin, string::const_iterator _end)
 | |
| {
 | |
| 	try
 | |
| 	{
 | |
| 		int m = boost::lexical_cast<int>(boost::make_iterator_range(_begin, _end));
 | |
| 		return m;
 | |
| 	}
 | |
| 	catch(boost::bad_lexical_cast const&)
 | |
| 	{
 | |
| 		return -1;
 | |
| 	}
 | |
| }
 | |
| 
 | |
| static Token keywordByName(string const& _name)
 | |
| {
 | |
| 	// The following macros are used inside TOKEN_LIST and cause non-keyword tokens to be ignored
 | |
| 	// and keywords to be put inside the keywords variable.
 | |
| #define KEYWORD(name, string, precedence) {string, Token::name},
 | |
| #define TOKEN(name, string, precedence)
 | |
| 	static map<string, Token> const keywords({TOKEN_LIST(TOKEN, KEYWORD)});
 | |
| #undef KEYWORD
 | |
| #undef TOKEN
 | |
| 	auto it = keywords.find(_name);
 | |
| 	return it == keywords.end() ? Token::Identifier : it->second;
 | |
| }
 | |
| 
 | |
| bool isYulKeyword(string const& _literal)
 | |
| {
 | |
| 	return _literal == "leave" || isYulKeyword(keywordByName(_literal));
 | |
| }
 | |
| 
 | |
| tuple<Token, unsigned int, unsigned int> fromIdentifierOrKeyword(string const& _literal)
 | |
| {
 | |
| 	auto positionM = find_if(_literal.begin(), _literal.end(), ::isdigit);
 | |
| 	if (positionM != _literal.end())
 | |
| 	{
 | |
| 		string baseType(_literal.begin(), positionM);
 | |
| 		auto positionX = find_if_not(positionM, _literal.end(), ::isdigit);
 | |
| 		int m = parseSize(positionM, positionX);
 | |
| 		Token keyword = keywordByName(baseType);
 | |
| 		if (keyword == Token::Bytes)
 | |
| 		{
 | |
| 			if (0 < m && m <= 32 && positionX == _literal.end())
 | |
| 				return make_tuple(Token::BytesM, m, 0);
 | |
| 		}
 | |
| 		else if (keyword == Token::UInt || keyword == Token::Int)
 | |
| 		{
 | |
| 			if (0 < m && m <= 256 && m % 8 == 0 && positionX == _literal.end())
 | |
| 			{
 | |
| 				if (keyword == Token::UInt)
 | |
| 					return make_tuple(Token::UIntM, m, 0);
 | |
| 				else
 | |
| 					return make_tuple(Token::IntM, m, 0);
 | |
| 			}
 | |
| 		}
 | |
| 		else if (keyword == Token::UFixed || keyword == Token::Fixed)
 | |
| 		{
 | |
| 			if (
 | |
| 				positionM < positionX &&
 | |
| 				positionX < _literal.end() &&
 | |
| 				*positionX == 'x' &&
 | |
| 				all_of(positionX + 1, _literal.end(), ::isdigit)
 | |
| 			) {
 | |
| 				int n = parseSize(positionX + 1, _literal.end());
 | |
| 				if (
 | |
| 					8 <= m && m <= 256 && m % 8 == 0 &&
 | |
| 					0 <= n && n <= 80
 | |
| 				) {
 | |
| 					if (keyword == Token::UFixed)
 | |
| 						return make_tuple(Token::UFixedMxN, m, n);
 | |
| 					else
 | |
| 						return make_tuple(Token::FixedMxN, m, n);
 | |
| 				}
 | |
| 			}
 | |
| 		}
 | |
| 		return make_tuple(Token::Identifier, 0, 0);
 | |
| 	}
 | |
| 
 | |
| 	return make_tuple(keywordByName(_literal), 0, 0);
 | |
| }
 | |
| 
 | |
| }
 | |
| }
 |