Merge pull request #666 from axic/feature/unicode-escape

Support unicode escape characters

docs/types.rst

@@ -214,7 +214,9 @@ a non-rational number).
 String Literals
 ---------------
 
-String Literals are written with double quotes (``"abc"``). As with integer literals, their type can vary, but they are implicitly convertible to ``bytes`` if they fit, to ``bytes`` and to ``string``.
+String Literals are written with double quotes (``"abc"``). As with integer literals, their type can vary, but they are implicitly convertible to ``bytes1``, ..., ``bytes32`` if they fit, to ``bytes`` and to ``string``.
+
+String Literals support escape characters, such as ``\n``, ``\xNN`` and ``\uNNNN``. ``\xNN`` takes a hex value and inserts the appropriate byte, while ``\uNNNN`` takes a Unicode codepoint and inserts an UTF8 sequence.
 
 .. index:: enum
 

libsolidity/parsing/Scanner.cpp

@@ -177,6 +177,41 @@ bool Scanner::scanHexByte(char& o_scannedByte)
 	return true;
 }
 
+bool Scanner::scanUnicode(unsigned & o_codepoint)
+{
+	unsigned x = 0;
+	for (int i = 0; i < 4; i++)
+	{
+		int d = hexValue(m_char);
+		if (d < 0)
+		{
+			rollback(i);
+			return false;
+		}
+		x = x * 16 + d;
+		advance();
+	}
+	o_codepoint = x;
+	return true;
+}
+
+// This supports codepoints between 0000 and FFFF.
+void Scanner::addUnicodeAsUTF8(unsigned codepoint)
+{
+	if (codepoint <= 0x7f)
+		addLiteralChar(codepoint);
+	else if (codepoint <= 0x7ff)
+	{
+		addLiteralChar(0xc0 | (codepoint >> 6));
+		addLiteralChar(0x80 | (codepoint & 0x3f));
+	}
+	else
+	{
+		addLiteralChar(0xe0 | (codepoint >> 12));
+		addLiteralChar(0x80 | ((codepoint >> 6) & 0x3f));
+		addLiteralChar(0x80 | (codepoint & 0x3f));
+	}
+}
 
 // Ensure that tokens can be stored in a byte.
 BOOST_STATIC_ASSERT(Token::NUM_TOKENS <= 0x100);
@@ -607,6 +642,14 @@ bool Scanner::scanEscape()
 	case 'v':
 		c = '\v';
 		break;
+	case 'u':
+	{
+		unsigned codepoint;
+		if (!scanUnicode(codepoint))
+			return false;
+		addUnicodeAsUTF8(codepoint);
+		return true;
+	}
 	case 'x':
 		if (!scanHexByte(c))
 			return false;

libsolidity/parsing/Scanner.h

@@ -175,6 +175,7 @@ private:
 	inline void addLiteralChar(char c) { m_nextToken.literal.push_back(c); }
 	inline void addCommentLiteralChar(char c) { m_nextSkippedComment.literal.push_back(c); }
 	inline void addLiteralCharAndAdvance() { addLiteralChar(m_char); advance(); }
+	void addUnicodeAsUTF8(unsigned codepoint);
 	///@}
 
 	bool advance() { m_char = m_source.advanceAndGet(); return !m_source.isPastEndOfInput(); }
@@ -185,6 +186,7 @@ private:
 	inline Token::Value selectToken(char _next, Token::Value _then, Token::Value _else);
 
 	bool scanHexByte(char& o_scannedByte);
+	bool scanUnicode(unsigned& o_codepoint);
 
 	/// Scans a single Solidity token.
 	void scanToken();

test/libsolidity/SolidityScanner.cpp

@@ -291,6 +291,46 @@ BOOST_AUTO_TEST_CASE(empty_comment)
 
 }
 
+BOOST_AUTO_TEST_CASE(valid_unicode_string_escape)
+{
+	Scanner scanner(CharStream("{ \"\\u00DAnicode\""));
+	BOOST_CHECK_EQUAL(scanner.currentToken(), Token::LBrace);
+	BOOST_CHECK_EQUAL(scanner.next(), Token::StringLiteral);
+	BOOST_CHECK_EQUAL(scanner.currentLiteral(), std::string("\xC3\x9Anicode", 8));
+}
+
+BOOST_AUTO_TEST_CASE(valid_unicode_string_escape_7f)
+{
+	Scanner scanner(CharStream("{ \"\\u007Fnicode\""));
+	BOOST_CHECK_EQUAL(scanner.currentToken(), Token::LBrace);
+	BOOST_CHECK_EQUAL(scanner.next(), Token::StringLiteral);
+	BOOST_CHECK_EQUAL(scanner.currentLiteral(), std::string("\x7Fnicode", 7));
+}
+
+BOOST_AUTO_TEST_CASE(valid_unicode_string_escape_7ff)
+{
+	Scanner scanner(CharStream("{ \"\\u07FFnicode\""));
+	BOOST_CHECK_EQUAL(scanner.currentToken(), Token::LBrace);
+	BOOST_CHECK_EQUAL(scanner.next(), Token::StringLiteral);
+	BOOST_CHECK_EQUAL(scanner.currentLiteral(), std::string("\xDF\xBFnicode", 8));
+}
+
+BOOST_AUTO_TEST_CASE(valid_unicode_string_escape_ffff)
+{
+	Scanner scanner(CharStream("{ \"\\uFFFFnicode\""));
+	BOOST_CHECK_EQUAL(scanner.currentToken(), Token::LBrace);
+	BOOST_CHECK_EQUAL(scanner.next(), Token::StringLiteral);
+	BOOST_CHECK_EQUAL(scanner.currentLiteral(), std::string("\xEF\xBF\xBFnicode", 9));
+}
+
+BOOST_AUTO_TEST_CASE(invalid_short_unicode_string_escape)
+{
+	Scanner scanner(CharStream("{ \"\\uFFnicode\""));
+	BOOST_CHECK_EQUAL(scanner.currentToken(), Token::LBrace);
+	BOOST_CHECK_EQUAL(scanner.next(), Token::Illegal);
+}
+
+
 BOOST_AUTO_TEST_SUITE_END()
 
 }