aboutsummaryrefslogtreecommitdiffstats
path: root/libsolidity/parsing
diff options
context:
space:
mode:
authorchriseth <chris@ethereum.org>2016-08-05 14:47:52 +0800
committerGitHub <noreply@github.com>2016-08-05 14:47:52 +0800
commit3c93a22d478b9439d6e226ad7954fe571117f439 (patch)
tree687cb5a48bfbd8ea356444cc9034978345229155 /libsolidity/parsing
parentff60ce988b29c792eacae6e0ec1501ae7fb50b19 (diff)
parentb881dbb29200558d93c642ef41950d9d43a7e679 (diff)
downloaddexon-solidity-3c93a22d478b9439d6e226ad7954fe571117f439.tar.gz
dexon-solidity-3c93a22d478b9439d6e226ad7954fe571117f439.tar.zst
dexon-solidity-3c93a22d478b9439d6e226ad7954fe571117f439.zip
Merge pull request #666 from axic/feature/unicode-escape
Support unicode escape characters
Diffstat (limited to 'libsolidity/parsing')
-rw-r--r--libsolidity/parsing/Scanner.cpp43
-rw-r--r--libsolidity/parsing/Scanner.h2
2 files changed, 45 insertions, 0 deletions
diff --git a/libsolidity/parsing/Scanner.cpp b/libsolidity/parsing/Scanner.cpp
index d630d0ab..d730210a 100644
--- a/libsolidity/parsing/Scanner.cpp
+++ b/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;
diff --git a/libsolidity/parsing/Scanner.h b/libsolidity/parsing/Scanner.h
index cd60aff8..708adf8f 100644
--- a/libsolidity/parsing/Scanner.h
+++ b/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();