[cfe-commits] r150295 - in /cfe/trunk: include/clang/Basic/DiagnosticLexKinds.td include/clang/Lex/LiteralSupport.h lib/Lex/LiteralSupport.cpp test/Lexer/char-literal-encoding-error.c test/Lexer/string-literal-encoding.c
Eli Friedman
eli.friedman at gmail.com
Fri Feb 10 21:08:10 PST 2012
Author: efriedma
Date: Fri Feb 10 23:08:10 2012
New Revision: 150295
URL: http://llvm.org/viewvc/llvm-project?rev=150295&view=rev
Log:
Implement warning for non-wide string literals with an unexpected encoding. Downgrade error for non-wide character literals with an unexpected encoding to a warning for compatibility with gcc and older versions of clang. <rdar://problem/10837678>.
Modified:
cfe/trunk/include/clang/Basic/DiagnosticLexKinds.td
cfe/trunk/include/clang/Lex/LiteralSupport.h
cfe/trunk/lib/Lex/LiteralSupport.cpp
cfe/trunk/test/Lexer/char-literal-encoding-error.c
cfe/trunk/test/Lexer/string-literal-encoding.c
Modified: cfe/trunk/include/clang/Basic/DiagnosticLexKinds.td
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/DiagnosticLexKinds.td?rev=150295&r1=150294&r2=150295&view=diff
==============================================================================
--- cfe/trunk/include/clang/Basic/DiagnosticLexKinds.td (original)
+++ cfe/trunk/include/clang/Basic/DiagnosticLexKinds.td Fri Feb 10 23:08:10 2012
@@ -136,9 +136,16 @@
"unsupported non-standard concatenation of string literals">;
def err_bad_string_encoding : Error<
"illegal character encoding in string literal">;
+def warn_bad_string_encoding : ExtWarn<
+ "illegal character encoding in string literal">,
+ InGroup<DiagGroup<"invalid-source-encoding">>;
def err_bad_character_encoding : Error<
"illegal character encoding in character literal">;
-
+def warn_bad_character_encoding : ExtWarn<
+ "illegal character encoding in character literal">,
+ InGroup<DiagGroup<"invalid-source-encoding">>;
+
+
//===----------------------------------------------------------------------===//
// PTH Diagnostics
//===----------------------------------------------------------------------===//
Modified: cfe/trunk/include/clang/Lex/LiteralSupport.h
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Lex/LiteralSupport.h?rev=150295&r1=150294&r2=150295&view=diff
==============================================================================
--- cfe/trunk/include/clang/Lex/LiteralSupport.h (original)
+++ cfe/trunk/include/clang/Lex/LiteralSupport.h Fri Feb 10 23:08:10 2012
@@ -199,6 +199,7 @@
private:
void init(const Token *StringToks, unsigned NumStringToks);
bool CopyStringFragment(StringRef Fragment);
+ bool DiagnoseBadString(const Token& Tok);
};
} // end namespace clang
Modified: cfe/trunk/lib/Lex/LiteralSupport.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Lex/LiteralSupport.cpp?rev=150295&r1=150294&r2=150295&view=diff
==============================================================================
--- cfe/trunk/lib/Lex/LiteralSupport.cpp (original)
+++ cfe/trunk/lib/Lex/LiteralSupport.cpp Fri Feb 10 23:08:10 2012
@@ -822,17 +822,32 @@
++begin;
} while (begin != end && *begin != '\\');
- uint32_t *tmp_begin = buffer_begin;
+ char const *tmp_in_start = start;
+ uint32_t *tmp_out_start = buffer_begin;
ConversionResult res =
ConvertUTF8toUTF32(reinterpret_cast<UTF8 const **>(&start),
reinterpret_cast<UTF8 const *>(begin),
&buffer_begin,buffer_end,strictConversion);
if (res!=conversionOK) {
- PP.Diag(Loc, diag::err_bad_character_encoding);
- HadError = true;
+ // If we see bad encoding for unprefixed character literals, warn and
+ // simply copy the byte values, for compatibility with gcc and
+ // older versions of clang.
+ bool NoErrorOnBadEncoding = isAscii();
+ unsigned Msg = diag::err_bad_character_encoding;
+ if (NoErrorOnBadEncoding)
+ Msg = diag::warn_bad_character_encoding;
+ PP.Diag(Loc, Msg);
+ if (NoErrorOnBadEncoding) {
+ start = tmp_in_start;
+ buffer_begin = tmp_out_start;
+ for ( ; start != begin; ++start, ++buffer_begin)
+ *buffer_begin = static_cast<uint8_t>(*start);
+ } else {
+ HadError = true;
+ }
} else {
- for (; tmp_begin<buffer_begin; ++tmp_begin) {
- if (*tmp_begin > largest_character_for_kind) {
+ for (; tmp_out_start <buffer_begin; ++tmp_out_start) {
+ if (*tmp_out_start > largest_character_for_kind) {
HadError = true;
PP.Diag(Loc, diag::err_character_too_large);
}
@@ -1097,10 +1112,8 @@
// Copy the string over
if (CopyStringFragment(StringRef(ThisTokBuf,ThisTokEnd-ThisTokBuf)))
{
- if (Diags)
- Diags->Report(FullSourceLoc(StringToks[i].getLocation(), SM),
- diag::err_bad_string_encoding);
- hadError = true;
+ if (DiagnoseBadString(StringToks[i]))
+ hadError = true;
}
} else {
@@ -1131,10 +1144,8 @@
// Copy the character span over.
if (CopyStringFragment(StringRef(InStart,ThisTokBuf-InStart)))
{
- if (Diags)
- Diags->Report(FullSourceLoc(StringToks[i].getLocation(), SM),
- diag::err_bad_string_encoding);
- hadError = true;
+ if (DiagnoseBadString(StringToks[i]))
+ hadError = true;
}
continue;
}
@@ -1219,6 +1230,9 @@
ConversionResult result = conversionOK;
// Copy the character span over.
if (CharByteWidth == 1) {
+ if (!isLegalUTF8Sequence(reinterpret_cast<const UTF8*>(Fragment.begin()),
+ reinterpret_cast<const UTF8*>(Fragment.end())))
+ result = sourceIllegal;
memcpy(ResultPtr, Fragment.data(), Fragment.size());
ResultPtr += Fragment.size();
} else if (CharByteWidth == 2) {
@@ -1226,7 +1240,7 @@
// FIXME: Make the type of the result buffer correct instead of
// using reinterpret_cast.
UTF16 *targetStart = reinterpret_cast<UTF16*>(ResultPtr);
- ConversionFlags flags = lenientConversion;
+ ConversionFlags flags = strictConversion;
result = ConvertUTF8toUTF16(
&sourceStart,sourceStart + Fragment.size(),
&targetStart,targetStart + 2*Fragment.size(),flags);
@@ -1237,7 +1251,7 @@
// FIXME: Make the type of the result buffer correct instead of
// using reinterpret_cast.
UTF32 *targetStart = reinterpret_cast<UTF32*>(ResultPtr);
- ConversionFlags flags = lenientConversion;
+ ConversionFlags flags = strictConversion;
result = ConvertUTF8toUTF32(
&sourceStart,sourceStart + Fragment.size(),
&targetStart,targetStart + 4*Fragment.size(),flags);
@@ -1249,6 +1263,17 @@
return result != conversionOK;
}
+bool StringLiteralParser::DiagnoseBadString(const Token &Tok) {
+ // If we see bad encoding for unprefixed string literals, warn and
+ // simply copy the byte values, for compatibility with gcc and older
+ // versions of clang.
+ bool NoErrorOnBadEncoding = isAscii();
+ unsigned Msg = NoErrorOnBadEncoding ? diag::warn_bad_string_encoding :
+ diag::err_bad_string_encoding;
+ if (Diags)
+ Diags->Report(FullSourceLoc(Tok.getLocation(), SM), Msg);
+ return !NoErrorOnBadEncoding;
+}
/// getOffsetOfStringByte - This function returns the offset of the
/// specified byte of the string data represented by Token. This handles
Modified: cfe/trunk/test/Lexer/char-literal-encoding-error.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Lexer/char-literal-encoding-error.c?rev=150295&r1=150294&r2=150295&view=diff
==============================================================================
--- cfe/trunk/test/Lexer/char-literal-encoding-error.c (original)
+++ cfe/trunk/test/Lexer/char-literal-encoding-error.c Fri Feb 10 23:08:10 2012
@@ -3,8 +3,13 @@
// This file is encoded using ISO-8859-1
int main() {
- 'é'; // expected-error {{illegal character encoding in character literal}}
- u'é'; // expected-error {{illegal character encoding in character literal}}
- U'é'; // expected-error {{illegal character encoding in character literal}}
- L'é'; // expected-error {{illegal character encoding in character literal}}
+ (void)'é'; // expected-warning {{illegal character encoding in character literal}}
+ (void)u'é'; // expected-error {{illegal character encoding in character literal}}
+ (void)U'é'; // expected-error {{illegal character encoding in character literal}}
+ (void)L'é'; // expected-error {{illegal character encoding in character literal}}
+
+ // For narrow character literals, since there is no error, make sure the
+ // encoding is correct
+ static_assert((unsigned char)'é' == 0xE9, ""); // expected-warning {{illegal character encoding in character literal}}
+ static_assert('éé' == 0xE9E9, ""); // expected-warning {{illegal character encoding in character literal}} expected-warning {{multi-character character constant}}
}
Modified: cfe/trunk/test/Lexer/string-literal-encoding.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Lexer/string-literal-encoding.c?rev=150295&r1=150294&r2=150295&view=diff
==============================================================================
--- cfe/trunk/test/Lexer/string-literal-encoding.c (original)
+++ cfe/trunk/test/Lexer/string-literal-encoding.c Fri Feb 10 23:08:10 2012
@@ -12,4 +12,7 @@
wchar_t const *d = LR"(Àéîõü)"; // expected-error {{illegal character encoding in string literal}}
char16_t const *e = uR"(Àéîõü)"; // expected-error {{illegal character encoding in string literal}}
char32_t const *f = UR"(Àéîõü)"; // expected-error {{illegal character encoding in string literal}}
+
+ char const *g = "Àéîõü"; // expected-warning {{illegal character encoding in string literal}}
+ char const *h = u8"Àéîõü"; // expected-error {{illegal character encoding in string literal}}
}
More information about the cfe-commits
mailing list