More "prep" work for handling UTF16 CFString.

Patch by Jean-Daniel Dupas. Thanks!


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@68203 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/include/clang/AST/Expr.h b/include/clang/AST/Expr.h
index 832f83c..92b3963 100644
--- a/include/clang/AST/Expr.h
+++ b/include/clang/AST/Expr.h
@@ -533,9 +533,9 @@
   const char *getStrData() const { return StrData; }
   unsigned getByteLength() const { return ByteLength; }
   bool isWide() const { return IsWide; }
-  bool containsNonAscii() const {
+  bool containsNonAsciiOrNull() const {
     for (unsigned i = 0; i < getByteLength(); ++i)
-      if (!isascii(getStrData()[i]))
+      if (!isascii(getStrData()[i]) || !getStrData()[i])
         return true;
     return false;
   }
diff --git a/lib/CodeGen/CGExprConstant.cpp b/lib/CodeGen/CGExprConstant.cpp
index 89b3319..51ad5b6 100644
--- a/lib/CodeGen/CGExprConstant.cpp
+++ b/lib/CodeGen/CGExprConstant.cpp
@@ -442,9 +442,8 @@
         break;
       const Expr *Arg = CE->getArg(0)->IgnoreParenCasts();
       const StringLiteral *Literal = cast<StringLiteral>(Arg);
-      std::string S(Literal->getStrData(), Literal->getByteLength());
       // FIXME: need to deal with UCN conversion issues.
-      return CGM.GetAddrOfConstantCFString(S);
+      return CGM.GetAddrOfConstantCFString(Literal);
     }
     case Expr::BlockExprClass: {
       std::string FunctionName;
diff --git a/lib/CodeGen/CGObjCMac.cpp b/lib/CodeGen/CGObjCMac.cpp
index 351f8e2..6b3ced0 100644
--- a/lib/CodeGen/CGObjCMac.cpp
+++ b/lib/CodeGen/CGObjCMac.cpp
@@ -900,12 +900,7 @@
 
 llvm::Constant *CGObjCCommonMac::GenerateConstantString(
   const ObjCStringLiteral *SL) {
-  std::string Str(SL->getString()->getStrData(), 
-                  SL->getString()->getByteLength());
-  if (SL->getString()->containsNonAscii()) {
-    // FIXME: Convert from UTF-8 to UTF-16.
-  }
-  return CGM.GetAddrOfConstantCFString(Str);
+  return CGM.GetAddrOfConstantCFString(SL->getString());
 }
 
 /// Generates a message send where the super is the receiver.  This is
diff --git a/lib/CodeGen/CodeGenModule.cpp b/lib/CodeGen/CodeGenModule.cpp
index 101b2e5..d428c83 100644
--- a/lib/CodeGen/CodeGenModule.cpp
+++ b/lib/CodeGen/CodeGenModule.cpp
@@ -1002,7 +1002,11 @@
 // We still need to work out the details of handling UTF-16. 
 // See: <rdr://2996215>
 llvm::Constant *CodeGenModule::
-GetAddrOfConstantCFString(const std::string &str) {
+GetAddrOfConstantCFString(const StringLiteral *Literal) {
+  //  if (Literal->containsNonAsciiOrNull()) {
+  //    // FIXME: Convert from UTF-8 to UTF-16.
+  //  }
+  std::string str(Literal->getStrData(), Literal->getByteLength());
   llvm::StringMapEntry<llvm::Constant *> &Entry = 
     CFConstantStringMap.GetOrCreateValue(&str[0], &str[str.length()]);
   
diff --git a/lib/CodeGen/CodeGenModule.h b/lib/CodeGen/CodeGenModule.h
index 8c7571f..c0dbbf8 100644
--- a/lib/CodeGen/CodeGenModule.h
+++ b/lib/CodeGen/CodeGenModule.h
@@ -187,7 +187,7 @@
 
   /// GetAddrOfConstantCFString - Return a pointer to a constant CFString object
   /// for the given string.
-  llvm::Constant *GetAddrOfConstantCFString(const std::string& str);
+  llvm::Constant *GetAddrOfConstantCFString(const StringLiteral *Literal);
 
   /// GetAddrOfConstantStringFromLiteral - Return a pointer to a constant array
   /// for the given string literal.