introduce a new CharSourceRange class, and enhance the diagnostics routines
to use them instead of SourceRange.  CharSourceRange is just a SourceRange
plus a bool that indicates whether the range has the end character resolved
or whether the end location is the start of the end token.  While most of
the compiler wants to think of ranges that have ends that are the start of
the end token, the printf diagnostic stuff wants to highlight ranges within
tokens.

This is transparent to the diagnostic stuff.  To start taking advantage of
the new capabilities, you can do something like this:
  Diag(..) << CharSourceRange::getCharRange(Begin,End)





git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106338 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Basic/Diagnostic.cpp b/lib/Basic/Diagnostic.cpp
index a2480b1..641d87b 100644
--- a/lib/Basic/Diagnostic.cpp
+++ b/lib/Basic/Diagnostic.cpp
@@ -1043,8 +1043,7 @@
 
 StoredDiagnostic::StoredDiagnostic(Diagnostic::Level Level, 
                                    const DiagnosticInfo &Info)
-  : Level(Level), Loc(Info.getLocation()) 
-{
+  : Level(Level), Loc(Info.getLocation()) {
   llvm::SmallString<64> Message;
   Info.FormatDiagnostic(Message);
   this->Message.assign(Message.begin(), Message.end());
@@ -1131,6 +1130,7 @@
       
       WriteSourceLocation(OS, SM, R->getBegin());
       WriteSourceLocation(OS, SM, R->getEnd());
+      WriteUnsigned(OS, R->isTokenRange());
     }
   }
 
@@ -1159,6 +1159,7 @@
   for (fixit_iterator F = fixit_begin(), FEnd = fixit_end(); F != FEnd; ++F) {
     WriteSourceLocation(OS, SM, F->RemoveRange.getBegin());
     WriteSourceLocation(OS, SM, F->RemoveRange.getEnd());
+    WriteUnsigned(OS, F->RemoveRange.isTokenRange());
     WriteSourceLocation(OS, SM, F->InsertionLoc);
     WriteString(OS, F->CodeToInsert);
   }
@@ -1272,11 +1273,14 @@
     return Diag;
   for (unsigned I = 0; I != NumSourceRanges; ++I) {
     SourceLocation Begin, End;
+    unsigned IsTokenRange;
     if (ReadSourceLocation(FM, SM, Memory, MemoryEnd, Begin) ||
-        ReadSourceLocation(FM, SM, Memory, MemoryEnd, End))
+        ReadSourceLocation(FM, SM, Memory, MemoryEnd, End) ||
+        ReadUnsigned(Memory, MemoryEnd, IsTokenRange))
       return Diag;
 
-    Diag.Ranges.push_back(SourceRange(Begin, End));
+    Diag.Ranges.push_back(CharSourceRange(SourceRange(Begin, End),
+                                          IsTokenRange));
   }
 
   // Read the fix-it hints.
@@ -1285,9 +1289,10 @@
     return Diag;
   for (unsigned I = 0; I != NumFixIts; ++I) {
     SourceLocation RemoveBegin, RemoveEnd, InsertionLoc;
-    unsigned InsertLen = 0;
+    unsigned InsertLen = 0, RemoveIsTokenRange;
     if (ReadSourceLocation(FM, SM, Memory, MemoryEnd, RemoveBegin) ||
         ReadSourceLocation(FM, SM, Memory, MemoryEnd, RemoveEnd) ||
+        ReadUnsigned(Memory, MemoryEnd, RemoveIsTokenRange) ||
         ReadSourceLocation(FM, SM, Memory, MemoryEnd, InsertionLoc) ||
         ReadUnsigned(Memory, MemoryEnd, InsertLen) ||
         Memory + InsertLen > MemoryEnd) {
@@ -1296,7 +1301,8 @@
     }
 
     FixItHint Hint;
-    Hint.RemoveRange = SourceRange(RemoveBegin, RemoveEnd);
+    Hint.RemoveRange = CharSourceRange(SourceRange(RemoveBegin, RemoveEnd),
+                                       RemoveIsTokenRange);
     Hint.InsertionLoc = InsertionLoc;
     Hint.CodeToInsert.assign(Memory, Memory + InsertLen);
     Memory += InsertLen;
diff --git a/lib/Checker/PathDiagnostic.cpp b/lib/Checker/PathDiagnostic.cpp
index 963923c..6e6c749 100644
--- a/lib/Checker/PathDiagnostic.cpp
+++ b/lib/Checker/PathDiagnostic.cpp
@@ -107,7 +107,7 @@
     new PathDiagnosticEventPiece(Info.getLocation(), StrC.str());
 
   for (unsigned i = 0, e = Info.getNumRanges(); i != e; ++i)
-    P->addRange(Info.getRange(i));
+    P->addRange(Info.getRange(i).getAsRange());
   for (unsigned i = 0, e = Info.getNumFixItHints(); i != e; ++i)
     P->addFixItHint(Info.getFixItHint(i));
   D->push_front(P);
diff --git a/lib/Frontend/TextDiagnosticPrinter.cpp b/lib/Frontend/TextDiagnosticPrinter.cpp
index 6ccf4f1..3f1eb82 100644
--- a/lib/Frontend/TextDiagnosticPrinter.cpp
+++ b/lib/Frontend/TextDiagnosticPrinter.cpp
@@ -70,7 +70,7 @@
 
 /// HighlightRange - Given a SourceRange and a line number, highlight (with ~'s)
 /// any characters in LineNo that intersect the SourceRange.
-void TextDiagnosticPrinter::HighlightRange(const SourceRange &R,
+void TextDiagnosticPrinter::HighlightRange(const CharSourceRange &R,
                                            const SourceManager &SM,
                                            unsigned LineNo, FileID FID,
                                            std::string &CaretLine,
@@ -112,8 +112,10 @@
     if (EndColNo) {
       --EndColNo;  // Zero base the col #.
 
-      // Add in the length of the token, so that we cover multi-char tokens.
-      EndColNo += Lexer::MeasureTokenLength(End, SM, *LangOpts);
+      // Add in the length of the token, so that we cover multi-char tokens if
+      // this is a token range.
+      if (R.isTokenRange())
+        EndColNo += Lexer::MeasureTokenLength(End, SM, *LangOpts);
     } else {
       EndColNo = CaretLine.size();
     }
@@ -281,7 +283,7 @@
 }
 
 void TextDiagnosticPrinter::EmitCaretDiagnostic(SourceLocation Loc,
-                                                SourceRange *Ranges,
+                                                CharSourceRange *Ranges,
                                                 unsigned NumRanges,
                                                 const SourceManager &SM,
                                                 const FixItHint *Hints,
@@ -312,10 +314,12 @@
 
     // Map the ranges.
     for (unsigned i = 0; i != NumRanges; ++i) {
-      SourceLocation S = Ranges[i].getBegin(), E = Ranges[i].getEnd();
-      if (S.isMacroID()) S = SM.getImmediateSpellingLoc(S);
-      if (E.isMacroID()) E = SM.getImmediateSpellingLoc(E);
-      Ranges[i] = SourceRange(S, E);
+      CharSourceRange &R = Ranges[i];
+      SourceLocation S = R.getBegin(), E = R.getEnd();
+      if (S.isMacroID())
+        R.setBegin(SM.getImmediateSpellingLoc(S));
+      if (E.isMacroID())
+        R.setEnd(SM.getImmediateSpellingLoc(E));
     }
 
     if (!Suppressed) {
@@ -777,7 +781,9 @@
             continue;
 
           // Add in the length of the token, so that we cover multi-char tokens.
-          unsigned TokSize = Lexer::MeasureTokenLength(E, SM, *LangOpts);
+          unsigned TokSize = 0;
+          if (Info.getRange(i).isTokenRange())
+            TokSize = Lexer::MeasureTokenLength(E, SM, *LangOpts);
 
           OS << '{' << SM.getLineNumber(BInfo.first, BInfo.second) << ':'
              << SM.getColumnNumber(BInfo.first, BInfo.second) << '-'
@@ -904,15 +910,15 @@
     LastCaretDiagnosticWasNote = (Level == Diagnostic::Note);
 
     // Get the ranges into a local array we can hack on.
-    SourceRange Ranges[20];
+    CharSourceRange Ranges[20];
     unsigned NumRanges = Info.getNumRanges();
     assert(NumRanges < 20 && "Out of space");
     for (unsigned i = 0; i != NumRanges; ++i)
       Ranges[i] = Info.getRange(i);
 
     unsigned NumHints = Info.getNumFixItHints();
-    for (unsigned idx = 0; idx < NumHints; ++idx) {
-      const FixItHint &Hint = Info.getFixItHint(idx);
+    for (unsigned i = 0; i != NumHints; ++i) {
+      const FixItHint &Hint = Info.getFixItHint(i);
       if (Hint.RemoveRange.isValid()) {
         assert(NumRanges < 20 && "Out of space");
         Ranges[NumRanges++] = Hint.RemoveRange;
diff --git a/lib/Rewrite/Rewriter.cpp b/lib/Rewrite/Rewriter.cpp
index 376678a..8b0bf91 100644
--- a/lib/Rewrite/Rewriter.cpp
+++ b/lib/Rewrite/Rewriter.cpp
@@ -72,7 +72,7 @@
 
 /// getRangeSize - Return the size in bytes of the specified range if they
 /// are in the same file.  If not, this returns -1.
-int Rewriter::getRangeSize(SourceRange Range) const {
+int Rewriter::getRangeSize(const CharSourceRange &Range) const {
   if (!isRewritable(Range.getBegin()) ||
       !isRewritable(Range.getEnd())) return -1;
 
@@ -97,12 +97,18 @@
 
 
   // Adjust the end offset to the end of the last token, instead of being the
-  // start of the last token.
-  EndOff += Lexer::MeasureTokenLength(Range.getEnd(), *SourceMgr, *LangOpts);
+  // start of the last token if this is a token range.
+  if (Range.isTokenRange())
+    EndOff += Lexer::MeasureTokenLength(Range.getEnd(), *SourceMgr, *LangOpts);
 
   return EndOff-StartOff;
 }
 
+int Rewriter::getRangeSize(SourceRange Range) const {
+  return getRangeSize(CharSourceRange::getTokenRange(Range));
+}
+
+
 /// getRewrittenText - Return the rewritten form of the text in the specified
 /// range.  If the start or end of the range was unrewritable or if they are
 /// in different buffers, this returns an empty string.