commit 71d95d6e51641e8475cf4d1664582c1ca5537924
author Alexander Kornienko <alexfh@google.com> Tue Nov 26 10:38:53 2013 +0000
committer Alexander Kornienko <alexfh@google.com> Tue Nov 26 10:38:53 2013 +0000
tree 0d4146d995b51c17822aedd0d6ed4fa8cbe11aad
parent b697b538dc07c72c729d1afd1603aa3708475b3b

Fix crash in getStringSplit.

Summary:
getStringSplit used to crash, when trying to split a long string
literal containing both printable and unprintable multi-byte UTF-8 characters.

Reviewers: djasper, klimek

Reviewed By: djasper

CC: cfe-commits, klimek

Differential Revision: http://llvm-reviews.chandlerc.com/D2268

llvm-svn: 195728

clang/lib/Format/BreakableToken.cpp

diff --git a/clang/lib/Format/BreakableToken.cpp b/clang/lib/Format/BreakableToken.cpp
index d720ce9..a08102a 100644
--- a/clang/lib/Format/BreakableToken.cpp
+++ b/clang/lib/Format/BreakableToken.cpp
@@ -92,9 +92,7 @@
     return BreakableToken::Split(StringRef::npos, 0);
   if (ColumnLimit <= UsedColumns)
     return BreakableToken::Split(StringRef::npos, 0);
-  unsigned MaxSplit = std::min<unsigned>(
-      ColumnLimit - UsedColumns,
-      encoding::columnWidthWithTabs(Text, UsedColumns, TabWidth, Encoding) - 1);
+  unsigned MaxSplit = ColumnLimit - UsedColumns;
   StringRef::size_type SpaceOffset = 0;
   StringRef::size_type SlashOffset = 0;
   StringRef::size_type WordStartOffset = 0;
@@ -110,7 +108,7 @@
           Text.substr(0, Advance), UsedColumns + Chars, TabWidth, Encoding);
     }
 
-    if (Chars > MaxSplit)
+    if (Chars > MaxSplit || Text.size() == Advance)
       break;
 
     if (IsBlank(Text[0]))

clang/lib/Format/Encoding.h

diff --git a/clang/lib/Format/Encoding.h b/clang/lib/Format/Encoding.h
index 356334d..dba5174 100644
--- a/clang/lib/Format/Encoding.h
+++ b/clang/lib/Format/Encoding.h
@@ -64,6 +64,10 @@
 inline unsigned columnWidth(StringRef Text, Encoding Encoding) {
   if (Encoding == Encoding_UTF8) {
     int ContentWidth = llvm::sys::unicode::columnWidthUTF8(Text);
+    // FIXME: Figure out the correct way to handle this in the presence of both
+    // printable and unprintable multi-byte UTF-8 characters. Falling back to
+    // returning the number of bytes may cause problems, as columnWidth suddenly
+    // becomes non-additive.
     if (ContentWidth >= 0)
       return ContentWidth;
   }
@@ -81,9 +85,7 @@
     StringRef::size_type TabPos = Tail.find('\t');
     if (TabPos == StringRef::npos)
       return TotalWidth + columnWidth(Tail, Encoding);
-    int Width = columnWidth(Tail.substr(0, TabPos), Encoding);
-    assert(Width >= 0);
-    TotalWidth += Width;
+    TotalWidth += columnWidth(Tail.substr(0, TabPos), Encoding);
     TotalWidth += TabWidth - (TotalWidth + StartColumn) % TabWidth;
     Tail = Tail.substr(TabPos + 1);
   }