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

1 files changed, 12 insertions, 0 deletions

diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index fc0e935037b..a7dce3b86f5 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -6991,6 +6991,16 @@ TEST_F(FormatTest, CountsUTF8CharactersProperly) {
 }
 
 TEST_F(FormatTest, SplitsUTF8Strings) {
+  // Non-printable characters' width is currently considered to be the length in
+  // bytes in UTF8. The characters can be displayed in very different manner
+  // (zero-width, single width with a substitution glyph, expanded to their code
+  // (e.g. "<8d>"), so there's no single correct way to handle them.
+  EXPECT_EQ("\"aaaaÄ\"\n"
+            "\"\";",
+            format("\"aaaač\";", getLLVMStyleWithColumns(10)));
+  EXPECT_EQ("\"aaaaaaaÄ\"\n"
+            "\"\";",
+            format("\"aaaaaaač\";", getLLVMStyleWithColumns(10)));
   EXPECT_EQ(
       "\"Однажды, в \"\n"
       "\"студёную \"\n"
@@ -7024,6 +7034,8 @@ TEST_F(FormatTest, HandlesDoubleWidthCharsInMultiLineStrings) {
 }
 
 TEST_F(FormatTest, SplitsUTF8LineComments) {
+  EXPECT_EQ("// aaaač",
+            format("// aaaač", getLLVMStyleWithColumns(10)));
   EXPECT_EQ("// Я из лесу\n"
             "// вышел; был\n"
             "// сильный\n"