Support for double width characters.

Summary: Only works for UTF-8-encoded files.

Reviewers: djasper

Reviewed By: djasper

CC: cfe-commits, klimek

Differential Revision: http://llvm-reviews.chandlerc.com/D1311

llvm-svn: 187935

2 files changed, 37 insertions, 26 deletions

diff --git a/clang/lib/Format/BreakableToken.cpp b/clang/lib/Format/BreakableToken.cpp
index 3b291377983..c2365f1402f 100644
--- a/clang/lib/Format/BreakableToken.cpp
+++ b/clang/lib/Format/BreakableToken.cpp
@@ -20,6 +20,7 @@
 #include "clang/Format/Format.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/Support/Debug.h"
+#include "llvm/Support/Locale.h"
 #include <algorithm>
 
 namespace clang {
@@ -38,6 +39,15 @@ static bool IsBlank(char C) {
   }
 }
 
+static unsigned columnWidth(StringRef Text, encoding::Encoding Encoding) {
+  if (Encoding == encoding::Encoding_UTF8) {
+    int ContentWidth = llvm::sys::locale::columnWidth(Text);
+    if (ContentWidth >= 0)
+      return ContentWidth;
+  }
+  return encoding::getCodePointCount(Text, Encoding);
+}
+
 static BreakableToken::Split getCommentSplit(StringRef Text,
                                              unsigned ContentStartColumn,
                                              unsigned ColumnLimit,
@@ -49,9 +59,12 @@ static BreakableToken::Split getCommentSplit(StringRef Text,
   unsigned MaxSplitBytes = 0;
 
   for (unsigned NumChars = 0;
-       NumChars < MaxSplit && MaxSplitBytes < Text.size(); ++NumChars)
-    MaxSplitBytes +=
+       NumChars < MaxSplit && MaxSplitBytes < Text.size();) {
+    unsigned NumBytes =
         encoding::getCodePointNumBytes(Text[MaxSplitBytes], Encoding);
+    NumChars += columnWidth(Text.substr(MaxSplitBytes, NumBytes), Encoding);
+    MaxSplitBytes += NumBytes;
+  }
 
   StringRef::size_type SpaceOffset = Text.find_last_of(Blanks, MaxSplitBytes);
   if (SpaceOffset == StringRef::npos ||
@@ -84,9 +97,8 @@ static BreakableToken::Split getStringSplit(StringRef Text,
     return BreakableToken::Split(StringRef::npos, 0);
   if (ColumnLimit <= ContentStartColumn)
     return BreakableToken::Split(StringRef::npos, 0);
-  unsigned MaxSplit =
-      std::min<unsigned>(ColumnLimit - ContentStartColumn,
-                         encoding::getCodePointCount(Text, Encoding) - 1);
+  unsigned MaxSplit = std::min<unsigned>(ColumnLimit - ContentStartColumn,
+                                         columnWidth(Text, Encoding) - 1);
   StringRef::size_type SpaceOffset = 0;
   StringRef::size_type SlashOffset = 0;
   StringRef::size_type WordStartOffset = 0;
@@ -98,7 +110,7 @@ static BreakableToken::Split getStringSplit(StringRef Text,
       Chars += Advance;
     } else {
       Advance = encoding::getCodePointNumBytes(Text[0], Encoding);
-      Chars += 1;
+      Chars += columnWidth(Text.substr(0, Advance), Encoding);
     }
 
     if (Chars > MaxSplit)
@@ -131,7 +143,7 @@ unsigned BreakableSingleLineToken::getLineCount() const { return 1; }
 unsigned BreakableSingleLineToken::getLineLengthAfterSplit(
     unsigned LineIndex, unsigned Offset, StringRef::size_type Length) const {
   return StartColumn + Prefix.size() + Postfix.size() +
-         encoding::getCodePointCount(Line.substr(Offset, Length), Encoding);
+         columnWidth(Line.substr(Offset, Length), Encoding);
 }
 
 BreakableSingleLineToken::BreakableSingleLineToken(
@@ -329,8 +341,7 @@ unsigned BreakableBlockComment::getLineCount() const { return Lines.size(); }
 unsigned BreakableBlockComment::getLineLengthAfterSplit(
     unsigned LineIndex, unsigned Offset, StringRef::size_type Length) const {
   return getContentStartColumn(LineIndex, Offset) +
-         encoding::getCodePointCount(Lines[LineIndex].substr(Offset, Length),
-                                     Encoding) +
+         columnWidth(Lines[LineIndex].substr(Offset, Length), Encoding) +
          // The last line gets a "*/" postfix.
          (LineIndex + 1 == Lines.size() ? 2 : 0);
 }
diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index 2969e568ebc..48b1f63fcad 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -5704,15 +5704,15 @@ TEST_F(FormatTest, CountsUTF8CharactersProperly) {
   verifyFormat("\"Однажды в студёную зимнюю пору...\"",
                getLLVMStyleWithColumns(35));
   verifyFormat("\"一 二 三 四 五 六 七 八 九 十\"",
-               getLLVMStyleWithColumns(21));
+               getLLVMStyleWithColumns(31));
   verifyFormat("// Однажды в студёную зимнюю пору...",
                getLLVMStyleWithColumns(36));
   verifyFormat("// 一 二 三 四 五 六 七 八 九 十",
-               getLLVMStyleWithColumns(22));
+               getLLVMStyleWithColumns(32));
   verifyFormat("/* Однажды в студёную зимнюю пору... */",
                getLLVMStyleWithColumns(39));
   verifyFormat("/* 一 二 三 四 五 六 七 八 九 十 */",
-               getLLVMStyleWithColumns(25));
+               getLLVMStyleWithColumns(35));
 }
 
 TEST_F(FormatTest, SplitsUTF8Strings) {
@@ -5723,11 +5723,12 @@ TEST_F(FormatTest, SplitsUTF8Strings) {
       "\"пору,\"",
       format("\"Однажды, в студёную зимнюю пору,\"",
              getLLVMStyleWithColumns(13)));
-  EXPECT_EQ("\"一 二 三 四 \"\n"
-            "\"五 六 七 八 \"\n"
-            "\"九 十\"",
-            format("\"一 二 三 四 五 六 七 八 九 十\"",
-                   getLLVMStyleWithColumns(10)));
+  EXPECT_EQ("\"一 二 三 \"\n"
+            "\"四 五六 \"\n"
+            "\"七 八 九 \"\n"
+            "\"十\"",
+            format("\"一 二 三 四 五六 七 八 九 十\"",
+                   getLLVMStyleWithColumns(11)));
 }
 
 TEST_F(FormatTest, SplitsUTF8LineComments) {
@@ -5739,9 +5740,9 @@ TEST_F(FormatTest, SplitsUTF8LineComments) {
                    getLLVMStyleWithColumns(13)));
   EXPECT_EQ("// 一二三\n"
             "// 四五六七\n"
-            "// 八\n"
-            "// 九 十",
-            format("// 一二三 四五六七 八  九 十", getLLVMStyleWithColumns(6)));
+            "// 八  九\n"
+            "// 十",
+            format("// 一二三 四五六七 八  九 十", getLLVMStyleWithColumns(9)));
 }
 
 TEST_F(FormatTest, SplitsUTF8BlockComments) {
@@ -5758,16 +5759,17 @@ TEST_F(FormatTest, SplitsUTF8BlockComments) {
                    getLLVMStyleWithColumns(13)));
   EXPECT_EQ("/* 一二三\n"
             " * 四五六七\n"
-            " * 八\n"
-            " * 九 十\n"
-            " */",
-            format("/* 一二三 四五六七 八  九 十 */", getLLVMStyleWithColumns(6)));
+            " * 八  九\n"
+            " * 十  */",
+            format("/* 一二三 四五六七 八  九 十  */", getLLVMStyleWithColumns(9)));
   EXPECT_EQ("/* 𝓣𝓮𝓼𝓽 𝔣𝔬𝔲𝔯\n"
             " * 𝕓𝕪𝕥𝕖\n"
             " * 𝖀𝕿𝕱-𝟠 */",
             format("/* 𝓣𝓮𝓼𝓽 𝔣𝔬𝔲𝔯 𝕓𝕪𝕥𝕖 𝖀𝕿𝕱-𝟠 */", getLLVMStyleWithColumns(12)));
 }
 
+#endif // _MSC_VER
+
 TEST_F(FormatTest, FormatsWithWebKitStyle) {
   FormatStyle Style = getWebKitStyle();
 
@@ -5847,7 +5849,5 @@ TEST_F(FormatTest, FormatsWithWebKitStyle) {
             format("if (aaaaaaaaaaaaaaa || bbbbbbbbbbbbbbb) { i++; }", Style));
 }
 
-#endif
-
 } // end namespace tooling
 } // end namespace clang