r197368 - clang-format: Improve handling of raw string literals.

[Daniel Jasper]

Author: djasper
Date: Mon Dec 16 01:23:08 2013
New Revision: 197368

URL: http://llvm.org/viewvc/llvm-project?rev=197368&view=rev
Log:
clang-format: Improve handling of raw string literals.

Especially try to keep existing line breaks before raw string literals,
as the code author might have aligned content to it.

Thereby, clang-format now keeps things like:

  parseStyle(R"(
      BasedOnStyle: Google,
      ColumnLimit: 100)");
  parseStyle(
      R"(BasedOnStyle: Google,
         ColumnLimit: 100)");

Modified:
    cfe/trunk/lib/Format/ContinuationIndenter.cpp
    cfe/trunk/lib/Format/ContinuationIndenter.h
    cfe/trunk/lib/Format/TokenAnnotator.cpp
    cfe/trunk/unittests/Format/FormatTest.cpp

Modified: cfe/trunk/lib/Format/ContinuationIndenter.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Format/ContinuationIndenter.cpp?rev=197368&r1=197367&r2=197368&view=diff
==============================================================================
--- cfe/trunk/lib/Format/ContinuationIndenter.cpp (original)
+++ cfe/trunk/lib/Format/ContinuationIndenter.cpp Mon Dec 16 01:23:08 2013
@@ -136,7 +136,7 @@ bool ContinuationIndenter::mustBreak(con
   if (Style.AlwaysBreakBeforeMultilineStrings &&
       State.Column > State.Stack.back().Indent && // Breaking saves columns.
       !Previous.isOneOf(tok::kw_return, tok::lessless, tok::at) &&
-      Previous.Type != TT_InlineASMColon && NextIsMultilineString(State))
+      Previous.Type != TT_InlineASMColon && nextIsMultilineString(State))
     return true;
   if (((Previous.Type == TT_DictLiteral && Previous.is(tok::l_brace)) ||
        Previous.Type == TT_ArrayInitializerLSquare) &&
@@ -865,12 +865,13 @@ unsigned ContinuationIndenter::getColumn
   return Style.ColumnLimit - (State.Line->InPPDirective ? 2 : 0);
 }
 
-bool ContinuationIndenter::NextIsMultilineString(const LineState &State) {
+bool ContinuationIndenter::nextIsMultilineString(const LineState &State) {
   const FormatToken &Current = *State.NextToken;
   if (!Current.is(tok::string_literal))
     return false;
   // We never consider raw string literals "multiline" for the purpose of
-  // AlwaysBreakBeforeMultilineStrings implementation.
+  // AlwaysBreakBeforeMultilineStrings implementation as they are special-cased
+  // (see TokenAnnotator::mustBreakBefore().
   if (Current.TokenText.startswith("R\""))
     return false;
   if (Current.IsMultiline)

Modified: cfe/trunk/lib/Format/ContinuationIndenter.h
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Format/ContinuationIndenter.h?rev=197368&r1=197367&r2=197368&view=diff
==============================================================================
--- cfe/trunk/lib/Format/ContinuationIndenter.h (original)
+++ cfe/trunk/lib/Format/ContinuationIndenter.h Mon Dec 16 01:23:08 2013
@@ -115,7 +115,7 @@ private:
   ///
   /// This includes implicitly concatenated strings, strings that will be broken
   /// by clang-format and string literals with escaped newlines.
-  bool NextIsMultilineString(const LineState &State);
+  bool nextIsMultilineString(const LineState &State);
 
   FormatStyle Style;
   SourceManager &SourceMgr;

Modified: cfe/trunk/lib/Format/TokenAnnotator.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Format/TokenAnnotator.cpp?rev=197368&r1=197367&r2=197368&view=diff
==============================================================================
--- cfe/trunk/lib/Format/TokenAnnotator.cpp (original)
+++ cfe/trunk/lib/Format/TokenAnnotator.cpp Mon Dec 16 01:23:08 2013
@@ -1412,6 +1412,12 @@ bool TokenAnnotator::mustBreakBefore(con
   } else if (Right.is(tok::l_brace) && (Right.BlockKind == BK_Block)) {
     return Style.BreakBeforeBraces == FormatStyle::BS_Allman ||
            Style.BreakBeforeBraces == FormatStyle::BS_GNU;
+  } else if (Right.is(tok::string_literal) &&
+             Right.TokenText.startswith("R\"")) {
+    // Raw string literals are special wrt. line breaks. The author has made a
+    // deliberate choice and might have aligned the contents of the string
+    // literal accordingly. Thus, we try keep existing line breaks.
+    return Right.NewlinesBefore > 0;
   }
   return false;
 }

Modified: cfe/trunk/unittests/Format/FormatTest.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/unittests/Format/FormatTest.cpp?rev=197368&r1=197367&r2=197368&view=diff
==============================================================================
--- cfe/trunk/unittests/Format/FormatTest.cpp (original)
+++ cfe/trunk/unittests/Format/FormatTest.cpp Mon Dec 16 01:23:08 2013
@@ -6187,7 +6187,7 @@ TEST_F(FormatTest, DontSplitStringLitera
 
 TEST_F(FormatTest, CountsCharactersInMultilineRawStringLiterals) {
   EXPECT_EQ("f(g(R\"x(raw literal)x\", a), b);",
-            format("f(g(R\"x(raw literal)x\", a), b);", getGoogleStyle()));
+            format("f(g(R\"x(raw literal)x\",   a), b);", getGoogleStyle()));
   EXPECT_EQ("fffffffffff(g(R\"x(\n"
             "multiline raw string literal xxxxxxxxxxxxxx\n"
             ")x\",\n"
@@ -6223,6 +6223,16 @@ TEST_F(FormatTest, CountsCharactersInMul
                    "multiline raw string literal xxxxxxxxxxxxxx\n"
                    ")x\" + bbbbbb);",
                    getGoogleStyleWithColumns(20)));
+  EXPECT_EQ("fffffffffff(\n"
+            "    R\"x(\n"
+            "multiline raw string literal xxxxxxxxxxxxxx\n"
+            ")x\" +\n"
+            "    bbbbbb);",
+            format("fffffffffff(\n"
+                   " R\"x(\n"
+                   "multiline raw string literal xxxxxxxxxxxxxx\n"
+                   ")x\" + bbbbbb);",
+                   getGoogleStyleWithColumns(20)));
 }
 
 TEST_F(FormatTest, SkipsUnknownStringLiterals) {