clang-format: Fix corner case for string splitting ..

.. in conjunction with Style.AlwaysBreakBeforeMultilineStrings. Also,
simplify the implementation by handling newly split strings and already
split strings by the same code.

llvm-svn: 189102

Author: djasper-gh

clang/lib/Format/ContinuationIndenter.cpp

@@ -116,12 +116,9 @@ bool ContinuationIndenter::mustBreak(const LineState &State) {
       !Current.isOneOf(tok::r_paren, tok::r_brace))
     return true;
   if (Style.AlwaysBreakBeforeMultilineStrings &&
-      State.Column > State.Stack.back().Indent &&
-      Current.is(tok::string_literal) && Previous.isNot(tok::lessless) &&
-      Previous.Type != TT_InlineASMColon &&
-      ((Current.getNextNonComment() &&
-        Current.getNextNonComment()->is(tok::string_literal)) ||
-       (Current.TokenText.find("\\\n") != StringRef::npos)))
+      State.Column > State.Stack.back().Indent && // Breaking saves columns.
+      Previous.isNot(tok::lessless) && Previous.Type != TT_InlineASMColon &&
+      NextIsMultilineString(State))
     return true;
 
   if (!Style.BreakBeforeBinaryOperators) {
@@ -547,13 +544,8 @@ unsigned ContinuationIndenter::moveStateToNextToken(LineState &State,
   }
 
   State.Column += Current.CodePointCount;
-
   State.NextToken = State.NextToken->Next;
-
-  unsigned Penalty = 0;
-  if (Newline || !Style.AlwaysBreakBeforeMultilineStrings ||
-      Current.isNot(tok::string_literal) || !Current.CanBreakBefore)
-    Penalty += breakProtrudingToken(Current, State, DryRun);
+  unsigned Penalty =  breakProtrudingToken(Current, State, DryRun);
 
   // If the previous has a special role, let it consume tokens as appropriate.
   // It is necessary to start at the previous token for the only implemented
@@ -688,5 +680,19 @@ unsigned ContinuationIndenter::getColumnLimit() const {
   return Style.ColumnLimit - (Line.InPPDirective ? 2 : 0);
 }
 
+bool ContinuationIndenter::NextIsMultilineString(const LineState &State) {
+  const FormatToken &Current = *State.NextToken;
+  if (!Current.is(tok::string_literal))
+    return false;
+  if (Current.getNextNonComment() &&
+      Current.getNextNonComment()->is(tok::string_literal))
+    return true; // Implicit concatenation.
+  if (State.Column + Current.CodePointCount + Current.UnbreakableTailLength >
+      Style.ColumnLimit)
+    return true; // String will be split.
+  // String literal might have escaped newlines.
+  return Current.TokenText.find("\\\n") != StringRef::npos;
+}
+
 } // namespace format
 } // namespace clang

clang/lib/Format/ContinuationIndenter.h

@@ -84,6 +84,13 @@ class ContinuationIndenter {
   unsigned breakProtrudingToken(const FormatToken &Current, LineState &State,
                                 bool DryRun);
 
+  /// \brief Returns \c true if the next token starts a multiline string
+  /// literal.
+  ///
+  /// This includes implicitly concatenated strings, strings that will be broken
+  /// by clang-format and string literals with escaped newlines.
+  bool NextIsMultilineString(const LineState &State);
+
   FormatStyle Style;
   SourceManager &SourceMgr;
   const AnnotatedLine &Line;

clang/unittests/Format/FormatTest.cpp

@@ -5265,6 +5265,12 @@ TEST_F(FormatTest, BreakStringLiterals) {
             format("llvm::outs() << "
                    "\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaa"
                    "aaaaaaaaaaaaaaaaaaa\";"));
+  EXPECT_EQ("ffff(\n"
+            "    {\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa \"\n"
+            "     \"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\"});",
+            format("ffff({\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa "
+                   "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\"});",
+                   getGoogleStyle()));
 
   FormatStyle AlignLeft = getLLVMStyleWithColumns(12);
   AlignLeft.AlignEscapedNewlinesLeft = true;