[clang-format] Add an option to insert a newline at EOF if missing

[owenpan]

Summary

Closes https://github.com/llvm/llvm-project/issues/38042.

[MyDeveloperDay]

I think I've died an gone to heaven!! LGTM... Happy New Year!

[owenpan]

In D141035#4028134, @MyDeveloperDay wrote:

I think I've died an gone to heaven!! LGTM... Happy New Year!

Happy New Year to you too! Thanks to your comment in D19031#1747112, I remembered that I had fixed it even before reporting the bug back in 2018!

clang/unittests/Format/FormatTest.cpp

25125–25126

Will remove them before landing.

clang/docs/ClangFormatStyleOptions.rst

while (i--)                      vs.      while (i--) {
         handleAttr(A);                            handleAttr(A);
                                                 }
                                               }
 
     do                               vs.      do {
       --i;                                      --i;
     while (i);                                } while (i);
 
+**InsertNewlineAtEOF** (``Boolean``) :versionbadge:`clang-format 16`
+  Insert a newline at end of file if missing.
+
 **InsertTrailingCommas** (``TrailingCommaStyle``) :versionbadge:`clang-format 11`
   If set to ``TCS_Wrapped`` will insert trailing commas in container
   literals (arrays and objects) that wrap across multiple lines.
   It is currently only available for JavaScript
   and disabled by default ``TCS_None``.
   ``InsertTrailingCommas`` cannot be used together with ``BinPackArguments``
   as inserting the comma disables bin-packing.
 

clang/docs/ReleaseNotes.rst

 - Add ``RemoveSemicolon`` option for removing ``;`` after a non-empty function definition.
 - Add ``RequiresExpressionIndentation`` option for configuring the alignment of requires-expressions.
   The default value of this option is ``OuterScope``, which differs in behavior from clang-format 15.
   To match the default behavior of clang-format 15, use the ``Keyword`` value.
 - Add ``IntegerLiteralSeparator`` option for fixing integer literal separators
   in C++, C#, Java, and JavaScript.
 - Add ``BreakAfterAttributes`` option for breaking after a group of C++11
   attributes before a function declaration/definition name.
+- Add ``InsertNewlineAtEOF`` option for inserting a newline at EOF if missing.
 
 clang-extdef-mapping
 --------------------
 
 libclang
 --------
 - Introduced the new function ``clang_getUnqualifiedType``, which mimics
   the behavior of ``QualType::getUnqualifiedType`` for ``CXType``.

clang/include/clang/Format/Format.h

   ///
   ///   do                               vs.      do {
   ///     --i;                                      --i;
   ///   while (i);                                } while (i);
   /// \endcode
   /// \version 15
   bool InsertBraces;
 
+  /// Insert a newline at end of file if missing.
+  /// \version 16
+  bool InsertNewlineAtEOF;
+
   /// The style of inserting trailing commas into container literals.
   enum TrailingCommaStyle : int8_t {
     /// Do not insert trailing commas.
     TCS_None,
     /// Insert trailing commas in container literals that were wrapped over
     /// multiple lines. Note that this is conceptually incompatible with
     /// bin-packing, because the trailing comma is used as an indicator
     /// that a container should be formatted one-per-line (i.e. not bin-packed).
return AccessModifierOffset == R.AccessModifierOffset &&
            IndentCaseLabels == R.IndentCaseLabels &&
            IndentExternBlock == R.IndentExternBlock &&
            IndentGotoLabels == R.IndentGotoLabels &&
            IndentPPDirectives == R.IndentPPDirectives &&
            IndentRequiresClause == R.IndentRequiresClause &&
            IndentWidth == R.IndentWidth &&
            IndentWrappedFunctionNames == R.IndentWrappedFunctionNames &&
            InsertBraces == R.InsertBraces &&
+           InsertNewlineAtEOF == R.InsertNewlineAtEOF &&
            IntegerLiteralSeparator.Binary == R.IntegerLiteralSeparator.Binary &&
            IntegerLiteralSeparator.Decimal ==
                R.IntegerLiteralSeparator.Decimal &&
            IntegerLiteralSeparator.Hex == R.IntegerLiteralSeparator.Hex &&
            JavaImportGroups == R.JavaImportGroups &&
            JavaScriptQuotes == R.JavaScriptQuotes &&
            JavaScriptWrapImports == R.JavaScriptWrapImports &&
            KeepEmptyLinesAtTheStartOfBlocks ==

clang/lib/Format/Format.cpp

static void mapping(IO &IO, FormatStyle &Style) {
     IO.mapOptional("IndentExternBlock", Style.IndentExternBlock);
     IO.mapOptional("IndentGotoLabels", Style.IndentGotoLabels);
     IO.mapOptional("IndentPPDirectives", Style.IndentPPDirectives);
     IO.mapOptional("IndentRequiresClause", Style.IndentRequiresClause);
     IO.mapOptional("IndentWidth", Style.IndentWidth);
     IO.mapOptional("IndentWrappedFunctionNames",
                    Style.IndentWrappedFunctionNames);
     IO.mapOptional("InsertBraces", Style.InsertBraces);
+    IO.mapOptional("InsertNewlineAtEOF", Style.InsertNewlineAtEOF);
     IO.mapOptional("InsertTrailingCommas", Style.InsertTrailingCommas);
     IO.mapOptional("IntegerLiteralSeparator", Style.IntegerLiteralSeparator);
     IO.mapOptional("JavaImportGroups", Style.JavaImportGroups);
     IO.mapOptional("JavaScriptQuotes", Style.JavaScriptQuotes);
     IO.mapOptional("JavaScriptWrapImports", Style.JavaScriptWrapImports);
     IO.mapOptional("KeepEmptyLinesAtTheStartOfBlocks",
                    Style.KeepEmptyLinesAtTheStartOfBlocks);
     IO.mapOptional("LambdaBodyIndentation", Style.LambdaBodyIndentation);
FormatStyle getLLVMStyle(FormatStyle::LanguageKind Language) {
   LLVMStyle.IndentCaseBlocks = false;
   LLVMStyle.IndentExternBlock = FormatStyle::IEBS_AfterExternBlock;
   LLVMStyle.IndentGotoLabels = true;
   LLVMStyle.IndentPPDirectives = FormatStyle::PPDIS_None;
   LLVMStyle.IndentRequiresClause = true;
   LLVMStyle.IndentWidth = 2;
   LLVMStyle.IndentWrappedFunctionNames = false;
   LLVMStyle.InsertBraces = false;
+  LLVMStyle.InsertNewlineAtEOF = false;
   LLVMStyle.InsertTrailingCommas = FormatStyle::TCS_None;
   LLVMStyle.IntegerLiteralSeparator = {/*Binary=*/0, /*Decimal=*/0, /*Hex=*/0};
   LLVMStyle.JavaScriptQuotes = FormatStyle::JSQS_Leave;
   LLVMStyle.JavaScriptWrapImports = true;
   LLVMStyle.KeepEmptyLinesAtTheStartOfBlocks = true;
   LLVMStyle.LambdaBodyIndentation = FormatStyle::LBI_Signature;
   LLVMStyle.MaxEmptyLinesToKeep = 1;
   LLVMStyle.NamespaceIndentation = FormatStyle::NI_None;

clang/lib/Format/TokenAnnotator.cpp

case tok::identifier:
       }
       break;
     case tok::arrow:
       if (Tok->isNot(TT_LambdaArrow) && Tok->Previous &&
           Tok->Previous->is(tok::kw_noexcept)) {
         Tok->setType(TT_TrailingReturnArrow);
       }
       break;
+    case tok::eof:
+      if (Style.InsertNewlineAtEOF && Tok->NewlinesBefore == 0)
+        Tok->NewlinesBefore = 1;
+      break;
     default:
       break;
     }
     return true;
   }
 
   void parseCSharpGenericTypeConstraint() {
     int OpenAngleBracketsCount = 0;

clang/unittests/Format/ConfigParseTest.cpp

TEST(ConfigParseTest, ParsesConfigurationBools) {
   CHECK_PARSE_BOOL(IndentAccessModifiers);
   CHECK_PARSE_BOOL(IndentCaseLabels);
   CHECK_PARSE_BOOL(IndentCaseBlocks);
   CHECK_PARSE_BOOL(IndentGotoLabels);
   CHECK_PARSE_BOOL_FIELD(IndentRequiresClause, "IndentRequires");
   CHECK_PARSE_BOOL(IndentRequiresClause);
   CHECK_PARSE_BOOL(IndentWrappedFunctionNames);
   CHECK_PARSE_BOOL(InsertBraces);
+  CHECK_PARSE_BOOL(InsertNewlineAtEOF);
   CHECK_PARSE_BOOL(KeepEmptyLinesAtTheStartOfBlocks);
   CHECK_PARSE_BOOL(ObjCSpaceAfterProperty);
   CHECK_PARSE_BOOL(ObjCSpaceBeforeProtocolList);
   CHECK_PARSE_BOOL(Cpp11BracedListStyle);
   CHECK_PARSE_BOOL(ReflowComments);
   CHECK_PARSE_BOOL(RemoveBracesLLVM);
   CHECK_PARSE_BOOL(RemoveSemicolon);
   CHECK_PARSE_BOOL(SortUsingDeclarations);

clang/unittests/Format/FormatTest.cpp

verifyFormat("void foo() {} //\n"
                "int bar;",
                "void foo() {}; //\n"
                "; int bar;",
                Style);
 #endif
 }
 
 TEST_F(FormatTest, BreakAfterAttributes) {
   FormatStyle Style = getLLVMStyle();
   EXPECT_EQ(Style.BreakAfterAttributes, FormatStyle::ABS_Never);
 
   const StringRef Code("[[nodiscard]] inline int f(int &i);\n"
                        "[[foo([[]])]] [[nodiscard]]\n"
                        "int g(int &i);\n"
                        "[[nodiscard]]\n"
                        "inline int f(int &i) {\n"
                        "  i = 1;\n"
                        "  return 0;\n"
verifyFormat("[[nodiscard]]\n"
                "  return 1;\n"
                "}",
                Code, Style);
 
   Style.BreakAfterAttributes = FormatStyle::ABS_Leave;
   EXPECT_EQ(Code, format(Code, Style));
 }
 
+TEST_F(FormatTest, InsertNewlineAtEOF) {
+  FormatStyle Style = getLLVMStyle();
+  Style.InsertNewlineAtEOF = true;
+
+  verifyFormat("int i;\n", Style);
+  verifyFormat("int i;\n", "int i;", Style);
+}
+
 } // namespace
 } // namespace format
 } // namespace clang