[clang-format] Add option to remove leading blank lines (#91221)

The options regarding which blank lines are kept are also aggregated.
The new option is `KeepEmptyLines`.

7 files changed, 124 insertions, 41 deletions

diff --git a/clang/docs/ClangFormatStyleOptions.rst b/clang/docs/ClangFormatStyleOptions.rst
index bb00c209..080cba9 100644
--- a/clang/docs/ClangFormatStyleOptions.rst
+++ b/clang/docs/ClangFormatStyleOptions.rst
@@ -4443,23 +4443,51 @@ the configuration (without a prefix: ``Auto``).
      false:
      import {VeryLongImportsAreAnnoying, VeryLongImportsAreAnnoying, VeryLongImportsAreAnnoying,} from "some/module.js"
 
+.. _KeepEmptyLines:
+
+**KeepEmptyLines** (``KeepEmptyLinesStyle``) :versionbadge:`clang-format 19` :ref:`¶ <KeepEmptyLines>`
+  Which empty lines are kept.  See ``MaxEmptyLinesToKeep`` for how many
+  consecutive empty lines are kept.
+
+  Nested configuration flags:
+
+  Options regarding which empty lines are kept.
+
+  For example, the config below will remove empty lines at start of the
+  file, end of the file, and start of blocks.
+
+
+  .. code-block:: c++
+
+    KeepEmptyLines:
+      AtEndOfFile: false
+      AtStartOfBlock: false
+      AtStartOfFile: false
+
+  * ``bool AtEndOfFile`` Keep empty lines at end of file.
+
+  * ``bool AtStartOfBlock`` Keep empty lines at start of a block.
+
+    .. code-block:: c++
+
+       true:                                  false:
+       if (foo) {                     vs.     if (foo) {
+                                                bar();
+         bar();                               }
+       }
+
+  * ``bool AtStartOfFile`` Keep empty lines at start of file.
+
+
 .. _KeepEmptyLinesAtEOF:
 
 **KeepEmptyLinesAtEOF** (``Boolean``) :versionbadge:`clang-format 17` :ref:`¶ <KeepEmptyLinesAtEOF>`
-  Keep empty lines (up to ``MaxEmptyLinesToKeep``) at end of file.
+  This option is deprecated. See ``AtEndOfFile`` of ``KeepEmptyLines``.
 
 .. _KeepEmptyLinesAtTheStartOfBlocks:
 
 **KeepEmptyLinesAtTheStartOfBlocks** (``Boolean``) :versionbadge:`clang-format 3.7` :ref:`¶ <KeepEmptyLinesAtTheStartOfBlocks>`
-  If true, the empty line at the start of blocks is kept.
-
-  .. code-block:: c++
-
-     true:                                  false:
-     if (foo) {                     vs.     if (foo) {
-                                              bar();
-       bar();                               }
-     }
+  This option is deprecated. See ``AtStartOfBlock`` of ``KeepEmptyLines``.
 
 .. _LambdaBodyIndentation:
 
diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index df579ae..9ed3ff4 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -1110,6 +1110,8 @@ clang-format
 - Adds ``AllowShortCaseExpressionOnASingleLine`` option.
 - Adds ``AlignCaseArrows`` suboption to ``AlignConsecutiveShortCaseStatements``.
 - Adds ``LeftWithLastLine`` suboption to ``AlignEscapedNewlines``.
+- Adds ``KeepEmptyLines`` option to deprecate ``KeepEmptyLinesAtEOF``
+  and ``KeepEmptyLinesAtTheStartOfBlocks``.
 
 libclang
 --------
diff --git a/clang/include/clang/Format/Format.h b/clang/include/clang/Format/Format.h
index 7d257be..efc2e450 100644
--- a/clang/include/clang/Format/Format.h
+++ b/clang/include/clang/Format/Format.h
@@ -3095,20 +3095,49 @@ struct FormatStyle {
   bool JavaScriptWrapImports;
   // clang-format on
 
-  /// Keep empty lines (up to ``MaxEmptyLinesToKeep``) at end of file.
-  /// \version 17
-  bool KeepEmptyLinesAtEOF;
-
-  /// If true, the empty line at the start of blocks is kept.
+  /// Options regarding which empty lines are kept.
+  ///
+  /// For example, the config below will remove empty lines at start of the
+  /// file, end of the file, and start of blocks.
+  ///
   /// \code
-  ///    true:                                  false:
-  ///    if (foo) {                     vs.     if (foo) {
-  ///                                             bar();
-  ///      bar();                               }
-  ///    }
+  ///   KeepEmptyLines:
+  ///     AtEndOfFile: false
+  ///     AtStartOfBlock: false
+  ///     AtStartOfFile: false
   /// \endcode
+  struct KeepEmptyLinesStyle {
+    /// Keep empty lines at end of file.
+    bool AtEndOfFile;
+    /// Keep empty lines at start of a block.
+    /// \code
+    ///    true:                                  false:
+    ///    if (foo) {                     vs.     if (foo) {
+    ///                                             bar();
+    ///      bar();                               }
+    ///    }
+    /// \endcode
+    bool AtStartOfBlock;
+    /// Keep empty lines at start of file.
+    bool AtStartOfFile;
+    bool operator==(const KeepEmptyLinesStyle &R) const {
+      return AtEndOfFile == R.AtEndOfFile &&
+             AtStartOfBlock == R.AtStartOfBlock &&
+             AtStartOfFile == R.AtStartOfFile;
+    }
+  };
+  /// Which empty lines are kept.  See ``MaxEmptyLinesToKeep`` for how many
+  /// consecutive empty lines are kept.
+  /// \version 19
+  KeepEmptyLinesStyle KeepEmptyLines;
+
+  /// This option is deprecated. See ``AtEndOfFile`` of ``KeepEmptyLines``.
+  /// \version 17
+  // bool KeepEmptyLinesAtEOF;
+
+  /// This option is deprecated. See ``AtStartOfBlock`` of ``KeepEmptyLines``.
   /// \version 3.7
-  bool KeepEmptyLinesAtTheStartOfBlocks;
+  // bool KeepEmptyLinesAtTheStartOfBlocks;
 
   /// Indentation logic for lambda bodies.
   enum LambdaBodyIndentationKind : int8_t {
@@ -5033,10 +5062,7 @@ struct FormatStyle {
            JavaImportGroups == R.JavaImportGroups &&
            JavaScriptQuotes == R.JavaScriptQuotes &&
            JavaScriptWrapImports == R.JavaScriptWrapImports &&
-           KeepEmptyLinesAtEOF == R.KeepEmptyLinesAtEOF &&
-           KeepEmptyLinesAtTheStartOfBlocks ==
-               R.KeepEmptyLinesAtTheStartOfBlocks &&
-           Language == R.Language &&
+           KeepEmptyLines == R.KeepEmptyLines && Language == R.Language &&
            LambdaBodyIndentation == R.LambdaBodyIndentation &&
            LineEnding == R.LineEnding && MacroBlockBegin == R.MacroBlockBegin &&
            MacroBlockEnd == R.MacroBlockEnd && Macros == R.Macros &&
diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp
index cd21fbb..259ef1d 100644
--- a/clang/lib/Format/Format.cpp
+++ b/clang/lib/Format/Format.cpp
@@ -369,6 +369,14 @@ template <> struct ScalarEnumerationTraits<FormatStyle::JavaScriptQuoteStyle> {
   }
 };
 
+template <> struct MappingTraits<FormatStyle::KeepEmptyLinesStyle> {
+  static void mapping(IO &IO, FormatStyle::KeepEmptyLinesStyle &Value) {
+    IO.mapOptional("AtEndOfFile", Value.AtEndOfFile);
+    IO.mapOptional("AtStartOfBlock", Value.AtStartOfBlock);
+    IO.mapOptional("AtStartOfFile", Value.AtStartOfFile);
+  }
+};
+
 template <> struct ScalarEnumerationTraits<FormatStyle::LanguageKind> {
   static void enumeration(IO &IO, FormatStyle::LanguageKind &Value) {
     IO.enumCase(Value, "Cpp", FormatStyle::LK_Cpp);
@@ -869,6 +877,9 @@ template <> struct MappingTraits<FormatStyle> {
                      OnCurrentLine);
       IO.mapOptional("DeriveLineEnding", DeriveLineEnding);
       IO.mapOptional("DerivePointerBinding", Style.DerivePointerAlignment);
+      IO.mapOptional("KeepEmptyLinesAtEOF", Style.KeepEmptyLines.AtEndOfFile);
+      IO.mapOptional("KeepEmptyLinesAtTheStartOfBlocks",
+                     Style.KeepEmptyLines.AtStartOfBlock);
       IO.mapOptional("IndentFunctionDeclarationAfterType",
                      Style.IndentWrappedFunctionNames);
       IO.mapOptional("IndentRequires", Style.IndentRequiresClause);
@@ -1004,9 +1015,7 @@ template <> struct MappingTraits<FormatStyle> {
     IO.mapOptional("JavaImportGroups", Style.JavaImportGroups);
     IO.mapOptional("JavaScriptQuotes", Style.JavaScriptQuotes);
     IO.mapOptional("JavaScriptWrapImports", Style.JavaScriptWrapImports);
-    IO.mapOptional("KeepEmptyLinesAtTheStartOfBlocks",
-                   Style.KeepEmptyLinesAtTheStartOfBlocks);
-    IO.mapOptional("KeepEmptyLinesAtEOF", Style.KeepEmptyLinesAtEOF);
+    IO.mapOptional("KeepEmptyLines", Style.KeepEmptyLines);
     IO.mapOptional("LambdaBodyIndentation", Style.LambdaBodyIndentation);
     IO.mapOptional("LineEnding", Style.LineEnding);
     IO.mapOptional("MacroBlockBegin", Style.MacroBlockBegin);
@@ -1517,8 +1526,11 @@ FormatStyle getLLVMStyle(FormatStyle::LanguageKind Language) {
       /*Hex=*/0,     /*HexMinDigits=*/0};
   LLVMStyle.JavaScriptQuotes = FormatStyle::JSQS_Leave;
   LLVMStyle.JavaScriptWrapImports = true;
-  LLVMStyle.KeepEmptyLinesAtEOF = false;
-  LLVMStyle.KeepEmptyLinesAtTheStartOfBlocks = true;
+  LLVMStyle.KeepEmptyLines = {
+      /*AtEndOfFile=*/false,
+      /*AtStartOfBlock=*/true,
+      /*AtStartOfFile=*/true,
+  };
   LLVMStyle.LambdaBodyIndentation = FormatStyle::LBI_Signature;
   LLVMStyle.Language = Language;
   LLVMStyle.LineEnding = FormatStyle::LE_DeriveLF;
@@ -1641,7 +1653,7 @@ FormatStyle getGoogleStyle(FormatStyle::LanguageKind Language) {
                                                 {".*", 3, 0, false}};
   GoogleStyle.IncludeStyle.IncludeIsMainRegex = "([-_](test|unittest))?$";
   GoogleStyle.IndentCaseLabels = true;
-  GoogleStyle.KeepEmptyLinesAtTheStartOfBlocks = false;
+  GoogleStyle.KeepEmptyLines.AtStartOfBlock = false;
   GoogleStyle.ObjCBinPackProtocolList = FormatStyle::BPS_Never;
   GoogleStyle.ObjCSpaceAfterProperty = false;
   GoogleStyle.ObjCSpaceBeforeProtocolList = true;
diff --git a/clang/lib/Format/UnwrappedLineFormatter.cpp b/clang/lib/Format/UnwrappedLineFormatter.cpp
index 729f3d7..1804c14 100644
--- a/clang/lib/Format/UnwrappedLineFormatter.cpp
+++ b/clang/lib/Format/UnwrappedLineFormatter.cpp
@@ -1478,11 +1478,13 @@ static auto computeNewlines(const AnnotatedLine &Line,
     Newlines = std::min(Newlines, 1u);
   if (Newlines == 0 && !RootToken.IsFirst)
     Newlines = 1;
-  if (RootToken.IsFirst && !RootToken.HasUnescapedNewline)
+  if (RootToken.IsFirst &&
+      (!Style.KeepEmptyLines.AtStartOfFile || !RootToken.HasUnescapedNewline)) {
     Newlines = 0;
+  }
 
   // Remove empty lines after "{".
-  if (!Style.KeepEmptyLinesAtTheStartOfBlocks && PreviousLine &&
+  if (!Style.KeepEmptyLines.AtStartOfBlock && PreviousLine &&
       PreviousLine->Last->is(tok::l_brace) &&
       !PreviousLine->startsWithNamespace() &&
       !(PrevPrevLine && PrevPrevLine->startsWithNamespace() &&
@@ -1554,9 +1556,9 @@ void UnwrappedLineFormatter::formatFirstToken(
     unsigned NewlineIndent) {
   FormatToken &RootToken = *Line.First;
   if (RootToken.is(tok::eof)) {
-    unsigned Newlines =
-        std::min(RootToken.NewlinesBefore,
-                 Style.KeepEmptyLinesAtEOF ? Style.MaxEmptyLinesToKeep + 1 : 1);
+    unsigned Newlines = std::min(
+        RootToken.NewlinesBefore,
+        Style.KeepEmptyLines.AtEndOfFile ? Style.MaxEmptyLinesToKeep + 1 : 1);
     unsigned TokenIndent = Newlines ? NewlineIndent : 0;
     Whitespaces->replaceWhitespace(RootToken, Newlines, TokenIndent,
                                    TokenIndent);
diff --git a/clang/unittests/Format/ConfigParseTest.cpp b/clang/unittests/Format/ConfigParseTest.cpp
index aded3ed..2466677a 100644
--- a/clang/unittests/Format/ConfigParseTest.cpp
+++ b/clang/unittests/Format/ConfigParseTest.cpp
@@ -178,8 +178,9 @@ TEST(ConfigParseTest, ParsesConfigurationBools) {
   CHECK_PARSE_BOOL(IndentWrappedFunctionNames);
   CHECK_PARSE_BOOL(InsertBraces);
   CHECK_PARSE_BOOL(InsertNewlineAtEOF);
-  CHECK_PARSE_BOOL(KeepEmptyLinesAtEOF);
-  CHECK_PARSE_BOOL(KeepEmptyLinesAtTheStartOfBlocks);
+  CHECK_PARSE_BOOL_FIELD(KeepEmptyLines.AtEndOfFile, "KeepEmptyLinesAtEOF");
+  CHECK_PARSE_BOOL_FIELD(KeepEmptyLines.AtStartOfBlock,
+                         "KeepEmptyLinesAtTheStartOfBlocks");
   CHECK_PARSE_BOOL(ObjCSpaceAfterProperty);
   CHECK_PARSE_BOOL(ObjCSpaceBeforeProtocolList);
   CHECK_PARSE_BOOL(Cpp11BracedListStyle);
@@ -226,6 +227,9 @@ TEST(ConfigParseTest, ParsesConfigurationBools) {
   CHECK_PARSE_NESTED_BOOL(BraceWrapping, SplitEmptyFunction);
   CHECK_PARSE_NESTED_BOOL(BraceWrapping, SplitEmptyRecord);
   CHECK_PARSE_NESTED_BOOL(BraceWrapping, SplitEmptyNamespace);
+  CHECK_PARSE_NESTED_BOOL(KeepEmptyLines, AtEndOfFile);
+  CHECK_PARSE_NESTED_BOOL(KeepEmptyLines, AtStartOfBlock);
+  CHECK_PARSE_NESTED_BOOL(KeepEmptyLines, AtStartOfFile);
   CHECK_PARSE_NESTED_BOOL(SpaceBeforeParensOptions, AfterControlStatements);
   CHECK_PARSE_NESTED_BOOL(SpaceBeforeParensOptions, AfterForeachMacros);
   CHECK_PARSE_NESTED_BOOL(SpaceBeforeParensOptions,
diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index db1decb..5276e79 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -45,6 +45,10 @@ TEST_F(FormatTest, FormatsGlobalStatementsAt0) {
   verifyFormat("\nint i;", " \n\t \v \f  int i;");
   verifyFormat("int i;\nint j;", "    int i; int j;");
   verifyFormat("int i;\nint j;", "    int i;\n  int j;");
+
+  auto Style = getLLVMStyle();
+  Style.KeepEmptyLines.AtStartOfFile = false;
+  verifyFormat("int i;", " \n\t \v \f  int i;", Style);
 }
 
 TEST_F(FormatTest, FormatsUnwrappedLinesAtFirstFormat) {
@@ -163,7 +167,7 @@ TEST_F(FormatTest, RemovesEmptyLines) {
   auto CustomStyle = getLLVMStyle();
   CustomStyle.BreakBeforeBraces = FormatStyle::BS_Custom;
   CustomStyle.BraceWrapping.AfterNamespace = true;
-  CustomStyle.KeepEmptyLinesAtTheStartOfBlocks = false;
+  CustomStyle.KeepEmptyLines.AtStartOfBlock = false;
   verifyFormat("namespace N\n"
                "{\n"
                "\n"
@@ -389,7 +393,7 @@ TEST_F(FormatTest, RemovesEmptyLines) {
   Style.BreakBeforeBraces = FormatStyle::BS_Custom;
   Style.BraceWrapping.AfterClass = true;
   Style.BraceWrapping.AfterFunction = true;
-  Style.KeepEmptyLinesAtTheStartOfBlocks = false;
+  Style.KeepEmptyLines.AtStartOfBlock = false;
 
   verifyFormat("class Foo\n"
                "{\n"
@@ -21956,6 +21960,11 @@ TEST_F(FormatTest, HandlesUTF8BOM) {
   verifyFormat("\xef\xbb\xbf");
   verifyFormat("\xef\xbb\xbf#include <iostream>");
   verifyFormat("\xef\xbb\xbf\n#include <iostream>");
+
+  auto Style = getLLVMStyle();
+  Style.KeepEmptyLines.AtStartOfFile = false;
+  verifyFormat("\xef\xbb\xbf#include <iostream>",
+               "\xef\xbb\xbf\n#include <iostream>", Style);
 }
 
 // FIXME: Encode Cyrillic and CJK characters below to appease MS compilers.
@@ -27230,7 +27239,7 @@ TEST_F(FormatTest, InsertNewlineAtEOF) {
 
 TEST_F(FormatTest, KeepEmptyLinesAtEOF) {
   FormatStyle Style = getLLVMStyle();
-  Style.KeepEmptyLinesAtEOF = true;
+  Style.KeepEmptyLines.AtEndOfFile = true;
 
   const StringRef Code{"int i;\n\n"};
   verifyNoChange(Code, Style);