[clang] [clang-format] Add the C language instead of treating it like C++ (PR #128287)
Owen Pan via cfe-commits
cfe-commits at lists.llvm.org
Fri Feb 21 21:45:21 PST 2025
https://github.com/owenca created https://github.com/llvm/llvm-project/pull/128287
Closes #128120
>From 84cfaec1f830463d2332f14eca9cfbb86919d3d5 Mon Sep 17 00:00:00 2001
From: Owen Pan <owenpiano at gmail.com>
Date: Fri, 21 Feb 2025 21:41:36 -0800
Subject: [PATCH] [clang-format] Add the C language instead of treating it like
C++
Closes #128120
---
clang/docs/ClangFormatStyleOptions.rst | 9 ++++++---
clang/docs/ReleaseNotes.rst | 7 ++++---
clang/include/clang/Format/Format.h | 14 ++++++++++----
clang/lib/Format/Format.cpp | 12 +++++++++++-
clang/lib/Format/FormatToken.cpp | 2 +-
clang/lib/Format/TokenAnnotator.cpp | 4 ++--
clang/lib/Format/TokenAnnotator.h | 2 +-
clang/lib/Format/UnwrappedLineParser.cpp | 2 +-
clang/unittests/Format/FormatTest.cpp | 4 ++++
9 files changed, 40 insertions(+), 16 deletions(-)
diff --git a/clang/docs/ClangFormatStyleOptions.rst b/clang/docs/ClangFormatStyleOptions.rst
index ba50eb2c7e89b..d157c07c9cef8 100644
--- a/clang/docs/ClangFormatStyleOptions.rst
+++ b/clang/docs/ClangFormatStyleOptions.rst
@@ -4786,8 +4786,8 @@ the configuration (without a prefix: ``Auto``).
.. note::
- You can also specify the language (``Cpp`` or ``ObjC``) for ``.h`` files
- by adding a ``// clang-format Language:`` line before the first
+ You can specify the language (``C``, ``Cpp``, or ``ObjC``) for ``.h``
+ files by adding a ``// clang-format Language:`` line before the first
non-comment (and non-empty) line, e.g. ``// clang-format Language: Cpp``.
Possible values:
@@ -4795,8 +4795,11 @@ the configuration (without a prefix: ``Auto``).
* ``LK_None`` (in configuration: ``None``)
Do not use.
+ * ``LK_C`` (in configuration: ``C``)
+ Should be used for C.
+
* ``LK_Cpp`` (in configuration: ``Cpp``)
- Should be used for C, C++.
+ Should be used for C++.
* ``LK_CSharp`` (in configuration: ``CSharp``)
Should be used for C#.
diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index ec9cab2eb657f..699cbb17edca7 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -271,9 +271,10 @@ clang-format
- Adds ``BreakBeforeTemplateCloser`` option.
- Adds ``BinPackLongBracedList`` option to override bin packing options in
long (20 item or more) braced list initializer lists.
-- Allow specifying the language (C++ or Objective-C) for a ``.h`` file by adding
- a special comment (e.g. ``// clang-format Language: ObjC``) near the top of
- the file.
+- Add the C language instead of treating it like C++.
+- Allow specifying the language (C, C++, or Objective-C) for a ``.h`` file by
+ adding a special comment (e.g. ``// clang-format Language: ObjC``) near the
+ top of the file.
libclang
--------
diff --git a/clang/include/clang/Format/Format.h b/clang/include/clang/Format/Format.h
index 2dc95c3c06d29..46fb1d52701b3 100644
--- a/clang/include/clang/Format/Format.h
+++ b/clang/include/clang/Format/Format.h
@@ -3318,7 +3318,9 @@ struct FormatStyle {
enum LanguageKind : int8_t {
/// Do not use.
LK_None,
- /// Should be used for C, C++.
+ /// Should be used for C.
+ LK_C,
+ /// Should be used for C++.
LK_Cpp,
/// Should be used for C#.
LK_CSharp,
@@ -3343,7 +3345,9 @@ struct FormatStyle {
/// https://sci-hub.st/10.1109/IEEESTD.2018.8299595
LK_Verilog
};
- bool isCpp() const { return Language == LK_Cpp || Language == LK_ObjC; }
+ bool isCpp() const {
+ return Language == LK_Cpp || Language == LK_C || Language == LK_ObjC;
+ }
bool isCSharp() const { return Language == LK_CSharp; }
bool isJson() const { return Language == LK_Json; }
bool isJavaScript() const { return Language == LK_JavaScript; }
@@ -3355,8 +3359,8 @@ struct FormatStyle {
/// The language that this format style targets.
/// \note
- /// You can also specify the language (``Cpp`` or ``ObjC``) for ``.h`` files
- /// by adding a ``// clang-format Language:`` line before the first
+ /// You can specify the language (``C``, ``Cpp``, or ``ObjC``) for ``.h``
+ /// files by adding a ``// clang-format Language:`` line before the first
/// non-comment (and non-empty) line, e.g. ``// clang-format Language: Cpp``.
/// \endnote
/// \version 3.5
@@ -5715,6 +5719,8 @@ FormatStyle::LanguageKind guessLanguage(StringRef FileName, StringRef Code);
// Returns a string representation of ``Language``.
inline StringRef getLanguageName(FormatStyle::LanguageKind Language) {
switch (Language) {
+ case FormatStyle::LK_C:
+ return "C";
case FormatStyle::LK_Cpp:
return "C++";
case FormatStyle::LK_CSharp:
diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp
index b063843078251..001bc4e522792 100644
--- a/clang/lib/Format/Format.cpp
+++ b/clang/lib/Format/Format.cpp
@@ -401,6 +401,7 @@ template <> struct MappingTraits<FormatStyle::KeepEmptyLinesStyle> {
template <> struct ScalarEnumerationTraits<FormatStyle::LanguageKind> {
static void enumeration(IO &IO, FormatStyle::LanguageKind &Value) {
+ IO.enumCase(Value, "C", FormatStyle::LK_C);
IO.enumCase(Value, "Cpp", FormatStyle::LK_Cpp);
IO.enumCase(Value, "Java", FormatStyle::LK_Java);
IO.enumCase(Value, "JavaScript", FormatStyle::LK_JavaScript);
@@ -3957,7 +3958,12 @@ LangOptions getFormattingLangOpts(const FormatStyle &Style) {
LangOpts.Digraphs = LexingStd >= FormatStyle::LS_Cpp11;
LangOpts.LineComment = 1;
- LangOpts.CXXOperatorNames = Style.isCpp();
+
+ const auto Language = Style.Language;
+ LangOpts.C11 = Language == FormatStyle::LK_C;
+ LangOpts.CXXOperatorNames =
+ Language == FormatStyle::LK_Cpp || Language == FormatStyle::LK_ObjC;
+
LangOpts.Bool = 1;
LangOpts.ObjC = 1;
LangOpts.MicrosoftExt = 1; // To get kw___try, kw___finally.
@@ -3982,6 +3988,8 @@ const char *StyleOptionHelpDescription =
" --style=\"{BasedOnStyle: llvm, IndentWidth: 8}\"";
static FormatStyle::LanguageKind getLanguageByFileName(StringRef FileName) {
+ if (FileName.ends_with(".c"))
+ return FormatStyle::LK_C;
if (FileName.ends_with(".java"))
return FormatStyle::LK_Java;
if (FileName.ends_with_insensitive(".js") ||
@@ -4039,6 +4047,8 @@ static FormatStyle::LanguageKind getLanguageByComment(const Environment &Env) {
continue;
Text = Text.trim();
+ if (Text == "C")
+ return FormatStyle::LK_C;
if (Text == "Cpp")
return FormatStyle::LK_Cpp;
if (Text == "ObjC")
diff --git a/clang/lib/Format/FormatToken.cpp b/clang/lib/Format/FormatToken.cpp
index fb040a0043602..97ee90155d870 100644
--- a/clang/lib/Format/FormatToken.cpp
+++ b/clang/lib/Format/FormatToken.cpp
@@ -44,7 +44,7 @@ static SmallVector<StringRef> CppNonKeywordTypes = {
bool FormatToken::isTypeName(const LangOptions &LangOpts) const {
if (is(TT_TypeName) || Tok.isSimpleTypeSpecifier(LangOpts))
return true;
- const bool IsCpp = LangOpts.CXXOperatorNames;
+ const bool IsCpp = LangOpts.CXXOperatorNames || LangOpts.C11;
return IsCpp && is(tok::identifier) &&
std::binary_search(CppNonKeywordTypes.begin(),
CppNonKeywordTypes.end(), TokenText);
diff --git a/clang/lib/Format/TokenAnnotator.cpp b/clang/lib/Format/TokenAnnotator.cpp
index e68daa422b7c4..24e66333b861b 100644
--- a/clang/lib/Format/TokenAnnotator.cpp
+++ b/clang/lib/Format/TokenAnnotator.cpp
@@ -129,7 +129,7 @@ class AnnotatingParser {
: Style(Style), Line(Line), CurrentToken(Line.First), AutoFound(false),
IsCpp(Style.isCpp()), LangOpts(getFormattingLangOpts(Style)),
Keywords(Keywords), Scopes(Scopes), TemplateDeclarationDepth(0) {
- assert(IsCpp == LangOpts.CXXOperatorNames);
+ assert(IsCpp == (LangOpts.CXXOperatorNames || LangOpts.C11));
Contexts.push_back(Context(tok::unknown, 1, /*IsExpression=*/false));
resetTokenMetadata();
}
@@ -3821,7 +3821,7 @@ static bool isFunctionDeclarationName(const LangOptions &LangOpts,
};
const auto *Next = Current.Next;
- const bool IsCpp = LangOpts.CXXOperatorNames;
+ const bool IsCpp = LangOpts.CXXOperatorNames || LangOpts.C11;
// Find parentheses of parameter list.
if (Current.is(tok::kw_operator)) {
diff --git a/clang/lib/Format/TokenAnnotator.h b/clang/lib/Format/TokenAnnotator.h
index 6aea310a56d69..7d065f748cbe5 100644
--- a/clang/lib/Format/TokenAnnotator.h
+++ b/clang/lib/Format/TokenAnnotator.h
@@ -225,7 +225,7 @@ class TokenAnnotator {
TokenAnnotator(const FormatStyle &Style, const AdditionalKeywords &Keywords)
: Style(Style), IsCpp(Style.isCpp()),
LangOpts(getFormattingLangOpts(Style)), Keywords(Keywords) {
- assert(IsCpp == LangOpts.CXXOperatorNames);
+ assert(IsCpp == (LangOpts.CXXOperatorNames || LangOpts.C11));
}
/// Adapts the indent levels of comment lines to the indent of the
diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
index 3a24d72d83e27..1ee06905aae43 100644
--- a/clang/lib/Format/UnwrappedLineParser.cpp
+++ b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -168,7 +168,7 @@ UnwrappedLineParser::UnwrappedLineParser(
: IG_Inited),
IncludeGuardToken(nullptr), FirstStartColumn(FirstStartColumn),
Macros(Style.Macros, SourceMgr, Style, Allocator, IdentTable) {
- assert(IsCpp == LangOpts.CXXOperatorNames);
+ assert(IsCpp == (LangOpts.CXXOperatorNames || LangOpts.C11));
}
void UnwrappedLineParser::reset() {
diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index 05febf12c17ba..fe0e47080a577 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -24971,6 +24971,7 @@ TEST_F(FormatTest, StructuredBindings) {
}
TEST_F(FormatTest, FileAndCode) {
+ EXPECT_EQ(FormatStyle::LK_C, guessLanguage("foo.c", ""));
EXPECT_EQ(FormatStyle::LK_Cpp, guessLanguage("foo.cc", ""));
EXPECT_EQ(FormatStyle::LK_ObjC, guessLanguage("foo.m", ""));
EXPECT_EQ(FormatStyle::LK_ObjC, guessLanguage("foo.mm", ""));
@@ -25137,6 +25138,9 @@ TEST_F(FormatTest, GuessLanguageWithChildLines) {
}
TEST_F(FormatTest, GetLanguageByComment) {
+ EXPECT_EQ(FormatStyle::LK_C,
+ guessLanguage("foo.h", "// clang-format Language: C\n"
+ "int i;"));
EXPECT_EQ(FormatStyle::LK_Cpp,
guessLanguage("foo.h", "// clang-format Language: Cpp\n"
"int DoStuff(CGRect rect);"));
More information about the cfe-commits
mailing list