[clang-format] Correctly annotate C++ alternative operators in C

[owenca]

PR #90161 uncovered a bug that annotates C++ xor as UnaryOperator if followed by a binary operator. This patch fixes that and all other C++ alternative operator keywords when followed by a binary operator in C.

Fixes #92688.

[llvmbot]

@llvm/pr-subscribers-clang-format

Author: Owen Pan (owenca)

Changes

PR #90161 uncovered a bug that annotates C++ xor as UnaryOperator if followed by a binary operator. This patch fixes that and all other C++ alternative operator keywords when followed by a binary operator in C.

Fixes #92688.

---

Full diff: https://github.com/llvm/llvm-project/pull/92880.diff

3 Files Affected:

• (modified) clang/lib/Format/UnwrappedLineParser.cpp (+17-4)
• (modified) clang/unittests/Format/FormatTest.cpp (+2)
• (modified) clang/unittests/Format/TokenAnnotatorTest.cpp (+54)

diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
index 3dd10f6bd2b31..b6f7567adc140 100644
--- a/clang/lib/Format/UnwrappedLineParser.cpp
+++ b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -1410,6 +1410,13 @@ void UnwrappedLineParser::readTokenWithJavaScriptASI() {
   }
 }
 
+static bool isAltOperator(const FormatToken &Tok) {
+  return isalpha(Tok.TokenText[0]) &&
+         Tok.isOneOf(tok::ampamp, tok::ampequal, tok::amp, tok::pipe,
+                     tok::tilde, tok::exclaim, tok::exclaimequal, tok::pipepipe,
+                     tok::pipeequal, tok::caret, tok::caretequal);
+}
+
 void UnwrappedLineParser::parseStructuralElement(
     const FormatToken *OpeningBrace, IfStmtKind *IfKind,
     FormatToken **IfLeftBrace, bool *HasDoWhile, bool *HasLabel) {
@@ -1689,9 +1696,15 @@ void UnwrappedLineParser::parseStructuralElement(
     break;
   }
 
-  const bool InRequiresExpression =
-      OpeningBrace && OpeningBrace->is(TT_RequiresExpressionLBrace);
-  do {
+  for (const bool InRequiresExpression =
+           OpeningBrace && OpeningBrace->is(TT_RequiresExpressionLBrace);
+       !eof();) {
+    if (IsCpp && isAltOperator(*FormatTok)) {
+      if (auto *Next = Tokens->peekNextToken(/*SkipComment=*/true);
+          Next && Next->isBinaryOperator()) {
+        FormatTok->Tok.setKind(tok::identifier);
+      }
+    }
     const FormatToken *Previous = FormatTok->Previous;
     switch (FormatTok->Tok.getKind()) {
     case tok::at:
@@ -2122,7 +2135,7 @@ void UnwrappedLineParser::parseStructuralElement(
       nextToken();
       break;
     }
-  } while (!eof());
+  }
 }
 
 bool UnwrappedLineParser::tryToParsePropertyAccessor() {
diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index 2f0c0f0266774..a9df994189f00 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -17340,12 +17340,14 @@ TEST_F(FormatTest, ConfigurableSpaceBeforeAssignmentOperators) {
   verifyFormat("int a = 5;");
   verifyFormat("a += 42;");
   verifyFormat("a or_eq 8;");
+  verifyFormat("xor = foo;");
 
   FormatStyle Spaces = getLLVMStyle();
   Spaces.SpaceBeforeAssignmentOperators = false;
   verifyFormat("int a= 5;", Spaces);
   verifyFormat("a+= 42;", Spaces);
   verifyFormat("a or_eq 8;", Spaces);
+  verifyFormat("xor= foo;", Spaces);
 }
 
 TEST_F(FormatTest, ConfigurableSpaceBeforeColon) {
diff --git a/clang/unittests/Format/TokenAnnotatorTest.cpp b/clang/unittests/Format/TokenAnnotatorTest.cpp
index 45c1554308c9b..6ea9c4a241dc5 100644
--- a/clang/unittests/Format/TokenAnnotatorTest.cpp
+++ b/clang/unittests/Format/TokenAnnotatorTest.cpp
@@ -3015,6 +3015,60 @@ TEST_F(TokenAnnotatorTest, SwitchExpression) {
   EXPECT_TOKEN(Tokens[20], tok::arrow, TT_CaseLabelArrow);
 }
 
+TEST_F(TokenAnnotatorTest, CppAltOperatorKeywords) {
+  auto Tokens = annotate("a = b and c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::ampamp, TT_BinaryOperator);
+
+  Tokens = annotate("a = b and_eq c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::ampequal, TT_BinaryOperator);
+
+  Tokens = annotate("a = b bitand c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::amp, TT_BinaryOperator);
+
+  Tokens = annotate("a = b bitor c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::pipe, TT_BinaryOperator);
+
+  Tokens = annotate("a = b compl c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::tilde, TT_UnaryOperator);
+
+  Tokens = annotate("a = b not c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::exclaim, TT_UnaryOperator);
+
+  Tokens = annotate("a = b not_eq c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::exclaimequal, TT_BinaryOperator);
+
+  Tokens = annotate("a = b or c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::pipepipe, TT_BinaryOperator);
+
+  Tokens = annotate("a = b or_eq c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::pipeequal, TT_BinaryOperator);
+
+  Tokens = annotate("a = b xor c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::caret, TT_BinaryOperator);
+
+  Tokens = annotate("a = b xor_eq c;");
+  ASSERT_EQ(Tokens.size(), 7u);
+  EXPECT_TOKEN(Tokens[3], tok::caretequal, TT_BinaryOperator);
+
+  Tokens = annotate("xor = foo;");
+  ASSERT_EQ(Tokens.size(), 5u);
+  EXPECT_TOKEN(Tokens[0], tok::identifier, TT_Unknown);
+
+  Tokens = annotate("int xor = foo;");
+  ASSERT_EQ(Tokens.size(), 6u);
+  EXPECT_TOKEN(Tokens[1], tok::identifier, TT_StartOfName);
+}
+
 } // namespace
 } // namespace format
 } // namespace clang