[clang] [clang-format] Handle Java record (PR #139215)

[via cfe-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-clang-format

Author: Owen Pan (owenca)

<details>
<summary>Changes</summary>

Fix #<!-- -->62089

---
Full diff: https://github.com/llvm/llvm-project/pull/139215.diff


4 Files Affected:

- (modified) clang/lib/Format/FormatToken.h (+2) 
- (modified) clang/lib/Format/UnwrappedLineParser.cpp (+10-3) 
- (modified) clang/lib/Format/UnwrappedLineParser.h (+1-1) 
- (modified) clang/unittests/Format/TokenAnnotatorTest.cpp (+11) 


``````````diff
diff --git a/clang/lib/Format/FormatToken.h b/clang/lib/Format/FormatToken.h
index b570171d032ac..766916911bb66 100644
--- a/clang/lib/Format/FormatToken.h
+++ b/clang/lib/Format/FormatToken.h
@@ -1068,6 +1068,7 @@ struct AdditionalKeywords {
     kw_interface = &IdentTable.get("interface");
     kw_native = &IdentTable.get("native");
     kw_package = &IdentTable.get("package");
+    kw_record = &IdentTable.get("record");
     kw_synchronized = &IdentTable.get("synchronized");
     kw_throws = &IdentTable.get("throws");
     kw___except = &IdentTable.get("__except");
@@ -1417,6 +1418,7 @@ struct AdditionalKeywords {
   IdentifierInfo *kw_interface;
   IdentifierInfo *kw_native;
   IdentifierInfo *kw_package;
+  IdentifierInfo *kw_record;
   IdentifierInfo *kw_synchronized;
   IdentifierInfo *kw_throws;
 
diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
index ae33f4f6987ca..2e138ff7ef59c 100644
--- a/clang/lib/Format/UnwrappedLineParser.cpp
+++ b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -1704,6 +1704,11 @@ void UnwrappedLineParser::parseStructuralElement(
         *HasLabel = true;
       return;
     }
+    if (Style.isJava() && FormatTok->is(Keywords.kw_record)) {
+      parseRecord(/*ParseAsExpr=*/false, /*IsJavaRecord=*/true);
+      addUnwrappedLine();
+      return;
+    }
     // In all other cases, parse the declaration.
     break;
   default:
@@ -3996,11 +4001,13 @@ void UnwrappedLineParser::parseJavaEnumBody() {
   addUnwrappedLine();
 }
 
-void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
+void UnwrappedLineParser::parseRecord(bool ParseAsExpr, bool IsJavaRecord) {
+  assert(!IsJavaRecord || FormatTok->is(Keywords.kw_record));
   const FormatToken &InitialToken = *FormatTok;
   nextToken();
 
-  FormatToken *ClassName = nullptr;
+  FormatToken *ClassName =
+      IsJavaRecord && FormatTok->is(tok::identifier) ? FormatTok : nullptr;
   bool IsDerived = false;
   auto IsNonMacroIdentifier = [](const FormatToken *Tok) {
     return Tok->is(tok::identifier) && Tok->TokenText != Tok->TokenText.upper();
@@ -4035,7 +4042,7 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
     switch (FormatTok->Tok.getKind()) {
     case tok::l_paren:
       // We can have macros in between 'class' and the class name.
-      if (!IsNonMacroIdentifier(Previous) ||
+      if (IsJavaRecord || !IsNonMacroIdentifier(Previous) ||
           // e.g. `struct macro(a) S { int i; };`
           Previous->Previous == &InitialToken) {
         parseParens();
diff --git a/clang/lib/Format/UnwrappedLineParser.h b/clang/lib/Format/UnwrappedLineParser.h
index 3c5d2c3e030e5..2d1492c1a6b8c 100644
--- a/clang/lib/Format/UnwrappedLineParser.h
+++ b/clang/lib/Format/UnwrappedLineParser.h
@@ -177,7 +177,7 @@ class UnwrappedLineParser {
   // Parses a record (aka class) as a top level element. If ParseAsExpr is true,
   // parses the record as a child block, i.e. if the class declaration is an
   // expression.
-  void parseRecord(bool ParseAsExpr = false);
+  void parseRecord(bool ParseAsExpr = false, bool IsJavaRecord = false);
   void parseObjCLightweightGenerics();
   void parseObjCMethod();
   void parseObjCProtocolList();
diff --git a/clang/unittests/Format/TokenAnnotatorTest.cpp b/clang/unittests/Format/TokenAnnotatorTest.cpp
index 7982ccb167b53..f39468424a393 100644
--- a/clang/unittests/Format/TokenAnnotatorTest.cpp
+++ b/clang/unittests/Format/TokenAnnotatorTest.cpp
@@ -3792,6 +3792,17 @@ TEST_F(TokenAnnotatorTest, SwitchExpression) {
   EXPECT_TOKEN(Tokens[20], tok::arrow, TT_CaseLabelArrow);
 }
 
+TEST_F(TokenAnnotatorTest, JavaRecord) {
+  auto Tokens = annotate("public record MyRecord() {}",
+                         getLLVMStyle(FormatStyle::LK_Java));
+  ASSERT_EQ(Tokens.size(), 8u) << Tokens;
+  EXPECT_TOKEN(Tokens[2], tok::identifier, TT_ClassHeadName);
+  // Not TT_FunctionDeclarationLParen.
+  EXPECT_TOKEN(Tokens[3], tok::l_paren, TT_Unknown);
+  EXPECT_TOKEN(Tokens[5], tok::l_brace, TT_RecordLBrace);
+  EXPECT_TOKEN(Tokens[6], tok::r_brace, TT_RecordRBrace);
+}
+
 TEST_F(TokenAnnotatorTest, CppAltOperatorKeywords) {
   auto Tokens = annotate("a = b and c;");
   ASSERT_EQ(Tokens.size(), 7u) << Tokens;

``````````

</details>


https://github.com/llvm/llvm-project/pull/139215