release/21.x: [clang-format] Fix bugs in annotating arrows and square brackets (llvm#160973)

Backport 4edda3d

Author: owenca

clang/lib/Format/TokenAnnotator.cpp

@@ -829,11 +829,6 @@ class AnnotatingParser {
           if (Parent && Parent->is(TT_PointerOrReference))
             Parent->overwriteFixedType(TT_BinaryOperator);
         }
-        // An arrow after an ObjC method expression is not a lambda arrow.
-        if (CurrentToken->is(TT_ObjCMethodExpr) && CurrentToken->Next &&
-            CurrentToken->Next->is(TT_LambdaArrow)) {
-          CurrentToken->Next->overwriteFixedType(TT_Unknown);
-        }
         Left->MatchingParen = CurrentToken;
         CurrentToken->MatchingParen = Left;
         // FirstObjCSelectorName is set when a colon is found. This does

clang/lib/Format/UnwrappedLineParser.cpp

@@ -2266,7 +2266,7 @@ bool UnwrappedLineParser::tryToParseLambda() {
   if (!tryToParseLambdaIntroducer())
     return false;
 
-  bool SeenArrow = false;
+  FormatToken *Arrow = nullptr;
   bool InTemplateParameterList = false;
 
   while (FormatTok->isNot(tok::l_brace)) {
@@ -2341,17 +2341,13 @@ bool UnwrappedLineParser::tryToParseLambda() {
     case tok::ellipsis:
     case tok::kw_true:
     case tok::kw_false:
-      if (SeenArrow || InTemplateParameterList) {
+      if (Arrow || InTemplateParameterList) {
         nextToken();
         break;
       }
       return true;
     case tok::arrow:
-      // This might or might not actually be a lambda arrow (this could be an
-      // ObjC method invocation followed by a dereferencing arrow). We might
-      // reset this back to TT_Unknown in TokenAnnotator.
-      FormatTok->setFinalizedType(TT_LambdaArrow);
-      SeenArrow = true;
+      Arrow = FormatTok;
       nextToken();
       break;
     case tok::kw_requires: {
@@ -2373,6 +2369,9 @@ bool UnwrappedLineParser::tryToParseLambda() {
   FormatTok->setFinalizedType(TT_LambdaLBrace);
   LSquare.setFinalizedType(TT_LambdaLSquare);
 
+  if (Arrow)
+    Arrow->setFinalizedType(TT_LambdaArrow);
+
   NestedLambdas.push_back(Line->SeenDecltypeAuto);
   parseChildBlock();
   assert(!NestedLambdas.empty());
@@ -2386,11 +2385,6 @@ bool UnwrappedLineParser::tryToParseLambdaIntroducer() {
   const FormatToken *LeftSquare = FormatTok;
   nextToken();
   if (Previous) {
-    if (Previous->Tok.getIdentifierInfo() &&
-        !Previous->isOneOf(tok::kw_return, tok::kw_co_await, tok::kw_co_yield,
-                           tok::kw_co_return)) {
-      return false;
-    }
     if (Previous->closesScope()) {
       // Not a potential C-style cast.
       if (Previous->isNot(tok::r_paren))
@@ -2400,6 +2394,13 @@ bool UnwrappedLineParser::tryToParseLambdaIntroducer() {
       // and `int (*)()`.
       if (!BeforeRParen || !BeforeRParen->isOneOf(tok::greater, tok::r_paren))
         return false;
+    } else if (Previous->is(tok::star)) {
+      Previous = Previous->getPreviousNonComment();
+    }
+    if (Previous && Previous->Tok.getIdentifierInfo() &&
+        !Previous->isOneOf(tok::kw_return, tok::kw_co_await, tok::kw_co_yield,
+                           tok::kw_co_return)) {
+      return false;
     }
   }
   if (LeftSquare->isCppStructuredBinding(IsCpp))

clang/unittests/Format/TokenAnnotatorTest.cpp

@@ -2229,6 +2229,12 @@ TEST_F(TokenAnnotatorTest, UnderstandsLambdas) {
   ASSERT_EQ(Tokens.size(), 21u) << Tokens;
   EXPECT_TOKEN(Tokens[11], tok::l_square, TT_LambdaLSquare);
   EXPECT_TOKEN(Tokens[13], tok::l_brace, TT_LambdaLBrace);
+
+  Tokens = annotate("SomeFunction({[]() -> int *[] { return {}; }});");
+  ASSERT_EQ(Tokens.size(), 22u) << Tokens;
+  EXPECT_TOKEN(Tokens[3], tok::l_square, TT_LambdaLSquare);
+  EXPECT_TOKEN(Tokens[5], tok::l_paren, TT_LambdaDefinitionLParen);
+  EXPECT_TOKEN(Tokens[10], tok::l_square, TT_ArraySubscriptLSquare);
 }
 
 TEST_F(TokenAnnotatorTest, UnderstandsFunctionAnnotations) {
@@ -4151,6 +4157,14 @@ TEST_F(TokenAnnotatorTest, LineCommentTrailingBackslash) {
   EXPECT_TOKEN(Tokens[1], tok::comment, TT_LineComment);
 }
 
+TEST_F(TokenAnnotatorTest, ArrowAfterSubscript) {
+  auto Tokens =
+      annotate("return (getStructType()->getElements())[eIdx]->getName();");
+  ASSERT_EQ(Tokens.size(), 19u) << Tokens;
+  // Not TT_LambdaArrow.
+  EXPECT_TOKEN(Tokens[13], tok::arrow, TT_Unknown);
+}
+
 } // namespace
 } // namespace format
 } // namespace clang