[clang] e77a01d - [Clang] Implement P3034R1 Module Declarations Shouldn’t be Macros (#90574)

via cfe-commits cfe-commits at lists.llvm.org
Sat Jul 20 05:41:06 PDT 2024


Author: yronglin
Date: 2024-07-20T20:41:00+08:00
New Revision: e77a01d79a48e15c94c89e4aa4bd27424a96b49b

URL: https://github.com/llvm/llvm-project/commit/e77a01d79a48e15c94c89e4aa4bd27424a96b49b
DIFF: https://github.com/llvm/llvm-project/commit/e77a01d79a48e15c94c89e4aa4bd27424a96b49b.diff

LOG: [Clang] Implement P3034R1 Module Declarations Shouldn’t be Macros (#90574)

This PR implement [P3034R1 Module Declarations Shouldn’t be
Macros](https://wg21.link/P3034R1), and refactor the convoluted state
machines in module name lexical analysis.

---------

Signed-off-by: yronglin <yronglin777 at gmail.com>
Co-authored-by: Aaron Ballman <aaron at aaronballman.com>
Co-authored-by: cor3ntin <corentinjabot at gmail.com>

Added: 
    clang/test/CXX/cpp/cpp.module/p2.cppm

Modified: 
    clang/docs/ReleaseNotes.rst
    clang/include/clang/Basic/DiagnosticLexKinds.td
    clang/include/clang/Basic/IdentifierTable.h
    clang/include/clang/Basic/TokenKinds.def
    clang/include/clang/Lex/Preprocessor.h
    clang/include/clang/Lex/Token.h
    clang/include/clang/Parse/Parser.h
    clang/lib/Basic/IdentifierTable.cpp
    clang/lib/Frontend/PrintPreprocessedOutput.cpp
    clang/lib/Lex/PPLexerChange.cpp
    clang/lib/Lex/Preprocessor.cpp
    clang/lib/Lex/TokenConcatenation.cpp
    clang/lib/Parse/ParseDecl.cpp
    clang/lib/Parse/Parser.cpp
    clang/test/CXX/module/basic/basic.link/module-declaration.cpp
    clang/test/CXX/module/dcl.dcl/dcl.module/dcl.module.import/p1.cppm
    clang/test/SemaCXX/modules.cppm
    clang/www/cxx_status.html

Removed: 
    


################################################################################
diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index aa91663260755..4638b91b48f95 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -285,6 +285,8 @@ C++2c Feature Support
 
 - Implemented `P2963R3 Ordering of constraints involving fold expressions <https://wg21.link/P2963R3>`_.
 
+- Implemented `P3034R1 Module Declarations Shouldn’t be Macros <https://wg21.link/P3034R1>`_.
+
 
 Resolutions to C++ Defect Reports
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

diff  --git a/clang/include/clang/Basic/DiagnosticLexKinds.td b/clang/include/clang/Basic/DiagnosticLexKinds.td
index 12d7b8c0205ee..08ece01009387 100644
--- a/clang/include/clang/Basic/DiagnosticLexKinds.td
+++ b/clang/include/clang/Basic/DiagnosticLexKinds.td
@@ -952,6 +952,11 @@ def warn_module_conflict : Warning<
   InGroup<ModuleConflict>;
 
 // C++20 modules
+def err_module_decl_cannot_be_macros : Error<
+  "the module name in a module%select{| partition}0 declaration cannot contain "
+  "an object-like macro %1">;
+def err_unxepected_paren_in_module_decl : Error<
+  "unexpected '(' after the module name in a module%select{| partition}0 declaration">;
 def err_header_import_semi_in_macro : Error<
   "semicolon terminating header import declaration cannot be produced "
   "by a macro">;

diff  --git a/clang/include/clang/Basic/IdentifierTable.h b/clang/include/clang/Basic/IdentifierTable.h
index ae9ebd9f59154..f40f74d0355ad 100644
--- a/clang/include/clang/Basic/IdentifierTable.h
+++ b/clang/include/clang/Basic/IdentifierTable.h
@@ -180,6 +180,10 @@ class alignas(IdentifierInfoAlignment) IdentifierInfo {
   LLVM_PREFERRED_TYPE(bool)
   unsigned IsModulesImport : 1;
 
+  // True if this is the 'module' contextual keyword.
+  LLVM_PREFERRED_TYPE(bool)
+  unsigned IsModulesDecl : 1;
+
   // True if this is a mangled OpenMP variant name.
   LLVM_PREFERRED_TYPE(bool)
   unsigned IsMangledOpenMPVariantName : 1;
@@ -196,7 +200,7 @@ class alignas(IdentifierInfoAlignment) IdentifierInfo {
   LLVM_PREFERRED_TYPE(bool)
   unsigned IsFinal : 1;
 
-  // 22 bits left in a 64-bit word.
+  // 21 bits left in a 64-bit word.
 
   // Managed by the language front-end.
   void *FETokenInfo = nullptr;
@@ -212,8 +216,8 @@ class alignas(IdentifierInfoAlignment) IdentifierInfo {
         IsCPPOperatorKeyword(false), NeedsHandleIdentifier(false),
         IsFromAST(false), ChangedAfterLoad(false), FEChangedAfterLoad(false),
         RevertedTokenID(false), OutOfDate(false), IsModulesImport(false),
-        IsMangledOpenMPVariantName(false), IsDeprecatedMacro(false),
-        IsRestrictExpansion(false), IsFinal(false) {}
+        IsModulesDecl(false), IsMangledOpenMPVariantName(false),
+        IsDeprecatedMacro(false), IsRestrictExpansion(false), IsFinal(false) {}
 
 public:
   IdentifierInfo(const IdentifierInfo &) = delete;
@@ -520,6 +524,18 @@ class alignas(IdentifierInfoAlignment) IdentifierInfo {
       RecomputeNeedsHandleIdentifier();
   }
 
+  /// Determine whether this is the contextual keyword \c module.
+  bool isModulesDeclaration() const { return IsModulesDecl; }
+
+  /// Set whether this identifier is the contextual keyword \c module.
+  void setModulesDeclaration(bool I) {
+    IsModulesDecl = I;
+    if (I)
+      NeedsHandleIdentifier = true;
+    else
+      RecomputeNeedsHandleIdentifier();
+  }
+
   /// Determine whether this is the mangled name of an OpenMP variant.
   bool isMangledOpenMPVariantName() const { return IsMangledOpenMPVariantName; }
 
@@ -740,6 +756,8 @@ class IdentifierTable {
     // If this is the 'import' contextual keyword, mark it as such.
     if (Name == "import")
       II->setModulesImport(true);
+    else if (Name == "module")
+      II->setModulesDeclaration(true);
 
     return *II;
   }

diff  --git a/clang/include/clang/Basic/TokenKinds.def b/clang/include/clang/Basic/TokenKinds.def
index 7f4912b9bcd96..8db18c049b6d0 100644
--- a/clang/include/clang/Basic/TokenKinds.def
+++ b/clang/include/clang/Basic/TokenKinds.def
@@ -1003,6 +1003,9 @@ ANNOTATION(module_include)
 ANNOTATION(module_begin)
 ANNOTATION(module_end)
 
+// Annotations for C++, Clang and Objective-C named modules.
+ANNOTATION(module_name)
+
 // Annotation for a header_name token that has been looked up and transformed
 // into the name of a header unit.
 ANNOTATION(header_unit)

diff  --git a/clang/include/clang/Lex/Preprocessor.h b/clang/include/clang/Lex/Preprocessor.h
index fc7d0053f2323..56aef99a3f38a 100644
--- a/clang/include/clang/Lex/Preprocessor.h
+++ b/clang/include/clang/Lex/Preprocessor.h
@@ -615,10 +615,6 @@ class Preprocessor {
 
   ModuleDeclSeq ModuleDeclState;
 
-  /// Whether the module import expects an identifier next. Otherwise,
-  /// it expects a '.' or ';'.
-  bool ModuleImportExpectsIdentifier = false;
-
   /// The identifier and source location of the currently-active
   /// \#pragma clang arc_cf_code_audited begin.
   std::pair<IdentifierInfo *, SourceLocation> PragmaARCCFCodeAuditedInfo;
@@ -1744,11 +1740,14 @@ class Preprocessor {
   /// Lex a token, forming a header-name token if possible.
   bool LexHeaderName(Token &Result, bool AllowMacroExpansion = true);
 
+  /// Lex a module name or a partition name.
+  bool LexModuleName(Token &Result, bool IsImport);
+
   /// Lex the parameters for an #embed directive, returns nullopt on error.
   std::optional<LexEmbedParametersResult> LexEmbedParameters(Token &Current,
                                                              bool ForHasEmbed);
-
   bool LexAfterModuleImport(Token &Result);
+  bool LexAfterModuleDecl(Token &Result);
   void CollectPpImportSuffix(SmallVectorImpl<Token> &Toks);
 
   void makeModuleVisible(Module *M, SourceLocation Loc);
@@ -3039,6 +3038,9 @@ class Preprocessor {
   static bool CLK_LexAfterModuleImport(Preprocessor &P, Token &Result) {
     return P.LexAfterModuleImport(Result);
   }
+  static bool CLK_LexAfterModuleDecl(Preprocessor &P, Token &Result) {
+    return P.LexAfterModuleDecl(Result);
+  }
 };
 
 /// Abstract base class that describes a handler that will receive
@@ -3071,6 +3073,77 @@ struct EmbedAnnotationData {
 /// Registry of pragma handlers added by plugins
 using PragmaHandlerRegistry = llvm::Registry<PragmaHandler>;
 
+/// Represents module or partition name token sequance.
+///
+///     module-name:
+///           module-name-qualifier[opt] identifier
+///
+///     partition-name: [C++20]
+///           : module-name-qualifier[opt] identifier
+///
+///     module-name-qualifier
+///           module-name-qualifier[opt] identifier .
+///
+/// This class can only be created by the preprocessor and guarantees that the
+/// two source array being contiguous in memory and only contains 3 kind of
+/// tokens (identifier, '.' and ':'). And only available when the preprocessor
+/// returns annot_module_name token.
+///
+/// For exmaple:
+///
+/// export module m.n:c.d
+///
+/// The module name array has 3 tokens ['m', '.', 'n'].
+/// The partition name array has 4 tokens [':', 'c', '.', 'd'].
+///
+/// When import a partition in a named module fragment (Eg. import :part1;),
+/// the module name array will be empty, and the partition name array has 2
+/// tokens.
+///
+/// When we meet a private-module-fragment (Eg. module :private;), preprocessor
+/// will not return a annot_module_name token, but will return 2 separate tokens
+/// [':', 'kw_private'].
+
+class ModuleNameInfo {
+  friend class Preprocessor;
+  ArrayRef<Token> ModuleName;
+  ArrayRef<Token> PartitionName;
+
+  ModuleNameInfo(ArrayRef<Token> AnnotToks, std::optional<unsigned> ColonIndex);
+
+public:
+  /// Return the contiguous token array.
+  ArrayRef<Token> getTokens() const {
+    if (ModuleName.empty())
+      return PartitionName;
+    if (PartitionName.empty())
+      return ModuleName;
+    return ArrayRef(ModuleName.begin(), PartitionName.end());
+  }
+  bool hasModuleName() const { return !ModuleName.empty(); }
+  bool hasPartitionName() const { return !PartitionName.empty(); }
+  ArrayRef<Token> getModuleName() const { return ModuleName; }
+  ArrayRef<Token> getPartitionName() const { return PartitionName; }
+  Token getColonToken() const {
+    assert(hasPartitionName() && "Do not have a partition name");
+    return getPartitionName().front();
+  }
+
+  /// Under the standard C++ Modules, the dot is just part of the module name,
+  /// and not a real hierarchy separator. Flatten such module names now.
+  std::string getFlatName() const;
+
+  /// Build a module id path from the contiguous token array, both include
+  /// module name and partition name.
+  void getModuleIdPath(
+      SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path) const;
+
+  /// Build a module id path from \param ModuleName.
+  static void getModuleIdPath(
+      ArrayRef<Token> ModuleName,
+      SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path);
+};
+
 } // namespace clang
 
 #endif // LLVM_CLANG_LEX_PREPROCESSOR_H

diff  --git a/clang/include/clang/Lex/Token.h b/clang/include/clang/Lex/Token.h
index 4f29fb7d11415..2be3ad39529f0 100644
--- a/clang/include/clang/Lex/Token.h
+++ b/clang/include/clang/Lex/Token.h
@@ -235,6 +235,9 @@ class Token {
     assert(isAnnotation() && "Used AnnotVal on non-annotation token");
     return PtrData;
   }
+  template <class T> T getAnnotationValueAs() const {
+    return static_cast<T>(getAnnotationValue());
+  }
   void setAnnotationValue(void *val) {
     assert(isAnnotation() && "Used AnnotVal on non-annotation token");
     PtrData = val;

diff  --git a/clang/include/clang/Parse/Parser.h b/clang/include/clang/Parse/Parser.h
index 93e60be512aae..afcdacf02583a 100644
--- a/clang/include/clang/Parse/Parser.h
+++ b/clang/include/clang/Parse/Parser.h
@@ -3876,7 +3876,7 @@ class Parser : public CodeCompletionHandler {
   }
 
   bool ParseModuleName(
-      SourceLocation UseLoc,
+      SourceLocation UseLoc, ArrayRef<Token> ModuleName,
       SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path,
       bool IsImport);
 

diff  --git a/clang/lib/Basic/IdentifierTable.cpp b/clang/lib/Basic/IdentifierTable.cpp
index 4f7ccaf4021d6..97d830214f890 100644
--- a/clang/lib/Basic/IdentifierTable.cpp
+++ b/clang/lib/Basic/IdentifierTable.cpp
@@ -322,8 +322,9 @@ void IdentifierTable::AddKeywords(const LangOptions &LangOpts) {
   if (LangOpts.IEEE128)
     AddKeyword("__ieee128", tok::kw___float128, KEYALL, LangOpts, *this);
 
-  // Add the 'import' contextual keyword.
+  // Add the 'import' and 'module' contextual keyword.
   get("import").setModulesImport(true);
+  get("module").setModulesDeclaration(true);
 }
 
 /// Checks if the specified token kind represents a keyword in the

diff  --git a/clang/lib/Frontend/PrintPreprocessedOutput.cpp b/clang/lib/Frontend/PrintPreprocessedOutput.cpp
index 0592423c12eca..1fff88ccf0405 100644
--- a/clang/lib/Frontend/PrintPreprocessedOutput.cpp
+++ b/clang/lib/Frontend/PrintPreprocessedOutput.cpp
@@ -758,9 +758,10 @@ void PrintPPOutputPPCallbacks::HandleWhitespaceBeforeTok(const Token &Tok,
   // These tokens are not expanded to anything and don't need whitespace before
   // them.
   if (Tok.is(tok::eof) ||
-      (Tok.isAnnotation() && !Tok.is(tok::annot_header_unit) &&
-       !Tok.is(tok::annot_module_begin) && !Tok.is(tok::annot_module_end) &&
-       !Tok.is(tok::annot_repl_input_end) && !Tok.is(tok::annot_embed)))
+      (Tok.isAnnotation() && Tok.isNot(tok::annot_header_unit) &&
+       Tok.isNot(tok::annot_module_begin) && Tok.isNot(tok::annot_module_end) &&
+       Tok.isNot(tok::annot_module_name) &&
+       Tok.isNot(tok::annot_repl_input_end) && Tok.isNot(tok::annot_embed)))
     return;
 
   // EmittedDirectiveOnThisLine takes priority over RequireSameLine.
@@ -951,6 +952,11 @@ static void PrintPreprocessedTokens(Preprocessor &PP, Token &Tok,
       PP.Lex(Tok);
       IsStartOfLine = true;
       continue;
+    } else if (Tok.is(tok::annot_module_name)) {
+      auto *Info = static_cast<ModuleNameInfo *>(Tok.getAnnotationValue());
+      *Callbacks->OS << Info->getFlatName();
+      PP.Lex(Tok);
+      continue;
     } else if (Tok.is(tok::annot_header_unit)) {
       // This is a header-name that has been (effectively) converted into a
       // module-name.

diff  --git a/clang/lib/Lex/PPLexerChange.cpp b/clang/lib/Lex/PPLexerChange.cpp
index 8221db46e06ac..c3a903917e9ce 100644
--- a/clang/lib/Lex/PPLexerChange.cpp
+++ b/clang/lib/Lex/PPLexerChange.cpp
@@ -122,7 +122,8 @@ void Preprocessor::EnterSourceFileWithLexer(Lexer *TheLexer,
   CurPPLexer = TheLexer;
   CurDirLookup = CurDir;
   CurLexerSubmodule = nullptr;
-  if (CurLexerCallback != CLK_LexAfterModuleImport)
+  if (CurLexerCallback != CLK_LexAfterModuleImport &&
+      CurLexerCallback != CLK_LexAfterModuleDecl)
     CurLexerCallback = TheLexer->isDependencyDirectivesLexer()
                            ? CLK_DependencyDirectivesLexer
                            : CLK_Lexer;
@@ -161,8 +162,7 @@ void Preprocessor::EnterMacro(Token &Tok, SourceLocation ILEnd,
   PushIncludeMacroStack();
   CurDirLookup = nullptr;
   CurTokenLexer = std::move(TokLexer);
-  if (CurLexerCallback != CLK_LexAfterModuleImport)
-    CurLexerCallback = CLK_TokenLexer;
+  CurLexerCallback = CLK_TokenLexer;
 }
 
 /// EnterTokenStream - Add a "macro" context to the top of the include stack,
@@ -216,7 +216,8 @@ void Preprocessor::EnterTokenStream(const Token *Toks, unsigned NumToks,
   PushIncludeMacroStack();
   CurDirLookup = nullptr;
   CurTokenLexer = std::move(TokLexer);
-  if (CurLexerCallback != CLK_LexAfterModuleImport)
+  if (CurLexerCallback != CLK_LexAfterModuleImport &&
+      CurLexerCallback != CLK_LexAfterModuleDecl)
     CurLexerCallback = CLK_TokenLexer;
 }
 

diff  --git a/clang/lib/Lex/Preprocessor.cpp b/clang/lib/Lex/Preprocessor.cpp
index 63e27e62cffc8..2726fae344337 100644
--- a/clang/lib/Lex/Preprocessor.cpp
+++ b/clang/lib/Lex/Preprocessor.cpp
@@ -860,9 +860,15 @@ bool Preprocessor::HandleIdentifier(Token &Identifier) {
     ModuleImportLoc = Identifier.getLocation();
     NamedModuleImportPath.clear();
     IsAtImport = true;
-    ModuleImportExpectsIdentifier = true;
     CurLexerCallback = CLK_LexAfterModuleImport;
   }
+
+  if ((II.isModulesDeclaration() || Identifier.is(tok::kw_module)) &&
+      !InMacroArgs && !DisableMacroExpansion &&
+      (getLangOpts().CPlusPlusModules || getLangOpts().DebuggerSupport) &&
+      CurLexerCallback != CLK_CachingLexer) {
+    CurLexerCallback = CLK_LexAfterModuleDecl;
+  }
   return true;
 }
 
@@ -905,6 +911,7 @@ void Preprocessor::Lex(Token &Result) {
     // This token is injected to represent the translation of '#include "a.h"'
     // into "import a.h;". Mimic the notional ';'.
     case tok::annot_module_include:
+    case tok::annot_repl_input_end:
     case tok::semi:
       TrackGMFState.handleSemi();
       StdCXXImportSeqState.handleSemi();
@@ -919,12 +926,30 @@ void Preprocessor::Lex(Token &Result) {
       StdCXXImportSeqState.handleExport();
       ModuleDeclState.handleExport();
       break;
-    case tok::colon:
-      ModuleDeclState.handleColon();
-      break;
-    case tok::period:
-      ModuleDeclState.handlePeriod();
+    case tok::annot_module_name: {
+      auto *Info = static_cast<ModuleNameInfo *>(Result.getAnnotationValue());
+      for (const auto &Tok : Info->getTokens()) {
+        switch (Tok.getKind()) {
+        case tok::identifier:
+          ModuleDeclState.handleIdentifier(Tok.getIdentifierInfo());
+          break;
+        case tok::period:
+          ModuleDeclState.handlePeriod();
+          break;
+        case tok::colon:
+          ModuleDeclState.handleColon();
+          break;
+        default:
+          llvm_unreachable("Unexpected token in module name");
+        }
+      }
+      if (ModuleDeclState.isModuleCandidate())
+        break;
+      TrackGMFState.handleMisc();
+      StdCXXImportSeqState.handleMisc();
+      ModuleDeclState.handleMisc();
       break;
+    }
     case tok::identifier:
       // Check "import" and "module" when there is no open bracket. The two
       // identifiers are not meaningful with open brackets.
@@ -936,17 +961,17 @@ void Preprocessor::Lex(Token &Result) {
             ModuleImportLoc = Result.getLocation();
             NamedModuleImportPath.clear();
             IsAtImport = false;
-            ModuleImportExpectsIdentifier = true;
             CurLexerCallback = CLK_LexAfterModuleImport;
           }
           break;
-        } else if (Result.getIdentifierInfo() == getIdentifierInfo("module")) {
+        }
+        if (Result.getIdentifierInfo()->isModulesDeclaration()) {
           TrackGMFState.handleModule(StdCXXImportSeqState.afterTopLevelSeq());
           ModuleDeclState.handleModule();
+          CurLexerCallback = CLK_LexAfterModuleDecl;
           break;
         }
       }
-      ModuleDeclState.handleIdentifier(Result.getIdentifierInfo());
       if (ModuleDeclState.isModuleCandidate())
         break;
       [[fallthrough]];
@@ -1121,6 +1146,151 @@ void Preprocessor::CollectPpImportSuffix(SmallVectorImpl<Token> &Toks) {
   }
 }
 
+ModuleNameInfo::ModuleNameInfo(ArrayRef<Token> AnnotToks,
+                               std::optional<unsigned> ColonIndex) {
+  assert(!AnnotToks.empty() && "Named module token cannot be empty.");
+  if (!ColonIndex.has_value())
+    ColonIndex = AnnotToks.size();
+  ModuleName = ArrayRef(AnnotToks.begin(), AnnotToks.begin() + *ColonIndex);
+  PartitionName = ArrayRef(AnnotToks.begin() + *ColonIndex, AnnotToks.end());
+  assert(ModuleName.end() == PartitionName.begin());
+}
+
+std::string ModuleNameInfo::getFlatName() const {
+  std::string FlatModuleName;
+  for (auto &Tok : getTokens()) {
+    switch (Tok.getKind()) {
+    case tok::identifier:
+      FlatModuleName += Tok.getIdentifierInfo()->getName();
+      break;
+    case tok::period:
+      FlatModuleName += '.';
+      break;
+    case tok::colon:
+      FlatModuleName += ':';
+      break;
+    default:
+      llvm_unreachable("Unexpected token in module name");
+    }
+  }
+  return FlatModuleName;
+}
+
+void ModuleNameInfo::getModuleIdPath(
+    SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path) const {
+  return getModuleIdPath(getTokens(), Path);
+}
+
+void ModuleNameInfo::getModuleIdPath(
+    ArrayRef<Token> ModuleName,
+    SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path) {
+  for (const auto &Tok : ModuleName) {
+    if (Tok.is(tok::identifier))
+      Path.push_back(
+          std::make_pair(Tok.getIdentifierInfo(), Tok.getLocation()));
+  }
+}
+
+/// Lex a module name or a partition name.
+///
+///     module-name:
+///           module-name-qualifier[opt] identifier
+///
+///     partition-name: [C++20]
+///           : module-name-qualifier[opt] identifier
+///
+///     module-name-qualifier
+///           module-name-qualifier[opt] identifier .
+bool Preprocessor::LexModuleName(Token &Result, bool IsImport) {
+  bool ExpectsIdentifier = true, IsLexingPartition = false;
+  SmallVector<Token, 8> ModuleName;
+  std::optional<unsigned> ColonTokIndex;
+  auto LexNextToken = [&](Token &Tok) {
+    if (IsImport)
+      Lex(Tok);
+    else
+      LexUnexpandedToken(Tok);
+  };
+
+  while (true) {
+    LexNextToken(Result);
+    if (ExpectsIdentifier && Result.is(tok::identifier)) {
+      auto *MI = getMacroInfo(Result.getIdentifierInfo());
+      if (getLangOpts().CPlusPlusModules && !IsImport && MI &&
+          MI->isObjectLike()) {
+        Diag(Result, diag::err_module_decl_cannot_be_macros)
+            << Result.getLocation() << IsLexingPartition
+            << Result.getIdentifierInfo();
+      }
+      ModuleName.push_back(Result);
+      ExpectsIdentifier = false;
+      continue;
+    }
+
+    if (!ExpectsIdentifier && Result.is(tok::period)) {
+      ModuleName.push_back(Result);
+      ExpectsIdentifier = true;
+      continue;
+    }
+
+    // Module partition only allowed in C++20 Modules.
+    if (getLangOpts().CPlusPlusModules && Result.is(tok::colon)) {
+      // Handle the form like: import :P;
+      // If the token after ':' is not an identifier, this is a invalid module
+      // name.
+      if (ModuleName.empty()) {
+        Token Tmp;
+        LexNextToken(Tmp);
+        EnterToken(Tmp, /*IsReiject=*/false);
+        // A private-module-fragment:
+        // export module :private;
+        if (!IsImport && Tmp.is(tok::kw_private))
+          return true;
+        // import :N;
+        if (IsImport && Tmp.isNot(tok::identifier))
+          return false;
+      } else if (!ExpectsIdentifier) {
+        ExpectsIdentifier = true;
+      }
+      IsLexingPartition = true;
+      ColonTokIndex = ModuleName.size();
+      ModuleName.push_back(Result);
+      continue;
+    }
+
+    // [cpp.module]/p2: where the pp-tokens (if any) shall not begin with a (
+    // preprocessing token [...]
+    //
+    // We only emit diagnostic in the preprocessor, and in the parser we skip
+    // invalid tokens and recover from errors.
+    if (getLangOpts().CPlusPlusModules && !ExpectsIdentifier &&
+        Result.is(tok::l_paren))
+      Diag(Result, diag::err_unxepected_paren_in_module_decl)
+          << IsLexingPartition;
+    break;
+  }
+
+  // Put the last token back to stream, it's not a valid part of module name.
+  // We lexed it unexpanded but it might be a valid macro expansion
+  Result.clearFlag(Token::DisableExpand);
+  auto ToksCopy = std::make_unique<Token[]>(1);
+  *ToksCopy.get() = Result;
+  EnterTokenStream(std::move(ToksCopy), 1,
+                   /*DisableMacroExpansion=*/false,
+                   /*IsReinject=*/false);
+
+  if (ModuleName.empty())
+    return false;
+  Result.startToken();
+  Result.setKind(tok::annot_module_name);
+  Result.setLocation(ModuleName.front().getLocation());
+  Result.setAnnotationEndLoc(ModuleName.back().getLocation());
+  auto AnnotToks = ArrayRef(ModuleName).copy(getPreprocessorAllocator());
+  ModuleNameInfo *Info =
+      new (getPreprocessorAllocator()) ModuleNameInfo(AnnotToks, ColonTokIndex);
+  Result.setAnnotationValue(static_cast<void *>(Info));
+  return true;
+}
 
 /// Lex a token following the 'import' contextual keyword.
 ///
@@ -1145,6 +1315,17 @@ bool Preprocessor::LexAfterModuleImport(Token &Result) {
   // Figure out what kind of lexer we actually have.
   recomputeCurLexerKind();
 
+  // Allocate a holding buffer for a sequence of tokens and introduce it into
+  // the token stream.
+  auto EnterTokens = [this](ArrayRef<Token> Toks) {
+    auto ToksCopy = std::make_unique<Token[]>(Toks.size());
+    std::copy(Toks.begin(), Toks.end(), ToksCopy.get());
+    EnterTokenStream(std::move(ToksCopy), Toks.size(),
+                     /*DisableMacroExpansion*/ true, /*IsReinject*/ false);
+  };
+
+  SmallVector<Token, 32> Suffix;
+
   // Lex the next token. The header-name lexing rules are used at the start of
   // a pp-import.
   //
@@ -1155,122 +1336,108 @@ bool Preprocessor::LexAfterModuleImport(Token &Result) {
     if (LexHeaderName(Result))
       return true;
 
-    if (Result.is(tok::colon) && ModuleDeclState.isNamedModule()) {
-      std::string Name = ModuleDeclState.getPrimaryName().str();
-      Name += ":";
-      NamedModuleImportPath.push_back(
-          {getIdentifierInfo(Name), Result.getLocation()});
-      CurLexerCallback = CLK_LexAfterModuleImport;
-      return true;
-    }
-  } else {
-    Lex(Result);
-  }
+    // Check for a header-name.
+    if (Result.is(tok::header_name)) {
+      // Enter the header-name token into the token stream; a Lex action cannot
+      // both return a token and cache tokens (doing so would corrupt the token
+      // cache if the call to Lex comes from CachingLex / PeekAhead).
+      Suffix.push_back(Result);
+
+      // Consume the pp-import-suffix and expand any macros in it now. We'll add
+      // it back into the token stream later.
+      CollectPpImportSuffix(Suffix);
+      if (Suffix.back().isNot(tok::semi)) {
+        // This is not a pp-import after all.
+        EnterTokens(Suffix);
+        return false;
+      }
 
-  // Allocate a holding buffer for a sequence of tokens and introduce it into
-  // the token stream.
-  auto EnterTokens = [this](ArrayRef<Token> Toks) {
-    auto ToksCopy = std::make_unique<Token[]>(Toks.size());
-    std::copy(Toks.begin(), Toks.end(), ToksCopy.get());
-    EnterTokenStream(std::move(ToksCopy), Toks.size(),
-                     /*DisableMacroExpansion*/ true, /*IsReinject*/ false);
-  };
+      // C++2a [cpp.module]p1:
+      //   The ';' preprocessing-token terminating a pp-import shall not have
+      //   been produced by macro replacement.
+      SourceLocation SemiLoc = Suffix.back().getLocation();
+      if (SemiLoc.isMacroID())
+        Diag(SemiLoc, diag::err_header_import_semi_in_macro);
+
+      // Reconstitute the import token.
+      Token ImportTok;
+      ImportTok.startToken();
+      ImportTok.setKind(tok::kw_import);
+      ImportTok.setLocation(ModuleImportLoc);
+      ImportTok.setIdentifierInfo(getIdentifierInfo("import"));
+      ImportTok.setLength(6);
+
+      auto Action = HandleHeaderIncludeOrImport(
+          /*HashLoc*/ SourceLocation(), ImportTok, Suffix.front(), SemiLoc);
+      switch (Action.Kind) {
+      case ImportAction::None:
+        break;
 
-  bool ImportingHeader = Result.is(tok::header_name);
-  // Check for a header-name.
-  SmallVector<Token, 32> Suffix;
-  if (ImportingHeader) {
-    // Enter the header-name token into the token stream; a Lex action cannot
-    // both return a token and cache tokens (doing so would corrupt the token
-    // cache if the call to Lex comes from CachingLex / PeekAhead).
-    Suffix.push_back(Result);
+      case ImportAction::ModuleBegin:
+        // Let the parser know we're textually entering the module.
+        Suffix.emplace_back();
+        Suffix.back().startToken();
+        Suffix.back().setKind(tok::annot_module_begin);
+        Suffix.back().setLocation(SemiLoc);
+        Suffix.back().setAnnotationEndLoc(SemiLoc);
+        Suffix.back().setAnnotationValue(Action.ModuleForHeader);
+        [[fallthrough]];
+
+      case ImportAction::ModuleImport:
+      case ImportAction::HeaderUnitImport:
+      case ImportAction::SkippedModuleImport:
+        // We chose to import (or textually enter) the file. Convert the
+        // header-name token into a header unit annotation token.
+        Suffix[0].setKind(tok::annot_header_unit);
+        Suffix[0].setAnnotationEndLoc(Suffix[0].getLocation());
+        Suffix[0].setAnnotationValue(Action.ModuleForHeader);
+        // FIXME: Call the moduleImport callback?
+        break;
+      case ImportAction::Failure:
+        assert(TheModuleLoader.HadFatalFailure &&
+               "This should be an early exit only to a fatal error");
+        Result.setKind(tok::eof);
+        CurLexer->cutOffLexing();
+        EnterTokens(Suffix);
+        return true;
+      }
 
-    // Consume the pp-import-suffix and expand any macros in it now. We'll add
-    // it back into the token stream later.
-    CollectPpImportSuffix(Suffix);
-    if (Suffix.back().isNot(tok::semi)) {
-      // This is not a pp-import after all.
       EnterTokens(Suffix);
       return false;
     }
+  } else {
+    Lex(Result);
+  }
 
-    // C++2a [cpp.module]p1:
-    //   The ';' preprocessing-token terminating a pp-import shall not have
-    //   been produced by macro replacement.
-    SourceLocation SemiLoc = Suffix.back().getLocation();
-    if (SemiLoc.isMacroID())
-      Diag(SemiLoc, diag::err_header_import_semi_in_macro);
-
-    // Reconstitute the import token.
-    Token ImportTok;
-    ImportTok.startToken();
-    ImportTok.setKind(tok::kw_import);
-    ImportTok.setLocation(ModuleImportLoc);
-    ImportTok.setIdentifierInfo(getIdentifierInfo("import"));
-    ImportTok.setLength(6);
-
-    auto Action = HandleHeaderIncludeOrImport(
-        /*HashLoc*/ SourceLocation(), ImportTok, Suffix.front(), SemiLoc);
-    switch (Action.Kind) {
-    case ImportAction::None:
-      break;
-
-    case ImportAction::ModuleBegin:
-      // Let the parser know we're textually entering the module.
-      Suffix.emplace_back();
-      Suffix.back().startToken();
-      Suffix.back().setKind(tok::annot_module_begin);
-      Suffix.back().setLocation(SemiLoc);
-      Suffix.back().setAnnotationEndLoc(SemiLoc);
-      Suffix.back().setAnnotationValue(Action.ModuleForHeader);
-      [[fallthrough]];
-
-    case ImportAction::ModuleImport:
-    case ImportAction::HeaderUnitImport:
-    case ImportAction::SkippedModuleImport:
-      // We chose to import (or textually enter) the file. Convert the
-      // header-name token into a header unit annotation token.
-      Suffix[0].setKind(tok::annot_header_unit);
-      Suffix[0].setAnnotationEndLoc(Suffix[0].getLocation());
-      Suffix[0].setAnnotationValue(Action.ModuleForHeader);
-      // FIXME: Call the moduleImport callback?
-      break;
-    case ImportAction::Failure:
-      assert(TheModuleLoader.HadFatalFailure &&
-             "This should be an early exit only to a fatal error");
-      Result.setKind(tok::eof);
-      CurLexer->cutOffLexing();
-      EnterTokens(Suffix);
+  if (Result.isOneOf(tok::identifier, tok::colon)) {
+    EnterToken(Result, /*IsReinject=*/false);
+    if (!LexModuleName(Result, /*IsImport=*/true))
       return true;
+    auto *Info = Result.getAnnotationValueAs<ModuleNameInfo *>();
+    if (getLangOpts().CPlusPlusModules) {
+      // Under the standard C++ Modules, the dot is just part of the module
+      // name, and not a real hierarchy separator. Flatten such module names
+      // now.
+      //
+      // FIXME: Is this the right level to be performing this transformation?
+      std::string FlatModuleName;
+      if (Info->getTokens().front().is(tok::colon)) {
+        // Import a module partition allowed in C++20 Modules.
+        // We can import a partition in named module TU.
+        if (NamedModuleImportPath.empty() && ModuleDeclState.isNamedModule())
+          FlatModuleName = llvm::Twine(ModuleDeclState.getPrimaryName())
+                               .concat(Info->getFlatName())
+                               .str();
+        else
+          return true;
+      } else {
+        FlatModuleName = Info->getFlatName();
+      }
+      NamedModuleImportPath.emplace_back(getIdentifierInfo(FlatModuleName),
+                                         Result.getLocation());
+    } else {
+      Info->getModuleIdPath(NamedModuleImportPath);
     }
-
-    EnterTokens(Suffix);
-    return false;
-  }
-
-  // The token sequence
-  //
-  //   import identifier (. identifier)*
-  //
-  // indicates a module import directive. We already saw the 'import'
-  // contextual keyword, so now we're looking for the identifiers.
-  if (ModuleImportExpectsIdentifier && Result.getKind() == tok::identifier) {
-    // We expected to see an identifier here, and we did; continue handling
-    // identifiers.
-    NamedModuleImportPath.push_back(
-        std::make_pair(Result.getIdentifierInfo(), Result.getLocation()));
-    ModuleImportExpectsIdentifier = false;
-    CurLexerCallback = CLK_LexAfterModuleImport;
-    return true;
-  }
-
-  // If we're expecting a '.' or a ';', and we got a '.', then wait until we
-  // see the next identifier. (We can also see a '[[' that begins an
-  // attribute-specifier-seq here under the Standard C++ Modules.)
-  if (!ModuleImportExpectsIdentifier && Result.getKind() == tok::period) {
-    ModuleImportExpectsIdentifier = true;
-    CurLexerCallback = CLK_LexAfterModuleImport;
-    return true;
   }
 
   // If we didn't recognize a module name at all, this is not a (valid) import.
@@ -1291,24 +1458,6 @@ bool Preprocessor::LexAfterModuleImport(Token &Result) {
     SemiLoc = Suffix.back().getLocation();
   }
 
-  // Under the standard C++ Modules, the dot is just part of the module name,
-  // and not a real hierarchy separator. Flatten such module names now.
-  //
-  // FIXME: Is this the right level to be performing this transformation?
-  std::string FlatModuleName;
-  if (getLangOpts().CPlusPlusModules) {
-    for (auto &Piece : NamedModuleImportPath) {
-      // If the FlatModuleName ends with colon, it implies it is a partition.
-      if (!FlatModuleName.empty() && FlatModuleName.back() != ':')
-        FlatModuleName += ".";
-      FlatModuleName += Piece.first->getName();
-    }
-    SourceLocation FirstPathLoc = NamedModuleImportPath[0].second;
-    NamedModuleImportPath.clear();
-    NamedModuleImportPath.push_back(
-        std::make_pair(getIdentifierInfo(FlatModuleName), FirstPathLoc));
-  }
-
   Module *Imported = nullptr;
   // We don't/shouldn't load the standard c++20 modules when preprocessing.
   if (getLangOpts().Modules && !isInImportingCXXNamedModules()) {
@@ -1330,6 +1479,33 @@ bool Preprocessor::LexAfterModuleImport(Token &Result) {
   return true;
 }
 
+/// Lex a token following the 'module' contextual keyword.
+///
+/// [cpp.module]/p2:
+/// The pp-tokens, if any, of a pp-module shall be of the form:
+///       pp-module-name pp-module-partition[opt] pp-tokens[opt]
+///
+/// where the pp-tokens (if any) shall not begin with a ( preprocessing token
+/// and the grammar non-terminals are defined as:
+///       pp-module-name:
+///             pp-module-name-qualifierp[opt] identifier
+///       pp-module-partition:
+///             : pp-module-name-qualifier[opt] identifier
+///       pp-module-name-qualifier:
+///             identifier .
+///             pp-module-name-qualifier identifier .
+/// No identifier in the pp-module-name or pp-module-partition shall currently
+/// be defined as an object-like macro.
+///
+/// [cpp.module]/p3:
+/// Any preprocessing tokens after the module preprocessing token in the module
+/// directive are processed just as in normal text.
+bool Preprocessor::LexAfterModuleDecl(Token &Result) {
+  // Figure out what kind of lexer we actually have.
+  recomputeCurLexerKind();
+  return LexModuleName(Result, /*IsImport=*/false);
+}
+
 void Preprocessor::makeModuleVisible(Module *M, SourceLocation Loc) {
   CurSubmoduleState->VisibleModules.setVisible(
       M, Loc, [](Module *) {},

diff  --git a/clang/lib/Lex/TokenConcatenation.cpp b/clang/lib/Lex/TokenConcatenation.cpp
index 865879d180533..cdb636923b9e9 100644
--- a/clang/lib/Lex/TokenConcatenation.cpp
+++ b/clang/lib/Lex/TokenConcatenation.cpp
@@ -160,6 +160,13 @@ static char GetFirstChar(const Preprocessor &PP, const Token &Tok) {
 bool TokenConcatenation::AvoidConcat(const Token &PrevPrevTok,
                                      const Token &PrevTok,
                                      const Token &Tok) const {
+  // If previous token is a module name, we need avoid concat it with current
+  // token, otherwise, there will has an extra space between 'M' and ';' for the
+  // following code:
+  //
+  // import M;
+  if (PrevTok.is(tok::annot_module_name))
+    return false;
   // Conservatively assume that every annotation token that has a printable
   // form requires whitespace.
   if (PrevTok.isAnnotation())
@@ -190,6 +197,9 @@ bool TokenConcatenation::AvoidConcat(const Token &PrevPrevTok,
       return true;
     ConcatInfo &= ~aci_avoid_equal;
   }
+
+  if (Tok.is(tok::annot_module_name))
+    return true;
   if (Tok.isAnnotation()) {
     // Modules annotation can show up when generated automatically for includes.
     assert(Tok.isOneOf(tok::annot_module_include, tok::annot_module_begin,

diff  --git a/clang/lib/Parse/ParseDecl.cpp b/clang/lib/Parse/ParseDecl.cpp
index 7ce9a9cea1c7a..577527d0318f2 100644
--- a/clang/lib/Parse/ParseDecl.cpp
+++ b/clang/lib/Parse/ParseDecl.cpp
@@ -3958,7 +3958,13 @@ void Parser::ParseDeclarationSpecifiers(
 
       // We're done with the declaration-specifiers.
       goto DoneWithDeclSpec;
-
+    case tok::annot_module_name: {
+      PP.EnterTokenStream(
+          Tok.getAnnotationValueAs<ModuleNameInfo *>()->getTokens(),
+          /*DisableMacroExpansion=*/true, /*IsReinject=*/false);
+      ConsumeAnyToken();
+      [[fallthrough]];
+    }
       // typedef-name
     case tok::kw___super:
     case tok::kw_decltype:

diff  --git a/clang/lib/Parse/Parser.cpp b/clang/lib/Parse/Parser.cpp
index 5ebe71e496a2e..afb2e1e416168 100644
--- a/clang/lib/Parse/Parser.cpp
+++ b/clang/lib/Parse/Parser.cpp
@@ -2511,18 +2511,28 @@ Parser::ParseModuleDecl(Sema::ModuleImportState &ImportState) {
   }
 
   SmallVector<std::pair<IdentifierInfo *, SourceLocation>, 2> Path;
-  if (ParseModuleName(ModuleLoc, Path, /*IsImport*/ false))
+  if (Tok.isNot(tok::annot_module_name)) {
+    Diag(Tok, diag::err_module_expected_ident) << /*IsImport=*/false;
+    SkipUntil(tok::semi, StopBeforeMatch);
+    return nullptr;
+  }
+
+  auto *Info = Tok.getAnnotationValueAs<ModuleNameInfo *>();
+  ConsumeAnnotationToken();
+  if (ParseModuleName(ModuleLoc, Info->getModuleName(), Path,
+                      /*IsImport=*/false))
     return nullptr;
 
   // Parse the optional module-partition.
   SmallVector<std::pair<IdentifierInfo *, SourceLocation>, 2> Partition;
-  if (Tok.is(tok::colon)) {
-    SourceLocation ColonLoc = ConsumeToken();
+  if (Info->hasPartitionName()) {
+    SourceLocation ColonLoc = Info->getColonToken().getLocation();
     if (!getLangOpts().CPlusPlusModules)
       Diag(ColonLoc, diag::err_unsupported_module_partition)
           << SourceRange(ColonLoc, Partition.back().second);
     // Recover by ignoring the partition name.
-    else if (ParseModuleName(ModuleLoc, Partition, /*IsImport*/ false))
+    else if (ParseModuleName(ModuleLoc, Info->getPartitionName(), Partition,
+                             /*IsImport=*/false))
       return nullptr;
   }
 
@@ -2581,18 +2591,32 @@ Decl *Parser::ParseModuleImport(SourceLocation AtLoc,
     // This is a header import that the preprocessor mapped to a module import.
     HeaderUnit = reinterpret_cast<Module *>(Tok.getAnnotationValue());
     ConsumeAnnotationToken();
-  } else if (Tok.is(tok::colon)) {
-    SourceLocation ColonLoc = ConsumeToken();
-    if (!getLangOpts().CPlusPlusModules)
-      Diag(ColonLoc, diag::err_unsupported_module_partition)
-          << SourceRange(ColonLoc, Path.back().second);
-    // Recover by leaving partition empty.
-    else if (ParseModuleName(ColonLoc, Path, /*IsImport*/ true))
-      return nullptr;
-    else
-      IsPartition = true;
   } else {
-    if (ParseModuleName(ImportLoc, Path, /*IsImport*/ true))
+    if (Tok.isNot(tok::annot_module_name)) {
+      if (Tok.is(tok::code_completion)) {
+        cutOffParsing();
+        Actions.CodeCompletion().CodeCompleteModuleImport(ImportLoc, Path);
+        return nullptr;
+      }
+      Diag(Tok, diag::err_module_expected_ident) << /*IsImport=*/true;
+      SkipUntil(tok::semi, StopBeforeMatch);
+      return nullptr;
+    }
+    auto *Info = Tok.getAnnotationValueAs<ModuleNameInfo *>();
+    ConsumeAnnotationToken();
+    if (Info->hasPartitionName()) {
+      SourceLocation ColonLoc = Info->getColonToken().getLocation();
+      if (!getLangOpts().CPlusPlusModules)
+        Diag(ColonLoc, diag::err_unsupported_module_partition)
+            << SourceRange(ColonLoc, Path.back().second);
+      // Recover by leaving partition empty.
+      else if (ParseModuleName(ColonLoc, Info->getPartitionName(), Path,
+                               /*IsImport=*/true))
+        return nullptr;
+      else
+        IsPartition = true;
+    } else if (ParseModuleName(ImportLoc, Info->getModuleName(), Path,
+                               /*IsImport=*/true))
       return nullptr;
   }
 
@@ -2689,32 +2713,31 @@ Decl *Parser::ParseModuleImport(SourceLocation AtLoc,
 ///         module-name-qualifier:
 ///           module-name-qualifier[opt] identifier '.'
 bool Parser::ParseModuleName(
-    SourceLocation UseLoc,
+    SourceLocation UseLoc, ArrayRef<Token> ModuleName,
     SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path,
     bool IsImport) {
-  // Parse the module path.
-  while (true) {
-    if (!Tok.is(tok::identifier)) {
-      if (Tok.is(tok::code_completion)) {
-        cutOffParsing();
-        Actions.CodeCompletion().CodeCompleteModuleImport(UseLoc, Path);
-        return true;
-      }
-
-      Diag(Tok, diag::err_module_expected_ident) << IsImport;
-      SkipUntil(tok::semi);
+  ModuleNameInfo::getModuleIdPath(ModuleName, Path);
+  // Eg. import A.B.
+  if (ModuleName.back().isNot(tok::identifier)) {
+    if (Tok.is(tok::code_completion)) {
+      cutOffParsing();
+      Actions.CodeCompletion().CodeCompleteModuleImport(UseLoc, Path);
       return true;
     }
+    Diag(ModuleName.back(), diag::err_module_expected_ident) << IsImport;
+    SkipUntil(tok::semi, StopBeforeMatch);
+    return true;
+  }
 
-    // Record this part of the module path.
-    Path.push_back(std::make_pair(Tok.getIdentifierInfo(), Tok.getLocation()));
-    ConsumeToken();
-
-    if (Tok.isNot(tok::period))
-      return false;
-
-    ConsumeToken();
+  // [cpp.module]/p2: where the pp-tokens (if any) shall not begin with a (
+  // preprocessing token [...]
+  //
+  // Skip unitl ';' to recovery.
+  if (getLangOpts().CPlusPlusModules && Tok.is(tok::l_paren)) {
+    SkipUntil(tok::semi, StopBeforeMatch);
+    return true;
   }
+  return false;
 }
 
 /// Try recover parser when module annotation appears where it must not

diff  --git a/clang/test/CXX/cpp/cpp.module/p2.cppm b/clang/test/CXX/cpp/cpp.module/p2.cppm
new file mode 100644
index 0000000000000..966a88ccfa972
--- /dev/null
+++ b/clang/test/CXX/cpp/cpp.module/p2.cppm
@@ -0,0 +1,88 @@
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file %s %t
+
+// RUN: %clang_cc1 -std=c++20 %t/A.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/B.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/C.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/D.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/E.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/F.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/G.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/H.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/I.cppm -triple x86_64-linux-gnu -verify
+// RUN: %clang_cc1 -std=c++20 %t/J.cppm -triple x86_64-linux-gnu -verify
+
+//--- version.h
+#ifndef VERSION_H
+#define VERSION_H
+
+#define VERSION libv5
+#define A a
+#define B b
+#define C c
+#define FUNC_LIKE(X) function_like_##X
+#define ATTRS [[]]
+#define SEMICOLON ;
+
+#endif // VERSION_H
+
+//--- A.cppm
+module;
+#include "version.h"
+export module VERSION;  // expected-error {{the module name in a module declaration cannot contain an object-like macro 'VERSION'}}
+
+//--- B.cppm
+module;
+#include "version.h"
+export module A.B;      // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                        // expected-error {{the module name in a module declaration cannot contain an object-like macro 'B'}}
+
+//--- C.cppm
+module;                             // expected-error {{missing 'module' declaration at end of global module fragment introduced here}}
+#include "version.h"
+export module A.FUNC_LIKE(foo):C;   // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                                    // expected-error {{unexpected '(' after the module name in a module declaration}}
+
+//--- D.cppm
+module;                               // expected-error {{missing 'module' declaration at end of global module fragment introduced here}}
+#include "version.h"
+export module B.A.FUNC_LIKE(bar):C;   // expected-error {{the module name in a module declaration cannot contain an object-like macro 'B'}} \
+                                      // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                                      // expected-error {{unexpected '(' after the module name in a module declaration}}
+
+//--- E.cppm
+module;
+#include "version.h"
+export module a.FUNC_LIKE:c; // OK, FUNC_LIKE would not be treated as a macro name.
+// expected-no-diagnostics
+
+//--- F.cppm
+module;
+#include "version.h"
+export module a.FUNC_LIKE:c ATTRS; // OK, FUNC_LIKE would not be treated as a macro name.
+// expected-no-diagnostics
+
+//--- G.cppm
+module;                               // expected-error {{missing 'module' declaration at end of global module fragment introduced here}}
+#include "version.h"
+export module A.FUNC_LIKE(B c:C ATTRS // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                                      // expected-error {{unexpected '(' after the module name in a module declaration}}
+
+//--- H.cppm
+module;                                   // expected-error {{missing 'module' declaration at end of global module fragment introduced here}}
+#include "version.h"
+export module A.FUNC_LIKE(B,). c:C ATTRS  // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                                          // expected-error {{unexpected '(' after the module name in a module declaration}}
+
+//--- I.cppm
+module;                                   // expected-error {{missing 'module' declaration at end of global module fragment introduced here}}
+#include "version.h"
+export module A.FUNC_LIKE(B,) c:C ATTRS   // expected-error {{the module name in a module declaration cannot contain an object-like macro 'A'}} \
+                                          // expected-error {{unexpected '(' after the module name in a module declaration}}
+
+//--- J.cppm
+module;
+#include "version.h"
+export module unexpanded : unexpanded ATTRS SEMICOLON // OK, ATTRS and SEMICOLON can be expanded.
+// expected-no-diagnostics

diff  --git a/clang/test/CXX/module/basic/basic.link/module-declaration.cpp b/clang/test/CXX/module/basic/basic.link/module-declaration.cpp
index d71358cc7a571..14bbc911febfc 100644
--- a/clang/test/CXX/module/basic/basic.link/module-declaration.cpp
+++ b/clang/test/CXX/module/basic/basic.link/module-declaration.cpp
@@ -8,27 +8,19 @@
 // RUN: %clang_cc1 -std=c++20 -emit-module-interface -fmodule-file=x=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm
 //
 // Module implementation for unknown and known module. (The former is ill-formed.)
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify -x c++ %t/M.cpp \
-// RUN:            -DTEST=1 -DEXPORT= -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x=%t/x.pcm -fmodule-file=x.y=%t/x.y.pcm -verify -x c++ %t/M.cpp \
-// RUN:            -DTEST=2 -DEXPORT= -DMODULE_NAME=x
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify -x c++ %t/M1.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x=%t/x.pcm -fmodule-file=x.y=%t/x.y.pcm -verify -x c++ %t/M2.cpp
 //
 // Module interface for unknown and known module. (The latter is ill-formed due to
 // redefinition.)
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=3 -DEXPORT=export -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=4 -DEXPORT=export -DMODULE_NAME=x
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M3.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M4.cpp
 //
 // Miscellaneous syntax.
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=7 -DEXPORT=export -DMODULE_NAME='z elderberry'
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=8 -DEXPORT=export -DMODULE_NAME='z [[]]'
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=9 -DEXPORT=export -DMODULE_NAME='z [[fancy]]'
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M.cpp \
-// RUN:            -DTEST=10 -DEXPORT=export -DMODULE_NAME='z [[maybe_unused]]'
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M5.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M6.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M7.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -verify %t/M8.cpp
 
 //--- x.cppm
 export module x;
@@ -38,17 +30,26 @@ int a, b;
 export module x.y;
 int c;
 
-//--- M.cpp
-
-EXPORT module MODULE_NAME;
-#if TEST == 7
-// expected-error at -2 {{expected ';'}} expected-error at -2 {{a type specifier is required}}
-#elif TEST == 9
-// expected-warning at -4 {{unknown attribute 'fancy' ignored}}
-#elif TEST == 10
-// expected-error-re at -6 {{'maybe_unused' attribute cannot be applied to a module{{$}}}}
-#elif TEST == 1
-// expected-error at -8 {{module 'z' not found}}
-#else
-// expected-no-diagnostics
-#endif
+//--- M1.cpp
+module z; // expected-error {{module 'z' not found}}
+
+//--- M2.cpp
+module x; // expected-no-diagnostics
+
+//--- M3.cpp
+export module z; // expected-no-diagnostics
+
+//--- M4.cpp
+export module x; // expected-no-diagnostics
+
+//--- M5.cpp
+export module z elderberry; // expected-error {{expected ';'}} expected-error {{a type specifier is required}}
+
+//--- M6.cpp
+export module z [[]]; // expected-no-diagnostics
+
+//--- M7.cpp
+export module z [[fancy]]; // expected-warning {{unknown attribute 'fancy' ignored}}
+
+//--- M8.cpp
+export module z [[maybe_unused]]; // expected-error-re {{'maybe_unused' attribute cannot be applied to a module{{$}}}}

diff  --git a/clang/test/CXX/module/dcl.dcl/dcl.module/dcl.module.import/p1.cppm b/clang/test/CXX/module/dcl.dcl/dcl.module/dcl.module.import/p1.cppm
index 873e4c0edeac2..ecad4db32a7e9 100644
--- a/clang/test/CXX/module/dcl.dcl/dcl.module/dcl.module.import/p1.cppm
+++ b/clang/test/CXX/module/dcl.dcl/dcl.module/dcl.module.import/p1.cppm
@@ -6,10 +6,12 @@
 // RUN: %clang_cc1 -std=c++20 -emit-module-interface -fmodule-file=x=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm
 // RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/a.b.cppm -o %t/a.b.pcm
 //
-// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -fmodule-file=x=%t/x.pcm -verify %t/test.cpp \
-// RUN:            -DMODULE_NAME=z -DINTERFACE
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -fmodule-file=x=%t/x.pcm -verify %t/test-interface.cpp \
+// RUN:            -DINTERFACE
 // RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -fmodule-file=x=%t/x.pcm \
-// RUN:            -fmodule-file=a.b=%t/a.b.pcm -verify %t/test.cpp -DMODULE_NAME=a.b
+// RUN:            -fmodule-file=a.b=%t/a.b.pcm -verify %t/test.cpp
+// RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -fmodule-file=x=%t/x.pcm \
+// RUN:            -verify %t/test-module-not-found.cpp
 // RUN: %clang_cc1 -std=c++20 -I%t -fmodule-file=x.y=%t/x.y.pcm -fmodule-file=x=%t/x.pcm -verify %t/test.x.cpp
 
 //--- x.cppm
@@ -34,11 +36,8 @@ int use_2 = b; // ok
 int use_3 = c; // expected-error {{use of undeclared identifier 'c'}}
 
 //--- test.cpp
-#ifdef INTERFACE
-export module MODULE_NAME;
-#else
-module MODULE_NAME;
-#endif
+module;
+module a.b;
 
 import x;
 
@@ -51,6 +50,28 @@ import x.y;
 import x.; // expected-error {{expected a module name after 'import'}}
 import .x; // expected-error {{expected a module name after 'import'}}
 
-import blarg; // expected-error {{module 'blarg' not found}}
+int use_4 = c; // ok
+
+
+//--- test-interface.cpp
+module;
+export module z;
+
+import x;
+
+import x [[]];
+import x [[foo]]; // expected-warning {{unknown attribute 'foo' ignored}}
+import x [[noreturn]]; // expected-error {{'noreturn' attribute cannot be applied to a module import}}
+import x [[blarg::noreturn]]; // expected-warning {{unknown attribute 'noreturn' ignored}}
+
+import x.y;
+import x.; // expected-error {{expected a module name after 'import'}}
+import .x; // expected-error {{expected a module name after 'import'}}
 
 int use_4 = c; // ok
+
+//--- test-module-not-found.cpp
+module;
+
+import blarg; // expected-error {{module 'blarg' not found}}
+

diff  --git a/clang/test/SemaCXX/modules.cppm b/clang/test/SemaCXX/modules.cppm
index 41204be76eafa..267417bf5da2c 100644
--- a/clang/test/SemaCXX/modules.cppm
+++ b/clang/test/SemaCXX/modules.cppm
@@ -1,19 +1,17 @@
-// RUN:     %clang_cc1 -std=c++20 -emit-module-interface %s -o %t.0.pcm -verify -DTEST=0
-// RUN:     %clang_cc1 -std=c++20 -emit-module-interface %s -o %t.1.pcm -verify -DTEST=1
-// RUN:     %clang_cc1 -std=c++20 -emit-module-interface %s -fmodule-file=foo=%t.0.pcm -o %t.2.pcm -verify -DTEST=2
-// RUN:     %clang_cc1 -std=c++20 -emit-module-interface %s -fmodule-file=foo=%t.0.pcm -o %t.3.pcm -verify -Dfoo=bar -DTEST=3
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file %s %t
 
-#if TEST == 0 || TEST == 2
-// expected-no-diagnostics
-#endif
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/A.cppm -o %t.0.pcm -verify
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/B.cppm -o %t.1.pcm -verify
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/C.cppm -fmodule-file=foo=%t.0.pcm -o %t.2.pcm -verify
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/D.cppm -fmodule-file=foo=%t.0.pcm -o %t.3.pcm -verify
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/E.cppm -fmodule-file=foo=%t.0.pcm -o %t.3.pcm -verify -Dfoo=bar
 
+//--- A.cppm
 export module foo;
-
 static int m;
-
 int n;
-
-#if TEST == 0
 export {
   int a;
   int b;
@@ -27,7 +25,43 @@ export void f() {}
 
 export struct T {
 } t;
-#elif TEST == 3
+// expected-no-diagnostics
+
+//--- B.cppm
+export module foo;
+static int m;
+int n;
+struct S {
+  export int n;        // expected-error {{expected member name or ';'}}
+  export static int n; // expected-error {{expected member name or ';'}}
+};
+
+// FIXME: Exports of declarations without external linkage are disallowed.
+// Exports of declarations with non-external-linkage types are disallowed.
+
+// Cannot export within another export. This isn't precisely covered by the
+// language rules right now, but (per personal correspondence between zygoloid
+// and gdr) is the intent.
+export { // expected-note {{export block begins here}}
+  extern "C++" {
+    namespace NestedExport {
+      export { // expected-error {{export declaration appears within another export declaration}}
+        int q;
+      }
+    } // namespace NestedExport
+  }
+}
+
+//--- C.cppm
+export module foo;
+static int m;
+int n;
+// expected-no-diagnostics
+
+//--- D.cppm
+export module foo;
+static int m;
+int n;
 int use_a = a; // expected-error {{use of undeclared identifier 'a'}}
 
 #undef foo
@@ -46,29 +80,12 @@ int use_n = n; // FIXME: this should not be visible, because it is not exported
 
 extern int n;
 static_assert(&n != p); // expected-error{{use of undeclared identifier 'p'}}
-#endif
 
-#if TEST == 1
-struct S {
-  export int n;        // expected-error {{expected member name or ';'}}
-  export static int n; // expected-error {{expected member name or ';'}}
-};
-#endif
-
-// FIXME: Exports of declarations without external linkage are disallowed.
-// Exports of declarations with non-external-linkage types are disallowed.
+//--- E.cppm
+export module foo; // expected-error {{the module name in a module declaration cannot contain an object-like macro 'foo'}}
+static int m;
+int n;
+int use_a = a; // expected-error {{use of undeclared identifier 'a'}}
 
-// Cannot export within another export. This isn't precisely covered by the
-// language rules right now, but (per personal correspondence between zygoloid
-// and gdr) is the intent.
-#if TEST == 1
-export { // expected-note {{export block begins here}}
-  extern "C++" {
-  namespace NestedExport {
-  export { // expected-error {{export declaration appears within another export declaration}}
-    int q;
-  }
-  } // namespace NestedExport
-  }
-}
-#endif
+#undef foo
+import foo; // expected-error {{imports must immediately follow the module declaration}}

diff  --git a/clang/www/cxx_status.html b/clang/www/cxx_status.html
index a6ded8be3ae9e..1f69a4e8a5620 100755
--- a/clang/www/cxx_status.html
+++ b/clang/www/cxx_status.html
@@ -182,7 +182,7 @@ <h2 id="cxx26">C++2c implementation status</h2>
  <tr>
   <td>Module Declarations Shouldn’t be Macros</td>
   <td><a href="https://wg21.link/P3034R1">P3034R1</a> (<a href="#dr">DR</a>)</td>
-  <td class="none" align="center">No</td>
+  <td class="unreleased" align="center">Clang 19</td>
  </tr>
  <tr>
   <td>Trivial infinite loops are not Undefined Behavior</td>


        


More information about the cfe-commits mailing list