When loading a module that involves submodules (e.g., std.vector),

check whether the named submodules themselves are actually
valid, and drill down to the named submodule (although we don't do
anything with it yet). Perform typo correction on the submodule names
when possible.



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@145477 91177308-0d34-0410-b5e6-96231b3b80d8

7 files changed, 187 insertions, 82 deletions

diff --git a/include/clang/Basic/DiagnosticFrontendKinds.td b/include/clang/Basic/DiagnosticFrontendKinds.td
index f48e9e3809..cd797fce47 100644
--- a/include/clang/Basic/DiagnosticFrontendKinds.td
+++ b/include/clang/Basic/DiagnosticFrontendKinds.td
@@ -180,4 +180,8 @@ def err_missing_module : Error<
   "no module named '%0' declared in module map file '%1'">, DefaultFatal;
 def err_missing_umbrella_header : Error<
   "cannot open umbrella header '%0': %1">, DefaultFatal;
+def err_no_submodule : Error<"no submodule named %0 in module '%1'">;
+def err_no_submodule_suggest : Error<
+  "no submodule named %0 in module '%1'; did you mean '%2'?">;
+
 }
diff --git a/include/clang/Frontend/CompilerInstance.h b/include/clang/Frontend/CompilerInstance.h
index 9f982c9265..90de3138a4 100644
--- a/include/clang/Frontend/CompilerInstance.h
+++ b/include/clang/Frontend/CompilerInstance.h
@@ -12,7 +12,9 @@
 
 #include "clang/Frontend/CompilerInvocation.h"
 #include "clang/Lex/ModuleLoader.h"
+#include "clang/Lex/ModuleMap.h"
 #include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/IntrusiveRefCntPtr.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/OwningPtr.h"
@@ -34,6 +36,7 @@ class CodeCompleteConsumer;
 class DiagnosticsEngine;
 class DiagnosticConsumer;
 class ExternalASTSource;
+class FileEntry;
 class FileManager;
 class FrontendAction;
 class Preprocessor;
@@ -96,6 +99,16 @@ class CompilerInstance : public ModuleLoader {
   /// \brief Non-owning reference to the ASTReader, if one exists.
   ASTReader *ModuleManager;
 
+  /// \brief A module that we have already attempted to load, which is known
+  /// by either a file entry (FIXME: a temporary measure) or via its module
+  /// definition.
+  typedef llvm::PointerUnion<const FileEntry *, ModuleMap::Module *> 
+    KnownModule;
+  
+  /// \brief The set of top-level modules that has already been loaded,
+  /// along with the module map
+  llvm::DenseMap<const IdentifierInfo *, KnownModule> KnownModules;
+  
   /// \brief Holds information about the output file.
   ///
   /// If TempFilename is not empty we must rename it to Filename at the end.
diff --git a/lib/Frontend/CompilerInstance.cpp b/lib/Frontend/CompilerInstance.cpp
index a72370312d..dfc98e8bfb 100644
--- a/lib/Frontend/CompilerInstance.cpp
+++ b/lib/Frontend/CompilerInstance.cpp
@@ -1080,99 +1080,167 @@ ModuleKey CompilerInstance::loadModule(SourceLocation ImportLoc,
 
   StringRef ModuleName = Path[0].first->getName();
   SourceLocation ModuleNameLoc = Path[0].second;
-  
-  // Search for a module with the given name.
-  ModuleMap::Module *Module = 0;
-  std::string ModuleFileName;
-  const FileEntry *ModuleFile
-    = PP->getHeaderSearchInfo().lookupModule(ModuleName, Module,
-                                             &ModuleFileName);
 
-  // FIXME: Verify that the rest of the module path actually corresponds to
-  // a submodule, and pass that information through.
+  ModuleMap::Module *Module = 0;
+  const FileEntry *ModuleFile = 0;
   
-  bool BuildingModule = false;
-  if (!ModuleFile && Module) {
-    // The module is not cached, but we have a module map from which we can
-    // build the module.
-
-    // Check whether there is a cycle in the module graph.
-    SmallVectorImpl<std::string> &ModuleBuildPath
-      = getPreprocessorOpts().ModuleBuildPath;
-    SmallVectorImpl<std::string>::iterator Pos
-      = std::find(ModuleBuildPath.begin(), ModuleBuildPath.end(), ModuleName);
-    if (Pos != ModuleBuildPath.end()) {
-      llvm::SmallString<256> CyclePath;
-      for (; Pos != ModuleBuildPath.end(); ++Pos) {
-        CyclePath += *Pos;
-        CyclePath += " -> ";
+  // If we don't already have information on this module, load the module now.
+  KnownModule &Known = KnownModules[Path[0].first];
+  if (Known.isNull()) {  
+    // Search for a module with the given name.
+    std::string ModuleFileName;
+    ModuleFile
+      = PP->getHeaderSearchInfo().lookupModule(ModuleName, Module,
+                                               &ModuleFileName);
+
+    bool BuildingModule = false;
+    if (!ModuleFile && Module) {
+      // The module is not cached, but we have a module map from which we can
+      // build the module.
+
+      // Check whether there is a cycle in the module graph.
+      SmallVectorImpl<std::string> &ModuleBuildPath
+        = getPreprocessorOpts().ModuleBuildPath;
+      SmallVectorImpl<std::string>::iterator Pos
+        = std::find(ModuleBuildPath.begin(), ModuleBuildPath.end(), ModuleName);
+      if (Pos != ModuleBuildPath.end()) {
+        llvm::SmallString<256> CyclePath;
+        for (; Pos != ModuleBuildPath.end(); ++Pos) {
+          CyclePath += *Pos;
+          CyclePath += " -> ";
+        }
+        CyclePath += ModuleName;
+
+        getDiagnostics().Report(ModuleNameLoc, diag::err_module_cycle)
+          << ModuleName << CyclePath;
+        return 0;
       }
-      CyclePath += ModuleName;
 
-      getDiagnostics().Report(ModuleNameLoc, diag::err_module_cycle)
-        << ModuleName << CyclePath;
-      return 0;
+      getDiagnostics().Report(ModuleNameLoc, diag::warn_module_build)
+        << ModuleName;
+      BuildingModule = true;
+      compileModule(*this, Module, ModuleFileName);
+      ModuleFile = FileMgr->getFile(ModuleFileName);
     }
 
-    getDiagnostics().Report(ModuleNameLoc, diag::warn_module_build)
-      << ModuleName;
-    BuildingModule = true;
-    compileModule(*this, Module, ModuleFileName);
-    ModuleFile = FileMgr->getFile(ModuleFileName);
-  }
-
-  if (!ModuleFile) {
-    getDiagnostics().Report(ModuleNameLoc,
-                            BuildingModule? diag::err_module_not_built
-                                          : diag::err_module_not_found)
-      << ModuleName
-      << SourceRange(ImportLoc, ModuleNameLoc);
-    return 0;
-  }
+    if (!ModuleFile) {
+      getDiagnostics().Report(ModuleNameLoc,
+                              BuildingModule? diag::err_module_not_built
+                                            : diag::err_module_not_found)
+        << ModuleName
+        << SourceRange(ImportLoc, ModuleNameLoc);
+      return 0;
+    }
 
-  // If we don't already have an ASTReader, create one now.
-  if (!ModuleManager) {
-    if (!hasASTContext())
-      createASTContext();
-
-    std::string Sysroot = getHeaderSearchOpts().Sysroot;
-    const PreprocessorOptions &PPOpts = getPreprocessorOpts();
-    ModuleManager = new ASTReader(getPreprocessor(), *Context,
-                                  Sysroot.empty() ? "" : Sysroot.c_str(),
-                                  PPOpts.DisablePCHValidation,
-                                  PPOpts.DisableStatCache);
-    if (hasASTConsumer()) {
-      ModuleManager->setDeserializationListener(
-        getASTConsumer().GetASTDeserializationListener());
-      getASTContext().setASTMutationListener(
-        getASTConsumer().GetASTMutationListener());
+    // If we don't already have an ASTReader, create one now.
+    if (!ModuleManager) {
+      if (!hasASTContext())
+        createASTContext();
+
+      std::string Sysroot = getHeaderSearchOpts().Sysroot;
+      const PreprocessorOptions &PPOpts = getPreprocessorOpts();
+      ModuleManager = new ASTReader(getPreprocessor(), *Context,
+                                    Sysroot.empty() ? "" : Sysroot.c_str(),
+                                    PPOpts.DisablePCHValidation,
+                                    PPOpts.DisableStatCache);
+      if (hasASTConsumer()) {
+        ModuleManager->setDeserializationListener(
+          getASTConsumer().GetASTDeserializationListener());
+        getASTContext().setASTMutationListener(
+          getASTConsumer().GetASTMutationListener());
+      }
+      llvm::OwningPtr<ExternalASTSource> Source;
+      Source.reset(ModuleManager);
+      getASTContext().setExternalSource(Source);
+      if (hasSema())
+        ModuleManager->InitializeSema(getSema());
+      if (hasASTConsumer())
+        ModuleManager->StartTranslationUnit(&getASTConsumer());
     }
-    llvm::OwningPtr<ExternalASTSource> Source;
-    Source.reset(ModuleManager);
-    getASTContext().setExternalSource(Source);
-    if (hasSema())
-      ModuleManager->InitializeSema(getSema());
-    if (hasASTConsumer())
-      ModuleManager->StartTranslationUnit(&getASTConsumer());
-  }
 
-  // Try to load the module we found.
-  switch (ModuleManager->ReadAST(ModuleFile->getName(),
-                                 serialization::MK_Module)) {
-  case ASTReader::Success:
-    break;
+    // Try to load the module we found.
+    switch (ModuleManager->ReadAST(ModuleFile->getName(),
+                                   serialization::MK_Module)) {
+    case ASTReader::Success:
+      break;
 
-  case ASTReader::IgnorePCH:
-    // FIXME: The ASTReader will already have complained, but can we showhorn
-    // that diagnostic information into a more useful form?
-    return 0;
+    case ASTReader::IgnorePCH:
+      // FIXME: The ASTReader will already have complained, but can we showhorn
+      // that diagnostic information into a more useful form?
+      return 0;
 
-  case ASTReader::Failure:
-    // Already complained.
-    return 0;
+    case ASTReader::Failure:
+      // Already complained.
+      return 0;
+    }
+    
+    if (Module)
+      Known = Module;
+    else
+      Known = ModuleFile;
+  } else {
+    Module = Known.dyn_cast<ModuleMap::Module *>();
   }
-
-  // FIXME: The module file's FileEntry makes a poor key indeed!
-  return (ModuleKey)ModuleFile;
+  
+  // Verify that the rest of the module path actually corresponds to
+  // a submodule.
+  ModuleMap::Module *Sub = 0;
+  if (Module && Path.size() > 1) {
+    Sub = Module;
+    for (unsigned I = 1, N = Path.size(); I != N; ++I) {
+      StringRef Name = Path[I].first->getName();
+      llvm::StringMap<ModuleMap::Module *>::iterator Pos
+        = Sub->SubModules.find(Name);
+      
+      if (Pos == Sub->SubModules.end()) {
+        // Attempt to perform typo correction to find a module name that works.
+        llvm::SmallVector<StringRef, 2> Best;
+        unsigned BestEditDistance = (std::numeric_limits<unsigned>::max)();
+        
+        for (llvm::StringMap<ModuleMap::Module *>::iterator 
+                  I = Sub->SubModules.begin(), 
+               IEnd = Sub->SubModules.end();
+             I != IEnd; ++I) {
+          unsigned ED = Name.edit_distance(I->getValue()->Name,
+                                           /*AllowReplacements=*/true,
+                                           BestEditDistance);
+          if (ED <= BestEditDistance) {
+            if (ED < BestEditDistance)
+              Best.clear();
+            Best.push_back(I->getValue()->Name);
+          }
+        }
+        
+        // If there was a clear winner, user it.
+        if (Best.size() == 1) {
+          getDiagnostics().Report(Path[I].second, 
+                                  diag::err_no_submodule_suggest)
+            << Path[I].first << Sub->getFullModuleName() << Best[0]
+            << SourceRange(Path[0].second, Path[I-1].second)
+            << FixItHint::CreateReplacement(SourceRange(Path[I].second),
+                                            Best[0]);
+          Pos = Sub->SubModules.find(Best[0]);
+        }
+      }
+      
+      if (Pos == Sub->SubModules.end()) {
+        // No submodule by this name. Complain, and don't look for further
+        // submodules.
+        getDiagnostics().Report(Path[I].second, diag::err_no_submodule)
+          << Path[I].first << Sub->getFullModuleName()
+          << SourceRange(Path[0].second, Path[I-1].second);
+        break;
+      }
+      
+      Sub = Pos->getValue();
+    }
+  }
+  
+  // FIXME: Tell the AST reader to make the named submodule visible.
+  
+  // FIXME: The module file's FileEntry makes a poor key indeed! Once we 
+  // eliminate the need for FileEntry here, the module itself will become the
+  // key (which does make sense).
+  return Known.getOpaqueValue();
 }
 
diff --git a/test/Modules/Inputs/submodules/module.map b/test/Modules/Inputs/submodules/module.map
new file mode 100644
index 0000000000..6d19975c96
--- /dev/null
+++ b/test/Modules/Inputs/submodules/module.map
@@ -0,0 +1,4 @@
+module std {
+  module vector { header "vector.h" }
+  module type_traits { header "type_traits.h" }
+}
diff --git a/test/Modules/Inputs/submodules/type_traits.h b/test/Modules/Inputs/submodules/type_traits.h
new file mode 100644
index 0000000000..9ef9006fbc
--- /dev/null
+++ b/test/Modules/Inputs/submodules/type_traits.h
@@ -0,0 +1,9 @@
+template<typename T>
+struct remove_reference {
+  typedef T type;
+};
+
+template<typename T>
+struct remove_reference<T&> {
+  typedef T type;
+};
diff --git a/test/Modules/Inputs/submodules/vector.h b/test/Modules/Inputs/submodules/vector.h
new file mode 100644
index 0000000000..3123dd1c84
--- /dev/null
+++ b/test/Modules/Inputs/submodules/vector.h
@@ -0,0 +1 @@
+template<typename T> class vector;
diff --git a/test/Modules/submodules.cpp b/test/Modules/submodules.cpp
new file mode 100644
index 0000000000..6d541f69e7
--- /dev/null
+++ b/test/Modules/submodules.cpp
@@ -0,0 +1,6 @@
+// RUN: rm -rf %t
+// RUN: %clang_cc1 -fmodule-cache-path %t -fauto-module-import -I %S/Inputs/submodules %s -verify
+
+__import_module__ std.vector;
+__import_module__ std.typetraits; // expected-error{{no submodule named 'typetraits' in module 'std'; did you mean 'type_traits'?}}
+__import_module__ std.vector.compare; // expected-error{{no submodule named 'compare' in module 'std.vector'}}