[VFS] Fix inconsistencies between relative paths and fallthrough in the RedirectingFileSystem

[JDevlieghere]

Summary

This patch addresses inconsistencies in the way fallthrough is handled in the RedirectingFileSystem. Rather than trying to change the working directory of the external filesystem, the RedirectingFileSystem will canonicalize every path before handing it down. This guarantees that relative paths are resolved relative to the RedirectingFileSystem's working directory. This allows us to have a strictly virtual working directory, and still fallthrough for absolute paths, but not for relative paths that would get resolved incorrectly at the lower layer (for example, in case of the RealFileSystem, because the strictly virtual path does not exist).

[dexonsmith]

This seems like a good simplification.

llvm/lib/Support/VirtualFileSystem.cpp

1085–1094

I think you need to call makeCanonical here.

1753–1755

Typo s/StirngRef/StringRef/

1754

Is there a reason you had to manually inline canonicalize() here? It'd be nice to reuse the logic, since it's not entirely obvious.

[JDevlieghere]

• Implement makeCanonical using canonicalize
• Add missing call to makeCanonical is isLocal

[JDevlieghere]

llvm/lib/Support/VirtualFileSystem.cpp

1754

I was trying to avoid the copy into the SmallString, but that didn't work out because you need a buffer to copy from (otherwise you're copying from yourself). We could add an overload to remove_leading_dotslash that performs the operation in-place, but that would be for a separate patch.

[dexonsmith]

LGTM.

llvm/include/llvm/Support/VirtualFileSystem.h

public:
 
     static bool classof(const Entry *E) { return E->getKind() == EK_File; }
   };
 
 private:
   friend class VFSFromYamlDirIterImpl;
   friend class RedirectingFileSystemParser;
 
-  bool shouldUseExternalFS() const {
-    return ExternalFSValidWD && IsFallthrough;
-  }
+  bool shouldUseExternalFS() const { return IsFallthrough; }
+
+  /// Canonicalize path by removing ".", "..", "./", components. This is
+  /// a VFS request, do not bother about symlinks in the path components
+  /// but canonicalize in order to perform the correct entry search.
+  std::error_code makeCanonical(SmallVectorImpl<char> &Path) const;
 
   // In a RedirectingFileSystem, keys can be specified in Posix or Windows
   // style (or even a mixture of both), so this comparison helper allows
   // slashes (representing a root) to match backslashes (and vice versa).  Note
   // that, other than the root, path components should not contain slashes or
   // backslashes.
   bool pathComponentMatches(llvm::StringRef lhs, llvm::StringRef rhs) const {
     if ((CaseSensitive ? lhs.equals(rhs) : lhs.equals_lower(rhs)))
       return true;
     return (lhs == "/" && rhs == "\\") || (lhs == "\\" && rhs == "/");
   }
 
   /// The root(s) of the virtual file system.
   std::vector<std::unique_ptr<Entry>> Roots;
 
   /// The current working directory of the file system.
   std::string WorkingDirectory;
 
-  /// Whether the current working directory is valid for the external FS.
-  bool ExternalFSValidWD = false;
-
   /// The file system to use for external references.
   IntrusiveRefCntPtr<FileSystem> ExternalFS;
 
   /// If IsRelativeOverlay is set, this represents the directory
   /// path that should be prefixed to each 'external-contents' entry
   /// when reading from YAML files.
   std::string ExternalContentsPrefixDir;
 
ErrorOr<Entry *> lookupPath(llvm::sys::path::const_iterator Start,
                               llvm::sys::path::const_iterator End,
                               Entry *From) const;
 
   /// Get the status of a given an \c Entry.
   ErrorOr<Status> status(const Twine &Path, Entry *E);
 
 public:
   /// Looks up \p Path in \c Roots.
-  ErrorOr<Entry *> lookupPath(const Twine &Path) const;
+  ErrorOr<Entry *> lookupPath(StringRef Path) const;
 
   /// Parses \p Buffer, which is expected to be in YAML format and
   /// returns a virtual file system representing its contents.
   static std::unique_ptr<RedirectingFileSystem>
   create(std::unique_ptr<MemoryBuffer> Buffer,
          SourceMgr::DiagHandlerTy DiagHandler, StringRef YAMLFilePath,
          void *DiagContext, IntrusiveRefCntPtr<FileSystem> ExternalFS);
 

llvm/lib/Support/VirtualFileSystem.cpp

 
 
 RedirectingFileSystem::RedirectingFileSystem(IntrusiveRefCntPtr<FileSystem> FS)
     : ExternalFS(std::move(FS)) {
   if (ExternalFS)
     if (auto ExternalWorkingDirectory =
             ExternalFS->getCurrentWorkingDirectory()) {
       WorkingDirectory = *ExternalWorkingDirectory;
-      ExternalFSValidWD = true;
     }
 }
 
 // FIXME: reuse implementation common with OverlayFSDirIterImpl as these
 // iterators are conceptually similar.
 class llvm::vfs::VFSFromYamlDirIterImpl
     : public llvm::vfs::detail::DirIterImpl {
   std::string Dir;

 }
 
 std::error_code
 RedirectingFileSystem::setCurrentWorkingDirectory(const Twine &Path) {
   // Don't change the working directory if the path doesn't exist.
   if (!exists(Path))
     return errc::no_such_file_or_directory;
 
-  // Always change the external FS but ignore its result.
-  if (ExternalFS) {
-    auto EC = ExternalFS->setCurrentWorkingDirectory(Path);
-    ExternalFSValidWD = !static_cast<bool>(EC);
-  }
-
   SmallString<128> AbsolutePath;
   Path.toVector(AbsolutePath);
   if (std::error_code EC = makeAbsolute(AbsolutePath))
     return EC;
   WorkingDirectory = std::string(AbsolutePath.str());
   return {};
 }
 
-std::error_code RedirectingFileSystem::isLocal(const Twine &Path,
+std::error_code RedirectingFileSystem::isLocal(const Twine &Path_,
                                                bool &Result) {
+  SmallString<256> Path;
+  Path_.toVector(Path);
+
+  if (std::error_code EC = makeCanonical(Path))
+    return {};
+
   return ExternalFS->isLocal(Path, Result);
 }
 
 std::error_code RedirectingFileSystem::makeAbsolute(SmallVectorImpl<char> &Path) const {
   if (llvm::sys::path::is_absolute(Path, llvm::sys::path::Style::posix) ||
       llvm::sys::path::is_absolute(Path, llvm::sys::path::Style::windows))
     return {};
 
   auto WorkingDir = getCurrentWorkingDirectory();
   if (!WorkingDir)

   Result.append(Path.data(), Path.size());
   Path.assign(Result.begin(), Result.end());
 
   return {};
 }
 
 directory_iterator RedirectingFileSystem::dir_begin(const Twine &Dir,
                                                     std::error_code &EC) {
-  ErrorOr<RedirectingFileSystem::Entry *> E = lookupPath(Dir);
+  SmallString<256> Path;
+  Dir.toVector(Path);
+
+  EC = makeCanonical(Path);
+  if (EC)
+    return {};
+
+  ErrorOr<RedirectingFileSystem::Entry *> E = lookupPath(Path);
   if (!E) {
     EC = E.getError();
     if (shouldUseExternalFS() && EC == errc::no_such_file_or_directory)
-      return ExternalFS->dir_begin(Dir, EC);
+      return ExternalFS->dir_begin(Path, EC);
     return {};
   }
-  ErrorOr<Status> S = status(Dir, *E);
+  ErrorOr<Status> S = status(Path, *E);
   if (!S) {
     EC = S.getError();
     return {};
   }
   if (!S->isDirectory()) {
     EC = std::error_code(static_cast<int>(errc::not_a_directory),
                          std::system_category());
     return {};
   }
 
   auto *D = cast<RedirectingFileSystem::RedirectingDirectoryEntry>(*E);
   return directory_iterator(std::make_shared<VFSFromYamlDirIterImpl>(
-      Dir, D->contents_begin(), D->contents_end(),
+      Path, D->contents_begin(), D->contents_end(),
       /*IterateExternalFS=*/shouldUseExternalFS(), *ExternalFS, EC));
 }
 
 void RedirectingFileSystem::setExternalContentsPrefixDir(StringRef PrefixDir) {
   ExternalContentsPrefixDir = PrefixDir.str();
 }
 
 StringRef RedirectingFileSystem::getExternalContentsPrefixDir() const {

     ToEntry = NewFile.get();
     cast<RedirectingFileSystem::RedirectingDirectoryEntry>(Parent)->addContent(
         std::move(NewFile));
   }
 
   return FS;
 }
 
-ErrorOr<RedirectingFileSystem::Entry *>
-RedirectingFileSystem::lookupPath(const Twine &Path_) const {
-  SmallString<256> Path;
-  Path_.toVector(Path);
-
-  // Handle relative paths
+std::error_code
+RedirectingFileSystem::makeCanonical(SmallVectorImpl<char> &Path) const {
   if (std::error_code EC = makeAbsolute(Path))
     return EC;
 
-  // Canonicalize path by removing ".", "..", "./", components. This is
-  // a VFS request, do not bother about symlinks in the path components
-  // but canonicalize in order to perform the correct entry search.
-  Path = canonicalize(Path);
-  if (Path.empty())
+  llvm::SmallString<256> CanonicalPath =
+      canonicalize(StringRef(Path.data(), Path.size()));
+  if (CanonicalPath.empty())
     return make_error_code(llvm::errc::invalid_argument);
 
+  Path.assign(CanonicalPath.begin(), CanonicalPath.end());
+  return {};
+}
+
+ErrorOr<RedirectingFileSystem::Entry *>
+RedirectingFileSystem::lookupPath(StringRef Path) const {
   sys::path::const_iterator Start = sys::path::begin(Path);
   sys::path::const_iterator End = sys::path::end(Path);
   for (const auto &Root : Roots) {
     ErrorOr<RedirectingFileSystem::Entry *> Result =
         lookupPath(Start, End, Root.get());
     if (Result || Result.getError() != llvm::errc::no_such_file_or_directory)
       return Result;
   }

                                      *S);
     return S;
   } else { // directory
     auto *DE = cast<RedirectingFileSystem::RedirectingDirectoryEntry>(E);
     return Status::copyWithNewName(DE->getStatus(), Path);
   }
 }
 
-ErrorOr<Status> RedirectingFileSystem::status(const Twine &Path) {
+ErrorOr<Status> RedirectingFileSystem::status(const Twine &Path_) {
+  SmallString<256> Path;
+  Path_.toVector(Path);
+
+  if (std::error_code EC = makeCanonical(Path))
+    return EC;
+
   ErrorOr<RedirectingFileSystem::Entry *> Result = lookupPath(Path);
   if (!Result) {
     if (shouldUseExternalFS() &&
         Result.getError() == llvm::errc::no_such_file_or_directory) {
       return ExternalFS->status(Path);
     }
     return Result.getError();
   }

   }
 
   std::error_code close() override { return InnerFile->close(); }
 };
 
 } // namespace
 
 ErrorOr<std::unique_ptr<File>>
-RedirectingFileSystem::openFileForRead(const Twine &Path) {
+RedirectingFileSystem::openFileForRead(const Twine &Path_) {
+  SmallString<256> Path;
+  Path_.toVector(Path);
+
+  if (std::error_code EC = makeCanonical(Path))
+    return EC;
+
   ErrorOr<RedirectingFileSystem::Entry *> E = lookupPath(Path);
   if (!E) {
     if (shouldUseExternalFS() &&
         E.getError() == llvm::errc::no_such_file_or_directory) {
       return ExternalFS->openFileForRead(Path);
     }
     return E.getError();
   }

   // FIXME: Update the status with the name and VFSMapped.
   Status S = getRedirectedFileStatus(Path, F->useExternalName(UseExternalNames),
                                      *ExternalStatus);
   return std::unique_ptr<File>(
       std::make_unique<FileWithFixedStatus>(std::move(*Result), S));
 }
 
 std::error_code
-RedirectingFileSystem::getRealPath(const Twine &Path,
+RedirectingFileSystem::getRealPath(const Twine &Path_,
                                    SmallVectorImpl<char> &Output) const {
+  SmallString<256> Path;
+  Path_.toVector(Path);
+
+  if (std::error_code EC = makeCanonical(Path))
+    return EC;
+
   ErrorOr<RedirectingFileSystem::Entry *> Result = lookupPath(Path);
   if (!Result) {
     if (shouldUseExternalFS() &&
         Result.getError() == llvm::errc::no_such_file_or_directory) {
       return ExternalFS->getRealPath(Path, Output);
     }
     return Result.getError();
   }

llvm/unittests/Support/VirtualFileSystemTest.cpp

   ASSERT_FALSE(EC);
   ASSERT_TRUE(FS.get() != nullptr);
 
   llvm::ErrorOr<vfs::Status> Status = FS->status("bar/a");
   ASSERT_FALSE(Status.getError());
   EXPECT_TRUE(Status->exists());
 
   Status = FS->status("foo/a");
-  ASSERT_TRUE(Status.getError());
+  ASSERT_FALSE(Status.getError());
+  EXPECT_TRUE(Status->exists());
+}
+
+TEST_F(VFSFromYAMLTest, VirtualWorkingDirectory) {
+  IntrusiveRefCntPtr<ErrorDummyFileSystem> Lower(new ErrorDummyFileSystem());
+  Lower->addDirectory("//root/");
+  Lower->addDirectory("//root/foo");
+  Lower->addRegularFile("//root/foo/a");
+  Lower->addRegularFile("//root/foo/b");
+  Lower->addRegularFile("//root/c");
+  IntrusiveRefCntPtr<vfs::FileSystem> FS = getFromYAMLString(
+      "{ 'use-external-names': false,\n"
+      "  'roots': [\n"
+      "{\n"
+      "  'type': 'directory',\n"
+      "  'name': '//root/bar',\n"
+      "  'contents': [ {\n"
+      "                  'type': 'file',\n"
+      "                  'name': 'a',\n"
+      "                  'external-contents': '//root/foo/a'\n"
+      "                }\n"
+      "              ]\n"
+      "}\n"
+      "]\n"
+      "}",
+      Lower);
+  ASSERT_TRUE(FS.get() != nullptr);
+  std::error_code EC = FS->setCurrentWorkingDirectory("//root/bar");
+  ASSERT_FALSE(EC);
+  ASSERT_TRUE(FS.get() != nullptr);
+
+  llvm::ErrorOr<vfs::Status> Status = FS->status("a");
+  ASSERT_FALSE(Status.getError());
+  EXPECT_TRUE(Status->exists());
 }
 
 TEST_F(VFSFromYAMLTest, YAMLVFSWriterTest) {
   TempDir TestDirectory("virtual-file-system-test", /*Unique*/ true);
   TempDir _a(TestDirectory.path("a"));
   TempFile _ab(TestDirectory.path("a, b"));
   TempDir _c(TestDirectory.path("c"));
   TempFile _cd(TestDirectory.path("c/d"));