[clangd] Add symbol origin for remote index

[kbobyrev]

Summary

Makes it easier to diagnose remote index issues with --debug-origins flag.

[sammccall]

clang-tools-extra/clangd/index/remote/marshalling/Marshalling.cpp

164–165

I think we should just overwrite rather than OR.
In practice the remote side is always going to be Static, but we don't really want to conflate this with the client-side static index.

Actually, do you think we should set this on the *outgoing* side (i.e. have the proto say remote)?

[kbobyrev]

Don't assign additional Remote Origin: overwrite it.

clang-tools-extra/clangd/index/SymbolOrigin.h

 // This is a bitfield as information can be combined from several sources.
 enum class SymbolOrigin : uint8_t {
   Unknown = 0,
   AST = 1 << 0,        // Directly from the AST (indexes should not set this).
   Dynamic = 1 << 1,    // From the dynamic index of opened files.
   Static = 1 << 2,     // From the static, externally-built index.
   Merge = 1 << 3,      // A non-trivial index merge was performed.
   Identifier = 1 << 4, // Raw identifiers in file.
+  Remote = 1 << 5,     // Remote index.
   // Remaining bits reserved for index implementations.
 };
 
 inline SymbolOrigin operator|(SymbolOrigin A, SymbolOrigin B) {
   return static_cast<SymbolOrigin>(static_cast<uint8_t>(A) |
                                    static_cast<uint8_t>(B));
 }
 inline SymbolOrigin &operator|=(SymbolOrigin &A, SymbolOrigin B) {

clang-tools-extra/clangd/index/SymbolOrigin.cpp

 //===--- SymbolOrigin.cpp ----------------------------------------*- C++-*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "SymbolOrigin.h"
 
 namespace clang {
 namespace clangd {
 
 llvm::raw_ostream &operator<<(llvm::raw_ostream &OS, SymbolOrigin O) {
   if (O == SymbolOrigin::Unknown)
     return OS << "unknown";
-  constexpr static char Sigils[] = "ADSMI567";
+  constexpr static char Sigils[] = "ADSMIR67";
   for (unsigned I = 0; I < sizeof(Sigils); ++I)
     if (static_cast<uint8_t>(O) & 1u << I)
       OS << Sigils[I];
   return OS;
 }
 
 } // namespace clangd
 } // namespace clang

clang-tools-extra/clangd/index/remote/marshalling/Marshalling.cpp

if (Message.has_definition()) {
     if (Definition)
       Result.Definition = *Definition;
   }
   auto Declaration = fromProtobuf(Message.canonical_declaration());
   if (!Declaration)
     return Declaration.takeError();
   Result.CanonicalDeclaration = *Declaration;
   Result.References = Message.references();
-  Result.Origin = static_cast<clangd::SymbolOrigin>(Message.origin());
+  // Overwrite symbol origin: it's coming from remote index.
+  Result.Origin = clangd::SymbolOrigin::Remote;
   Result.Signature = Message.signature();
   Result.TemplateSpecializationArgs = Message.template_specialization_args();
   Result.CompletionSnippetSuffix = Message.completion_snippet_suffix();
   Result.Documentation = Message.documentation();
   Result.ReturnType = Message.return_type();
   Result.Type = Message.type();
   for (const auto &Header : Message.headers()) {
     auto SerializedHeader = fromProtobuf(Header);

clang-tools-extra/clangd/unittests/remote/MarshallingTests.cpp

 //===--- MarshallingTests.cpp ------------------------------------*- C++-*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "../TestTU.h"
 #include "Index.pb.h"
 #include "TestFS.h"
 #include "index/Index.h"
 #include "index/Ref.h"
 #include "index/Relation.h"
 #include "index/Serialization.h"
 #include "index/Symbol.h"
 #include "index/SymbolID.h"
 #include "index/SymbolLocation.h"
+#include "index/SymbolOrigin.h"
 #include "index/remote/marshalling/Marshalling.h"
 #include "clang/Index/IndexSymbol.h"
 #include "llvm/ADT/SmallString.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/Twine.h"
 #include "llvm/Support/Error.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/StringSaver.h"
TEST(RemoteMarshallingTest, SymbolSerialization) {
   Marshaller ProtobufMarshaller(testPath("home/"), testPath("home/"));
 
   // Check that symbols are exactly the same if the path to indexed project is
   // the same on indexing machine and the client.
   auto Serialized = ProtobufMarshaller.toProtobuf(Sym);
   ASSERT_TRUE(bool(Serialized));
   auto Deserialized = ProtobufMarshaller.fromProtobuf(*Serialized);
   ASSERT_TRUE(bool(Deserialized));
+  // Origin is overwritten when deserializing.
+  Sym.Origin = SymbolOrigin::Remote;
   EXPECT_EQ(toYAML(Sym), toYAML(*Deserialized));
   // Serialized paths are relative and have UNIX slashes.
   EXPECT_EQ(convert_to_slash(Serialized->definition().file_path(),
                              llvm::sys::path::Style::posix),
             Serialized->definition().file_path());
   EXPECT_TRUE(
       llvm::sys::path::is_relative(Serialized->definition().file_path()));
 
TEST(RemoteMarshallingTest, IncludeHeaderURIs) {
   Marshaller ProtobufMarshaller(testPath(""), testPath(""));
 
   auto Serialized = ProtobufMarshaller.toProtobuf(Sym);
   ASSERT_TRUE(bool(Serialized));
   EXPECT_EQ(static_cast<size_t>(Serialized->headers_size()),
             Sym.IncludeHeaders.size());
   auto Deserialized = ProtobufMarshaller.fromProtobuf(*Serialized);
   ASSERT_TRUE(bool(Deserialized));
+  Sym.Origin = SymbolOrigin::Remote;
   EXPECT_EQ(toYAML(Sym), toYAML(*Deserialized));
 
   // This is an absolute path to a header: can not be transmitted over the wire.
   Header.IncludeHeader = Strings.save(testPath("project/include/Common.h"));
   Header.References = 42;
   Sym.IncludeHeaders.push_back(Header);
   Serialized = ProtobufMarshaller.toProtobuf(Sym);
   EXPECT_FALSE(bool(Serialized));