[clang] [clang][DependencyScanning] Implementation of `CompilerInstanceWithContext` to Improve By-Name Queries (PR #164345)

[Jan Svoboda via cfe-commits]

================
@@ -705,3 +707,185 @@ bool DependencyScanningAction::runInvocation(
 
   return Result;
 }
+
+const std::string CompilerInstanceWithContext::FakeFileBuffer =
+    std::string(MAX_NUM_NAMES, ' ');
+
+llvm::Error CompilerInstanceWithContext::initialize() {
+  std::tie(OverlayFS, CommandLine) = initVFSForByNameScanning(
+      Worker.BaseFS, CommandLine, CWD, "ScanningByName");
+
+  DiagPrinterWithOS =
+      std::make_unique<TextDiagnosticsPrinterWithOutput>(CommandLine);
+  DiagEngineWithCmdAndOpts = std::make_unique<DignosticsEngineWithDiagOpts>(
+      CommandLine, OverlayFS, DiagPrinterWithOS->DiagPrinter);
+
+  std::tie(Driver, Compilation) = buildCompilation(
+      CommandLine, *DiagEngineWithCmdAndOpts->DiagEngine, OverlayFS);
+
+  if (!Compilation) {
+    return llvm::make_error<llvm::StringError>("Failed to build compilation",
+                                               llvm::inconvertibleErrorCode());
+  }
+
+  const driver::Command &Command = *(Compilation->getJobs().begin());
+  const auto &CommandArgs = Command.getArguments();
+  size_t ArgSize = CommandArgs.size();
+  assert(ArgSize >= 1 && "Cannot have a command with 0 args");
+  const char *FirstArg = CommandArgs[0];
+  if (strcmp(FirstArg, "-cc1"))
+    return llvm::make_error<llvm::StringError>(
+        "Incorrect compilation command, missing cc1",
+        llvm::inconvertibleErrorCode());
+  Invocation = std::make_unique<CompilerInvocation>();
+
+  if (!CompilerInvocation::CreateFromArgs(*Invocation, Command.getArguments(),
+                                          *DiagEngineWithCmdAndOpts->DiagEngine,
+                                          Command.getExecutable())) {
+    DiagEngineWithCmdAndOpts->DiagEngine->Report(
+        diag::err_fe_expected_compiler_job)
+        << llvm::join(CommandLine, " ");
+    return llvm::make_error<llvm::StringError>(
+        "Cannot create CompilerInvocation from Args",
+        llvm::inconvertibleErrorCode());
+  }
+
+  // TODO: CMDArgsStrVector is making string copies. We should optimize later
+  // and avoid the copies.
+  std::vector<std::string> CMDArgsStrVector(ArgSize + 1);
+  CMDArgsStrVector.push_back(Command.getExecutable());
+  llvm::transform(CommandArgs, CMDArgsStrVector.begin() + 1,
+                  [](const char *s) { return std::string(s); });
+
+  Invocation = createCompilerInvocation(CMDArgsStrVector,
+                                        *DiagEngineWithCmdAndOpts->DiagEngine);
+  if (!Invocation) {
+    DiagEngineWithCmdAndOpts->DiagEngine->Report(
+        diag::err_fe_expected_compiler_job)
+        << llvm::join(CommandLine, " ");
+    return llvm::make_error<llvm::StringError>(
+        "Cannot create CompilerInvocation from Args",
+        llvm::inconvertibleErrorCode());
+  }
+
+  Invocation->getFrontendOpts().DisableFree = false;
+  Invocation->getCodeGenOpts().DisableFree = false;
+
+  if (any(Worker.Service.getOptimizeArgs() & ScanningOptimizations::Macros))
+    canonicalizeDefines(Invocation->getPreprocessorOpts());
+
+  // Create the CompilerInstance.
+  IntrusiveRefCntPtr<ModuleCache> ModCache =
+      makeInProcessModuleCache(Worker.Service.getModuleCacheEntries());
+  CIPtr = std::make_unique<CompilerInstance>(
+      std::make_shared<CompilerInvocation>(*Invocation), Worker.PCHContainerOps,
----------------
jansvoboda11 wrote:

Ah, now I see that this so that the invocation we store as member on `CompilerInstanceWithContext` is used for the consumer. Can we call it `OriginalInvocation` or something?

https://github.com/llvm/llvm-project/pull/164345