[clang][deps] Give the fake file a unique name in by-module-name scans

[jansvoboda11]

Summary

When scanning dependencies of a module, the command line we're given doesn't have an input file, which the driver needs to be happy. We've been creating a fake in-memory input file named after the module. However, this can hide files/directories on the actual filesystem, leading to errors.

This patch works around that issue by generating a unique file name, which won't collide with the actual file system.

We could also change the driver APIs so that we're able to specify an "assumed" input file. This would be more work, though, since the driver assumes the input name comes from the actual command-line.

Depends on D140176.

[jansvoboda11]

Improve tests

clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp

bool DependencyScanningWorker::computeDependencies(
     StringRef WorkingDirectory, const std::vector<std::string> &CommandLine,
     DependencyConsumer &Consumer, DiagnosticConsumer &DC,
     std::optional<StringRef> ModuleName) {
   // Reset what might have been modified in the previous worker invocation.
   BaseFS->setCurrentWorkingDirectory(WorkingDirectory);
 
   std::optional<std::vector<std::string>> ModifiedCommandLine;
   llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> ModifiedFS;
-  if (ModuleName) {
-    ModifiedCommandLine = CommandLine;
-    ModifiedCommandLine->emplace_back(*ModuleName);
 
+  // If we're scanning based on a module name alone, we don't expect the client
+  // to provide us with an input file. However, the driver really wants to have
+  // one. Let's just make it up to make the driver happy.
+  if (ModuleName) {
     auto OverlayFS =
         llvm::makeIntrusiveRefCnt<llvm::vfs::OverlayFileSystem>(BaseFS);
     auto InMemoryFS =
         llvm::makeIntrusiveRefCnt<llvm::vfs::InMemoryFileSystem>();
     InMemoryFS->setCurrentWorkingDirectory(WorkingDirectory);
-    InMemoryFS->addFile(*ModuleName, 0, llvm::MemoryBuffer::getMemBuffer(""));
     OverlayFS->pushOverlay(InMemoryFS);
     ModifiedFS = OverlayFS;
+
+    SmallString<128> FakeInputPath;
+    // TODO: We should retry the creation if the path already exists.
+    llvm::sys::fs::createUniquePath(*ModuleName + "-%%%%%%%%.input",
+                                    FakeInputPath,
+                                    /*MakeAbsolute=*/false);
+    InMemoryFS->addFile(FakeInputPath, 0, llvm::MemoryBuffer::getMemBuffer(""));
+
+    ModifiedCommandLine = CommandLine;
+    ModifiedCommandLine->emplace_back(FakeInputPath);
   }
 
   const std::vector<std::string> &FinalCommandLine =
       ModifiedCommandLine ? *ModifiedCommandLine : CommandLine;
+  auto &FinalFS = ModifiedFS ? ModifiedFS : BaseFS;
 
   FileSystemOptions FSOpts;
   FSOpts.WorkingDir = WorkingDirectory.str();
-  auto FileMgr = llvm::makeIntrusiveRefCnt<FileManager>(
-      FSOpts, ModifiedFS ? ModifiedFS : BaseFS);
+  auto FileMgr = llvm::makeIntrusiveRefCnt<FileManager>(FSOpts, FinalFS);
 
-  std::vector<const char *> FinalCCommandLine(CommandLine.size(), nullptr);
-  llvm::transform(CommandLine, FinalCCommandLine.begin(),
+  std::vector<const char *> FinalCCommandLine(FinalCommandLine.size(), nullptr);
+  llvm::transform(FinalCommandLine, FinalCCommandLine.begin(),
                   [](const std::string &Str) { return Str.c_str(); });
 
   auto DiagOpts = CreateAndPopulateDiagOpts(FinalCCommandLine);
   sanitizeDiagOpts(*DiagOpts);
   IntrusiveRefCntPtr<DiagnosticsEngine> Diags =
       CompilerInstance::createDiagnostics(DiagOpts.release(), &DC,
                                           /*ShouldOwnClient=*/false);
 

clang/test/ClangScanDeps/Inputs/modules_cdb_by_mod_name.json

-[
-{
-  "directory": "DIR",
-  "command": "clang -E -IInputs -D INCLUDE_HEADER2 -MD -MF DIR/modules_cdb2.d -fmodules -fcxx-modules -fmodules-cache-path=DIR/module-cache -fimplicit-modules -fimplicit-module-maps -gmodules -x c++",
-  "file": ""
-},
-{
-  "directory": "DIR",
-  "command": "clang -E -IInputs -fmodules -fcxx-modules -fmodules-cache-path=DIR/module-cache -fimplicit-modules -fimplicit-module-maps -x c++",
-  "file": ""
-},
-]

clang/test/ClangScanDeps/Inputs/modules_cdb_clangcl_by_mod_name.json

-[
-{
-  "directory": "DIR",
-  "command": "clang-cl /E /IInputs /D INCLUDE_HEADER2 /clang:-MD /clang:-MF /clang:DIR/modules_cdb2_clangcl.d /clang:-fmodules /clang:-fcxx-modules /clang:-fmodules-cache-path=DIR/module-cache_clangcl /clang:-fimplicit-modules /clang:-fimplicit-module-maps /clang:-x /clang:c++ --",
-  "file": ""
-},
-{
-  "directory": "DIR",
-  "command": "clang-cl /E /IInputs /clang:-fmodules /clang:-fcxx-modules /clang:-fmodules-cache-path=DIR/module-cache_clangcl /clang:-fimplicit-modules /clang:-fimplicit-module-maps /clang:-x /clang:c++ --",
-  "file": ""
-},
-]

clang/test/ClangScanDeps/modules-full-by-mod-name.c

+// UNSUPPORTED: target=powerpc64-ibm-aix{{.*}}
+
+// RUN: rm -rf %t
+// RUN: split-file %s %t
+
+//--- module.modulemap
+module root { header "root.h" }
+module direct { header "direct.h" }
+module transitive { header "transitive.h" }
+//--- root.h
+#include "direct.h"
+#include "root/textual.h"
+//--- direct.h
+#include "transitive.h"
+//--- transitive.h
+// empty
+
+//--- root/textual.h
+// This is here to verify that the "root" directory doesn't clash with name of
+// the "root" module.
+
+//--- cdb.json.template
+[{
+  "file": "",
+  "directory": "DIR",
+  "command": "clang -fmodules -fmodules-cache-path=DIR/cache -I DIR -x c"
+}]
+
+// RUN: sed "s|DIR|%/t|g" %t/cdb.json.template > %t/cdb.json
+// RUN: clang-scan-deps -compilation-database %t/cdb.json -format experimental-full -module-name=root > %t/result.json
+// RUN: cat %t/result.json | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t %s
+
+// CHECK:      {
+// CHECK-NEXT:   "modules": [
+// CHECK-NEXT:     {
+// CHECK-NEXT:       "clang-module-deps": [
+// CHECK-NEXT:         {
+// CHECK-NEXT:           "context-hash": "{{.*}}",
+// CHECK-NEXT:           "module-name": "transitive"
+// CHECK-NEXT:         }
+// CHECK-NEXT:       ],
+// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/module.modulemap",
+// CHECK-NEXT:       "command-line": [
+// CHECK:            ],
+// CHECK-NEXT:       "context-hash": "{{.*}}",
+// CHECK-NEXT:       "file-deps": [
+// CHECK-NEXT:         "[[PREFIX]]/direct.h"
+// CHECK-NEXT:         "[[PREFIX]]/module.modulemap"
+// CHECK-NEXT:       ],
+// CHECK-NEXT:       "name": "direct"
+// CHECK-NEXT:     },
+// CHECK-NEXT:     {
+// CHECK-NEXT:       "clang-module-deps": [
+// CHECK-NEXT:         {
+// CHECK-NEXT:           "context-hash": "{{.*}}",
+// CHECK-NEXT:           "module-name": "direct"
+// CHECK-NEXT:         }
+// CHECK-NEXT:       ],
+// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/module.modulemap",
+// CHECK-NEXT:       "command-line": [
+// CHECK:            ],
+// CHECK-NEXT:       "context-hash": "{{.*}}",
+// CHECK-NEXT:       "file-deps": [
+// CHECK-NEXT:         "[[PREFIX]]/module.modulemap"
+// CHECK-NEXT:         "[[PREFIX]]/root.h"
+// CHECK-NEXT:         "[[PREFIX]]/root/textual.h"
+// CHECK-NEXT:       ],
+// CHECK-NEXT:       "name": "root"
+// CHECK-NEXT:     },
+// CHECK-NEXT:     {
+// CHECK-NEXT:       "clang-module-deps": [],
+// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/module.modulemap",
+// CHECK-NEXT:       "command-line": [
+// CHECK:            ],
+// CHECK-NEXT:       "context-hash": "{{.*}}",
+// CHECK-NEXT:       "file-deps": [
+// CHECK-NEXT:         "[[PREFIX]]/module.modulemap"
+// CHECK-NEXT:         "[[PREFIX]]/transitive.h"
+// CHECK-NEXT:       ],
+// CHECK-NEXT:       "name": "transitive"
+// CHECK-NEXT:     }
+// CHECK-NEXT:   ],
+// CHECK-NEXT:   "translation-units": []
+// CHECK-NEXT: }

clang/test/ClangScanDeps/modules-full-by-mod-name.cpp

-// UNSUPPORTED: target=powerpc64-ibm-aix{{.*}}
-// RUN: rm -rf %t.dir
-// RUN: rm -rf %t.cdb
-// RUN: mkdir -p %t.dir
-// RUN: cp %s %t.dir/modules_cdb_input.cpp
-// RUN: cp %s %t.dir/modules_cdb_input2.cpp
-// RUN: mkdir %t.dir/Inputs
-// RUN: cp %S/Inputs/header.h %t.dir/Inputs/header.h
-// RUN: cp %S/Inputs/header2.h %t.dir/Inputs/header2.h
-// RUN: cp %S/Inputs/module.modulemap %t.dir/Inputs/module.modulemap
-// RUN: sed -e "s|DIR|%/t.dir|g" %S/Inputs/modules_cdb_by_mod_name.json > %t.cdb
-// RUN: sed -e "s|DIR|%/t.dir|g" %S/Inputs/modules_cdb_clangcl_by_mod_name.json > %t_clangcl.cdb
-//
-// RUN: clang-scan-deps -compilation-database %t.cdb -j 1 -format experimental-full \
-// RUN:   -mode preprocess-dependency-directives -module-name=header1 > %t.result
-// RUN: cat %t.result | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t.dir --check-prefixes=CHECK %s
-//
-// RUN: clang-scan-deps -compilation-database %t_clangcl.cdb -j 4 -format experimental-full \
-// RUN:   -mode preprocess-dependency-directives -module-name=header1 > %t_clangcl.result
-// RUN: cat %t_clangcl.result | sed 's:\\\\\?:/:g' | FileCheck -DPREFIX=%/t.dir --check-prefixes=CHECK %s
-
-// CHECK:      {
-// CHECK-NEXT:   "modules": [
-// CHECK-NEXT:     {
-// CHECK-NEXT:       "clang-module-deps": [
-// CHECK-NEXT:         {
-// CHECK-NEXT:           "context-hash": "[[HASH_H2_DINCLUDE:[A-Z0-9]+]]",
-// CHECK-NEXT:           "module-name": "header2"
-// CHECK-NEXT:         }
-// CHECK-NEXT:       ],
-// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/Inputs/module.modulemap",
-// CHECK-NEXT:       "command-line": [
-// CHECK-NEXT:         "-cc1"
-// CHECK:              "-emit-module"
-// CHECK:              "-fmodule-name=header1"
-// CHECK:              "-fno-implicit-modules"
-// CHECK:            ],
-// CHECK-NEXT:       "context-hash": "[[HASH_H1_DINCLUDE:[A-Z0-9]+]]",
-// CHECK-NEXT:       "file-deps": [
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/header.h",
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/module.modulemap"
-// CHECK-NEXT:       ],
-// CHECK-NEXT:       "name": "header1"
-// CHECK-NEXT:     },
-// CHECK-NEXT:     {
-// CHECK-NEXT:       "clang-module-deps": [],
-// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/Inputs/module.modulemap",
-// CHECK-NEXT:       "command-line": [
-// CHECK-NEXT:         "-cc1",
-// CHECK:              "-emit-module",
-// CHECK:              "-fmodule-name=header1",
-// CHECK:              "-fno-implicit-modules",
-// CHECK:            ],
-// CHECK-NEXT:       "context-hash": "[[HASH_H1:[A-Z0-9]+]]",
-// CHECK-NEXT:       "file-deps": [
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/header.h",
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/module.modulemap"
-// CHECK-NEXT:       ],
-// CHECK-NEXT:       "name": "header1"
-// CHECK-NEXT:     },
-// CHECK-NEXT:     {
-// CHECK-NEXT:       "clang-module-deps": [],
-// CHECK-NEXT:       "clang-modulemap-file": "[[PREFIX]]/Inputs/module.modulemap",
-// CHECK-NEXT:       "command-line": [
-// CHECK-NEXT:         "-cc1",
-// CHECK:              "-emit-module",
-// CHECK:              "-fmodule-name=header2",
-// CHECK:              "-fno-implicit-modules",
-// CHECK:            ],
-// CHECK-NEXT:       "context-hash": "[[HASH_H2_DINCLUDE]]",
-// CHECK-NEXT:       "file-deps": [
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/header2.h",
-// CHECK-NEXT:         "[[PREFIX]]/Inputs/module.modulemap"
-// CHECK-NEXT:       ],
-// CHECK-NEXT:       "name": "header2"
-// CHECK-NEXT:     }
-// CHECK-NEXT:   ],
-// CHECK-NEXT:   "translation-units": []
-// CHECK-NEXT: }