[Serialized diagnostics] Record generated buffer contents in diagnostics files

clang/include/clang-c/CXDiagnostic.h

@@ -148,6 +148,30 @@ CINDEX_LINKAGE void clang_disposeDiagnosticSet(CXDiagnosticSet Diags);
  */
 CINDEX_LINKAGE CXDiagnosticSet clang_getChildDiagnostics(CXDiagnostic D);
 
+/**
+ * Get the contents if the given file that was provided via diagnostics.
+ *
+ * \param diags the diagnostics set to query for the contents of the file.
+ * \param file the file to get the contents of.
+ * \param outFileSize if non-null, set to the file size on success.
+ * \returns on success, a pointer to the file contents. Otherwise, NULL.
+ */
+CINDEX_LINKAGE const char *clang_getDiagnosticFileContents(
+    CXDiagnosticSet diags, CXFile file, size_t *outFileSize);
+
+/**
+ * Retrieve the original source range if the given file was provided via
+ * diagnostics and is conceptually a replacement for the original source range.
+ *
+ * \param diags the diagnostics set to query for the contents of the file.
+ * \param file the file to get the contents of.
+ * \returns on success, the source range (into another file) that is
+ * conceptually replaced by the contents of the given file (available via
+ * \c clang_getDiagnosticFileContents).
+ */
+CINDEX_LINKAGE CXSourceRange clang_getDiagnosticFileOriginalSourceRange(
+    CXDiagnosticSet diags, CXFile file);
+
 /**
  * Destroy a diagnostic.
  */

clang/include/clang-c/Index.h

@@ -34,7 +34,7 @@
  * compatible, thus CINDEX_VERSION_MAJOR is expected to remain stable.
  */
 #define CINDEX_VERSION_MAJOR 0
-#define CINDEX_VERSION_MINOR 62
+#define CINDEX_VERSION_MINOR 63
 
 #define CINDEX_VERSION_ENCODE(major, minor) (((major)*10000) + ((minor)*1))
 

clang/include/clang/Frontend/SerializedDiagnosticReader.h

@@ -109,6 +109,16 @@ class SerializedDiagnosticReader {
     return {};
   }
 
+  /// Visit file contents. This associates the file's \c ID with the
+  /// contents of
+  virtual std::error_code visitSourceFileContentsRecord(
+      unsigned ID,
+      const Location &OriginalStartLoc,
+      const Location &OriginalEndLoc,
+      StringRef Contents) {
+    return {};
+  }
+
   /// Visit a fixit hint.
   virtual std::error_code
   visitFixitRecord(const Location &Start, const Location &End, StringRef Text) {

clang/include/clang/Frontend/SerializedDiagnostics.h

@@ -32,8 +32,9 @@ enum RecordIDs {
   RECORD_CATEGORY,
   RECORD_FILENAME,
   RECORD_FIXIT,
+  RECORD_SOURCE_FILE_CONTENTS,
   RECORD_FIRST = RECORD_VERSION,
-  RECORD_LAST = RECORD_FIXIT
+  RECORD_LAST = RECORD_SOURCE_FILE_CONTENTS
 };
 
 /// A stable version of DiagnosticIDs::Level.

clang/lib/Frontend/SerializedDiagnosticPrinter.cpp

@@ -93,6 +93,7 @@ class SDiagsMerger : SerializedDiagnosticReader {
   AbbrevLookup FileLookup;
   AbbrevLookup CategoryLookup;
   AbbrevLookup DiagFlagLookup;
+  llvm::DenseSet<unsigned> ContentsWritten;
 
 public:
   SDiagsMerger(SDiagsWriter &Writer) : Writer(Writer) {}
@@ -112,6 +113,12 @@ class SDiagsMerger : SerializedDiagnosticReader {
   std::error_code visitFilenameRecord(unsigned ID, unsigned Size,
                                       unsigned Timestamp,
                                       StringRef Name) override;
+  std::error_code visitSourceFileContentsRecord(
+      unsigned ID,
+      const Location &OriginalStartLoc,
+      const Location &OriginalEndLoc,
+      StringRef Contents) override;
+
   std::error_code visitFixitRecord(const serialized_diags::Location &Start,
                                    const serialized_diags::Location &End,
                                    StringRef CodeToInsert) override;
@@ -462,6 +469,8 @@ void SDiagsWriter::EmitBlockInfoBlock() {
   EmitRecordID(RECORD_DIAG_FLAG, "DiagFlag", Stream, Record);
   EmitRecordID(RECORD_FILENAME, "FileName", Stream, Record);
   EmitRecordID(RECORD_FIXIT, "FixIt", Stream, Record);
+  EmitRecordID(
+      RECORD_SOURCE_FILE_CONTENTS, "SourceFileContents", Stream, Record);
 
   // Emit abbreviation for RECORD_DIAG.
   Abbrev = std::make_shared<BitCodeAbbrev>();
@@ -518,6 +527,16 @@ void SDiagsWriter::EmitBlockInfoBlock() {
   Abbrevs.set(RECORD_FIXIT, Stream.EmitBlockInfoAbbrev(BLOCK_DIAG,
                                                        Abbrev));
 
+  // Emit the abbreviation for RECORD_SOURCE_FILE_CONTENTS.
+  Abbrev = std::make_shared<BitCodeAbbrev>();
+  Abbrev->Add(BitCodeAbbrevOp(RECORD_SOURCE_FILE_CONTENTS));
+  Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 10)); // File ID.
+  AddRangeLocationAbbrev(*Abbrev);
+  Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 16)); // File size.
+  Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Blob)); // File contents.
+  Abbrevs.set(RECORD_SOURCE_FILE_CONTENTS,
+              Stream.EmitBlockInfoAbbrev(BLOCK_DIAG, Abbrev));
+
   Stream.ExitBlock();
 }
 
@@ -890,6 +909,28 @@ std::error_code SDiagsMerger::visitFilenameRecord(unsigned ID, unsigned Size,
   return std::error_code();
 }
 
+std::error_code SDiagsMerger::visitSourceFileContentsRecord(
+    unsigned ID,
+    const Location &OriginalStartLoc,
+    const Location &OriginalEndLoc,
+    StringRef Contents) {
+  unsigned MappedID = FileLookup[ID];
+  if (!ContentsWritten.insert(MappedID).second)
+    return std::error_code();
+
+  RecordData::value_type Record[] = {
+      RECORD_SOURCE_FILE_CONTENTS, MappedID,
+      FileLookup[OriginalStartLoc.FileID],
+      OriginalStartLoc.Line, OriginalStartLoc.Col, OriginalStartLoc.Offset,
+      FileLookup[OriginalEndLoc.FileID], OriginalEndLoc.Line,
+      OriginalEndLoc.Col, OriginalEndLoc.Offset,
+    Contents.size()};
+
+  Writer.State->Stream.EmitRecordWithBlob(
+      Writer.State->Abbrevs.get(RECORD_SOURCE_FILE_CONTENTS), Record, Contents);
+  return std::error_code();
+}
+
 std::error_code SDiagsMerger::visitCategoryRecord(unsigned ID, StringRef Name) {
   CategoryLookup[ID] = Writer.getEmitCategory(ID);
   return std::error_code();

clang/lib/Frontend/SerializedDiagnosticReader.cpp

@@ -302,6 +302,16 @@ SerializedDiagnosticReader::readDiagnosticBlock(llvm::BitstreamCursor &Stream) {
                Location(Record[4], Record[5], Record[6], Record[7]), Blob)))
         return EC;
       continue;
+    case RECORD_SOURCE_FILE_CONTENTS:
+      if (Record.size() != 10 || Record[9] != Blob.size())
+        return SDError::MalformedDiagnosticRecord;
+      if ((EC = visitSourceFileContentsRecord(
+               Record[0],
+               Location(Record[1], Record[2], Record[3], Record[4]),
+               Location(Record[5], Record[6], Record[7], Record[8]),
+               Blob)))
+        return EC;
+      continue;
     case RECORD_SOURCE_RANGE:
       // A source range is two locations (4 each).
       if (Record.size() != 8)

clang/tools/c-index-test/c-index-test.c

@@ -4716,7 +4716,8 @@ static void printFixIts(CXDiagnostic D, unsigned indent) {
   }  
 }
 
-static void printDiagnosticSet(CXDiagnosticSet Diags, unsigned indent) {
+static void printDiagnosticSet(
+    CXDiagnosticSet Diags, unsigned indent, CXDiagnosticSet TopDiags) {
   unsigned i, n;
 
   if (!Diags)
@@ -4730,7 +4731,8 @@ static void printDiagnosticSet(CXDiagnosticSet Diags, unsigned indent) {
     CXString FileName, DiagSpelling, DiagOption, DiagCat;
     unsigned line, column, offset;
     const char *FileNameStr = 0, *DiagOptionStr = 0, *DiagCatStr = 0;
-
+    const char *FileContents = 0;
+
     D = clang_getDiagnosticInSet(Diags, i);
     DiagLoc = clang_getDiagnosticLocation(D);
     clang_getExpansionLocation(DiagLoc, &File, &line, &column, &offset);
@@ -4763,15 +4765,37 @@ static void printDiagnosticSet(CXDiagnosticSet Diags, unsigned indent) {
 
     printRanges(D, indent);
     printFixIts(D, indent);
-
+
+    // If we have the source file contents for this file, print them now.
+    FileContents = clang_getDiagnosticFileContents(TopDiags, File, 0);
+    if (FileContents) {
+      CXSourceRange OriginalSourceRange;
+
+      fprintf(stderr, "CONTENTS OF FILE %s:\n",
+              FileNameStr ? FileNameStr : "(null)");
+
+      OriginalSourceRange = clang_getDiagnosticFileOriginalSourceRange(
+          TopDiags, File);
+      if (!clang_equalRanges(clang_getNullRange(), OriginalSourceRange)) {
+        printIndent(indent);
+        fprintf(stderr, "Original source range: ");
+        printLocation(clang_getRangeStart(OriginalSourceRange));
+        fprintf(stderr, " - ");
+        printLocation(clang_getRangeEnd(OriginalSourceRange));
+        fprintf(stderr, "\n");
+      }
+
+      fprintf(stderr, "%s\nEND CONTENTS OF FILE\n", FileContents);
+    }
+
     /* Print subdiagnostics. */
-    printDiagnosticSet(clang_getChildDiagnostics(D), indent+2);
+    printDiagnosticSet(clang_getChildDiagnostics(D), indent+2, TopDiags);
 
     clang_disposeString(FileName);
     clang_disposeString(DiagSpelling);
     clang_disposeString(DiagOption);
     clang_disposeString(DiagCat);
-  }  
+  }
 }
 
 static int read_diagnostics(const char *filename) {
@@ -4788,7 +4812,7 @@ static int read_diagnostics(const char *filename) {
     return 1;
   }
 
-  printDiagnosticSet(Diags, 0);
+  printDiagnosticSet(Diags, 0, Diags);
   fprintf(stderr, "Number of diagnostics: %d\n",
           clang_getNumDiagnosticsInSet(Diags));
   clang_disposeDiagnosticSet(Diags);

clang/tools/libclang/CIndexDiagnostic.cpp

@@ -34,6 +34,11 @@ CXDiagnosticSetImpl::appendDiagnostic(std::unique_ptr<CXDiagnosticImpl> D) {
   Diagnostics.push_back(std::move(D));
 }
 
+void CXDiagnosticSetImpl::recordSourceFileContents(
+    CXFile file, StringRef contents, CXSourceRange originalSourceRange) {
+  FileContents[file] = CXSourceFileContents{contents, originalSourceRange};
+}
+
 CXDiagnosticImpl::~CXDiagnosticImpl() {}
 
 namespace {
@@ -479,6 +484,30 @@ CXDiagnosticSet clang_getChildDiagnostics(CXDiagnostic Diag) {
   return nullptr;
 }
 
+const char *clang_getDiagnosticFileContents(
+    CXDiagnosticSet diags, CXFile file, size_t *outFileSize) {
+  if (CXDiagnosticSetImpl *D = static_cast<CXDiagnosticSetImpl *>(diags)) {
+    CXSourceRange originalSourceRange;
+    if (auto contents = D->getSourceFileContents(file, originalSourceRange)) {
+      if (outFileSize)
+        *outFileSize = contents->size();
+      return contents->data();
+    }
+  }
+  return nullptr;
+}
+
+CXSourceRange clang_getDiagnosticFileOriginalSourceRange(
+    CXDiagnosticSet diags, CXFile file) {
+  if (CXDiagnosticSetImpl *D = static_cast<CXDiagnosticSetImpl *>(diags)) {
+    CXSourceRange originalSourceRange;
+    if (auto contents = D->getSourceFileContents(file, originalSourceRange)) {
+      return originalSourceRange;
+    }
+  }
+  return clang_getNullRange();
+}
+
 unsigned clang_getNumDiagnosticsInSet(CXDiagnosticSet Diags) {
   if (CXDiagnosticSetImpl *D = static_cast<CXDiagnosticSetImpl*>(Diags))
     return D->getNumDiagnostics();

clang/tools/libclang/CIndexDiagnostic.h

@@ -15,6 +15,9 @@
 
 #include "clang-c/Index.h"
 #include "clang/Basic/LLVM.h"
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/Optional.h"
+#include "llvm/ADT/StringRef.h"
 #include <memory>
 #include <vector>
 #include <assert.h>
@@ -24,9 +27,15 @@ namespace clang {
 class LangOptions;
 class StoredDiagnostic;
 class CXDiagnosticImpl;
-  
+
 class CXDiagnosticSetImpl {
+  struct CXSourceFileContents {
+    StringRef Contents;
+    CXSourceRange OriginalSourceRange;
+  };
+
   std::vector<std::unique_ptr<CXDiagnosticImpl>> Diagnostics;
+  llvm::DenseMap<CXFile, CXSourceFileContents> FileContents;
   const bool IsExternallyManaged;
 public:
   CXDiagnosticSetImpl(bool isManaged = false)
@@ -45,6 +54,19 @@ class CXDiagnosticSetImpl {
 
   void appendDiagnostic(std::unique_ptr<CXDiagnosticImpl> D);
 
+  void recordSourceFileContents(
+      CXFile file, StringRef contents, CXSourceRange originalSourceRange);
+
+  Optional<StringRef> getSourceFileContents(
+      CXFile file, CXSourceRange &originalSourceRange) {
+    auto found = FileContents.find(file);
+    if (found == FileContents.end())
+      return None;
+
+    originalSourceRange = found->second.OriginalSourceRange;
+    return found->second.Contents;
+  }
+
   bool empty() const {
     return Diagnostics.empty();
   }

clang/tools/libclang/CXLoadedDiagnostic.cpp

@@ -225,6 +225,12 @@ class DiagLoader : serialized_diags::SerializedDiagnosticReader {
                                       unsigned Timestamp,
                                       StringRef Name) override;
 
+  std::error_code visitSourceFileContentsRecord(
+      unsigned ID,
+      const serialized_diags::Location &OriginalStartLoc,
+      const serialized_diags::Location &OriginalEndLoc,
+      StringRef Contents) override;
+
   std::error_code visitFixitRecord(const serialized_diags::Location &Start,
                                    const serialized_diags::Location &End,
                                    StringRef CodeToInsert) override;
@@ -347,6 +353,28 @@ std::error_code DiagLoader::visitFilenameRecord(unsigned ID, unsigned Size,
   return std::error_code();
 }
 
+std::error_code DiagLoader::visitSourceFileContentsRecord(
+    unsigned ID,
+    const serialized_diags::Location &OriginalStartLoc,
+    const serialized_diags::Location &OriginalEndLoc,
+    StringRef Contents
+) {
+  CXSourceRange OriginalSourceRange;
+  if (std::error_code EC = readRange(
+          OriginalStartLoc, OriginalEndLoc, OriginalSourceRange))
+    return EC;
+
+  auto file = const_cast<FileEntry *>(TopDiags->Files[ID]);
+  if (!file)
+    return reportInvalidFile("Source file contents for unknown file ID");
+
+  StringRef CopiedContents(TopDiags->copyString(Contents),
+                           Contents.size());
+
+  TopDiags->recordSourceFileContents(file, CopiedContents, OriginalSourceRange);
+  return std::error_code();
+}
+
 std::error_code
 DiagLoader::visitSourceRangeRecord(const serialized_diags::Location &Start,
                                    const serialized_diags::Location &End) {

clang/tools/libclang/libclang.map

@@ -484,6 +484,8 @@ LLVM_16 {
     clang_experimental_DependencyScannerServiceOptions_setObjectStore;
     clang_experimental_DependencyScannerWorker_getFileDependencies_v5;
     clang_createAPISet;
+    clang_getDiagnosticFileContents;
+    clang_getDiagnosticFileOriginalSourceRange;
     clang_disposeAPISet;
     clang_getSymbolGraphForCursor;
     clang_getSymbolGraphForUSR;