[clang] Unify SourceLocation and IdentifierInfo* pair-like data structures to IdentifierLoc (llvm#135808)

I found this issue when I working on
llvm#107168.

Currently we have many similiar data structures like:
 - `std::pair<IdentifierInfo *, SourceLocation>`.
 - Element type of `ModuleIdPath`.
 - `IdentifierLocPair`.
 - `IdentifierLoc`.
 
This PR unify these data structures to `IdentifierLoc`, moved
`IdentifierLoc` definition to SourceLocation.h, and deleted other
similer data structures.

---------

Signed-off-by: yronglin <[email protected]>

Author: yronglin

clang-tools-extra/pp-trace/PPCallbacksTracker.cpp

@@ -547,8 +547,8 @@ void PPCallbacksTracker::appendArgument(const char *Name, ModuleIdPath Value) {
     if (I)
       SS << ", ";
     SS << "{"
-       << "Name: " << Value[I].first->getName() << ", "
-       << "Loc: " << getSourceLocationString(PP, Value[I].second) << "}";
+       << "Name: " << Value[I].getIdentifierInfo()->getName() << ", "
+       << "Loc: " << getSourceLocationString(PP, Value[I].getLoc()) << "}";
   }
   SS << "]";
   appendArgument(Name, SS.str());

clang/include/clang/AST/OpenACCClause.h

@@ -258,7 +258,7 @@ inline bool operator!=(const OpenACCBindClause &LHS,
   return !(LHS == RHS);
 }
 
-using DeviceTypeArgument = std::pair<IdentifierInfo *, SourceLocation>;
+using DeviceTypeArgument = IdentifierLoc;
 /// A 'device_type' or 'dtype' clause, takes a list of either an 'asterisk' or
 /// an identifier. The 'asterisk' means 'the rest'.
 class OpenACCDeviceTypeClause final
@@ -280,16 +280,16 @@ class OpenACCDeviceTypeClause final
         "Invalid clause kind for device-type");
 
     assert(!llvm::any_of(Archs, [](const DeviceTypeArgument &Arg) {
-      return Arg.second.isInvalid();
+      return Arg.getLoc().isInvalid();
     }) && "Invalid SourceLocation for an argument");
 
-    assert(
-        (Archs.size() == 1 || !llvm::any_of(Archs,
-                                            [](const DeviceTypeArgument &Arg) {
-                                              return Arg.first == nullptr;
-                                            })) &&
-        "Only a single asterisk version is permitted, and must be the "
-        "only one");
+    assert((Archs.size() == 1 ||
+            !llvm::any_of(Archs,
+                          [](const DeviceTypeArgument &Arg) {
+                            return Arg.getIdentifierInfo() == nullptr;
+                          })) &&
+           "Only a single asterisk version is permitted, and must be the "
+           "only one");
 
     std::uninitialized_copy(Archs.begin(), Archs.end(),
                             getTrailingObjects<DeviceTypeArgument>());
@@ -302,7 +302,7 @@ class OpenACCDeviceTypeClause final
   }
   bool hasAsterisk() const {
     return getArchitectures().size() > 0 &&
-           getArchitectures()[0].first == nullptr;
+           getArchitectures()[0].getIdentifierInfo() == nullptr;
   }
 
   ArrayRef<DeviceTypeArgument> getArchitectures() const {

clang/include/clang/Basic/IdentifierTable.h

@@ -18,6 +18,7 @@
 #include "clang/Basic/Builtins.h"
 #include "clang/Basic/DiagnosticIDs.h"
 #include "clang/Basic/LLVM.h"
+#include "clang/Basic/SourceLocation.h"
 #include "clang/Basic/TokenKinds.h"
 #include "llvm/ADT/DenseMapInfo.h"
 #include "llvm/ADT/FoldingSet.h"
@@ -76,9 +77,6 @@ inline bool isReservedInAllContexts(ReservedIdentifierStatus Status) {
          Status != ReservedIdentifierStatus::StartsWithUnderscoreAndIsExternC;
 }
 
-/// A simple pair of identifier info and location.
-using IdentifierLocPair = std::pair<IdentifierInfo *, SourceLocation>;
-
 /// IdentifierInfo and other related classes are aligned to
 /// 8 bytes so that DeclarationName can use the lower 3 bits
 /// of a pointer to one of these classes.
@@ -1165,6 +1163,28 @@ class SelectorTable {
   static std::string getPropertyNameFromSetterSelector(Selector Sel);
 };
 
+/// A simple pair of identifier info and location.
+class IdentifierLoc {
+  SourceLocation Loc;
+  IdentifierInfo *II = nullptr;
+
+public:
+  IdentifierLoc() = default;
+  IdentifierLoc(SourceLocation L, IdentifierInfo *Ident) : Loc(L), II(Ident) {}
+
+  void setLoc(SourceLocation L) { Loc = L; }
+  void setIdentifierInfo(IdentifierInfo *Ident) { II = Ident; }
+  SourceLocation getLoc() const { return Loc; }
+  IdentifierInfo *getIdentifierInfo() const { return II; }
+
+  bool operator==(const IdentifierLoc &X) const {
+    return Loc == X.Loc && II == X.II;
+  }
+
+  bool operator!=(const IdentifierLoc &X) const {
+    return Loc != X.Loc || II != X.II;
+  }
+};
 }  // namespace clang
 
 namespace llvm {

clang/include/clang/Lex/ModuleLoader.h

@@ -14,6 +14,7 @@
 #ifndef LLVM_CLANG_LEX_MODULELOADER_H
 #define LLVM_CLANG_LEX_MODULELOADER_H
 
+#include "clang/Basic/IdentifierTable.h"
 #include "clang/Basic/LLVM.h"
 #include "clang/Basic/Module.h"
 #include "clang/Basic/SourceLocation.h"
@@ -29,7 +30,7 @@ class IdentifierInfo;
 
 /// A sequence of identifier/location pairs used to describe a particular
 /// module or submodule, e.g., std.vector.
-using ModuleIdPath = ArrayRef<std::pair<IdentifierInfo *, SourceLocation>>;
+using ModuleIdPath = ArrayRef<IdentifierLoc>;
 
 /// Describes the result of attempting to load a module.
 class ModuleLoadResult {

clang/include/clang/Lex/PPCallbacks.h

@@ -15,6 +15,7 @@
 #define LLVM_CLANG_LEX_PPCALLBACKS_H
 
 #include "clang/Basic/DiagnosticIDs.h"
+#include "clang/Basic/IdentifierTable.h"
 #include "clang/Basic/SourceLocation.h"
 #include "clang/Basic/SourceManager.h"
 #include "clang/Lex/ModuleLoader.h"

clang/include/clang/Lex/Preprocessor.h

@@ -327,7 +327,7 @@ class Preprocessor {
   SourceLocation ModuleImportLoc;
 
   /// The import path for named module that we're currently processing.
-  SmallVector<std::pair<IdentifierInfo *, SourceLocation>, 2> NamedModuleImportPath;
+  SmallVector<IdentifierLoc, 2> NamedModuleImportPath;
 
   llvm::DenseMap<FileID, SmallVector<const char *>> CheckPoints;
   unsigned CheckPointCounter = 0;
@@ -622,7 +622,7 @@ class Preprocessor {
 
   /// The identifier and source location of the currently-active
   /// \#pragma clang arc_cf_code_audited begin.
-  std::pair<IdentifierInfo *, SourceLocation> PragmaARCCFCodeAuditedInfo;
+  IdentifierLoc PragmaARCCFCodeAuditedInfo;
 
   /// The source location of the currently-active
   /// \#pragma clang assume_nonnull begin.
@@ -1998,16 +1998,15 @@ class Preprocessor {
   /// arc_cf_code_audited begin.
   ///
   /// Returns an invalid location if there is no such pragma active.
-  std::pair<IdentifierInfo *, SourceLocation>
-  getPragmaARCCFCodeAuditedInfo() const {
+  IdentifierLoc getPragmaARCCFCodeAuditedInfo() const {
     return PragmaARCCFCodeAuditedInfo;
   }
 
   /// Set the location of the currently-active \#pragma clang
   /// arc_cf_code_audited begin.  An invalid location ends the pragma.
   void setPragmaARCCFCodeAuditedInfo(IdentifierInfo *Ident,
                                      SourceLocation Loc) {
-    PragmaARCCFCodeAuditedInfo = {Ident, Loc};
+    PragmaARCCFCodeAuditedInfo = IdentifierLoc(Loc, Ident);
   }
 
   /// The location of the currently-active \#pragma clang

clang/include/clang/Parse/LoopHint.h

@@ -9,12 +9,12 @@
 #ifndef LLVM_CLANG_PARSE_LOOPHINT_H
 #define LLVM_CLANG_PARSE_LOOPHINT_H
 
+#include "clang/Basic/IdentifierTable.h"
 #include "clang/Basic/SourceLocation.h"
 
 namespace clang {
 
 class Expr;
-struct IdentifierLoc;
 
 /// Loop optimization hint for loop and unroll pragmas.
 struct LoopHint {

clang/include/clang/Parse/Parser.h

@@ -1725,8 +1725,8 @@ class Parser : public CodeCompletionHandler {
   ObjCTypeParamList *parseObjCTypeParamList();
   ObjCTypeParamList *parseObjCTypeParamListOrProtocolRefs(
       ObjCTypeParamListScope &Scope, SourceLocation &lAngleLoc,
-      SmallVectorImpl<IdentifierLocPair> &protocolIdents,
-      SourceLocation &rAngleLoc, bool mayBeProtocolList = true);
+      SmallVectorImpl<IdentifierLoc> &protocolIdents, SourceLocation &rAngleLoc,
+      bool mayBeProtocolList = true);
 
   void HelperActionsForIvarDeclarations(ObjCContainerDecl *interfaceDecl,
                                         SourceLocation atLoc,
@@ -3818,8 +3818,7 @@ class Parser : public CodeCompletionHandler {
                                SourceLocation Loc,
                                llvm::SmallVectorImpl<Expr *> &IntExprs);
   /// Parses the 'device-type-list', which is a list of identifiers.
-  bool ParseOpenACCDeviceTypeList(
-      llvm::SmallVector<std::pair<IdentifierInfo *, SourceLocation>> &Archs);
+  bool ParseOpenACCDeviceTypeList(llvm::SmallVector<IdentifierLoc> &Archs);
   /// Parses the 'async-argument', which is an integral value with two
   /// 'special' values that are likely negative (but come from Macros).
   OpenACCIntExprParseResult ParseOpenACCAsyncArgument(OpenACCDirectiveKind DK,
@@ -3951,10 +3950,8 @@ class Parser : public CodeCompletionHandler {
     return false;
   }
 
-  bool ParseModuleName(
-      SourceLocation UseLoc,
-      SmallVectorImpl<std::pair<IdentifierInfo *, SourceLocation>> &Path,
-      bool IsImport);
+  bool ParseModuleName(SourceLocation UseLoc,
+                       SmallVectorImpl<IdentifierLoc> &Path, bool IsImport);
 
   //===--------------------------------------------------------------------===//
   // C++11/G++: Type Traits [Type-Traits.html in the GCC manual]

clang/include/clang/Sema/ParsedAttr.h

@@ -40,7 +40,6 @@ class LangOptions;
 class Sema;
 class Stmt;
 class TargetInfo;
-struct IdentifierLoc;
 
 /// Represents information about a change in availability for
 /// an entity, which is part of the encoding of the 'availability'
@@ -99,15 +98,6 @@ struct PropertyData {
 
 } // namespace detail
 
-/// Wraps an identifier and optional source location for the identifier.
-struct IdentifierLoc {
-  SourceLocation Loc;
-  IdentifierInfo *Ident;
-
-  static IdentifierLoc *create(ASTContext &Ctx, SourceLocation Loc,
-                               IdentifierInfo *Ident);
-};
-
 /// A union of the various pointer types that can be passed to an
 /// ParsedAttr as an argument.
 using ArgsUnion = llvm::PointerUnion<Expr *, IdentifierLoc *>;

clang/include/clang/Sema/Sema.h

@@ -143,7 +143,7 @@ enum class LangAS : unsigned int;
 class LocalInstantiationScope;
 class LookupResult;
 class MangleNumberingContext;
-typedef ArrayRef<std::pair<IdentifierInfo *, SourceLocation>> ModuleIdPath;
+typedef ArrayRef<IdentifierLoc> ModuleIdPath;
 class ModuleLoader;
 class MultiLevelTemplateArgumentList;
 struct NormalizedConstraint;