[clang] Add LangOpts.ScanDeps to fix #88896

Author: tsfn

clang/include/clang/Basic/LangOptions.def

@@ -438,6 +438,8 @@ LANGOPT(CXXAssumptions, 1, 1, NotCompatible, "Enable or disable codegen and comp
 
 LANGOPT(RawStringLiterals, 1, 1, NotCompatible, "Enable or disable raw string literals")
 
+LANGOPT(ScanDeps, 1, 0, NotCompatible, "True if we are scanning by DependencyDirectivesScanner")
+
 ENUM_LANGOPT(StrictFlexArraysLevel, StrictFlexArraysLevelKind, 2,
              StrictFlexArraysLevelKind::Default, NotCompatible,
              "Rely on strict definition of flexible arrays")

clang/lib/Lex/DependencyDirectivesScanner.cpp

@@ -74,6 +74,7 @@ struct Scanner {
     LangOpts.ObjC = true;
     LangOpts.LineComment = true;
     LangOpts.RawStringLiterals = true;
+    LangOpts.ScanDeps = true;
     // FIXME: we do not enable C11 or C++11, so we are missing u/u8/U"".
     return LangOpts;
   }

clang/lib/Lex/Lexer.cpp

@@ -2087,7 +2087,7 @@ bool Lexer::LexNumericConstant(Token &Result, const char *CurPtr) {
   }
 
   // If we have a digit separator, continue.
-  if (C == '\'' && (LangOpts.CPlusPlus14 || LangOpts.C23)) {
+  if (C == '\'' && (LangOpts.CPlusPlus14 || LangOpts.C23 || LangOpts.ScanDeps)) {
     auto [Next, NextSize] = getCharAndSizeNoWarn(CurPtr + Size, LangOpts);
     if (isAsciiIdentifierContinue(Next)) {
       if (!isLexingRawMode())

clang/unittests/Lex/DependencyDirectivesScannerTest.cpp

@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "clang/Lex/DependencyDirectivesScanner.h"
+#include "clang/Basic/TokenKinds.h"
 #include "llvm/ADT/SmallString.h"
 #include "gtest/gtest.h"
 
@@ -1000,6 +1001,21 @@ int z = 128'78;
   EXPECT_STREQ("#include <test.h>\n", Out.data());
 }
 
+TEST(MinimizeSourceToDependencyDirectivesTest, CharacterLiteralInPreprocessor) {
+  SmallVector<char, 128> Out;
+  SmallVector<dependency_directives_scan::Token, 8> Tokens;
+  SmallVector<Directive, 4> Directives;
+
+  StringRef Source = R"(
+    #if 1'2 == 12
+    #endif
+    )";
+  ASSERT_FALSE(minimizeSourceToDependencyDirectives(Source, Out, Tokens, Directives));
+  ASSERT_GE(Tokens.size(), 4u);
+  EXPECT_EQ(Tokens[2].Kind, tok::numeric_constant);
+  EXPECT_EQ(Tokens[3].Kind, tok::equalequal);
+}
+
 TEST(MinimizeSourceToDependencyDirectivesTest, PragmaOnce) {
   SmallVector<char, 128> Out;
   SmallVector<dependency_directives_scan::Token, 4> Tokens;