Googletest export

Print std::u8string, std::u16string, and std::u32string as string literals

Previously, these types were printed as "{ U+123, U+456, U+789 }". However,
printed output in that form is difficult to compare against any literals that
might be defined in code. Instead, just treat these types like std::string
and std::wstring, escaping non-ASCII characters with a hexadecimal escape
sequence.

The tests have also been updated to cover the new functionality: as a bonus,
the tests now also pass with the MSVC toolchain.

Internally, the code has been reorganized to primarily operate in terms of
char32_t, under the assumption that char32_t will always be at least as big
as wchar_t. While that assumption is currently true, perhaps it won't be in
the future...

PiperOrigin-RevId: 364033132

Author: Abseil Team

BUILD.bazel

@@ -43,6 +43,14 @@ config_setting(
     constraint_values = ["@platforms//os:windows"],
 )
 
+config_setting(
+    name = "msvc_compiler",
+    flag_values = {
+        "@bazel_tools//tools/cpp:compiler": "msvc-cl",
+    },
+    visibility = [":__subpackages__"],
+)
+
 config_setting(
     name = "has_absl",
     values = {"define": "absl=1"},

googletest/cmake/internal_utils.cmake

@@ -81,6 +81,8 @@ macro(config_compiler_and_linker)
     # Suppress "unreachable code" warning
     # http://stackoverflow.com/questions/3232669 explains the issue.
     set(cxx_base_flags "${cxx_base_flags} -wd4702")
+    # Ensure MSVC treats source files as UTF-8 encoded.
+    set(cxx_base_flags "${cxx_base_flags} -utf-8")
   elseif (CMAKE_CXX_COMPILER_ID STREQUAL "Clang")
     set(cxx_base_flags "-Wall -Wshadow -Werror -Wconversion")
     set(cxx_exception_flags "-fexceptions")

googletest/include/gtest/gtest-printers.h

@@ -505,24 +505,21 @@ inline void PrintTo(unsigned char* s, ::std::ostream* os) {
   PrintTo(ImplicitCast_<const void*>(s), os);
 }
 #ifdef __cpp_char8_t
-inline void PrintTo(const char8_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
-}
+// Overloads for u8 strings.
+void PrintTo(const char8_t* s, ::std::ostream* os);
 inline void PrintTo(char8_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
+  PrintTo(ImplicitCast_<const char8_t*>(s), os);
 }
 #endif
-inline void PrintTo(const char16_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
-}
+// Overloads for u16 strings.
+void PrintTo(const char16_t* s, ::std::ostream* os);
 inline void PrintTo(char16_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
-}
-inline void PrintTo(const char32_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
+  PrintTo(ImplicitCast_<const char16_t*>(s), os);
 }
+// Overloads for u32 strings.
+void PrintTo(const char32_t* s, ::std::ostream* os);
 inline void PrintTo(char32_t* s, ::std::ostream* os) {
-  PrintTo(ImplicitCast_<const void*>(s), os);
+  PrintTo(ImplicitCast_<const char32_t*>(s), os);
 }
 
 // MSVC can be configured to define wchar_t as a typedef of unsigned
@@ -558,6 +555,26 @@ inline void PrintTo(const ::std::string& s, ::std::ostream* os) {
   PrintStringTo(s, os);
 }
 
+// Overloads for ::std::u8string
+#ifdef __cpp_char8_t
+GTEST_API_ void PrintU8StringTo(const ::std::u8string& s, ::std::ostream* os);
+inline void PrintTo(const ::std::u8string& s, ::std::ostream* os) {
+  PrintU8StringTo(s, os);
+}
+#endif
+
+// Overloads for ::std::u16string
+GTEST_API_ void PrintU16StringTo(const ::std::u16string& s, ::std::ostream* os);
+inline void PrintTo(const ::std::u16string& s, ::std::ostream* os) {
+  PrintU16StringTo(s, os);
+}
+
+// Overloads for ::std::u32string
+GTEST_API_ void PrintU32StringTo(const ::std::u32string& s, ::std::ostream* os);
+inline void PrintTo(const ::std::u32string& s, ::std::ostream* os) {
+  PrintU32StringTo(s, os);
+}
+
 // Overloads for ::std::wstring.
 #if GTEST_HAS_STD_WSTRING
 GTEST_API_ void PrintWideStringTo(const ::std::wstring&s, ::std::ostream* os);
@@ -805,6 +822,20 @@ void UniversalPrintArray(const T* begin, size_t len, ::std::ostream* os) {
 GTEST_API_ void UniversalPrintArray(
     const char* begin, size_t len, ::std::ostream* os);
 
+#ifdef __cpp_char8_t
+// This overload prints a (const) char8_t array compactly.
+GTEST_API_ void UniversalPrintArray(const char8_t* begin, size_t len,
+                                    ::std::ostream* os);
+#endif
+
+// This overload prints a (const) char16_t array compactly.
+GTEST_API_ void UniversalPrintArray(const char16_t* begin, size_t len,
+                                    ::std::ostream* os);
+
+// This overload prints a (const) char32_t array compactly.
+GTEST_API_ void UniversalPrintArray(const char32_t* begin, size_t len,
+                                    ::std::ostream* os);
+
 // This overload prints a (const) wchar_t array compactly.
 GTEST_API_ void UniversalPrintArray(
     const wchar_t* begin, size_t len, ::std::ostream* os);
@@ -877,12 +908,55 @@ class UniversalTersePrinter<const char*> {
   }
 };
 template <>
-class UniversalTersePrinter<char*> {
+class UniversalTersePrinter<char*> : public UniversalTersePrinter<const char*> {
+};
+
+#ifdef __cpp_char8_t
+template <>
+class UniversalTersePrinter<const char8_t*> {
+ public:
+  static void Print(const char8_t* str, ::std::ostream* os) {
+    if (str == nullptr) {
+      *os << "NULL";
+    } else {
+      UniversalPrint(::std::u8string(str), os);
+    }
+  }
+};
+template <>
+class UniversalTersePrinter<char8_t*>
+    : public UniversalTersePrinter<const char8_t*> {};
+#endif
+
+template <>
+class UniversalTersePrinter<const char16_t*> {
+ public:
+  static void Print(const char16_t* str, ::std::ostream* os) {
+    if (str == nullptr) {
+      *os << "NULL";
+    } else {
+      UniversalPrint(::std::u16string(str), os);
+    }
+  }
+};
+template <>
+class UniversalTersePrinter<char16_t*>
+    : public UniversalTersePrinter<const char16_t*> {};
+
+template <>
+class UniversalTersePrinter<const char32_t*> {
  public:
-  static void Print(char* str, ::std::ostream* os) {
-    UniversalTersePrinter<const char*>::Print(str, os);
+  static void Print(const char32_t* str, ::std::ostream* os) {
+    if (str == nullptr) {
+      *os << "NULL";
+    } else {
+      UniversalPrint(::std::u32string(str), os);
+    }
   }
 };
+template <>
+class UniversalTersePrinter<char32_t*>
+    : public UniversalTersePrinter<const char32_t*> {};
 
 #if GTEST_HAS_STD_WSTRING
 template <>

googletest/include/gtest/internal/gtest-port.h

@@ -1936,6 +1936,19 @@ inline bool IsUpper(char ch) {
 inline bool IsXDigit(char ch) {
   return isxdigit(static_cast<unsigned char>(ch)) != 0;
 }
+#ifdef __cpp_char8_t
+inline bool IsXDigit(char8_t ch) {
+  return isxdigit(static_cast<unsigned char>(ch)) != 0;
+}
+#endif
+inline bool IsXDigit(char16_t ch) {
+  const unsigned char low_byte = static_cast<unsigned char>(ch);
+  return ch == low_byte && isxdigit(low_byte) != 0;
+}
+inline bool IsXDigit(char32_t ch) {
+  const unsigned char low_byte = static_cast<unsigned char>(ch);
+  return ch == low_byte && isxdigit(low_byte) != 0;
+}
 inline bool IsXDigit(wchar_t ch) {
   const unsigned char low_byte = static_cast<unsigned char>(ch);
   return ch == low_byte && isxdigit(low_byte) != 0;