commit 182d9c05e78b1ddb1cb8242cd3628a7855a0336f

Author: Andrey Kosyakov <caseq@chromium.org>

Date:   2023-08-17T13:50:11-07:00

    Define UChar as char16_t

    We used to have UChar defined as uint16_t which does not go along

    with STL these days if you try to have an std::basic_string<> of it,

    as there are no standard std::char_traits<> specialization for uint16_t.

    This switches UChar to char16_t where practical, introducing a few

    compatibility shims to keep CL size small, as (1) this would likely

    have to be back-ported and (2) crdtp extensively uses uint16_t for

    wide chars.

    Bug: b:296390693

    Change-Id: I66a32d8f0050915225b187de56896c26dd76163d

    Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/4789966

    Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>

    Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>

    Auto-Submit: Andrey Kosyakov <caseq@chromium.org>

    Cr-Commit-Position: refs/heads/main@{#89559}

diff --git deps/v8/src/inspector/string-16.cc deps/v8/src/inspector/string-16.cc

index a8b786a8166..6df9963e970 100644

--- deps/v8/src/inspector/string-16.cc

+++ deps/v8/src/inspector/string-16.cc

@@ -27,7 +27,7 @@ bool isSpaceOrNewLine(UChar c) {

   return isASCII(c) && c <= ' ' && (c == ' ' || (c <= 0xD && c >= 0x9));

 }

-int64_t charactersToInteger(const UChar* characters, size_t length,

+int64_t charactersToInteger(const uint16_t* characters, size_t length,

                             bool* ok = nullptr) {

   std::vector<char> buffer;

   buffer.reserve(length + 1);

@@ -50,6 +50,8 @@ int64_t charactersToInteger(const UChar* characters, size_t length,

 String16::String16(const UChar* characters, size_t size)

     : m_impl(characters, size) {}

+String16::String16(const uint16_t* characters, size_t size)

+    : m_impl(reinterpret_cast<const UChar*>(characters), size) {}

 String16::String16(const UChar* characters) : m_impl(characters) {}

@@ -241,6 +243,10 @@ String16 String16::fromUTF16LE(const UChar* stringStart, size_t length) {

 #endif  // V8_TARGET_BIG_ENDIAN

 }

+String16 String16::fromUTF16LE(const uint16_t* stringStart, size_t length) {

+  return fromUTF16LE(reinterpret_cast<const UChar*>(stringStart), length);

+}

+

 std::string String16::utf8() const {

   return UTF16ToUTF8(m_impl.data(), m_impl.size());

 }

diff --git deps/v8/src/inspector/string-16.h deps/v8/src/inspector/string-16.h

index 1678ffb2e1e..d9f6c466ab1 100644

--- deps/v8/src/inspector/string-16.h

+++ deps/v8/src/inspector/string-16.h

@@ -6,6 +6,7 @@

 #define V8_INSPECTOR_STRING_16_H_

 #include <stdint.h>

+#include <uchar.h>

 #include <cctype>

 #include <climits>

@@ -18,7 +19,7 @@

 namespace v8_inspector {

-using UChar = uint16_t;

+using UChar = char16_t;

 class String16 {

  public:

@@ -28,6 +29,7 @@ class String16 {

   String16(const String16&) V8_NOEXCEPT = default;

   String16(String16&&) V8_NOEXCEPT = default;

   String16(const UChar* characters, size_t size);

+  String16(const uint16_t* characters, size_t size);

   V8_EXPORT String16(const UChar* characters);

   V8_EXPORT String16(const char* characters);

   String16(const char* characters, size_t size);

@@ -49,7 +51,9 @@ class String16 {

   int toInteger(bool* ok = nullptr) const;

   std::pair<size_t, size_t> getTrimmedOffsetAndLength() const;

   String16 stripWhiteSpace() const;

-  const UChar* characters16() const { return m_impl.c_str(); }

+  const uint16_t* characters16() const {

+    return reinterpret_cast<const uint16_t*>(m_impl.c_str());

+  }

   size_t length() const { return m_impl.length(); }

   bool isEmpty() const { return !m_impl.length(); }

   UChar operator[](size_t index) const { return m_impl[index]; }

@@ -79,6 +83,8 @@ class String16 {

   // On Big endian architectures, byte order needs to be flipped.

   V8_EXPORT static String16 fromUTF16LE(const UChar* stringStart,

                                         size_t length);

+  V8_EXPORT static String16 fromUTF16LE(const uint16_t* stringStart,

+                                        size_t length);

   std::size_t hash() const {

     if (!hash_code) {

diff --git deps/v8/src/inspector/v8-string-conversions.cc deps/v8/src/inspector/v8-string-conversions.cc

index 0c75e66b972..8cf19be816c 100644

--- deps/v8/src/inspector/v8-string-conversions.cc

+++ deps/v8/src/inspector/v8-string-conversions.cc

@@ -12,7 +12,7 @@

 namespace v8_inspector {

 namespace {

-using UChar = uint16_t;

+using UChar = char16_t;

 using UChar32 = uint32_t;

 bool isASCII(UChar c) { return !(c & ~0x7F); }

@@ -386,7 +386,7 @@ std::string UTF16ToUTF8(const UChar* stringStart, size_t length) {

 std::basic_string<UChar> UTF8ToUTF16(const char* stringStart, size_t length) {

   if (!stringStart || !length) return std::basic_string<UChar>();

-  std::vector<uint16_t> buffer(length);

+  std::vector<UChar> buffer(length);

   UChar* bufferStart = buffer.data();

   UChar* bufferCurrent = bufferStart;

@@ -395,7 +395,7 @@ std::basic_string<UChar> UTF8ToUTF16(const char* stringStart, size_t length) {

                          reinterpret_cast<const char*>(stringStart + length),

                          &bufferCurrent, bufferCurrent + buffer.size(), nullptr,

                          true) != conversionOK)

-    return std::basic_string<uint16_t>();

+    return std::basic_string<UChar>();

   size_t utf16Length = bufferCurrent - bufferStart;

   return std::basic_string<UChar>(bufferStart, bufferStart + utf16Length);

 }

diff --git deps/v8/src/inspector/v8-string-conversions.h deps/v8/src/inspector/v8-string-conversions.h

index eb33c6816a5..1126255dac2 100644

--- deps/v8/src/inspector/v8-string-conversions.h

+++ deps/v8/src/inspector/v8-string-conversions.h

@@ -5,14 +5,16 @@

 #ifndef V8_INSPECTOR_V8_STRING_CONVERSIONS_H_

 #define V8_INSPECTOR_V8_STRING_CONVERSIONS_H_

+#include <uchar.h>

+

 #include <cstdint>

 #include <string>

 // Conversion routines between UT8 and UTF16, used by string-16.{h,cc}. You may

 // want to use string-16.h directly rather than these.

 namespace v8_inspector {

-std::basic_string<uint16_t> UTF8ToUTF16(const char* stringStart, size_t length);

-std::string UTF16ToUTF8(const uint16_t* stringStart, size_t length);

+std::basic_string<char16_t> UTF8ToUTF16(const char* stringStart, size_t length);

+std::string UTF16ToUTF8(const char16_t* stringStart, size_t length);

 }  // namespace v8_inspector

 #endif  // V8_INSPECTOR_V8_STRING_CONVERSIONS_H_

diff --git deps/v8/third_party/inspector_protocol/crdtp/test_platform_v8.cc deps/v8/third_party/inspector_protocol/crdtp/test_platform_v8.cc

index c9d89eaa42f..1a46d781b89 100644

--- deps/v8/third_party/inspector_protocol/crdtp/test_platform_v8.cc

+++ deps/v8/third_party/inspector_protocol/crdtp/test_platform_v8.cc

@@ -11,13 +11,16 @@

 namespace v8_crdtp {

 std::string UTF16ToUTF8(span<uint16_t> in) {

-  return v8_inspector::UTF16ToUTF8(in.data(), in.size());

+  return v8_inspector::UTF16ToUTF8(reinterpret_cast<const char16_t*>(in.data()),

+                                   in.size());

 }

 std::vector<uint16_t> UTF8ToUTF16(span<uint8_t> in) {

-  std::basic_string<uint16_t> utf16 = v8_inspector::UTF8ToUTF16(

+  std::basic_string<char16_t> utf16 = v8_inspector::UTF8ToUTF16(

       reinterpret_cast<const char*>(in.data()), in.size());

-  return std::vector<uint16_t>(utf16.begin(), utf16.end());

+  return std::vector<uint16_t>(

+      reinterpret_cast<const uint16_t*>(utf16.data()),

+      reinterpret_cast<const uint16_t*>(utf16.data()) + utf16.size());

 }

 }  // namespace v8_crdtp