From 26a029d407be480d791972afb5975cf62c9360a6 Mon Sep 17 00:00:00 2001
From: Daniel Baumann <daniel.baumann@progress-linux.org>
Date: Fri, 19 Apr 2024 02:47:55 +0200
Subject: Adding upstream version 124.0.1.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
---
 xpcom/tests/gtest/TestEncoding.cpp | 108 +++++++++++++++++++++++++++++++++++++
 1 file changed, 108 insertions(+)
 create mode 100644 xpcom/tests/gtest/TestEncoding.cpp

(limited to 'xpcom/tests/gtest/TestEncoding.cpp')

diff --git a/xpcom/tests/gtest/TestEncoding.cpp b/xpcom/tests/gtest/TestEncoding.cpp
new file mode 100644
index 0000000000..2abbc5d708
--- /dev/null
+++ b/xpcom/tests/gtest/TestEncoding.cpp
@@ -0,0 +1,108 @@
+/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
+/* vim: set ts=8 sts=2 et sw=2 tw=80: */
+/* This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
+
+#include <stdlib.h>
+#include "nsString.h"
+#include "gtest/gtest.h"
+
+TEST(Encoding, GoodSurrogatePair)
+{
+  // When this string is decoded, the surrogate pair is U+10302 and the rest of
+  // the string is specified by indexes 2 onward.
+  const char16_t goodPairData[] = {0xD800, 0xDF02, 0x65, 0x78, 0x0};
+  nsDependentString goodPair16(goodPairData);
+
+  uint32_t byteCount = 0;
+  char* goodPair8 = ToNewUTF8String(goodPair16, &byteCount);
+  EXPECT_TRUE(!!goodPair8);
+
+  EXPECT_EQ(byteCount, 6u);
+
+  const unsigned char expected8[] = {0xF0, 0x90, 0x8C, 0x82, 0x65, 0x78, 0x0};
+  EXPECT_EQ(0, memcmp(expected8, goodPair8, sizeof(expected8)));
+
+  // This takes a different code path from the above, so test it to make sure
+  // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
+  nsDependentCString expected((const char*)expected8);
+  EXPECT_EQ(0, CompareUTF8toUTF16(expected, goodPair16));
+
+  free(goodPair8);
+}
+
+TEST(Encoding, BackwardsSurrogatePair)
+{
+  // When this string is decoded, the two surrogates are wrongly ordered and
+  // must each be interpreted as U+FFFD.
+  const char16_t backwardsPairData[] = {0xDDDD, 0xD863, 0x65, 0x78, 0x0};
+  nsDependentString backwardsPair16(backwardsPairData);
+
+  uint32_t byteCount = 0;
+  char* backwardsPair8 = ToNewUTF8String(backwardsPair16, &byteCount);
+  EXPECT_TRUE(!!backwardsPair8);
+
+  EXPECT_EQ(byteCount, 8u);
+
+  const unsigned char expected8[] = {0xEF, 0xBF, 0xBD, 0xEF, 0xBF,
+                                     0xBD, 0x65, 0x78, 0x0};
+  EXPECT_EQ(0, memcmp(expected8, backwardsPair8, sizeof(expected8)));
+
+  // This takes a different code path from the above, so test it to make sure
+  // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
+  nsDependentCString expected((const char*)expected8);
+  EXPECT_EQ(0, CompareUTF8toUTF16(expected, backwardsPair16));
+
+  free(backwardsPair8);
+}
+
+TEST(Encoding, MalformedUTF16OrphanHighSurrogate)
+{
+  // When this string is decoded, the high surrogate should be replaced and the
+  // rest of the string is specified by indexes 1 onward.
+  const char16_t highSurrogateData[] = {0xD863, 0x74, 0x65, 0x78, 0x74, 0x0};
+  nsDependentString highSurrogate16(highSurrogateData);
+
+  uint32_t byteCount = 0;
+  char* highSurrogate8 = ToNewUTF8String(highSurrogate16, &byteCount);
+  EXPECT_TRUE(!!highSurrogate8);
+
+  EXPECT_EQ(byteCount, 7u);
+
+  const unsigned char expected8[] = {0xEF, 0xBF, 0xBD, 0x74,
+                                     0x65, 0x78, 0x74, 0x0};
+  EXPECT_EQ(0, memcmp(expected8, highSurrogate8, sizeof(expected8)));
+
+  // This takes a different code path from the above, so test it to make sure
+  // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
+  nsDependentCString expected((const char*)expected8);
+  EXPECT_EQ(0, CompareUTF8toUTF16(expected, highSurrogate16));
+
+  free(highSurrogate8);
+}
+
+TEST(Encoding, MalformedUTF16OrphanLowSurrogate)
+{
+  // When this string is decoded, the low surrogate should be replaced and the
+  // rest of the string is specified by indexes 1 onward.
+  const char16_t lowSurrogateData[] = {0xDDDD, 0x74, 0x65, 0x78, 0x74, 0x0};
+  nsDependentString lowSurrogate16(lowSurrogateData);
+
+  uint32_t byteCount = 0;
+  char* lowSurrogate8 = ToNewUTF8String(lowSurrogate16, &byteCount);
+  EXPECT_TRUE(!!lowSurrogate8);
+
+  EXPECT_EQ(byteCount, 7u);
+
+  const unsigned char expected8[] = {0xEF, 0xBF, 0xBD, 0x74,
+                                     0x65, 0x78, 0x74, 0x0};
+  EXPECT_EQ(0, memcmp(expected8, lowSurrogate8, sizeof(expected8)));
+
+  // This takes a different code path from the above, so test it to make sure
+  // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
+  nsDependentCString expected((const char*)expected8);
+  EXPECT_EQ(0, CompareUTF8toUTF16(expected, lowSurrogate16));
+
+  free(lowSurrogate8);
+}
-- 
cgit v1.2.3