1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/xpcom/tests/TestEncoding.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,203 @@
1.4 +/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +#include "TestHarness.h"
1.10 +
1.11 +nsresult TestGoodSurrogatePair()
1.12 +{
1.13 + // When this string is decoded, the surrogate pair is U+10302 and the rest of
1.14 + // the string is specified by indexes 2 onward.
1.15 + const char16_t goodPairData[] = { 0xD800, 0xDF02, 0x65, 0x78, 0x0 };
1.16 + nsDependentString goodPair16(goodPairData);
1.17 +
1.18 + uint32_t byteCount = 0;
1.19 + char* goodPair8 = ToNewUTF8String(goodPair16, &byteCount);
1.20 + if (!goodPair8)
1.21 + {
1.22 + fail("out of memory creating goodPair8");
1.23 + return NS_ERROR_OUT_OF_MEMORY;
1.24 + }
1.25 +
1.26 + if (byteCount != 6)
1.27 + {
1.28 + fail("wrong number of bytes; expected 6, got %lu", byteCount);
1.29 + return NS_ERROR_FAILURE;
1.30 + }
1.31 +
1.32 + const unsigned char expected8[] =
1.33 + { 0xF0, 0x90, 0x8C, 0x82, 0x65, 0x78, 0x0 };
1.34 + if (0 != memcmp(expected8, goodPair8, sizeof(expected8)))
1.35 + {
1.36 + fail("wrong translation to UTF8");
1.37 + return NS_ERROR_FAILURE;
1.38 + }
1.39 +
1.40 + // This takes a different code path from the above, so test it to make sure
1.41 + // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
1.42 + nsDependentCString expected((const char*)expected8);
1.43 + if (0 != CompareUTF8toUTF16(expected, goodPair16))
1.44 + {
1.45 + fail("bad comparison between UTF-8 and equivalent UTF-16");
1.46 + return NS_ERROR_FAILURE;
1.47 + }
1.48 +
1.49 + NS_Free(goodPair8);
1.50 +
1.51 + passed("TestGoodSurrogatePair");
1.52 + return NS_OK;
1.53 +}
1.54 +
1.55 +nsresult TestBackwardsSurrogatePair()
1.56 +{
1.57 + // When this string is decoded, the two surrogates are wrongly ordered and
1.58 + // must each be interpreted as U+FFFD.
1.59 + const char16_t backwardsPairData[] = { 0xDDDD, 0xD863, 0x65, 0x78, 0x0 };
1.60 + nsDependentString backwardsPair16(backwardsPairData);
1.61 +
1.62 + uint32_t byteCount = 0;
1.63 + char* backwardsPair8 = ToNewUTF8String(backwardsPair16, &byteCount);
1.64 + if (!backwardsPair8)
1.65 + {
1.66 + fail("out of memory creating backwardsPair8");
1.67 + return NS_ERROR_OUT_OF_MEMORY;
1.68 + }
1.69 +
1.70 + if (byteCount != 8)
1.71 + {
1.72 + fail("wrong number of bytes; expected 8, got %lu", byteCount);
1.73 + return NS_ERROR_FAILURE;
1.74 + }
1.75 +
1.76 + const unsigned char expected8[] =
1.77 + { 0xEF, 0xBF, 0xBD, 0xEF, 0xBF, 0xBD, 0x65, 0x78, 0x0 };
1.78 + if (0 != memcmp(expected8, backwardsPair8, sizeof(expected8)))
1.79 + {
1.80 + fail("wrong translation to UTF8");
1.81 + return NS_ERROR_FAILURE;
1.82 + }
1.83 +
1.84 + // This takes a different code path from the above, so test it to make sure
1.85 + // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
1.86 + nsDependentCString expected((const char*)expected8);
1.87 + if (0 != CompareUTF8toUTF16(expected, backwardsPair16))
1.88 + {
1.89 + fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
1.90 + return NS_ERROR_FAILURE;
1.91 + }
1.92 +
1.93 + NS_Free(backwardsPair8);
1.94 +
1.95 + passed("TestBackwardsSurrogatePair");
1.96 + return NS_OK;
1.97 +}
1.98 +
1.99 +nsresult TestMalformedUTF16OrphanHighSurrogate()
1.100 +{
1.101 + // When this string is decoded, the high surrogate should be replaced and the
1.102 + // rest of the string is specified by indexes 1 onward.
1.103 + const char16_t highSurrogateData[] = { 0xD863, 0x74, 0x65, 0x78, 0x74, 0x0 };
1.104 + nsDependentString highSurrogate16(highSurrogateData);
1.105 +
1.106 + uint32_t byteCount = 0;
1.107 + char* highSurrogate8 = ToNewUTF8String(highSurrogate16, &byteCount);
1.108 + if (!highSurrogate8)
1.109 + {
1.110 + fail("out of memory creating highSurrogate8");
1.111 + return NS_ERROR_OUT_OF_MEMORY;
1.112 + }
1.113 +
1.114 + if (byteCount != 7)
1.115 + {
1.116 + fail("wrong number of bytes; expected 7, got %lu", byteCount);
1.117 + return NS_ERROR_FAILURE;
1.118 + }
1.119 +
1.120 + const unsigned char expected8[] =
1.121 + { 0xEF, 0xBF, 0xBD, 0x74, 0x65, 0x78, 0x74, 0x0 };
1.122 + if (0 != memcmp(expected8, highSurrogate8, sizeof(expected8)))
1.123 + {
1.124 + fail("wrong translation to UTF8");
1.125 + return NS_ERROR_FAILURE;
1.126 + }
1.127 +
1.128 + // This takes a different code path from the above, so test it to make sure
1.129 + // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
1.130 + nsDependentCString expected((const char*)expected8);
1.131 + if (0 != CompareUTF8toUTF16(expected, highSurrogate16))
1.132 + {
1.133 + fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
1.134 + return NS_ERROR_FAILURE;
1.135 + }
1.136 +
1.137 + NS_Free(highSurrogate8);
1.138 +
1.139 + passed("TestMalformedUTF16OrphanHighSurrogate");
1.140 + return NS_OK;
1.141 +}
1.142 +
1.143 +nsresult TestMalformedUTF16OrphanLowSurrogate()
1.144 +{
1.145 + // When this string is decoded, the low surrogate should be replaced and the
1.146 + // rest of the string is specified by indexes 1 onward.
1.147 + const char16_t lowSurrogateData[] = { 0xDDDD, 0x74, 0x65, 0x78, 0x74, 0x0 };
1.148 + nsDependentString lowSurrogate16(lowSurrogateData);
1.149 +
1.150 + uint32_t byteCount = 0;
1.151 + char* lowSurrogate8 = ToNewUTF8String(lowSurrogate16, &byteCount);
1.152 + if (!lowSurrogate8)
1.153 + {
1.154 + fail("out of memory creating lowSurrogate8");
1.155 + return NS_ERROR_OUT_OF_MEMORY;
1.156 + }
1.157 +
1.158 + if (byteCount != 7)
1.159 + {
1.160 + fail("wrong number of bytes; expected 7, got %lu", byteCount);
1.161 + return NS_ERROR_FAILURE;
1.162 + }
1.163 +
1.164 + const unsigned char expected8[] =
1.165 + { 0xEF, 0xBF, 0xBD, 0x74, 0x65, 0x78, 0x74, 0x0 };
1.166 + if (0 != memcmp(expected8, lowSurrogate8, sizeof(expected8)))
1.167 + {
1.168 + fail("wrong translation to UTF8");
1.169 + return NS_ERROR_FAILURE;
1.170 + }
1.171 +
1.172 + // This takes a different code path from the above, so test it to make sure
1.173 + // the UTF-16 enumeration remains in sync with the UTF-8 enumeration.
1.174 + nsDependentCString expected((const char*)expected8);
1.175 + if (0 != CompareUTF8toUTF16(expected, lowSurrogate16))
1.176 + {
1.177 + fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
1.178 + return NS_ERROR_FAILURE;
1.179 + }
1.180 +
1.181 + NS_Free(lowSurrogate8);
1.182 +
1.183 + passed("TestMalformedUTF16OrphanLowSurrogate");
1.184 + return NS_OK;
1.185 +}
1.186 +
1.187 +
1.188 +int main(int argc, char** argv)
1.189 +{
1.190 + ScopedXPCOM xpcom("TestEncoding");
1.191 + if (xpcom.failed())
1.192 + return 1;
1.193 +
1.194 + int rv = 0;
1.195 +
1.196 + if (NS_FAILED(TestGoodSurrogatePair()))
1.197 + rv = 1;
1.198 + if (NS_FAILED(TestBackwardsSurrogatePair()))
1.199 + rv = 1;
1.200 + if (NS_FAILED(TestMalformedUTF16OrphanHighSurrogate()))
1.201 + rv = 1;
1.202 + if (NS_FAILED(TestMalformedUTF16OrphanLowSurrogate()))
1.203 + rv = 1;
1.204 +
1.205 + return rv;
1.206 +}
