gecko/xpcom/tests/TestEncoding.cpp
Ehsan Akhgari 0fd9123eac Bug 579517 - Part 1: Automated conversion of NSPR numeric types to stdint types in Gecko; r=bsmedberg
This patch was generated by a script.  Here's the source of the script for
future reference:

function convert() {
echo "Converting $1 to $2..."
find . ! -wholename "*nsprpub*" \
       ! -wholename "*security/nss*" \
       ! -wholename "*/.hg*" \
       ! -wholename "obj-ff-dbg*" \
       ! -name nsXPCOMCID.h \
       ! -name prtypes.h \
         -type f \
      \( -iname "*.cpp" \
         -o -iname "*.h" \
         -o -iname "*.c" \
         -o -iname "*.cc" \
         -o -iname "*.idl" \
         -o -iname "*.ipdl" \
         -o -iname "*.ipdlh" \
         -o -iname "*.mm" \) | \
    xargs -n 1 sed -i -e "s/\b$1\b/$2/g"
}

convert PRInt8 int8_t
convert PRUint8 uint8_t
convert PRInt16 int16_t
convert PRUint16 uint16_t
convert PRInt32 int32_t
convert PRUint32 uint32_t
convert PRInt64 int64_t
convert PRUint64 uint64_t

convert PRIntn int
convert PRUintn unsigned

convert PRSize size_t

convert PROffset32 int32_t
convert PROffset64 int64_t

convert PRPtrdiff ptrdiff_t

convert PRFloat64 double
2012-08-22 11:56:38 -04:00

204 lines
5.9 KiB
C++

/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* 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 "TestHarness.h"
nsresult TestGoodSurrogatePair()
{
// When this string is decoded, the surrogate pair is U+10302 and the rest of
// the string is specified by indexes 2 onward.
const PRUnichar goodPairData[] = { 0xD800, 0xDF02, 0x65, 0x78, 0x0 };
nsDependentString goodPair16(goodPairData);
uint32_t byteCount = 0;
char* goodPair8 = ToNewUTF8String(goodPair16, &byteCount);
if (!goodPair8)
{
fail("out of memory creating goodPair8");
return NS_ERROR_OUT_OF_MEMORY;
}
if (byteCount != 6)
{
fail("wrong number of bytes; expected 6, got %lu", byteCount);
return NS_ERROR_FAILURE;
}
const unsigned char expected8[] =
{ 0xF0, 0x90, 0x8C, 0x82, 0x65, 0x78, 0x0 };
if (0 != memcmp(expected8, goodPair8, sizeof(expected8)))
{
fail("wrong translation to UTF8");
return NS_ERROR_FAILURE;
}
// 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);
if (0 != CompareUTF8toUTF16(expected, goodPair16))
{
fail("bad comparison between UTF-8 and equivalent UTF-16");
return NS_ERROR_FAILURE;
}
NS_Free(goodPair8);
passed("TestGoodSurrogatePair");
return NS_OK;
}
nsresult TestBackwardsSurrogatePair()
{
// When this string is decoded, the two surrogates are wrongly ordered and
// must each be interpreted as U+FFFD.
const PRUnichar backwardsPairData[] = { 0xDDDD, 0xD863, 0x65, 0x78, 0x0 };
nsDependentString backwardsPair16(backwardsPairData);
uint32_t byteCount = 0;
char* backwardsPair8 = ToNewUTF8String(backwardsPair16, &byteCount);
if (!backwardsPair8)
{
fail("out of memory creating backwardsPair8");
return NS_ERROR_OUT_OF_MEMORY;
}
if (byteCount != 8)
{
fail("wrong number of bytes; expected 8, got %lu", byteCount);
return NS_ERROR_FAILURE;
}
const unsigned char expected8[] =
{ 0xEF, 0xBF, 0xBD, 0xEF, 0xBF, 0xBD, 0x65, 0x78, 0x0 };
if (0 != memcmp(expected8, backwardsPair8, sizeof(expected8)))
{
fail("wrong translation to UTF8");
return NS_ERROR_FAILURE;
}
// 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);
if (0 != CompareUTF8toUTF16(expected, backwardsPair16))
{
fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
return NS_ERROR_FAILURE;
}
NS_Free(backwardsPair8);
passed("TestBackwardsSurrogatePair");
return NS_OK;
}
nsresult TestMalformedUTF16OrphanHighSurrogate()
{
// When this string is decoded, the high surrogate should be replaced and the
// rest of the string is specified by indexes 1 onward.
const PRUnichar highSurrogateData[] = { 0xD863, 0x74, 0x65, 0x78, 0x74, 0x0 };
nsDependentString highSurrogate16(highSurrogateData);
uint32_t byteCount = 0;
char* highSurrogate8 = ToNewUTF8String(highSurrogate16, &byteCount);
if (!highSurrogate8)
{
fail("out of memory creating highSurrogate8");
return NS_ERROR_OUT_OF_MEMORY;
}
if (byteCount != 7)
{
fail("wrong number of bytes; expected 7, got %lu", byteCount);
return NS_ERROR_FAILURE;
}
const unsigned char expected8[] =
{ 0xEF, 0xBF, 0xBD, 0x74, 0x65, 0x78, 0x74, 0x0 };
if (0 != memcmp(expected8, highSurrogate8, sizeof(expected8)))
{
fail("wrong translation to UTF8");
return NS_ERROR_FAILURE;
}
// 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);
if (0 != CompareUTF8toUTF16(expected, highSurrogate16))
{
fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
return NS_ERROR_FAILURE;
}
NS_Free(highSurrogate8);
passed("TestMalformedUTF16OrphanHighSurrogate");
return NS_OK;
}
nsresult TestMalformedUTF16OrphanLowSurrogate()
{
// When this string is decoded, the low surrogate should be replaced and the
// rest of the string is specified by indexes 1 onward.
const PRUnichar lowSurrogateData[] = { 0xDDDD, 0x74, 0x65, 0x78, 0x74, 0x0 };
nsDependentString lowSurrogate16(lowSurrogateData);
uint32_t byteCount = 0;
char* lowSurrogate8 = ToNewUTF8String(lowSurrogate16, &byteCount);
if (!lowSurrogate8)
{
fail("out of memory creating lowSurrogate8");
return NS_ERROR_OUT_OF_MEMORY;
}
if (byteCount != 7)
{
fail("wrong number of bytes; expected 7, got %lu", byteCount);
return NS_ERROR_FAILURE;
}
const unsigned char expected8[] =
{ 0xEF, 0xBF, 0xBD, 0x74, 0x65, 0x78, 0x74, 0x0 };
if (0 != memcmp(expected8, lowSurrogate8, sizeof(expected8)))
{
fail("wrong translation to UTF8");
return NS_ERROR_FAILURE;
}
// 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);
if (0 != CompareUTF8toUTF16(expected, lowSurrogate16))
{
fail("bad comparison between UTF-8 and malformed but equivalent UTF-16");
return NS_ERROR_FAILURE;
}
NS_Free(lowSurrogate8);
passed("TestMalformedUTF16OrphanLowSurrogate");
return NS_OK;
}
int main(int argc, char** argv)
{
ScopedXPCOM xpcom("TestEncoding");
if (xpcom.failed())
return 1;
int rv = 0;
if (NS_FAILED(TestGoodSurrogatePair()))
rv = 1;
if (NS_FAILED(TestBackwardsSurrogatePair()))
rv = 1;
if (NS_FAILED(TestMalformedUTF16OrphanHighSurrogate()))
rv = 1;
if (NS_FAILED(TestMalformedUTF16OrphanLowSurrogate()))
rv = 1;
return rv;
}