2016-11-22 09:27:41 +00:00
|
|
|
// Copyright 2016 the V8 project authors. All rights reserved.
|
|
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
|
|
// found in the LICENSE file.
|
|
|
|
|
|
|
|
#include <memory>
|
|
|
|
#include <string>
|
2017-09-29 12:43:07 +00:00
|
|
|
#include <vector>
|
2016-11-22 09:27:41 +00:00
|
|
|
|
|
|
|
#include "src/unicode-decoder.h"
|
2017-09-29 12:43:07 +00:00
|
|
|
#include "src/unicode-inl.h"
|
2016-11-22 09:27:41 +00:00
|
|
|
#include "testing/gtest/include/gtest/gtest.h"
|
|
|
|
|
|
|
|
namespace v8 {
|
|
|
|
namespace internal {
|
|
|
|
|
|
|
|
namespace {
|
|
|
|
|
|
|
|
using Utf8Decoder = unibrow::Utf8Decoder<512>;
|
|
|
|
|
|
|
|
void Decode(Utf8Decoder* decoder, const std::string& str) {
|
|
|
|
// Put the string in its own buffer on the heap to make sure that
|
|
|
|
// AddressSanitizer's heap-buffer-overflow logic can see what's going on.
|
|
|
|
std::unique_ptr<char[]> buffer(new char[str.length()]);
|
|
|
|
memcpy(buffer.get(), str.data(), str.length());
|
|
|
|
decoder->Reset(buffer.get(), str.length());
|
|
|
|
}
|
|
|
|
|
2017-09-29 12:43:07 +00:00
|
|
|
void DecodeNormally(const std::vector<byte>& bytes,
|
|
|
|
std::vector<unibrow::uchar>* output) {
|
|
|
|
size_t cursor = 0;
|
|
|
|
while (cursor < bytes.size()) {
|
|
|
|
output->push_back(
|
|
|
|
unibrow::Utf8::ValueOf(bytes.data() + cursor, bytes.size(), &cursor));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void DecodeIncrementally(const std::vector<byte>& bytes,
|
|
|
|
std::vector<unibrow::uchar>* output) {
|
|
|
|
unibrow::Utf8::Utf8IncrementalBuffer buffer = 0;
|
|
|
|
for (auto b : bytes) {
|
|
|
|
unibrow::uchar result = unibrow::Utf8::ValueOfIncremental(b, &buffer);
|
|
|
|
if (result != unibrow::Utf8::kIncomplete) {
|
|
|
|
output->push_back(result);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
unibrow::uchar result = unibrow::Utf8::ValueOfIncrementalFinish(&buffer);
|
|
|
|
if (result != unibrow::Utf8::kBufferEmpty) {
|
|
|
|
output->push_back(result);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-11-22 09:27:41 +00:00
|
|
|
} // namespace
|
|
|
|
|
|
|
|
TEST(UnicodeTest, ReadOffEndOfUtf8String) {
|
|
|
|
Utf8Decoder decoder;
|
|
|
|
|
|
|
|
// Not enough continuation bytes before string ends.
|
|
|
|
Decode(&decoder, "\xE0");
|
|
|
|
Decode(&decoder, "\xED");
|
|
|
|
Decode(&decoder, "\xF0");
|
|
|
|
Decode(&decoder, "\xF4");
|
|
|
|
}
|
|
|
|
|
2017-09-29 12:43:07 +00:00
|
|
|
TEST(UnicodeTest, IncrementalUTF8DecodingVsNonIncrementalUtf8Decoding) {
|
|
|
|
// Unfortunately, V8 has two UTF-8 decoders. This test checks that they
|
|
|
|
// produce the same result. This test was inspired by
|
|
|
|
// https://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt .
|
|
|
|
typedef struct {
|
|
|
|
std::vector<byte> bytes;
|
|
|
|
std::vector<unibrow::uchar> unicode_expected;
|
|
|
|
} TestCase;
|
|
|
|
|
|
|
|
TestCase data[] = {
|
|
|
|
// Correct UTF-8 text.
|
|
|
|
{{0xce, 0xba, 0xe1, 0xbd, 0xb9, 0xcf, 0x83, 0xce, 0xbc, 0xce, 0xb5},
|
|
|
|
{0x3ba, 0x1f79, 0x3c3, 0x3bc, 0x3b5}},
|
|
|
|
|
|
|
|
// First possible sequence of a certain length:
|
|
|
|
// 1 byte
|
|
|
|
{{0x00}, {0x0}},
|
|
|
|
// 2 bytes
|
|
|
|
{{0xc2, 0x80}, {0x80}},
|
|
|
|
// 3 bytes
|
|
|
|
{{0xe0, 0xa0, 0x80}, {0x800}},
|
|
|
|
// 4 bytes
|
|
|
|
{{0xf0, 0x90, 0x80, 0x80}, {0x10000}},
|
|
|
|
// 5 bytes (not supported)
|
|
|
|
{{0xf8, 0x88, 0x80, 0x80, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6 bytes (not supported)
|
|
|
|
{{0xfc, 0x84, 0x80, 0x80, 0x80, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Last possible sequence of certain length:
|
|
|
|
// 1 byte
|
|
|
|
{{0x7f}, {0x7f}},
|
|
|
|
// 2 bytes
|
|
|
|
{{0xdf, 0xbf}, {0x7ff}},
|
|
|
|
// 3 bytes
|
|
|
|
{{0xef, 0xbf, 0xbf}, {0xffff}},
|
|
|
|
// 4 bytes (this sequence is not a valid code point)
|
|
|
|
{{0xf7, 0xbf, 0xbf, 0xbf}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5 bytes (not supported)
|
|
|
|
{{0xfb, 0xbf, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6 bytes (not supported)
|
|
|
|
{{0xfd, 0xbf, 0xbf, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// Other boundary conditions:
|
|
|
|
{{0xed, 0x9f, 0xbf}, {0xd7ff}},
|
|
|
|
{{0xee, 0x80, 0x80}, {0xe000}},
|
|
|
|
// U+fffd (invalid code point)
|
|
|
|
{{0xef, 0xbf, 0xbd}, {0xfffd}},
|
|
|
|
// U+10ffff (last valid code point)
|
|
|
|
{{0xf4, 0x8f, 0xbf, 0xbf}, {0x10ffff}},
|
|
|
|
// First invalid (too large) code point
|
|
|
|
{{0xf4, 0x90, 0x80, 0x80}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Malformed sequences:
|
|
|
|
// Unexpected continuation bytes:
|
|
|
|
// First continuation byte
|
|
|
|
{{0x80}, {0xfffd}},
|
|
|
|
// Last continuation byte
|
|
|
|
{{0xbf}, {0xfffd}},
|
|
|
|
// 2 continuation bytes
|
|
|
|
{{0x80, 0xbf}, {0xfffd, 0xfffd}},
|
|
|
|
// 3 continuation bytes
|
|
|
|
{{0x80, 0xbf, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 4 continuation bytes
|
|
|
|
{{0x80, 0xbf, 0x80, 0xbf}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5 continuation bytes
|
|
|
|
{{0x80, 0xbf, 0x80, 0xbf, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6 continuation bytes
|
|
|
|
{{0x80, 0xbf, 0x80, 0xbf, 0x80, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 7 continuation bytes
|
|
|
|
{{0x80, 0xbf, 0x80, 0xbf, 0x80, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// Sequence of all 64 possible continuation bytes
|
|
|
|
{{0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a,
|
|
|
|
0x8b, 0x8c, 0x8d, 0x8e, 0x8f, 0x90, 0x91, 0x92, 0x93, 0x94, 0x95,
|
|
|
|
0x96, 0x97, 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, 0xa0,
|
|
|
|
0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xab,
|
|
|
|
0xac, 0xad, 0xae, 0xaf, 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6,
|
|
|
|
0xb7, 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// Using each possible continuation byte in a two-byte sequence:
|
|
|
|
{{0xd0, 0x80, 0xd0, 0x81, 0xd0, 0x82, 0xd0, 0x83, 0xd0, 0x84, 0xd0, 0x85,
|
|
|
|
0xd0, 0x86, 0xd0, 0x87, 0xd0, 0x88, 0xd0, 0x89, 0xd0, 0x8a, 0xd0, 0x8b,
|
|
|
|
0xd0, 0x8c, 0xd0, 0x8d, 0xd0, 0x8e, 0xd0, 0x8f, 0xd0, 0x90, 0xd0, 0x91,
|
|
|
|
0xd0, 0x92, 0xd0, 0x93, 0xd0, 0x94, 0xd0, 0x95, 0xd0, 0x96, 0xd0, 0x97,
|
|
|
|
0xd0, 0x98, 0xd0, 0x99, 0xd0, 0x9a, 0xd0, 0x9b, 0xd0, 0x9c, 0xd0, 0x9d,
|
|
|
|
0xd0, 0x9e, 0xd0, 0x9f, 0xd0, 0xa0, 0xd0, 0xa1, 0xd0, 0xa2, 0xd0, 0xa3,
|
|
|
|
0xd0, 0xa4, 0xd0, 0xa5, 0xd0, 0xa6, 0xd0, 0xa7, 0xd0, 0xa8, 0xd0, 0xa9,
|
|
|
|
0xd0, 0xaa, 0xd0, 0xab, 0xd0, 0xac, 0xd0, 0xad, 0xd0, 0xae, 0xd0, 0xaf,
|
|
|
|
0xd0, 0xb0, 0xd0, 0xb1, 0xd0, 0xb2, 0xd0, 0xb3, 0xd0, 0xb4, 0xd0, 0xb5,
|
|
|
|
0xd0, 0xb6, 0xd0, 0xb7, 0xd0, 0xb8, 0xd0, 0xb9, 0xd0, 0xba, 0xd0, 0xbb,
|
|
|
|
0xd0, 0xbc, 0xd0, 0xbd, 0xd0, 0xbe, 0xd0, 0xbf},
|
|
|
|
{0x400, 0x401, 0x402, 0x403, 0x404, 0x405, 0x406, 0x407, 0x408, 0x409,
|
|
|
|
0x40a, 0x40b, 0x40c, 0x40d, 0x40e, 0x40f, 0x410, 0x411, 0x412, 0x413,
|
|
|
|
0x414, 0x415, 0x416, 0x417, 0x418, 0x419, 0x41a, 0x41b, 0x41c, 0x41d,
|
|
|
|
0x41e, 0x41f, 0x420, 0x421, 0x422, 0x423, 0x424, 0x425, 0x426, 0x427,
|
|
|
|
0x428, 0x429, 0x42a, 0x42b, 0x42c, 0x42d, 0x42e, 0x42f, 0x430, 0x431,
|
|
|
|
0x432, 0x433, 0x434, 0x435, 0x436, 0x437, 0x438, 0x439, 0x43a, 0x43b,
|
|
|
|
0x43c, 0x43d, 0x43e, 0x43f}},
|
|
|
|
|
|
|
|
// Lonely first bytes:
|
|
|
|
// All 32 first bytes of 32-byte sequences, each followed by a space
|
|
|
|
// (generates 32 invalid char + space sequences.
|
|
|
|
{{0xc0, 0x20, 0xc1, 0x20, 0xc2, 0x20, 0xc3, 0x20, 0xc4, 0x20, 0xc5,
|
|
|
|
0x20, 0xc6, 0x20, 0xc7, 0x20, 0xc8, 0x20, 0xc9, 0x20, 0xca, 0x20,
|
|
|
|
0xcb, 0x20, 0xcc, 0x20, 0xcd, 0x20, 0xce, 0x20, 0xcf, 0x20, 0xd0,
|
|
|
|
0x20, 0xd1, 0x20, 0xd2, 0x20, 0xd3, 0x20, 0xd4, 0x20, 0xd5, 0x20,
|
|
|
|
0xd6, 0x20, 0xd7, 0x20, 0xd8, 0x20, 0xd9, 0x20, 0xda, 0x20, 0xdb,
|
|
|
|
0x20, 0xdc, 0x20, 0xdd, 0x20, 0xde, 0x20, 0xdf, 0x20},
|
|
|
|
{0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20}},
|
|
|
|
// All 16 first bytes of 3-byte sequences, each followed by a space
|
|
|
|
// (generates 16 invalid char + space sequences):
|
|
|
|
{{0xe0, 0x20, 0xe1, 0x20, 0xe2, 0x20, 0xe3, 0x20, 0xe4, 0x20, 0xe5,
|
|
|
|
0x20, 0xe6, 0x20, 0xe7, 0x20, 0xe8, 0x20, 0xe9, 0x20, 0xea, 0x20,
|
|
|
|
0xeb, 0x20, 0xec, 0x20, 0xed, 0x20, 0xee, 0x20, 0xef, 0x20},
|
|
|
|
{0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20}},
|
|
|
|
// All 8 first bytes of 4-byte sequences, each followed by a space
|
|
|
|
// (generates 8 invalid char + space sequences):
|
|
|
|
{{0xf0, 0x20, 0xf1, 0x20, 0xf2, 0x20, 0xf3, 0x20, 0xf4, 0x20, 0xf5, 0x20,
|
|
|
|
0xf6, 0x20, 0xf7, 0x20},
|
|
|
|
{0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20,
|
|
|
|
0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20}},
|
|
|
|
// All 4 first bytes of 5-byte sequences (not supported), each followed by
|
|
|
|
// a space (generates 4 invalid char + space sequences):
|
|
|
|
{{0xf8, 0x20, 0xf9, 0x20, 0xfa, 0x20, 0xfb, 0x20},
|
|
|
|
{0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20, 0xfffd, 0x20}},
|
|
|
|
// All 2 first bytes of 6-byte sequences (not supported), each followed by
|
|
|
|
// a space (generates 2 invalid char + space sequences):
|
|
|
|
{{0xfc, 0x20, 0xfd, 0x20}, {0xfffd, 0x20, 0xfffd, 0x20}},
|
|
|
|
|
|
|
|
// Sequences with last continuation byte missing. Normally the whole
|
|
|
|
// incomplete sequence generates a single invalid character (exceptions
|
|
|
|
// explained below).
|
|
|
|
|
|
|
|
// 2-byte sequences with last byte missing
|
|
|
|
{{0xc0}, {0xfffd}},
|
|
|
|
{{0xdf}, {0xfffd}},
|
|
|
|
// 3-byte sequences with last byte missing.
|
|
|
|
{{0xe8, 0x80}, {0xfffd}},
|
|
|
|
{{0xe0, 0xbf}, {0xfffd}},
|
|
|
|
{{0xef, 0xbf}, {0xfffd}},
|
|
|
|
// Start of an overlong sequence. The first "maximal subpart" is the first
|
|
|
|
// byte; it creates an invalid character. Each following byte generates an
|
|
|
|
// invalid character too.
|
|
|
|
{{0xe0, 0x80}, {0xfffd, 0xfffd}},
|
|
|
|
// 4-byte sequences with last byte missing
|
|
|
|
{{0xf1, 0x80, 0x80}, {0xfffd}},
|
|
|
|
{{0xf4, 0x8f, 0xbf}, {0xfffd}},
|
|
|
|
// Start of an overlong sequence. The first "maximal subpart" is the first
|
|
|
|
// byte; it creates an invalid character. Each following byte generates an
|
|
|
|
// invalid character too.
|
|
|
|
{{0xf0, 0x80, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5-byte sequences (not supported) with last byte missing
|
|
|
|
{{0xf8, 0x80, 0x80, 0x80}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfb, 0xbf, 0xbf, 0xbf}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6-byte sequences (not supported) with last byte missing
|
|
|
|
{{0xfc, 0x80, 0x80, 0x80, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfd, 0xbf, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Concatenation of incomplete sequences: above incomplete sequences
|
|
|
|
// concatenated.
|
|
|
|
{{0xc0, 0xdf, 0xe8, 0x80, 0xe0, 0xbf, 0xef, 0xbf, 0xe0, 0x80,
|
|
|
|
0xf1, 0x80, 0x80, 0xf4, 0x8f, 0xbf, 0xf0, 0x80, 0x80, 0xf8,
|
|
|
|
0x80, 0x80, 0x80, 0xfb, 0xbf, 0xbf, 0xbf, 0xfc, 0x80, 0x80,
|
|
|
|
0x80, 0x80, 0xfd, 0xbf, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd,
|
|
|
|
0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Incomplete sequence tests repeated with a space after the incomplete
|
|
|
|
// sequence.
|
|
|
|
|
|
|
|
// 2-byte sequences with last byte missing
|
|
|
|
{{0xc0, 0x20}, {0xfffd, 0x20}},
|
|
|
|
{{0xdf, 0x20}, {0xfffd, 0x20}},
|
|
|
|
// 3-byte sequences with last byte missing
|
|
|
|
{{0xe8, 0x80, 0x20}, {0xfffd, 0x20}},
|
|
|
|
{{0xe0, 0xbf, 0x20}, {0xfffd, 0x20}},
|
|
|
|
{{0xef, 0xbf, 0x20}, {0xfffd, 0x20}},
|
|
|
|
// Start of overlong 3-byte sequence with last byte missing
|
|
|
|
{{0xe0, 0x80, 0x20}, {0xfffd, 0xfffd, 0x20}},
|
|
|
|
// 4-byte sequences with last byte missing
|
|
|
|
{{0xf1, 0x80, 0x80, 0x20}, {0xfffd, 0x20}},
|
|
|
|
{{0xf4, 0x8f, 0xbf, 0x20}, {0xfffd, 0x20}},
|
|
|
|
// Start of overlong 4-byte sequence with last byte missing
|
|
|
|
{{0xf0, 0x80, 0x80, 0x20}, {0xfffd, 0xfffd, 0xfffd, 0x20}},
|
|
|
|
// 5-byte sequences (not supported) with last byte missing
|
|
|
|
{{0xf8, 0x80, 0x80, 0x80, 0x20}, {0xfffd, 0xfffd, 0xfffd, 0xfffd, 0x20}},
|
|
|
|
{{0xfb, 0xbf, 0xbf, 0xbf, 0x20}, {0xfffd, 0xfffd, 0xfffd, 0xfffd, 0x20}},
|
|
|
|
// 6-byte sequences (not supported) with last byte missing
|
|
|
|
{{0xfc, 0x80, 0x80, 0x80, 0x80, 0x20},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0x20}},
|
|
|
|
{{0xfd, 0xbf, 0xbf, 0xbf, 0xbf, 0x20},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0x20}},
|
|
|
|
|
|
|
|
// Impossible bytes
|
|
|
|
{{0xfe}, {0xfffd}},
|
|
|
|
{{0xff}, {0xfffd}},
|
|
|
|
{{0xfe, 0xfe, 0xff, 0xff}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// Lead-byte-like bytes which aren't valid lead bytes.
|
|
|
|
{{0xc0}, {0xfffd}},
|
|
|
|
{{0xc0, 0xaa}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xc1}, {0xfffd}},
|
|
|
|
{{0xc1, 0xaa}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xf5}, {0xfffd}},
|
|
|
|
{{0xf5, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf6}, {0xfffd}},
|
|
|
|
{{0xf6, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf7}, {0xfffd}},
|
|
|
|
{{0xf7, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf8}, {0xfffd}},
|
|
|
|
{{0xf8, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf9}, {0xfffd}},
|
|
|
|
{{0xf9, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfa}, {0xfffd}},
|
|
|
|
{{0xfa, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfb}, {0xfffd}},
|
|
|
|
{{0xfb, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfc}, {0xfffd}},
|
|
|
|
{{0xfc, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfd}, {0xfffd}},
|
|
|
|
{{0xfd, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xfe}, {0xfffd}},
|
|
|
|
{{0xfe, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xff}, {0xfffd}},
|
|
|
|
{{0xff, 0xaa, 0xaa, 0xaa}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Overlong sequences:
|
|
|
|
|
|
|
|
// Overlong encodings for "/"
|
|
|
|
{{0xc0, 0xaf}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xe0, 0x80, 0xaf}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf0, 0x80, 0x80, 0xaf}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5-byte sequence (not supported anyway)
|
|
|
|
{{0xf8, 0x80, 0x80, 0x80, 0xaf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6-byte sequence (not supported anyway)
|
|
|
|
{{0xfc, 0x80, 0x80, 0x80, 0x80, 0xaf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Maximum overlong sequences
|
|
|
|
{{0xc1, 0xbf}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xe0, 0x9f, 0xbf}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf0, 0x8f, 0xbf, 0xbf}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5-byte sequence (not supported anyway)
|
|
|
|
{{0xf8, 0x87, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6-byte sequence (not supported anyway)
|
|
|
|
{{0xfc, 0x83, 0xbf, 0xbf, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Overlong encodings for 0
|
|
|
|
{{0xc0, 0x80}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xe0, 0x80, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xf0, 0x80, 0x80, 0x80}, {0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 5-byte sequence (not supported anyway)
|
|
|
|
{{0xf8, 0x80, 0x80, 0x80, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
// 6-byte sequence (not supported anyway)
|
|
|
|
{{0xfc, 0x80, 0x80, 0x80, 0x80, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Illegal code positions:
|
|
|
|
|
|
|
|
// Single UTF-16 surrogates
|
|
|
|
{{0xed, 0xa0, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xa0, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xad, 0xbf}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xae, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xaf, 0xbf}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xb0, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xbe, 0x80}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xbf, 0xbf}, {0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Paired surrogates
|
|
|
|
{{0xed, 0xa0, 0x80, 0xed, 0xb0, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xa0, 0x80, 0xed, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xad, 0xbf, 0xed, 0xb0, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xad, 0xbf, 0xed, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xae, 0x80, 0xed, 0xb0, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xae, 0x80, 0xed, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xaf, 0xbf, 0xed, 0xb0, 0x80},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xaf, 0xbf, 0xed, 0xbf, 0xbf},
|
|
|
|
{0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Surrogates with the last byte missing.
|
|
|
|
{{0xed, 0xa0}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xa0}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xad}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xae}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xaf}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xb0}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xbe}, {0xfffd, 0xfffd}},
|
|
|
|
{{0xed, 0xbf}, {0xfffd, 0xfffd}},
|
|
|
|
|
|
|
|
// Other non-characters
|
|
|
|
{{0xef, 0xbf, 0xbe}, {0xfffe}},
|
|
|
|
{{0xef, 0xbf, 0xbf}, {0xffff}},
|
|
|
|
{{0xef, 0xb7, 0x90, 0xef, 0xb7, 0x91, 0xef, 0xb7, 0x92, 0xef, 0xb7, 0x93,
|
|
|
|
0xef, 0xb7, 0x94, 0xef, 0xb7, 0x95, 0xef, 0xb7, 0x96, 0xef, 0xb7, 0x97,
|
|
|
|
0xef, 0xb7, 0x98, 0xef, 0xb7, 0x99, 0xef, 0xb7, 0x9a, 0xef, 0xb7, 0x9b,
|
|
|
|
0xef, 0xb7, 0x9c, 0xef, 0xb7, 0x9d, 0xef, 0xb7, 0x9e, 0xef, 0xb7, 0x9f,
|
|
|
|
0xef, 0xb7, 0xa0, 0xef, 0xb7, 0xa1, 0xef, 0xb7, 0xa2, 0xef, 0xb7, 0xa3,
|
|
|
|
0xef, 0xb7, 0xa4, 0xef, 0xb7, 0xa5, 0xef, 0xb7, 0xa6, 0xef, 0xb7, 0xa7,
|
|
|
|
0xef, 0xb7, 0xa8, 0xef, 0xb7, 0xa9, 0xef, 0xb7, 0xaa, 0xef, 0xb7, 0xab,
|
|
|
|
0xef, 0xb7, 0xac, 0xef, 0xb7, 0xad, 0xef, 0xb7, 0xae, 0xef, 0xb7, 0xaf},
|
|
|
|
{0xfdd0, 0xfdd1, 0xfdd2, 0xfdd3, 0xfdd4, 0xfdd5, 0xfdd6, 0xfdd7,
|
|
|
|
0xfdd8, 0xfdd9, 0xfdda, 0xfddb, 0xfddc, 0xfddd, 0xfdde, 0xfddf,
|
|
|
|
0xfde0, 0xfde1, 0xfde2, 0xfde3, 0xfde4, 0xfde5, 0xfde6, 0xfde7,
|
|
|
|
0xfde8, 0xfde9, 0xfdea, 0xfdeb, 0xfdec, 0xfded, 0xfdee, 0xfdef}},
|
|
|
|
{{0xf0, 0x9f, 0xbf, 0xbe, 0xf0, 0x9f, 0xbf, 0xbf, 0xf0, 0xaf, 0xbf,
|
|
|
|
0xbe, 0xf0, 0xaf, 0xbf, 0xbf, 0xf0, 0xbf, 0xbf, 0xbe, 0xf0, 0xbf,
|
|
|
|
0xbf, 0xbf, 0xf1, 0x8f, 0xbf, 0xbe, 0xf1, 0x8f, 0xbf, 0xbf, 0xf1,
|
|
|
|
0x9f, 0xbf, 0xbe, 0xf1, 0x9f, 0xbf, 0xbf, 0xf1, 0xaf, 0xbf, 0xbe,
|
|
|
|
0xf1, 0xaf, 0xbf, 0xbf, 0xf1, 0xbf, 0xbf, 0xbe, 0xf1, 0xbf, 0xbf,
|
|
|
|
0xbf, 0xf2, 0x8f, 0xbf, 0xbe, 0xf2, 0x8f, 0xbf, 0xbf},
|
|
|
|
{0x1fffe, 0x1ffff, 0x2fffe, 0x2ffff, 0x3fffe, 0x3ffff, 0x4fffe, 0x4ffff,
|
|
|
|
0x5fffe, 0x5ffff, 0x6fffe, 0x6ffff, 0x7fffe, 0x7ffff, 0x8fffe,
|
|
|
|
0x8ffff}},
|
|
|
|
};
|
|
|
|
|
|
|
|
for (auto test : data) {
|
|
|
|
// For figuring out which test fails:
|
|
|
|
fprintf(stderr, "test: ");
|
|
|
|
for (auto b : test.bytes) {
|
|
|
|
fprintf(stderr, "%x ", b);
|
|
|
|
}
|
|
|
|
fprintf(stderr, "\n");
|
|
|
|
|
|
|
|
std::vector<unibrow::uchar> output_normal;
|
|
|
|
DecodeNormally(test.bytes, &output_normal);
|
|
|
|
|
|
|
|
CHECK_EQ(output_normal.size(), test.unicode_expected.size());
|
|
|
|
for (size_t i = 0; i < output_normal.size(); ++i) {
|
|
|
|
CHECK_EQ(output_normal[i], test.unicode_expected[i]);
|
|
|
|
}
|
|
|
|
|
|
|
|
std::vector<unibrow::uchar> output_incremental;
|
|
|
|
DecodeIncrementally(test.bytes, &output_incremental);
|
|
|
|
|
|
|
|
CHECK_EQ(output_incremental.size(), test.unicode_expected.size());
|
|
|
|
for (size_t i = 0; i < output_incremental.size(); ++i) {
|
|
|
|
CHECK_EQ(output_incremental[i], test.unicode_expected[i]);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-11-22 09:27:41 +00:00
|
|
|
} // namespace internal
|
|
|
|
} // namespace v8
|