/*
* Copyright (c) 2022, Michiel Visser <opensource@webmichiel.nl>
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#include <AK/BinarySearch.h>
#include <AK/QuickSort.h>
#include <LibCompress/Brotli.h>
#include <LibCompress/BrotliDictionary.h>
namespace Compress {
ErrorOr<size_t> BrotliDecompressionStream::CanonicalCode::read_symbol(LittleEndianInputBitStream& input_stream)
{
size_t code_bits = 1;
while (code_bits < (1 << 16)) {
// FIXME: This is very inefficient and could greatly be improved by implementing this
// algorithm: https://www.hanshq.net/zip.html#huffdec
size_t index;
if (binary_search(m_symbol_codes.span(), code_bits, &index))
return m_symbol_values[index];
code_bits = (code_bits << 1) | TRY(input_stream.read_bit());
}
return Error::from_string_literal("no matching code found");
}
BrotliDecompressionStream::BrotliDecompressionStream(Stream& stream)
: m_input_stream(MaybeOwned(stream))
{
}
ErrorOr<size_t> BrotliDecompressionStream::read_window_length()
{
if (TRY(m_input_stream.read_bit())) {
switch (TRY(m_input_stream.read_bits(3))) {
case 0: {
switch (TRY(m_input_stream.read_bits(3))) {
case 0:
return 17;
case 1:
return Error::from_string_literal("invalid window length");
case 2:
return 10;
case 3:
return 11;
case 4:
return 12;
case 5:
return 13;
case 6:
return 14;
case 7:
return 15;
default:
VERIFY_NOT_REACHED();
}
}
case 1:
return 18;
case 2:
return 19;
case 3:
return 20;
case 4:
return 21;
case 5:
return 22;
case 6:
return 23;
case 7:
return 24;
default:
VERIFY_NOT_REACHED();
}
} else {
return 16;
}
}
ErrorOr<size_t> BrotliDecompressionStream::read_size_number_of_nibbles()
{
switch (TRY(m_input_stream.read_bits(2))) {
case 0:
return 4;
case 1:
return 5;
case 2:
return 6;
case 3:
return 0;
default:
VERIFY_NOT_REACHED();
}
}
ErrorOr<size_t> BrotliDecompressionStream::read_variable_length()
{
// Value Bit Pattern
// ----- -----------
// 1 0
// 2 0001
// 3..4 x0011
// 5..8 xx0101
// 9..16 xxx0111
// 17..32 xxxx1001
// 33..64 xxxxx1011
// 65..128 xxxxxx1101
// 129..256 xxxxxxx1111
if (TRY(m_input_stream.read_bit())) {
switch (TRY(m_input_stream.read_bits(3))) {
case 0:
return 2;
case 1:
return 3 + TRY(m_input_stream.read_bits(1));
case 2:
return 5 + TRY(m_input_stream.read_bits(2));
case 3:
return 9 + TRY(m_input_stream.read_bits(3));
case 4:
return 17 + TRY(m_input_stream.read_bits(4));
case 5:
return 33 + TRY(m_input_stream.read_bits(5));
case 6:
return 65 + TRY(m_input_stream.read_bits(6));
case 7:
return 129 + TRY(m_input_stream.read_bits(7));
default:
VERIFY_NOT_REACHED();
}
} else {
return 1;
}
}
ErrorOr<size_t> BrotliDecompressionStream::read_complex_prefix_code_length()
{
// Symbol Code
// ------ ----
// 0 00
// 1 0111
// 2 011
// 3 10
// 4 01
// 5 1111
switch (TRY(m_input_stream.read_bits(2))) {
case 0:
return 0;
case 1:
return 4;
case 2:
return 3;
case 3: {
if (TRY(m_input_stream.read_bit()) == 0) {
return 2;
} else {
if (TRY(m_input_stream.read_bit()) == 0) {
return 1;
} else {
return 5;
}
}
}
default:
VERIFY_NOT_REACHED();
}
}
ErrorOr<void> BrotliDecompressionStream::read_prefix_code(CanonicalCode& code, size_t alphabet_size)
{
size_t hskip = TRY(m_input_stream.read_bits(2));
if (hskip == 1) {
TRY(read_simple_prefix_code(code, alphabet_size));
} else {
TRY(read_complex_prefix_code(code, alphabet_size, hskip));
}
return {};
}
ErrorOr<void> BrotliDecompressionStream::read_simple_prefix_code(CanonicalCode& code, size_t alphabet_size)
{
VERIFY(code.m_symbol_codes.is_empty());
VERIFY(code.m_symbol_values.is_empty());
size_t number_of_symbols = 1 + TRY(m_input_stream.read_bits(2));
size_t symbol_size = 0;
while ((1u << symbol_size) < alphabet_size)
symbol_size++;
Vector<size_t> symbols;
for (size_t i = 0; i < number_of_symbols; i++) {
size_t symbol = TRY(m_input_stream.read_bits(symbol_size));
symbols.append(symbol);
if (symbol >= alphabet_size)
return Error::from_string_literal("symbol larger than alphabet");
}
if (number_of_symbols == 1) {
code.m_symbol_codes.append(0b1);
code.m_symbol_values = move(symbols);
} else if (number_of_symbols == 2) {
code.m_symbol_codes.extend({ 0b10, 0b11 });
if (symbols[0] > symbols[1])
swap(symbols[0], symbols[1]);
code.m_symbol_values = move(symbols);
} else if (number_of_symbols == 3) {
code.m_symbol_codes.extend({ 0b10, 0b110, 0b111 });
if (symbols[1] > symbols[2])
swap(symbols[1], symbols[2]);
code.m_symbol_values = move(symbols);
} else if (number_of_symbols == 4) {
bool tree_select = TRY(m_input_stream.read_bit());
if (tree_select) {
code.m_symbol_codes.extend({ 0b10, 0b110, 0b1110, 0b1111 });
if (symbols[2] > symbols[3])
swap(symbols[2], symbols[3]);
code.m_symbol_values = move(symbols);
} else {
code.m_symbol_codes.extend({ 0b100, 0b101, 0b110, 0b111 });
quick_sort(symbols);
code.m_symbol_values = move(symbols);
}
}
return {};
}
ErrorOr<void> BrotliDecompressionStream::read_complex_prefix_code(CanonicalCode& code, size_t alphabet_size, size_t hskip)
{
// hskip should only be 0, 2 or 3
VERIFY(hskip != 1);
VERIFY(hskip <= 3);
// Read the prefix code_value that is used to encode the actual prefix code_value
size_t const symbol_mapping[18] = { 1, 2, 3, 4, 0, 5, 17, 6, 16, 7, 8, 9, 10, 11, 12, 13, 14, 15 };
size_t code_length[18] { 0 };
size_t code_length_counts[6] { 0 };
size_t sum = 0;
size_t number_of_non_zero_symbols = 0;
for (size_t i = hskip; i < 18; i++) {
size_t len = TRY(read_complex_prefix_code_length());
code_length[symbol_mapping[i]] = len;
if (len != 0) {
code_length_counts[len]++;
sum += (32 >> len);
number_of_non_zero_symbols++;
}
if (sum == 32)
break;
else if (sum > 32)
return Error::from_string_literal("invalid prefix code");
}
BrotliDecompressionStream::CanonicalCode temp_code;
if (number_of_non_zero_symbols > 1) {
size_t code_value = 0;
for (size_t bits = 1; bits <= 5; bits++) {
code_value = (code_value + code_length_counts[bits - 1]) << 1;
size_t current_code_value = code_value;
for (size_t i = 0; i < 18; i++) {
size_t len = code_length[i];
if (len == bits) {
temp_code.m_symbol_codes.append((1 << bits) | current_code_value);
temp_code.m_symbol_values.append(i);
current_code_value++;
}
}
}
} else {
for (size_t i = 0; i < 18; i++) {
size_t len = code_length[i];
if (len != 0) {
temp_code.m_symbol_codes.append(1);
temp_code.m_symbol_values.append(i);
break;
}
}
}
// Read the actual prefix code_value
sum = 0;
size_t i = 0;
size_t previous_non_zero_code_length = 8;
size_t last_symbol = 0;
size_t last_repeat = 0;
Vector<size_t> result_symbols;
Vector<size_t> result_lengths;
size_t result_lengths_count[16] { 0 };
while (i < alphabet_size) {
auto symbol = TRY(temp_code.read_symbol(m_input_stream));
if (symbol < 16) {
result_symbols.append(i);
result_lengths.append(symbol);
result_lengths_count[symbol]++;
if (symbol != 0) {
previous_non_zero_code_length = symbol;
sum += (32768 >> symbol);
if (sum == 32768)
break;
else if (sum > 32768)
return Error::from_string_literal("invalid prefix code");
}
last_repeat = 0;
i++;
} else if (symbol == 16) {
size_t repeat_count = 0;
if (last_symbol == 16 && last_repeat != 0) {
repeat_count = (4 * (last_repeat - 2));
} else {
last_repeat = 0;
}
repeat_count += 3 + TRY(m_input_stream.read_bits(2));
for (size_t rep = 0; rep < (repeat_count - last_repeat); rep++) {
result_symbols.append(i);
result_lengths.append(previous_non_zero_code_length);
result_lengths_count[previous_non_zero_code_length]++;
if (previous_non_zero_code_length != 0) {
sum += (32768 >> previous_non_zero_code_length);
if (sum == 32768)
break;
else if (sum > 32768)
return Error::from_string_literal("invalid prefix code");
}
i++;
if (i >= alphabet_size)
break;
}
if (sum == 32768)
break;
VERIFY(sum < 32768);
last_repeat = repeat_count;
} else if (symbol == 17) {
size_t repeat_count = 0;
if (last_symbol == 17 && last_repeat != 0) {
repeat_count = (8 * (last_repeat - 2));
} else {
last_repeat = 0;
}
repeat_count += 3 + TRY(m_input_stream.read_bits(3));
i += (repeat_count - last_repeat);
last_repeat = repeat_count;
}
last_symbol = symbol;
}
result_lengths_count[0] = 0;
size_t code_value = 0;
for (size_t bits = 1; bits < 16; bits++) {
code_value = (code_value + result_lengths_count[bits - 1]) << 1;
size_t current_code_value = code_value;
for (size_t n = 0; n < result_symbols.size(); n++) {
size_t len = result_lengths[n];
if (len == bits) {
code.m_symbol_codes.append((1 << bits) | current_code_value);
code.m_symbol_values.append(result_symbols[n]);
current_code_value++;
}
}
}
return {};
}
static void inverse_move_to_front_transform(Span<u8> v)
{
// RFC 7932 section 7.3
u8 mtf[256];
for (size_t i = 0; i < 256; ++i) {
mtf[i] = (u8)i;
}
for (size_t i = 0; i < v.size(); ++i) {
u8 index = v[i];
u8 value = mtf[index];
v[i] = value;
for (; index; --index) {
mtf[index] = mtf[index - 1];
}
mtf[0] = value;
}
}
ErrorOr<void> BrotliDecompressionStream::read_context_map(size_t number_of_codes, Vector<u8>& context_map, size_t context_map_size)
{
bool use_run_length_encoding = TRY(m_input_stream.read_bit());
size_t run_length_encoding_max = 0;
if (use_run_length_encoding) {
run_length_encoding_max = 1 + TRY(m_input_stream.read_bits(4));
}
BrotliDecompressionStream::CanonicalCode code;
TRY(read_prefix_code(code, number_of_codes + run_length_encoding_max));
size_t i = 0;
while (i < context_map_size) {
size_t symbol = TRY(code.read_symbol(m_input_stream));
if (symbol <= run_length_encoding_max) {
size_t repeat_base = 1 << symbol;
size_t repeat_additional = TRY(m_input_stream.read_bits(symbol));
size_t repeat_count = repeat_base + repeat_additional;
while (repeat_count--) {
context_map.append(0);
i++;
}
} else {
size_t value = symbol - run_length_encoding_max;
context_map.append(value);
i++;
}
}
bool inverse_move_to_front = TRY(m_input_stream.read_bit());
if (inverse_move_to_front)
inverse_move_to_front_transform(context_map.span());
return {};
}
ErrorOr<void> BrotliDecompressionStream::read_block_configuration(Block& block)
{
size_t blocks_of_type = TRY(read_variable_length());
block.type = 0;
block.type_previous = 1;
block.number_of_types = blocks_of_type;
block.type_code.clear();
block.length_code.clear();
if (blocks_of_type == 1) {
block.length = 16 * MiB;
} else {
TRY(read_prefix_code(block.type_code, 2 + blocks_of_type));
TRY(read_prefix_code(block.length_code, 26));
TRY(block_update_length(block));
}
return {};
}
ErrorOr<void> BrotliDecompressionStream::block_update_length(Block& block)
{
size_t const block_length_code_base[26] { 1, 5, 9, 13, 17, 25, 33, 41, 49, 65, 81, 97, 113, 145, 177, 209, 241, 305, 369, 497, 753, 1265, 2289, 4337, 8433, 16625 };
size_t const block_length_code_extra[26] { 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 7, 8, 9, 10, 11, 12, 13, 24 };
size_t symbol = TRY(block.length_code.read_symbol(m_input_stream));
size_t block_length = block_length_code_base[symbol] + TRY(m_input_stream.read_bits(block_length_code_extra[symbol]));
block.length = block_length;
return {};
}
ErrorOr<void> BrotliDecompressionStream::block_read_new_state(Block& block)
{
size_t block_type_symbol = TRY(block.type_code.read_symbol(m_input_stream));
TRY(block_update_length(block));
if (block_type_symbol == 0) {
swap(block.type, block.type_previous);
} else if (block_type_symbol == 1) {
block.type_previous = block.type;
block.type = (block.type + 1) % block.number_of_types;
} else {
block.type_previous = block.type;
block.type = block_type_symbol - 2;
}
return {};
}
size_t BrotliDecompressionStream::literal_code_index_from_context()
{
size_t const context_id_lut0[256] {
0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 4, 0, 0, 4, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
8, 12, 16, 12, 12, 20, 12, 16, 24, 28, 12, 12, 32, 12, 36, 12,
44, 44, 44, 44, 44, 44, 44, 44, 44, 44, 32, 32, 24, 40, 28, 12,
12, 48, 52, 52, 52, 48, 52, 52, 52, 48, 52, 52, 52, 52, 52, 48,
52, 52, 52, 52, 52, 48, 52, 52, 52, 52, 52, 24, 12, 28, 12, 12,
12, 56, 60, 60, 60, 56, 60, 60, 60, 56, 60, 60, 60, 60, 60, 56,
60, 60, 60, 60, 60, 56, 60, 60, 60, 60, 60, 24, 12, 28, 12, 0,
0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,
0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,
0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,
0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,
2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3,
2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3,
2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3,
2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3
};
size_t const context_id_lut1[256] {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1,
1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1,
1, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 1, 1, 1, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2
};
size_t const context_id_lut2[256] {
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 7
};
size_t context_mode = m_literal_context_modes[m_literal_block.type];
size_t context_id;
switch (context_mode) {
case 0:
context_id = m_lookback_buffer.value().lookback(1, 0) & 0x3f;
break;
case 1:
context_id = m_lookback_buffer.value().lookback(1, 0) >> 2;
break;
case 2:
context_id = context_id_lut0[m_lookback_buffer.value().lookback(1, 0)] | context_id_lut1[m_lookback_buffer.value().lookback(2, 0)];
break;
case 3:
context_id = (context_id_lut2[m_lookback_buffer.value().lookback(1, 0)] << 3) | context_id_lut2[m_lookback_buffer.value().lookback(2, 0)];
break;
default:
VERIFY_NOT_REACHED();
}
size_t literal_code_index = m_context_mapping_literal[64 * m_literal_block.type + context_id];
return literal_code_index;
}
ErrorOr<Bytes> BrotliDecompressionStream::read_some(Bytes output_buffer)
{
size_t bytes_read = 0;
while (bytes_read < output_buffer.size()) {
if (m_current_state == State::WindowSize) {
size_t window_bits = TRY(read_window_length());
m_window_size = (1 << window_bits) - 16;
m_lookback_buffer = TRY(LookbackBuffer::try_create(m_window_size));
m_current_state = State::Idle;
} else if (m_current_state == State::Idle) {
// If the final block was read, we are done decompressing
if (m_read_final_block)
break;
// RFC 7932 section 9.1
//
// 1 bit: ISLAST, set to 1 if this is the last meta-block
m_read_final_block = TRY(m_input_stream.read_bit());
if (m_read_final_block) {
// 1 bit: ISLASTEMPTY, if set to 1, the meta-block is empty; this
// field is only present if ISLAST bit is set -- if it is 1,
// then the meta-block and the brotli stream ends at that
// bit, with any remaining bits in the last byte of the
// compressed stream filled with zeros (if the fill bits are
// not zero, then the stream should be rejected as invalid)
bool is_last_block_empty = TRY(m_input_stream.read_bit());
// If the last block is empty we are done decompressing
if (is_last_block_empty)
break;
}
// 2 bits: MNIBBLES, number of nibbles to represent the uncompressed
// length
size_t size_number_of_nibbles = TRY(read_size_number_of_nibbles());
// If MNIBBLES is 0, the meta-block is empty, i.e., it does
// not generate any uncompressed data. In this case, the
// rest of the meta-block has the following format:
if (size_number_of_nibbles == 0) {
// 1 bit: reserved, must be zero
bool reserved = TRY(m_input_stream.read_bit());
if (reserved)
return Error::from_string_literal("invalid reserved bit");
// 2 bits: MSKIPBYTES, number of bytes to represent
// metadata length
//
// MSKIPBYTES * 8 bits: MSKIPLEN - 1, where MSKIPLEN is
// the number of metadata bytes; this field is
// only present if MSKIPBYTES is positive;
// otherwise, MSKIPLEN is 0 (if MSKIPBYTES is
// greater than 1, and the last byte is all
// zeros, then the stream should be rejected as
// invalid)
size_t skip_bytes = TRY(m_input_stream.read_bits(2));
if (skip_bytes == 0) {
// 0..7 bits: fill bits until the next byte boundary,
// must be all zeros
u8 remainder = m_input_stream.align_to_byte_boundary();
if (remainder != 0)
return Error::from_string_literal("remainder bits are non-zero");
continue;
}
// MSKIPLEN bytes of metadata, not part of the
// uncompressed data or the sliding window
size_t skip_length = 1 + TRY(m_input_stream.read_bits(8 * skip_bytes));
u8 remainder = m_input_stream.align_to_byte_boundary();
if (remainder != 0)
return Error::from_string_literal("remainder bits are non-zero");
// Discard meta-data bytes
u8 temp_buffer[4096];
Bytes temp_bytes { temp_buffer, 4096 };
while (skip_length > 0) {
Bytes temp_bytes_slice = temp_bytes.slice(0, min(4096, skip_length));
auto metadata_bytes = TRY(m_input_stream.read_some(temp_bytes_slice));
if (metadata_bytes.is_empty())
return Error::from_string_literal("eof");
if (metadata_bytes.last() == 0)
return Error::from_string_literal("invalid stream");
skip_length -= metadata_bytes.size();
}
continue;
}
size_t uncompressed_size = 1 + TRY(m_input_stream.read_bits(4 * size_number_of_nibbles));
// 1 bit: ISUNCOMPRESSED, if set to 1, any bits of compressed data
// up to the next byte boundary are ignored, and the rest of
// the meta-block contains MLEN bytes of literal data; this
// field is only present if the ISLAST bit is not set (if the
// ignored bits are not all zeros, the stream should be
// rejected as invalid)
bool is_uncompressed = false;
if (!m_read_final_block)
is_uncompressed = TRY(m_input_stream.read_bit());
m_bytes_left = uncompressed_size;
if (is_uncompressed) {
u8 remainder = m_input_stream.align_to_byte_boundary();
if (remainder != 0)
return Error::from_string_literal("remainder is non-zero");
m_current_state = State::UncompressedData;
} else {
TRY(read_block_configuration(m_literal_block));
TRY(read_block_configuration(m_insert_and_copy_block));
TRY(read_block_configuration(m_distance_block));
m_postfix_bits = TRY(m_input_stream.read_bits(2));
m_direct_distances = TRY(m_input_stream.read_bits(4)) << m_postfix_bits;
m_literal_context_modes.clear();
for (size_t i = 0; i < m_literal_block.number_of_types; i++) {
size_t context_mode = TRY(m_input_stream.read_bits(2));
m_literal_context_modes.append(context_mode);
}
m_context_mapping_literal.clear();
size_t number_of_literal_codes = TRY(read_variable_length());
if (number_of_literal_codes == 1) {
for (size_t i = 0; i < 64 * m_literal_block.number_of_types; i++)
m_context_mapping_literal.append(0);
} else {
TRY(read_context_map(number_of_literal_codes, m_context_mapping_literal, 64 * m_literal_block.number_of_types));
}
m_context_mapping_distance.clear();
size_t number_of_distance_codes = TRY(read_variable_length());
if (number_of_distance_codes == 1) {
for (size_t i = 0; i < 4 * m_distance_block.number_of_types; i++)
m_context_mapping_distance.append(0);
} else {
TRY(read_context_map(number_of_distance_codes, m_context_mapping_distance, 4 * m_distance_block.number_of_types));
}
m_literal_codes.clear();
for (size_t i = 0; i < number_of_literal_codes; i++) {
CanonicalCode code;
TRY(read_prefix_code(code, 256));
m_literal_codes.append(move(code));
}
m_insert_and_copy_codes.clear();
for (size_t i = 0; i < m_insert_and_copy_block.number_of_types; i++) {
CanonicalCode code;
TRY(read_prefix_code(code, 704));
m_insert_and_copy_codes.append(move(code));
}
m_distance_codes.clear();
for (size_t i = 0; i < number_of_distance_codes; i++) {
CanonicalCode code;
TRY(read_prefix_code(code, 16 + m_direct_distances + (48 << m_postfix_bits)));
m_distance_codes.append(move(code));
}
m_current_state = State::CompressedCommand;
}
} else if (m_current_state == State::UncompressedData) {
size_t number_of_fitting_bytes = min(output_buffer.size() - bytes_read, m_bytes_left);
VERIFY(number_of_fitting_bytes > 0);
auto uncompressed_bytes = TRY(m_input_stream.read_some(output_buffer.slice(bytes_read, number_of_fitting_bytes)));
if (uncompressed_bytes.is_empty())
return Error::from_string_literal("eof");
m_bytes_left -= uncompressed_bytes.size();
bytes_read += uncompressed_bytes.size();
// If all bytes were read, return to the idle state
if (m_bytes_left == 0)
m_current_state = State::Idle;
} else if (m_current_state == State::CompressedCommand) {
if (m_insert_and_copy_block.length == 0) {
TRY(block_read_new_state(m_insert_and_copy_block));
}
m_insert_and_copy_block.length--;
size_t insert_and_copy_symbol = TRY(m_insert_and_copy_codes[m_insert_and_copy_block.type].read_symbol(m_input_stream));
size_t const insert_length_code_base[11] { 0, 0, 0, 0, 8, 8, 0, 16, 8, 16, 16 };
size_t const copy_length_code_base[11] { 0, 8, 0, 8, 0, 8, 16, 0, 16, 8, 16 };
bool const implicit_zero_distance[11] { true, true, false, false, false, false, false, false, false, false, false };
size_t insert_and_copy_index = insert_and_copy_symbol >> 6;
size_t insert_length_code_offset = (insert_and_copy_symbol >> 3) & 0b111;
size_t copy_length_code_offset = insert_and_copy_symbol & 0b111;
size_t insert_length_code = insert_length_code_base[insert_and_copy_index] + insert_length_code_offset;
size_t copy_length_code = copy_length_code_base[insert_and_copy_index] + copy_length_code_offset;
m_implicit_zero_distance = implicit_zero_distance[insert_and_copy_index];
size_t const insert_length_base[24] { 0, 1, 2, 3, 4, 5, 6, 8, 10, 14, 18, 26, 34, 50, 66, 98, 130, 194, 322, 578, 1090, 2114, 6210, 22594 };
size_t const insert_length_extra[24] { 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 7, 8, 9, 10, 12, 14, 24 };
size_t const copy_length_base[24] { 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 18, 22, 30, 38, 54, 70, 102, 134, 198, 326, 582, 1094, 2118 };
size_t const copy_length_extra[24] { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 7, 8, 9, 10, 24 };
m_insert_length = insert_length_base[insert_length_code] + TRY(m_input_stream.read_bits(insert_length_extra[insert_length_code]));
m_copy_length = copy_length_base[copy_length_code] + TRY(m_input_stream.read_bits(copy_length_extra[copy_length_code]));
if (m_insert_length > 0) {
m_current_state = State::CompressedLiteral;
} else {
m_current_state = State::CompressedDistance;
}
} else if (m_current_state == State::CompressedLiteral) {
if (m_literal_block.length == 0) {
TRY(block_read_new_state(m_literal_block));
}
m_literal_block.length--;
size_t literal_code_index = literal_code_index_from_context();
size_t literal_value = TRY(m_literal_codes[literal_code_index].read_symbol(m_input_stream));
output_buffer[bytes_read] = literal_value;
m_lookback_buffer.value().write(literal_value);
bytes_read++;
m_insert_length--;
m_bytes_left--;
if (m_bytes_left == 0)
m_current_state = State::Idle;
else if (m_insert_length == 0)
m_current_state = State::CompressedDistance;
} else if (m_current_state == State::CompressedDistance) {
size_t distance_symbol;
if (m_implicit_zero_distance) {
distance_symbol = 0;
} else {
if (m_distance_block.length == 0) {
TRY(block_read_new_state(m_distance_block));
}
m_distance_block.length--;
size_t context_id = clamp(m_copy_length - 2, 0, 3);
size_t distance_code_index = m_context_mapping_distance[4 * m_distance_block.type + context_id];
distance_symbol = TRY(m_distance_codes[distance_code_index].read_symbol(m_input_stream));
}
size_t distance;
bool reuse_previous_distance = false;
if (distance_symbol < 16) {
switch (distance_symbol) {
case 0:
distance = m_distances[0];
reuse_previous_distance = true;
break;
case 1:
distance = m_distances[1];
break;
case 2:
distance = m_distances[2];
break;
case 3:
distance = m_distances[3];
break;
case 4:
distance = m_distances[0] - 1;
break;
case 5:
distance = m_distances[0] + 1;
break;
case 6:
distance = m_distances[0] - 2;
break;
case 7:
distance = m_distances[0] + 2;
break;
case 8:
distance = m_distances[0] - 3;
break;
case 9:
distance = m_distances[0] + 3;
break;
case 10:
distance = m_distances[1] - 1;
break;
case 11:
distance = m_distances[1] + 1;
break;
case 12:
distance = m_distances[1] - 2;
break;
case 13:
distance = m_distances[1] + 2;
break;
case 14:
distance = m_distances[1] - 3;
break;
case 15:
distance = m_distances[1] + 3;
break;
}
} else if (distance_symbol < 16 + m_direct_distances) {
distance = distance_symbol - 15;
} else {
size_t POSTFIX_MASK = (1 << m_postfix_bits) - 1;
size_t ndistbits = 1 + ((distance_symbol - m_direct_distances - 16) >> (m_postfix_bits + 1));
size_t dextra = TRY(m_input_stream.read_bits(ndistbits));
size_t hcode = (distance_symbol - m_direct_distances - 16) >> m_postfix_bits;
size_t lcode = (distance_symbol - m_direct_distances - 16) & POSTFIX_MASK;
size_t offset = ((2 + (hcode & 1)) << ndistbits) - 4;
distance = ((offset + dextra) << m_postfix_bits) + lcode + m_direct_distances + 1;
}
m_distance = distance;
size_t total_written = m_lookback_buffer.value().total_written();
size_t max_lookback = min(total_written, m_window_size);
if (distance > max_lookback) {
size_t word_index = distance - (max_lookback + 1);
m_dictionary_data = TRY(BrotliDictionary::lookup_word(word_index, m_copy_length));
m_copy_length = m_dictionary_data.size();
if (m_copy_length == 0)
m_current_state = State::CompressedCommand;
else
m_current_state = State::CompressedDictionary;
} else {
if (!reuse_previous_distance) {
m_distances[3] = m_distances[2];
m_distances[2] = m_distances[1];
m_distances[1] = m_distances[0];
m_distances[0] = distance;
}
m_current_state = State::CompressedCopy;
}
} else if (m_current_state == State::CompressedCopy) {
u8 copy_value = m_lookback_buffer.value().lookback(m_distance);
output_buffer[bytes_read] = copy_value;
m_lookback_buffer.value().write(copy_value);
bytes_read++;
m_copy_length--;
m_bytes_left--;
if (m_bytes_left == 0)
m_current_state = State::Idle;
else if (m_copy_length == 0)
m_current_state = State::CompressedCommand;
} else if (m_current_state == State::CompressedDictionary) {
size_t offset = m_dictionary_data.size() - m_copy_length;
u8 dictionary_value = m_dictionary_data[offset];
output_buffer[bytes_read] = dictionary_value;
m_lookback_buffer.value().write(dictionary_value);
bytes_read++;
m_copy_length--;
m_bytes_left--;
if (m_bytes_left == 0)
m_current_state = State::Idle;
else if (m_copy_length == 0)
m_current_state = State::CompressedCommand;
}
}
return output_buffer.slice(0, bytes_read);
}
bool BrotliDecompressionStream::is_eof() const
{
return m_read_final_block && m_current_state == State::Idle;
}
}