mirror of
https://github.com/LadybirdBrowser/ladybird.git
synced 2024-11-22 07:30:19 +00:00
242 lines
8.8 KiB
C++
242 lines
8.8 KiB
C++
/*
|
|
* Copyright (c) 2021, kleines Filmröllchen <filmroellchen@serenityos.org>.
|
|
*
|
|
* SPDX-License-Identifier: BSD-2-Clause
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include <AK/ByteBuffer.h>
|
|
#include <AK/Concepts.h>
|
|
#include <AK/Error.h>
|
|
#include <AK/NonnullOwnPtr.h>
|
|
#include <AK/NonnullRefPtr.h>
|
|
#include <AK/OwnPtr.h>
|
|
#include <AK/Span.h>
|
|
#include <AK/StdLibExtraDetails.h>
|
|
#include <AK/Types.h>
|
|
#include <LibCore/Stream.h>
|
|
|
|
namespace Core::Stream {
|
|
|
|
/// A stream wrapper class that allows you to read arbitrary amounts of bits
|
|
/// in big-endian order from another stream.
|
|
/// Note that this stream does not own its underlying stream, it merely takes a reference.
|
|
class BigEndianInputBitStream : public Stream {
|
|
public:
|
|
static ErrorOr<NonnullOwnPtr<BigEndianInputBitStream>> construct(Stream& stream)
|
|
{
|
|
return adopt_nonnull_own_or_enomem<BigEndianInputBitStream>(new BigEndianInputBitStream(stream));
|
|
}
|
|
|
|
// ^Stream
|
|
virtual ErrorOr<Bytes> read(Bytes bytes) override
|
|
{
|
|
if (m_current_byte.has_value() && is_aligned_to_byte_boundary()) {
|
|
bytes[0] = m_current_byte.release_value();
|
|
return m_stream.read(bytes.slice(1));
|
|
}
|
|
align_to_byte_boundary();
|
|
return m_stream.read(bytes);
|
|
}
|
|
virtual ErrorOr<size_t> write(ReadonlyBytes bytes) override { return m_stream.write(bytes); }
|
|
virtual ErrorOr<void> write_entire_buffer(ReadonlyBytes bytes) override { return m_stream.write_entire_buffer(bytes); }
|
|
virtual bool is_eof() const override { return m_stream.is_eof() && !m_current_byte.has_value(); }
|
|
virtual bool is_open() const override { return m_stream.is_open(); }
|
|
virtual void close() override
|
|
{
|
|
m_stream.close();
|
|
align_to_byte_boundary();
|
|
}
|
|
|
|
ErrorOr<bool> read_bit()
|
|
{
|
|
return read_bits<bool>(1);
|
|
}
|
|
/// Depending on the number of bits to read, the return type can be chosen appropriately.
|
|
/// This avoids a bunch of static_cast<>'s for the user.
|
|
// TODO: Support u128, u256 etc. as well: The concepts would be quite complex.
|
|
template<Unsigned T = u64>
|
|
ErrorOr<T> read_bits(size_t count)
|
|
{
|
|
if constexpr (IsSame<bool, T>) {
|
|
VERIFY(count == 1);
|
|
}
|
|
T result = 0;
|
|
|
|
size_t nread = 0;
|
|
while (nread < count) {
|
|
if (m_current_byte.has_value()) {
|
|
if constexpr (!IsSame<bool, T> && !IsSame<u8, T>) {
|
|
// read as many bytes as possible directly
|
|
if (((count - nread) >= 8) && is_aligned_to_byte_boundary()) {
|
|
// shift existing data over
|
|
result <<= 8;
|
|
result |= m_current_byte.value();
|
|
nread += 8;
|
|
m_current_byte.clear();
|
|
} else {
|
|
auto const bit = (m_current_byte.value() >> (7 - m_bit_offset)) & 1;
|
|
result <<= 1;
|
|
result |= bit;
|
|
++nread;
|
|
if (m_bit_offset++ == 7)
|
|
m_current_byte.clear();
|
|
}
|
|
} else {
|
|
// Always take this branch for booleans or u8: there's no purpose in reading more than a single bit
|
|
auto const bit = (m_current_byte.value() >> (7 - m_bit_offset)) & 1;
|
|
if constexpr (IsSame<bool, T>)
|
|
result = bit;
|
|
else {
|
|
result <<= 1;
|
|
result |= bit;
|
|
}
|
|
++nread;
|
|
if (m_bit_offset++ == 7)
|
|
m_current_byte.clear();
|
|
}
|
|
} else {
|
|
auto temp_buffer = TRY(ByteBuffer::create_uninitialized(1));
|
|
TRY(m_stream.read(temp_buffer.bytes()));
|
|
m_current_byte = temp_buffer[0];
|
|
m_bit_offset = 0;
|
|
}
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
/// Discards any sub-byte stream positioning the input stream may be keeping track of.
|
|
/// Non-bitwise reads will implicitly call this.
|
|
void align_to_byte_boundary()
|
|
{
|
|
m_current_byte.clear();
|
|
m_bit_offset = 0;
|
|
}
|
|
|
|
/// Whether we are (accidentally or intentionally) at a byte boundary right now.
|
|
ALWAYS_INLINE bool is_aligned_to_byte_boundary() const { return m_bit_offset == 0; }
|
|
|
|
private:
|
|
BigEndianInputBitStream(Stream& stream)
|
|
: m_stream(stream)
|
|
{
|
|
}
|
|
|
|
Optional<u8> m_current_byte;
|
|
size_t m_bit_offset { 0 };
|
|
Stream& m_stream;
|
|
};
|
|
|
|
/// A stream wrapper class that allows you to read arbitrary amounts of bits
|
|
/// in little-endian order from another stream.
|
|
/// Note that this stream does not own its underlying stream, it merely takes a reference.
|
|
class LittleEndianInputBitStream : public Stream {
|
|
public:
|
|
static ErrorOr<NonnullOwnPtr<LittleEndianInputBitStream>> construct(Stream& stream)
|
|
{
|
|
return adopt_nonnull_own_or_enomem<LittleEndianInputBitStream>(new LittleEndianInputBitStream(stream));
|
|
}
|
|
|
|
LittleEndianInputBitStream(Stream& stream)
|
|
: m_stream(stream)
|
|
{
|
|
}
|
|
|
|
// ^Stream
|
|
virtual ErrorOr<Bytes> read(Bytes bytes) override
|
|
{
|
|
if (m_current_byte.has_value() && is_aligned_to_byte_boundary()) {
|
|
bytes[0] = m_current_byte.release_value();
|
|
return m_stream.read(bytes.slice(1));
|
|
}
|
|
align_to_byte_boundary();
|
|
return m_stream.read(bytes);
|
|
}
|
|
virtual ErrorOr<size_t> write(ReadonlyBytes bytes) override { return m_stream.write(bytes); }
|
|
virtual ErrorOr<void> write_entire_buffer(ReadonlyBytes bytes) override { return m_stream.write_entire_buffer(bytes); }
|
|
virtual bool is_eof() const override { return m_stream.is_eof() && !m_current_byte.has_value(); }
|
|
virtual bool is_open() const override { return m_stream.is_open(); }
|
|
virtual void close() override
|
|
{
|
|
m_stream.close();
|
|
align_to_byte_boundary();
|
|
}
|
|
|
|
ErrorOr<bool> read_bit()
|
|
{
|
|
return read_bits<bool>(1);
|
|
}
|
|
/// Depending on the number of bits to read, the return type can be chosen appropriately.
|
|
/// This avoids a bunch of static_cast<>'s for the user.
|
|
// TODO: Support u128, u256 etc. as well: The concepts would be quite complex.
|
|
template<Unsigned T = u64>
|
|
ErrorOr<T> read_bits(size_t count)
|
|
{
|
|
if constexpr (IsSame<bool, T>) {
|
|
VERIFY(count == 1);
|
|
}
|
|
T result = 0;
|
|
|
|
size_t nread = 0;
|
|
while (nread < count) {
|
|
if (m_current_byte.has_value()) {
|
|
if constexpr (!IsSame<bool, T> && !IsSame<u8, T>) {
|
|
// read as many bytes as possible directly
|
|
if (((count - nread) >= 8) && is_aligned_to_byte_boundary()) {
|
|
// shift existing data over
|
|
result |= (m_current_byte.value() << nread);
|
|
nread += 8;
|
|
m_current_byte.clear();
|
|
} else {
|
|
auto const bit = (m_current_byte.value() >> m_bit_offset) & 1;
|
|
result |= (bit << nread);
|
|
++nread;
|
|
if (m_bit_offset++ == 7)
|
|
m_current_byte.clear();
|
|
}
|
|
} else {
|
|
// Always take this branch for booleans or u8: there's no purpose in reading more than a single bit
|
|
auto const bit = (m_current_byte.value() >> m_bit_offset) & 1;
|
|
if constexpr (IsSame<bool, T>)
|
|
result = bit;
|
|
else
|
|
result |= (bit << nread);
|
|
++nread;
|
|
if (m_bit_offset++ == 7)
|
|
m_current_byte.clear();
|
|
}
|
|
} else {
|
|
auto temp_buffer = TRY(ByteBuffer::create_uninitialized(1));
|
|
auto read_bytes = TRY(m_stream.read(temp_buffer.bytes()));
|
|
if (read_bytes.is_empty())
|
|
return Error::from_string_literal("eof");
|
|
m_current_byte = temp_buffer[0];
|
|
m_bit_offset = 0;
|
|
}
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
/// Discards any sub-byte stream positioning the input stream may be keeping track of.
|
|
/// Non-bitwise reads will implicitly call this.
|
|
u8 align_to_byte_boundary()
|
|
{
|
|
u8 remaining_bits = m_current_byte.value_or(0) >> m_bit_offset;
|
|
m_current_byte.clear();
|
|
m_bit_offset = 0;
|
|
return remaining_bits;
|
|
}
|
|
|
|
/// Whether we are (accidentally or intentionally) at a byte boundary right now.
|
|
ALWAYS_INLINE bool is_aligned_to_byte_boundary() const { return m_bit_offset == 0; }
|
|
|
|
private:
|
|
Optional<u8> m_current_byte;
|
|
size_t m_bit_offset { 0 };
|
|
Stream& m_stream;
|
|
};
|
|
|
|
}
|