mirror of
https://github.com/LadybirdBrowser/ladybird.git
synced 2024-11-22 23:50:19 +00:00
d5871f5717
Similar to POSIX read, the basic read and write functions of AK::Stream do not have a lower limit of how much data they read or write (apart from "none at all"). Rename the functions to "read some [data]" and "write some [data]" (with "data" being omitted, since everything here is reading and writing data) to make them sufficiently distinct from the functions that ensure to use the entire buffer (which should be the go-to function for most usages). No functional changes, just a lot of new FIXMEs.
328 lines
11 KiB
C++
328 lines
11 KiB
C++
/*
|
|
* Copyright (c) 2021, sin-ack <sin-ack@protonmail.com>
|
|
* Copyright (c) 2022, the SerenityOS developers.
|
|
*
|
|
* SPDX-License-Identifier: BSD-2-Clause
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include <AK/CircularBuffer.h>
|
|
#include <AK/OwnPtr.h>
|
|
#include <AK/Stream.h>
|
|
|
|
namespace AK {
|
|
|
|
template<typename T>
|
|
concept StreamLike = IsBaseOf<Stream, T>;
|
|
template<typename T>
|
|
concept SeekableStreamLike = IsBaseOf<SeekableStream, T>;
|
|
|
|
template<typename T>
|
|
class BufferedHelper {
|
|
AK_MAKE_NONCOPYABLE(BufferedHelper);
|
|
|
|
public:
|
|
template<StreamLike U>
|
|
BufferedHelper(Badge<U>, NonnullOwnPtr<T> stream, CircularBuffer buffer)
|
|
: m_stream(move(stream))
|
|
, m_buffer(move(buffer))
|
|
{
|
|
}
|
|
|
|
BufferedHelper(BufferedHelper&& other)
|
|
: m_stream(move(other.m_stream))
|
|
, m_buffer(move(other.m_buffer))
|
|
{
|
|
}
|
|
|
|
BufferedHelper& operator=(BufferedHelper&& other)
|
|
{
|
|
m_stream = move(other.m_stream);
|
|
m_buffer = move(other.m_buffer);
|
|
return *this;
|
|
}
|
|
|
|
template<template<typename> typename BufferedType>
|
|
static ErrorOr<NonnullOwnPtr<BufferedType<T>>> create_buffered(NonnullOwnPtr<T> stream, size_t buffer_size)
|
|
{
|
|
if (!buffer_size)
|
|
return Error::from_errno(EINVAL);
|
|
if (!stream->is_open())
|
|
return Error::from_errno(ENOTCONN);
|
|
|
|
auto buffer = TRY(CircularBuffer::create_empty(buffer_size));
|
|
|
|
return adopt_nonnull_own_or_enomem(new BufferedType<T>(move(stream), move(buffer)));
|
|
}
|
|
|
|
T& stream() { return *m_stream; }
|
|
T const& stream() const { return *m_stream; }
|
|
|
|
ErrorOr<Bytes> read(Bytes buffer)
|
|
{
|
|
if (!stream().is_open())
|
|
return Error::from_errno(ENOTCONN);
|
|
if (buffer.is_empty())
|
|
return buffer;
|
|
|
|
// Fill the internal buffer if it has run dry.
|
|
if (m_buffer.used_space() == 0)
|
|
TRY(populate_read_buffer());
|
|
|
|
// Let's try to take all we can from the buffer first.
|
|
return m_buffer.read(buffer);
|
|
}
|
|
|
|
// Reads into the buffer until \n is encountered.
|
|
// The size of the Bytes object is the maximum amount of bytes that will be
|
|
// read. Returns the bytes read as a StringView.
|
|
ErrorOr<StringView> read_line(Bytes buffer)
|
|
{
|
|
return StringView { TRY(read_until(buffer, "\n"sv)) };
|
|
}
|
|
|
|
ErrorOr<Bytes> read_until(Bytes buffer, StringView candidate)
|
|
{
|
|
return read_until_any_of(buffer, Array { candidate });
|
|
}
|
|
|
|
template<size_t N>
|
|
ErrorOr<Bytes> read_until_any_of(Bytes buffer, Array<StringView, N> candidates)
|
|
{
|
|
if (!stream().is_open())
|
|
return Error::from_errno(ENOTCONN);
|
|
|
|
if (buffer.is_empty())
|
|
return buffer;
|
|
|
|
auto const candidate = TRY(find_and_populate_until_any_of(candidates, buffer.size()));
|
|
|
|
if (stream().is_eof()) {
|
|
if (buffer.size() < m_buffer.used_space()) {
|
|
// Normally, reading from an EOFed stream and receiving bytes
|
|
// would mean that the stream is no longer EOF. However, it's
|
|
// possible with a buffered stream that the user is able to read
|
|
// the buffer contents even when the underlying stream is EOF.
|
|
// We already violate this invariant once by giving the user the
|
|
// chance to read the remaining buffer contents, but if the user
|
|
// doesn't give us a big enough buffer, then we would be
|
|
// violating the invariant twice the next time the user attempts
|
|
// to read, which is No Good. So let's give a descriptive error
|
|
// to the caller about why it can't read.
|
|
return Error::from_errno(EMSGSIZE);
|
|
}
|
|
}
|
|
|
|
if (candidate.has_value()) {
|
|
auto const read_bytes = m_buffer.read(buffer.trim(candidate->offset));
|
|
TRY(m_buffer.discard(candidate->size));
|
|
return read_bytes;
|
|
}
|
|
|
|
// If we still haven't found anything, then it's most likely the case
|
|
// that the delimiter ends beyond the length of the caller-passed
|
|
// buffer. Let's just fill the caller's buffer up.
|
|
return m_buffer.read(buffer);
|
|
}
|
|
|
|
struct Match {
|
|
size_t offset {};
|
|
size_t size {};
|
|
};
|
|
|
|
template<size_t N>
|
|
ErrorOr<Optional<Match>> find_and_populate_until_any_of(Array<StringView, N> const& candidates, Optional<size_t> max_offset = {})
|
|
{
|
|
Optional<size_t> longest_candidate;
|
|
for (auto& candidate : candidates) {
|
|
if (candidate.length() >= longest_candidate.value_or(candidate.length()))
|
|
longest_candidate = candidate.length();
|
|
}
|
|
|
|
// The intention here is to try to match all the possible
|
|
// delimiter candidates and try to find the longest one we can
|
|
// remove from the buffer after copying up to the delimiter to the
|
|
// user buffer.
|
|
|
|
auto const find_candidates = [this, &candidates, &longest_candidate](size_t min_offset, Optional<size_t> max_offset = {}) -> Optional<Match> {
|
|
auto const corrected_minimum_offset = *longest_candidate > min_offset ? 0 : min_offset - *longest_candidate;
|
|
max_offset = max_offset.value_or(m_buffer.used_space());
|
|
|
|
Optional<size_t> longest_match;
|
|
size_t match_size = 0;
|
|
for (auto& candidate : candidates) {
|
|
auto const result = m_buffer.offset_of(candidate, corrected_minimum_offset, *max_offset);
|
|
|
|
if (result.has_value()) {
|
|
auto previous_match = longest_match.value_or(*result);
|
|
if ((previous_match < *result) || (previous_match == *result && match_size < candidate.length())) {
|
|
longest_match = result;
|
|
match_size = candidate.length();
|
|
}
|
|
}
|
|
}
|
|
|
|
if (longest_match.has_value())
|
|
return Match { *longest_match, match_size };
|
|
|
|
return {};
|
|
};
|
|
|
|
if (auto first_find = find_candidates(0, max_offset); first_find.has_value())
|
|
return first_find;
|
|
|
|
auto last_size = m_buffer.used_space();
|
|
|
|
while (m_buffer.used_space() < max_offset.value_or(m_buffer.capacity())) {
|
|
auto const read_bytes = TRY(populate_read_buffer());
|
|
if (read_bytes == 0)
|
|
break;
|
|
|
|
if (auto first_find = find_candidates(last_size, max_offset); first_find.has_value())
|
|
return first_find;
|
|
last_size = m_buffer.used_space();
|
|
}
|
|
|
|
return Optional<Match> {};
|
|
}
|
|
|
|
// Returns whether a line can be read, populating the buffer in the process.
|
|
ErrorOr<bool> can_read_line()
|
|
{
|
|
if (stream().is_eof())
|
|
return m_buffer.used_space() > 0;
|
|
|
|
return TRY(find_and_populate_until_any_of(Array<StringView, 1> { "\n"sv })).has_value();
|
|
}
|
|
|
|
bool is_eof() const
|
|
{
|
|
if (m_buffer.used_space() > 0) {
|
|
return false;
|
|
}
|
|
|
|
return stream().is_eof();
|
|
}
|
|
|
|
size_t buffer_size() const
|
|
{
|
|
return m_buffer.capacity();
|
|
}
|
|
|
|
size_t buffered_data_size() const
|
|
{
|
|
return m_buffer.used_space();
|
|
}
|
|
|
|
void clear_buffer()
|
|
{
|
|
m_buffer.clear();
|
|
}
|
|
|
|
ErrorOr<void> discard_bytes(size_t count)
|
|
{
|
|
return m_buffer.discard(count);
|
|
}
|
|
|
|
private:
|
|
ErrorOr<size_t> populate_read_buffer()
|
|
{
|
|
if (m_buffer.empty_space() == 0)
|
|
return 0;
|
|
|
|
// TODO: Figure out if we can do direct writes in a comfortable way.
|
|
Array<u8, 1024> temporary_buffer;
|
|
auto const fillable_slice = temporary_buffer.span().trim(min(temporary_buffer.size(), m_buffer.empty_space()));
|
|
size_t nread = 0;
|
|
do {
|
|
auto result = stream().read_some(fillable_slice);
|
|
if (result.is_error()) {
|
|
if (!result.error().is_errno())
|
|
return result.release_error();
|
|
if (result.error().code() == EINTR)
|
|
continue;
|
|
if (result.error().code() == EAGAIN)
|
|
break;
|
|
return result.release_error();
|
|
}
|
|
auto const filled_slice = result.value();
|
|
VERIFY(m_buffer.write(filled_slice) == filled_slice.size());
|
|
nread += filled_slice.size();
|
|
break;
|
|
} while (true);
|
|
return nread;
|
|
}
|
|
|
|
NonnullOwnPtr<T> m_stream;
|
|
CircularBuffer m_buffer;
|
|
};
|
|
|
|
// NOTE: A Buffered which accepts any Stream could be added here, but it is not
|
|
// needed at the moment.
|
|
|
|
template<SeekableStreamLike T>
|
|
class BufferedSeekable final : public SeekableStream {
|
|
friend BufferedHelper<T>;
|
|
|
|
public:
|
|
static ErrorOr<NonnullOwnPtr<BufferedSeekable<T>>> create(NonnullOwnPtr<T> stream, size_t buffer_size = 16384)
|
|
{
|
|
return BufferedHelper<T>::template create_buffered<BufferedSeekable>(move(stream), buffer_size);
|
|
}
|
|
|
|
BufferedSeekable(BufferedSeekable&& other) = default;
|
|
BufferedSeekable& operator=(BufferedSeekable&& other) = default;
|
|
|
|
virtual ErrorOr<Bytes> read_some(Bytes buffer) override { return m_helper.read(move(buffer)); }
|
|
virtual ErrorOr<size_t> write_some(ReadonlyBytes buffer) override { return m_helper.stream().write_some(buffer); }
|
|
virtual bool is_eof() const override { return m_helper.is_eof(); }
|
|
virtual bool is_open() const override { return m_helper.stream().is_open(); }
|
|
virtual void close() override { m_helper.stream().close(); }
|
|
virtual ErrorOr<size_t> seek(i64 offset, SeekMode mode) override
|
|
{
|
|
if (mode == SeekMode::FromCurrentPosition) {
|
|
// If possible, seek using the buffer alone.
|
|
if (0 <= offset && static_cast<u64>(offset) <= m_helper.buffered_data_size()) {
|
|
MUST(m_helper.discard_bytes(offset));
|
|
return TRY(m_helper.stream().tell()) - m_helper.buffered_data_size();
|
|
}
|
|
|
|
offset = offset - m_helper.buffered_data_size();
|
|
}
|
|
|
|
auto result = TRY(m_helper.stream().seek(offset, mode));
|
|
m_helper.clear_buffer();
|
|
|
|
return result;
|
|
}
|
|
virtual ErrorOr<void> truncate(size_t length) override
|
|
{
|
|
return m_helper.stream().truncate(length);
|
|
}
|
|
|
|
ErrorOr<StringView> read_line(Bytes buffer) { return m_helper.read_line(move(buffer)); }
|
|
ErrorOr<Bytes> read_until(Bytes buffer, StringView candidate) { return m_helper.read_until(move(buffer), move(candidate)); }
|
|
template<size_t N>
|
|
ErrorOr<Bytes> read_until_any_of(Bytes buffer, Array<StringView, N> candidates) { return m_helper.read_until_any_of(move(buffer), move(candidates)); }
|
|
ErrorOr<bool> can_read_line() { return m_helper.can_read_line(); }
|
|
|
|
size_t buffer_size() const { return m_helper.buffer_size(); }
|
|
|
|
virtual ~BufferedSeekable() override = default;
|
|
|
|
private:
|
|
BufferedSeekable(NonnullOwnPtr<T> stream, CircularBuffer buffer)
|
|
: m_helper(Badge<BufferedSeekable<T>> {}, move(stream), move(buffer))
|
|
{
|
|
}
|
|
|
|
BufferedHelper<T> m_helper;
|
|
};
|
|
|
|
}
|
|
|
|
#if USING_AK_GLOBALLY
|
|
using AK::BufferedHelper;
|
|
using AK::BufferedSeekable;
|
|
#endif
|