LucidCube/Tools/MCADefrag/MCADefrag.cpp
Rebekah Rowe 6c4b2e9186
Implement GPL3+ and Apache2.0 Dual License.
Commit is being made to allow additions of GPL3+ code previously
un-addable. With these changes, contributions back to cuberite are
possible with the backporting exemtion, as well as adding stuff in
minetest with minetest code properly being read through and implimented
to upgrade it to GPL3 from GPL2.

project still has Apache2.0 license and credits to all its contributers, but now has the freedom of GPL3+ and all the code that can be implimented and shared with it.
2023-03-20 11:49:56 -04:00

468 lines
10 KiB
C++

/*
* Copyright 2011-2022 Cuberite Contributors
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
// MCADefrag.cpp
// Implements the main app entrypoint and the cMCADefrag class representing the entire app
#include "Globals.h"
#include "MCADefrag.h"
#include "Logger.h"
#include "LoggerSimple.h"
#include "LoggerListeners.h"
// An array of 4096 zero bytes, used for writing the padding
static const Byte g_Zeroes[4096] = {0};
int main(int argc, char ** argv)
{
auto consoleLogListener = MakeConsoleListener(false);
auto consoleAttachment = cLogger::GetInstance().AttachListener(std::move(consoleLogListener));
auto fileLogListenerRet = MakeFileListener();
if (!fileLogListenerRet.first)
{
LOGERROR("Failed to open log file, aborting");
return EXIT_FAILURE;
}
auto fileAttachment = cLogger::GetInstance().AttachListener(std::move(fileLogListenerRet.second));
cLogger::InitiateMultithreading();
cMCADefrag Defrag;
if (!Defrag.Init(argc, argv))
{
return EXIT_FAILURE;
}
Defrag.Run();
return 0;
}
////////////////////////////////////////////////////////////////////////////////
// cMCADefrag:
cMCADefrag::cMCADefrag(void) :
m_NumThreads(4),
m_ShouldRecompress(true)
{
}
bool cMCADefrag::Init(int argc, char ** argv)
{
// Nothing needed yet
return true;
}
void cMCADefrag::Run(void)
{
// Fill the queue with MCA files
m_Queue = cFile::GetFolderContents(".");
// Start the processing threads:
for (int i = 0; i < m_NumThreads; i++)
{
StartThread();
}
// Wait for all the threads to finish:
while (!m_Threads.empty())
{
m_Threads.front()->Stop();
delete m_Threads.front();
m_Threads.pop_front();
}
}
void cMCADefrag::StartThread(void)
{
cThread * Thread = new cThread(*this);
m_Threads.push_back(Thread);
Thread->Start();
}
AString cMCADefrag::GetNextFileName(void)
{
cCSLock Lock(m_CS);
if (m_Queue.empty())
{
return AString();
}
AString res = m_Queue.back();
m_Queue.pop_back();
return res;
}
////////////////////////////////////////////////////////////////////////////////
// cMCADefrag::cThread:
cMCADefrag::cThread::cThread(cMCADefrag & a_Parent) :
super("MCA Defragmentor"),
m_Parent(a_Parent),
m_IsChunkUncompressed(false),
m_Compressor(12) // Set the highest compression factor
{
}
void cMCADefrag::cThread::Execute(void)
{
for (;;)
{
AString FileName = m_Parent.GetNextFileName();
if (FileName.empty())
{
return;
}
ProcessFile(FileName);
}
}
void cMCADefrag::cThread::ProcessFile(const AString & a_FileName)
{
// Filter out non-MCA files:
if ((a_FileName.length() < 4) || (a_FileName.substr(a_FileName.length() - 4, 4) != ".mca"))
{
return;
}
LOGINFO("%s", a_FileName.c_str());
// Open input and output files:
AString OutFileName = a_FileName + ".new";
cFile In, Out;
if (!In.Open(a_FileName, cFile::fmRead))
{
LOGWARNING("Cannot open file %s for reading, skipping file.", a_FileName.c_str());
return;
}
if (!Out.Open(OutFileName.c_str(), cFile::fmWrite))
{
LOGWARNING("Cannot open file %s for writing, skipping file.", OutFileName.c_str());
return;
}
// Read the Locations and Timestamps from the input file:
Byte Locations[4096];
UInt32 Timestamps[1024];
if (In.Read(Locations, sizeof(Locations)) != sizeof(Locations))
{
LOGWARNING("Cannot read Locations in file %s, skipping file.", a_FileName.c_str());
return;
}
if (In.Read(Timestamps, sizeof(Timestamps)) != sizeof(Timestamps))
{
LOGWARNING("Cannot read Timestamps in file %s, skipping file.", a_FileName.c_str());
return;
}
// Write dummy Locations to the Out file (will be overwritten once the correct ones are known)
if (Out.Write(Locations, sizeof(Locations)) != sizeof(Locations))
{
LOGWARNING("Cannot write Locations to file %s, skipping file.", OutFileName.c_str());
return;
}
m_CurrentSectorOut = 2;
// Write a copy of the Timestamps into the Out file:
if (Out.Write(Timestamps, sizeof(Timestamps)) != sizeof(Timestamps))
{
LOGWARNING("Cannot write Timestamps to file %s, skipping file.", OutFileName.c_str());
return;
}
// Process each chunk:
for (size_t i = 0; i < 1024; i++)
{
size_t idx = i * 4;
if (
(Locations[idx] == 0) &&
(Locations[idx + 1] == 0) &&
(Locations[idx + 2] == 0) &&
(Locations[idx + 3] == 0)
)
{
// Chunk not present
continue;
}
m_IsChunkUncompressed = false;
if (!ReadChunk(In, Locations + idx))
{
LOGWARNING("Cannot read chunk #%d from file %s. Skipping file.", i, a_FileName.c_str());
return;
}
if (!WriteChunk(Out, Locations + idx))
{
LOGWARNING("Cannot write chunk #%d to file %s. Skipping file.", i, OutFileName.c_str());
return;
}
}
// Write the new Locations into the MCA header:
Out.Seek(0);
if (Out.Write(Locations, sizeof(Locations)) != sizeof(Locations))
{
LOGWARNING("Cannot write updated Locations to file %s, skipping file.", OutFileName.c_str());
return;
}
// Close the files, delete orig, rename new:
In.Close();
Out.Close();
cFile::Delete(a_FileName);
cFile::Rename(OutFileName, a_FileName);
}
bool cMCADefrag::cThread::ReadChunk(cFile & a_File, const Byte * a_LocationRaw)
{
int SectorNum = (a_LocationRaw[0] << 16) | (a_LocationRaw[1] << 8) | a_LocationRaw[2];
int SizeInSectors = a_LocationRaw[3] * (4 KiB);
if (a_File.Seek(SectorNum * (4 KiB)) < 0)
{
LOGWARNING("Failed to seek to chunk data - file pos %llu (%d KiB, %.02f MiB)!",
static_cast<Int64>(SectorNum) * (4 KiB), SectorNum * 4,
static_cast<double>(SectorNum) / 256
);
return false;
}
// Read the exact size:
Byte Buf[4];
if (a_File.Read(Buf, 4) != 4)
{
LOGWARNING("Failed to read chunk data length");
return false;
}
m_CompressedChunkDataSize = (Buf[0] << 24) | (Buf[1] << 16) | (Buf[2] << 8) | Buf[3];
if ((m_CompressedChunkDataSize > SizeInSectors) || (m_CompressedChunkDataSize < 0))
{
LOGWARNING("Invalid chunk data - SizeInSectors (%d) smaller that RealSize (%d)", SizeInSectors, m_CompressedChunkDataSize);
return false;
}
// Read the data:
if (a_File.Read(m_CompressedChunkData, static_cast<size_t>(m_CompressedChunkDataSize)) != m_CompressedChunkDataSize)
{
LOGWARNING("Failed to read chunk data!");
return false;
}
// Uncompress the data if recompression is active
if (m_Parent.m_ShouldRecompress)
{
m_IsChunkUncompressed = UncompressChunk();
if (!m_IsChunkUncompressed)
{
LOGINFO("Chunk failed to uncompress, will be copied verbatim instead.");
}
}
return true;
}
bool cMCADefrag::cThread::WriteChunk(cFile & a_File, Byte * a_LocationRaw)
{
// Recompress the data if recompression is active:
if (m_Parent.m_ShouldRecompress)
{
if (!CompressChunk())
{
LOGINFO("Chunk failed to recompress, will be coped verbatim instead.");
}
}
// Update the Location:
a_LocationRaw[0] = static_cast<Byte>(m_CurrentSectorOut >> 16);
a_LocationRaw[1] = (m_CurrentSectorOut >> 8) & 0xff;
a_LocationRaw[2] = m_CurrentSectorOut & 0xff;
a_LocationRaw[3] = static_cast<Byte>((m_CompressedChunkDataSize + (4 KiB) + 3) / (4 KiB)); // +3 because the m_CompressedChunkDataSize doesn't include the exact-length
m_CurrentSectorOut += a_LocationRaw[3];
// Write the data length:
Byte Buf[4];
Buf[0] = static_cast<Byte>(m_CompressedChunkDataSize >> 24);
Buf[1] = (m_CompressedChunkDataSize >> 16) & 0xff;
Buf[2] = (m_CompressedChunkDataSize >> 8) & 0xff;
Buf[3] = m_CompressedChunkDataSize & 0xff;
if (a_File.Write(Buf, 4) != 4)
{
LOGWARNING("Failed to write chunk length!");
return false;
}
// Write the data:
if (a_File.Write(m_CompressedChunkData, static_cast<size_t>(m_CompressedChunkDataSize)) != m_CompressedChunkDataSize)
{
LOGWARNING("Failed to write chunk data!");
return false;
}
// Pad onto the next sector:
int NumPadding = a_LocationRaw[3] * 4096 - (m_CompressedChunkDataSize + 4);
ASSERT(NumPadding >= 0);
if ((NumPadding > 0) && (a_File.Write(g_Zeroes, static_cast<size_t>(NumPadding)) != NumPadding))
{
LOGWARNING("Failed to write padding");
return false;
}
return true;
}
bool cMCADefrag::cThread::UncompressChunk(void)
{
switch (m_CompressedChunkData[0])
{
case COMPRESSION_GZIP: return UncompressChunkGzip();
case COMPRESSION_ZLIB: return UncompressChunkZlib();
}
LOGINFO("Chunk is compressed with in an unknown algorithm");
return false;
}
bool cMCADefrag::cThread::UncompressChunkGzip(void)
{
// TODO
// This format is not used in practice
return false;
}
bool cMCADefrag::cThread::UncompressChunkZlib(void)
{
try
{
// Uncompress the data
const auto ExtractedData = m_Extractor.ExtractZLib(
{
reinterpret_cast<const std::byte *>(m_CompressedChunkData + 1), // The first byte is the compression method, skip it
static_cast<size_t>(m_CompressedChunkDataSize - 1)
});
const auto Extracted = ExtractedData.GetView();
if (Extracted.size() > MAX_RAW_CHUNK_DATA_SIZE)
{
LOGINFO("Too much data for the internal decompression buffer!");
return false;
}
std::copy(Extracted.begin(), Extracted.end(), reinterpret_cast<std::byte *>(m_RawChunkData));
m_RawChunkDataSize = static_cast<int>(Extracted.size());
return true;
}
catch (const std::exception & Oops)
{
LOGWARNING("Failed to uncompress chunk data. %s", Oops.what());
return false;
}
}
bool cMCADefrag::cThread::CompressChunk(void)
{
try
{
// Compress the data (using the highest compression factor, as set in the constructor)
const auto CompressedData = m_Compressor.CompressZLib(
{
reinterpret_cast<const std::byte *>(m_RawChunkData),
static_cast<size_t>(m_RawChunkDataSize)
});
const auto Compressed = CompressedData.GetView();
// Check that the compressed data can fit:
if (Compressed.size() > MAX_COMPRESSED_CHUNK_DATA_SIZE)
{
LOGINFO("Too much data for the internal compression buffer!");
return false;
}
m_CompressedChunkData[0] = COMPRESSION_ZLIB;
std::copy(Compressed.begin(), Compressed.end(), reinterpret_cast<std::byte *>(m_CompressedChunkData + 1));
m_CompressedChunkDataSize = static_cast<int>(Compressed.size()) + 1;
return true;
}
catch (const std::exception & Oops)
{
LOGWARNING("Recompression failed. %s", Oops.what());
return false;
}
}