1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
|
/**************************************************************************
*
* Copyright 2011 Zack Rusin
* All Rights Reserved.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*
**************************************************************************/
/*
* Snappy file format.
* -------------------
*
* Snappy at its core is just a compressoin algorithm so we're
* creating a new file format which uses snappy compression
* to hold the trace data.
*
* The file is composed of a number of chunks, they are:
* chunk {
* uint32 - specifying the length of the compressed data
* compressed data, in little endian
* }
* File can contain any number of such chunks.
* The default size of an uncompressed chunk is specified in
* SNAPPY_CHUNK_SIZE.
*
* Note:
* Currently the default size for a a to-be-compressed data is
* 1mb, meaning that the compressed data will be <= 1mb.
* The reason it's 1mb is because it seems
* to offer a pretty good compression/disk io speed ratio
* but that might change.
*
*/
#include <snappy.h>
#include <snappy-sinksource.h>
#include <iostream>
#include <algorithm>
#include <assert.h>
#include <string.h>
#include "trace_file.hpp"
#include "trace_snappy.hpp"
#define SNAPPY_CHUNK_SIZE (1 * 1024 * 1024)
using namespace trace;
class SnappyFile : public File {
public:
SnappyFile(void);
virtual ~SnappyFile();
virtual bool supportsOffsets(void) const override;
virtual File::Offset currentOffset(void) const override;
virtual void setCurrentOffset(const File::Offset &offset) override;
protected:
virtual bool rawOpen(const char *filename) override;
virtual size_t rawRead(void *buffer, size_t length) override;
virtual int rawGetc(void) override;
virtual void rawClose(void) override;
virtual bool rawSkip(size_t length) override;
size_t containerSizeInBytes(void) const override;
size_t containerBytesRead(void) const override;
size_t dataBytesRead(void) const override;
const char* containerType() const override;
private:
inline size_t usedCacheSize(void) const
{
assert(m_cachePtr >= m_cache);
return m_cachePtr - m_cache;
}
inline size_t freeCacheSize(void) const
{
assert(m_cacheSize >= usedCacheSize());
if (m_cacheSize > 0) {
return m_cacheSize - usedCacheSize();
} else {
return 0;
}
}
inline bool endOfData(void) const
{
return m_stream.eof() && freeCacheSize() == 0;
}
void flushWriteCache(void);
void flushReadCache(size_t skipLength = 0);
void createCache(size_t size);
size_t readCompressedLength();
private:
mutable std::ifstream m_stream;
size_t m_cacheMaxSize;
size_t m_cacheSize;
char *m_cache;
char *m_cachePtr;
char *m_compressedCache;
uint64_t m_currentChunkOffset = 0;
std::streampos m_endPos = 0;
size_t m_dataBytesRead = 0;
};
SnappyFile::SnappyFile(void)
: File(),
m_cacheMaxSize(SNAPPY_CHUNK_SIZE),
m_cacheSize(m_cacheMaxSize),
m_cache(new char [m_cacheMaxSize]),
m_cachePtr(m_cache)
{
size_t maxCompressedLength =
snappy::MaxCompressedLength(SNAPPY_CHUNK_SIZE);
m_compressedCache = new char[maxCompressedLength];
}
SnappyFile::~SnappyFile()
{
close();
delete [] m_compressedCache;
delete [] m_cache;
}
bool SnappyFile::rawOpen(const char *filename)
{
std::ios_base::openmode fmode = std::fstream::binary
| std::fstream::in;
m_stream.open(filename, fmode);
//read in the initial buffer if we're reading
if (m_stream.is_open()) {
m_stream.seekg(0, std::ios::end);
m_endPos = m_stream.tellg();
m_stream.seekg(0, std::ios::beg);
m_dataBytesRead = 0;
// read the snappy file identifier
unsigned char byte1, byte2;
m_stream >> byte1;
m_stream >> byte2;
assert(byte1 == SNAPPY_BYTE1 && byte2 == SNAPPY_BYTE2);
flushReadCache();
}
return m_stream.is_open();
}
size_t SnappyFile::rawRead(void *buffer, size_t length)
{
if (endOfData()) {
return 0;
}
if (freeCacheSize() >= length) {
memcpy(buffer, m_cachePtr, length);
m_cachePtr += length;
m_dataBytesRead += length;
} else {
size_t sizeToRead = length;
size_t offset = 0;
while (sizeToRead) {
size_t chunkSize = std::min(freeCacheSize(), sizeToRead);
offset = length - sizeToRead;
memcpy((char*)buffer + offset, m_cachePtr, chunkSize);
m_cachePtr += chunkSize;
m_dataBytesRead += chunkSize;
sizeToRead -= chunkSize;
if (sizeToRead > 0) {
flushReadCache();
}
if (!m_cacheSize) {
return length - sizeToRead;
}
}
}
return length;
}
int SnappyFile::rawGetc(void)
{
unsigned char c = 0;
if (rawRead(&c, 1) != 1)
return -1;
return c;
}
void SnappyFile::rawClose(void)
{
m_stream.close();
delete [] m_cache;
m_cache = NULL;
m_cachePtr = NULL;
}
void SnappyFile::flushReadCache(size_t skipLength)
{
//assert(m_cachePtr == m_cache + m_cacheSize);
m_currentChunkOffset = m_stream.tellg();
size_t compressedLength;
compressedLength = readCompressedLength();
if (!compressedLength) {
// Reached end of file
createCache(0);
return;
}
m_stream.read((char*)m_compressedCache, compressedLength);
if (m_stream.fail()) {
std::cerr << "warning: unexpected end of file while reading trace\n";
compressedLength = m_stream.gcount();
if (!snappy::GetUncompressedLength(m_compressedCache, compressedLength,
&m_cacheSize)) {
createCache(0);
return;
}
createCache(m_cacheSize);
snappy::ByteArraySource source(m_compressedCache, compressedLength);
snappy::UncheckedByteArraySink sink(m_cache);
m_cacheSize = snappy::UncompressAsMuchAsPossible(&source, &sink);
return;
}
if (!snappy::GetUncompressedLength(m_compressedCache, compressedLength,
&m_cacheSize)) {
createCache(0);
return;
}
createCache(m_cacheSize);
if (skipLength < m_cacheSize) {
snappy::RawUncompress(m_compressedCache, compressedLength,
m_cache);
}
}
void SnappyFile::createCache(size_t size)
{
if (size > m_cacheMaxSize) {
do {
m_cacheMaxSize <<= 1;
} while (size > m_cacheMaxSize);
delete [] m_cache;
m_cache = new char[size];
m_cacheMaxSize = size;
}
m_cachePtr = m_cache;
m_cacheSize = size;
}
size_t SnappyFile::readCompressedLength()
{
unsigned char buf[4];
size_t length;
m_stream.read((char *)buf, sizeof buf);
if (m_stream.fail()) {
length = 0;
} else {
length = (size_t)buf[0];
length |= ((size_t)buf[1] << 8);
length |= ((size_t)buf[2] << 16);
length |= ((size_t)buf[3] << 24);
}
return length;
}
bool SnappyFile::supportsOffsets(void) const
{
return true;
}
File::Offset SnappyFile::currentOffset(void) const
{
File::Offset offset;
offset.chunk = m_currentChunkOffset;
offset.offsetInChunk = m_cachePtr - m_cache;
return offset;
}
void SnappyFile::setCurrentOffset(const File::Offset &offset)
{
// to remove eof bit
m_stream.clear();
// seek to the start of a chunk
m_stream.seekg(offset.chunk, std::ios::beg);
// load the chunk
flushReadCache();
assert(m_cacheSize >= offset.offsetInChunk);
// seek within our cache to the correct location within the chunk
m_cachePtr = m_cache + offset.offsetInChunk;
}
bool SnappyFile::rawSkip(size_t length)
{
if (endOfData()) {
return false;
}
if (freeCacheSize() >= length) {
m_cachePtr += length;
m_dataBytesRead += length;
} else {
size_t sizeToRead = length;
while (sizeToRead) {
size_t chunkSize = std::min(freeCacheSize(), sizeToRead);
m_cachePtr += chunkSize;
m_dataBytesRead += chunkSize;
sizeToRead -= chunkSize;
if (sizeToRead > 0) {
flushReadCache(sizeToRead);
}
if (!m_cacheSize) {
break;
}
}
}
return true;
}
size_t SnappyFile::containerSizeInBytes(void) const {
return static_cast<size_t>(m_endPos);
}
size_t SnappyFile::containerBytesRead(void) const {
return m_currentChunkOffset;
}
size_t SnappyFile::dataBytesRead(void) const {
return static_cast<size_t>(m_dataBytesRead);
}
const char *SnappyFile::containerType(void) const {
return "Snappy";
}
File* File::createSnappy(void) {
return new SnappyFile;
}
|