Mercurial > hg > orthanc
annotate Core/Compression/GzipCompressor.cpp @ 3384:2b645caab9ee
todo
| author | Sebastien Jodogne <s.jodogne@gmail.com> |
|---|---|
| date | Wed, 29 May 2019 18:12:28 +0200 |
| parents | 596cfabd72c5 |
| children | 94f4a18a79cc |
| rev | line source |
|---|---|
| 1513 | 1 /** |
| 2 * Orthanc - A Lightweight, RESTful DICOM Store | |
| 1900 | 3 * Copyright (C) 2012-2016 Sebastien Jodogne, Medical Physics |
| 1513 | 4 * Department, University Hospital of Liege, Belgium |
|
3060
4e43e67f8ecf
preparing for 2019
Sebastien Jodogne <s.jodogne@gmail.com>
parents:
2954
diff
changeset
|
5 * Copyright (C) 2017-2019 Osimis S.A., Belgium |
| 1513 | 6 * |
| 7 * This program is free software: you can redistribute it and/or | |
| 8 * modify it under the terms of the GNU General Public License as | |
| 9 * published by the Free Software Foundation, either version 3 of the | |
| 10 * License, or (at your option) any later version. | |
| 11 * | |
| 12 * In addition, as a special exception, the copyright holders of this | |
| 13 * program give permission to link the code of its release with the | |
| 14 * OpenSSL project's "OpenSSL" library (or with modified versions of it | |
| 15 * that use the same license as the "OpenSSL" library), and distribute | |
| 16 * the linked executables. You must obey the GNU General Public License | |
| 17 * in all respects for all of the code used other than "OpenSSL". If you | |
| 18 * modify file(s) with this exception, you may extend this exception to | |
| 19 * your version of the file(s), but you are not obligated to do so. If | |
| 20 * you do not wish to do so, delete this exception statement from your | |
| 21 * version. If you delete this exception statement from all source files | |
| 22 * in the program, then also delete it here. | |
| 23 * | |
| 24 * This program is distributed in the hope that it will be useful, but | |
| 25 * WITHOUT ANY WARRANTY; without even the implied warranty of | |
| 26 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
| 27 * General Public License for more details. | |
| 28 * | |
| 29 * You should have received a copy of the GNU General Public License | |
| 30 * along with this program. If not, see <http://www.gnu.org/licenses/>. | |
| 31 **/ | |
| 32 | |
| 33 | |
| 34 #include "../PrecompiledHeaders.h" | |
| 35 #include "GzipCompressor.h" | |
| 36 | |
| 37 #include <stdio.h> | |
| 38 #include <string.h> | |
| 39 #include <zlib.h> | |
| 40 | |
| 41 #include "../OrthancException.h" | |
| 42 #include "../Logging.h" | |
| 43 | |
| 44 namespace Orthanc | |
| 45 { | |
| 46 uint64_t GzipCompressor::GuessUncompressedSize(const void* compressed, | |
| 47 size_t compressedSize) | |
| 48 { | |
| 49 /** | |
| 50 * "Is there a way to find out the size of the original file which | |
| 51 * is inside a GZIP file? [...] There is no truly reliable way, | |
| 52 * other than gunzipping the stream. You do not need to save the | |
| 53 * result of the decompression, so you can determine the size by | |
| 54 * simply reading and decoding the entire file without taking up | |
| 55 * space with the decompressed result. | |
| 56 * | |
| 57 * There is an unreliable way to determine the uncompressed size, | |
| 58 * which is to look at the last four bytes of the gzip file, which | |
| 59 * is the uncompressed length of that entry modulo 232 in little | |
| 60 * endian order. | |
| 61 * | |
| 62 * It is unreliable because a) the uncompressed data may be longer | |
| 63 * than 2^32 bytes, and b) the gzip file may consist of multiple | |
| 64 * gzip streams, in which case you would find the length of only | |
| 65 * the last of those streams. | |
| 66 * | |
| 67 * If you are in control of the source of the gzip files, you know | |
| 68 * that they consist of single gzip streams, and you know that | |
| 69 * they are less than 2^32 bytes uncompressed, then and only then | |
| 70 * can you use those last four bytes with confidence." | |
| 71 * | |
| 72 * http://stackoverflow.com/a/9727599/881731 | |
| 73 **/ | |
| 74 | |
| 75 if (compressedSize < 4) | |
| 76 { | |
| 77 throw OrthancException(ErrorCode_BadFileFormat); | |
| 78 } | |
| 79 | |
| 80 const uint8_t* p = reinterpret_cast<const uint8_t*>(compressed) + compressedSize - 4; | |
| 81 | |
| 82 return ((static_cast<uint32_t>(p[0]) << 0) + | |
| 83 (static_cast<uint32_t>(p[1]) << 8) + | |
| 84 (static_cast<uint32_t>(p[2]) << 16) + | |
| 85 (static_cast<uint32_t>(p[3]) << 24)); | |
| 86 } | |
| 87 | |
| 88 | |
| 89 | |
| 90 void GzipCompressor::Compress(std::string& compressed, | |
| 91 const void* uncompressed, | |
| 92 size_t uncompressedSize) | |
| 93 { | |
|
3378
596cfabd72c5
Fixed a couple of truncation warnings
Benjamin Golinvaux <bgo@osimis.io>
parents:
3060
diff
changeset
|
94 uLongf compressedSize = compressBound(static_cast<uLong>(uncompressedSize)) |
|
596cfabd72c5
Fixed a couple of truncation warnings
Benjamin Golinvaux <bgo@osimis.io>
parents:
3060
diff
changeset
|
95 + 1024 /* security margin */; |
|
596cfabd72c5
Fixed a couple of truncation warnings
Benjamin Golinvaux <bgo@osimis.io>
parents:
3060
diff
changeset
|
96 |
| 1513 | 97 if (compressedSize == 0) |
| 98 { | |
| 99 compressedSize = 1; | |
| 100 } | |
| 101 | |
| 102 uint8_t* target; | |
| 103 if (HasPrefixWithUncompressedSize()) | |
| 104 { | |
| 105 compressed.resize(compressedSize + sizeof(uint64_t)); | |
| 106 target = reinterpret_cast<uint8_t*>(&compressed[0]) + sizeof(uint64_t); | |
| 107 } | |
| 108 else | |
| 109 { | |
| 110 compressed.resize(compressedSize); | |
| 111 target = reinterpret_cast<uint8_t*>(&compressed[0]); | |
| 112 } | |
| 113 | |
| 114 z_stream stream; | |
| 115 memset(&stream, 0, sizeof(stream)); | |
| 116 | |
| 117 stream.next_in = const_cast<Bytef*>(reinterpret_cast<const Bytef*>(uncompressed)); | |
| 118 stream.next_out = reinterpret_cast<Bytef*>(target); | |
| 119 | |
| 120 stream.avail_in = static_cast<uInt>(uncompressedSize); | |
| 121 stream.avail_out = static_cast<uInt>(compressedSize); | |
| 122 | |
| 123 // Ensure no overflow (if the buffer is too large for the current archicture) | |
| 124 if (static_cast<size_t>(stream.avail_in) != uncompressedSize || | |
| 125 static_cast<size_t>(stream.avail_out) != compressedSize) | |
| 126 { | |
| 127 throw OrthancException(ErrorCode_NotEnoughMemory); | |
| 128 } | |
| 129 | |
| 130 // Initialize the compression engine | |
| 131 int error = deflateInit2(&stream, | |
| 132 GetCompressionLevel(), | |
| 133 Z_DEFLATED, | |
| 134 MAX_WBITS + 16, // ask for gzip output | |
| 135 8, // default memory level | |
| 136 Z_DEFAULT_STRATEGY); | |
| 137 | |
| 138 if (error != Z_OK) | |
| 139 { | |
| 140 // Cannot initialize zlib | |
| 141 compressed.clear(); | |
| 142 throw OrthancException(ErrorCode_InternalError); | |
| 143 } | |
| 144 | |
| 145 // Compress the input buffer | |
| 146 error = deflate(&stream, Z_FINISH); | |
| 147 | |
| 148 if (error != Z_STREAM_END) | |
| 149 { | |
| 150 deflateEnd(&stream); | |
| 151 compressed.clear(); | |
| 152 | |
| 153 switch (error) | |
| 154 { | |
| 155 case Z_MEM_ERROR: | |
| 156 throw OrthancException(ErrorCode_NotEnoughMemory); | |
| 157 | |
| 158 default: | |
| 159 throw OrthancException(ErrorCode_InternalError); | |
| 160 } | |
| 161 } | |
| 162 | |
| 163 size_t size = stream.total_out; | |
| 164 | |
| 165 if (deflateEnd(&stream) != Z_OK) | |
| 166 { | |
| 167 throw OrthancException(ErrorCode_InternalError); | |
| 168 } | |
| 169 | |
| 170 // The compression was successful | |
| 171 if (HasPrefixWithUncompressedSize()) | |
| 172 { | |
| 173 uint64_t s = static_cast<uint64_t>(uncompressedSize); | |
| 174 memcpy(&compressed[0], &s, sizeof(uint64_t)); | |
| 175 compressed.resize(size + sizeof(uint64_t)); | |
| 176 } | |
| 177 else | |
| 178 { | |
| 179 compressed.resize(size); | |
| 180 } | |
| 181 } | |
| 182 | |
| 183 | |
| 184 void GzipCompressor::Uncompress(std::string& uncompressed, | |
| 185 const void* compressed, | |
| 186 size_t compressedSize) | |
| 187 { | |
| 188 uint64_t uncompressedSize; | |
| 189 const uint8_t* source = reinterpret_cast<const uint8_t*>(compressed); | |
| 190 | |
| 191 if (HasPrefixWithUncompressedSize()) | |
| 192 { | |
| 193 uncompressedSize = ReadUncompressedSizePrefix(compressed, compressedSize); | |
| 194 source += sizeof(uint64_t); | |
| 195 compressedSize -= sizeof(uint64_t); | |
| 196 } | |
| 197 else | |
| 198 { | |
| 199 uncompressedSize = GuessUncompressedSize(compressed, compressedSize); | |
| 200 } | |
| 201 | |
| 202 try | |
| 203 { | |
| 1545 | 204 uncompressed.resize(static_cast<size_t>(uncompressedSize)); |
| 1513 | 205 } |
| 206 catch (...) | |
| 207 { | |
| 208 throw OrthancException(ErrorCode_NotEnoughMemory); | |
| 209 } | |
| 210 | |
| 211 z_stream stream; | |
| 212 memset(&stream, 0, sizeof(stream)); | |
| 213 | |
| 214 char dummy = '\0'; // zlib does not like NULL output buffers (even if the uncompressed data is empty) | |
| 215 stream.next_in = const_cast<Bytef*>(source); | |
| 216 stream.next_out = reinterpret_cast<Bytef*>(uncompressedSize == 0 ? &dummy : &uncompressed[0]); | |
| 217 | |
| 218 stream.avail_in = static_cast<uInt>(compressedSize); | |
| 219 stream.avail_out = static_cast<uInt>(uncompressedSize); | |
| 220 | |
| 221 // Ensure no overflow (if the buffer is too large for the current archicture) | |
| 222 if (static_cast<size_t>(stream.avail_in) != compressedSize || | |
| 223 static_cast<size_t>(stream.avail_out) != uncompressedSize) | |
| 224 { | |
| 225 throw OrthancException(ErrorCode_NotEnoughMemory); | |
| 226 } | |
| 227 | |
| 228 // Initialize the compression engine | |
| 229 int error = inflateInit2(&stream, | |
| 230 MAX_WBITS + 16); // this is a gzip input | |
| 231 | |
| 232 if (error != Z_OK) | |
| 233 { | |
| 234 // Cannot initialize zlib | |
| 235 uncompressed.clear(); | |
| 236 throw OrthancException(ErrorCode_InternalError); | |
| 237 } | |
| 238 | |
| 239 // Uncompress the input buffer | |
| 240 error = inflate(&stream, Z_FINISH); | |
| 241 | |
| 242 if (error != Z_STREAM_END) | |
| 243 { | |
| 244 inflateEnd(&stream); | |
| 245 uncompressed.clear(); | |
| 246 | |
| 247 switch (error) | |
| 248 { | |
| 249 case Z_MEM_ERROR: | |
| 250 throw OrthancException(ErrorCode_NotEnoughMemory); | |
| 251 | |
| 252 case Z_BUF_ERROR: | |
| 253 case Z_NEED_DICT: | |
| 254 throw OrthancException(ErrorCode_BadFileFormat); | |
| 255 | |
| 256 default: | |
| 257 throw OrthancException(ErrorCode_InternalError); | |
| 258 } | |
| 259 } | |
| 260 | |
| 261 size_t size = stream.total_out; | |
| 262 | |
| 263 if (inflateEnd(&stream) != Z_OK) | |
| 264 { | |
| 265 uncompressed.clear(); | |
| 266 throw OrthancException(ErrorCode_InternalError); | |
| 267 } | |
| 268 | |
| 269 if (size != uncompressedSize) | |
| 270 { | |
| 271 uncompressed.clear(); | |
| 272 | |
| 273 // The uncompressed size was not that properly guess, presumably | |
| 274 // because of a file size over 4GB. Should fallback to | |
| 275 // stream-based decompression. | |
|
2954
d924f9bb61cc
taking advantage of details in OrthancException
Sebastien Jodogne <s.jodogne@gmail.com>
parents:
2447
diff
changeset
|
276 throw OrthancException(ErrorCode_NotImplemented, |
|
d924f9bb61cc
taking advantage of details in OrthancException
Sebastien Jodogne <s.jodogne@gmail.com>
parents:
2447
diff
changeset
|
277 "The uncompressed size of a gzip-encoded buffer was not properly guessed"); |
| 1513 | 278 } |
| 279 } | |
| 280 } |