contrib/libarchive/archive_write_set_compression_gzip.c

   1 /*-
   2  * Copyright (c) 2003-2004 Tim Kientzle
   3  * All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer
  10  *    in this position and unchanged.
  11  * 2. Redistributions in binary form must reproduce the above copyright
  12  *    notice, this list of conditions and the following disclaimer in the
  13  *    documentation and/or other materials provided with the distribution.
  14  *
  15  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
  16  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  17  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  18  * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
  19  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  20  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  21  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  22  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  23  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  24  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  25  */
  26
  27 #include "archive_platform.h"
  28
  29 /* Don't compile this if we don't have zlib. */
  30 #if HAVE_ZLIB_H
  31
  32 __FBSDID("$FreeBSD: src/lib/libarchive/archive_write_set_compression_gzip.c,v 1.9 2004/11/06 05:25:53 kientzle Exp $");
  33
  34 #include <errno.h>
  35 #include <stdlib.h>
  36 #include <string.h>
  37 #include <time.h>
  38 #include <zlib.h>
  39
  40 #include "archive.h"
  41 #include "archive_private.h"
  42
  43 struct private_data {
  44         z_stream         stream;
  45         int64_t          total_in;
  46         unsigned char   *compressed;
  47         size_t           compressed_buffer_size;
  48         unsigned long    crc;
  49 };
  50
  51
  52 /*
  53  * Yuck.  zlib.h is not const-correct, so I need this one bit
  54  * of ugly hackery to convert a const * pointer to a non-const pointer.
  55  */
  56 #define SET_NEXT_IN(st,src)                                     \
  57         (st)->stream.next_in = (void *)(uintptr_t)(const void *)(src)
  58
  59 static int      archive_compressor_gzip_finish(struct archive *);
  60 static int      archive_compressor_gzip_init(struct archive *);
  61 static int      archive_compressor_gzip_write(struct archive *, const void *,
  62                     size_t);
  63 static int      drive_compressor(struct archive *, struct private_data *,
  64                     int finishing);
  65
  66
  67 /*
  68  * Allocate, initialize and return a archive object.
  69  */
  70 int
  71 archive_write_set_compression_gzip(struct archive *a)
  72 {
  73         archive_check_magic(a, ARCHIVE_WRITE_MAGIC, ARCHIVE_STATE_NEW);
  74         a->compression_init = &archive_compressor_gzip_init;
  75         a->compression_code = ARCHIVE_COMPRESSION_GZIP;
  76         a->compression_name = "gzip";
  77         return (ARCHIVE_OK);
  78 }
  79
  80 /*
  81  * Setup callback.
  82  */
  83 static int
  84 archive_compressor_gzip_init(struct archive *a)
  85 {
  86         int ret;
  87         struct private_data *state;
  88         time_t t;
  89
  90         a->compression_code = ARCHIVE_COMPRESSION_GZIP;
  91         a->compression_name = "gzip";
  92
  93         if (a->client_opener != NULL) {
  94                 ret = (a->client_opener)(a, a->client_data);
  95                 if (ret != ARCHIVE_OK)
  96                         return (ret);
  97         }
  98
  99         state = (struct private_data *)malloc(sizeof(*state));
 100         if (state == NULL) {
 101                 archive_set_error(a, ENOMEM,
 102                     "Can't allocate data for compression");
 103                 return (ARCHIVE_FATAL);
 104         }
 105         memset(state, 0, sizeof(*state));
 106
 107         state->compressed_buffer_size = a->bytes_per_block;
 108         state->compressed = malloc(state->compressed_buffer_size);
 109         state->crc = crc32(0L, NULL, 0);
 110
 111         if (state->compressed == NULL) {
 112                 archive_set_error(a, ENOMEM,
 113                     "Can't allocate data for compression buffer");
 114                 free(state);
 115                 return (ARCHIVE_FATAL);
 116         }
 117
 118         state->stream.next_out = state->compressed;
 119         state->stream.avail_out = state->compressed_buffer_size;
 120
 121         /* Prime output buffer with a gzip header. */
 122         t = time(NULL);
 123         state->compressed[0] = 0x1f; /* GZip signature bytes */
 124         state->compressed[1] = 0x8b;
 125         state->compressed[2] = 0x08; /* "Deflate" compression */
 126         state->compressed[3] = 0; /* No options */
 127         state->compressed[4] = (t)&0xff;  /* Timestamp */
 128         state->compressed[5] = (t>>8)&0xff;
 129         state->compressed[6] = (t>>16)&0xff;
 130         state->compressed[7] = (t>>24)&0xff;
 131         state->compressed[8] = 0; /* No deflate options */
 132         state->compressed[9] = 3; /* OS=Unix */
 133         state->stream.next_out += 10;
 134         state->stream.avail_out -= 10;
 135
 136         a->compression_write = archive_compressor_gzip_write;
 137         a->compression_finish = archive_compressor_gzip_finish;
 138
 139         /* Initialize compression library. */
 140         ret = deflateInit2(&(state->stream),
 141             Z_DEFAULT_COMPRESSION,
 142             Z_DEFLATED,
 143             -15 /* < 0 to suppress zlib header */,
 144             8,
 145             Z_DEFAULT_STRATEGY);
 146
 147         if (ret == Z_OK) {
 148                 a->compression_data = state;
 149                 return (0);
 150         }
 151
 152         /* Library setup failed: clean up. */
 153         archive_set_error(a, ARCHIVE_ERRNO_MISC, "Internal error "
 154             "initializing compression library");
 155         free(state->compressed);
 156         free(state);
 157
 158         /* Override the error message if we know what really went wrong. */
 159         switch (ret) {
 160         case Z_STREAM_ERROR:
 161                 archive_set_error(a, ARCHIVE_ERRNO_MISC,
 162                     "Internal error initializing "
 163                     "compression library: invalid setup parameter");
 164                 break;
 165         case Z_MEM_ERROR:
 166                 archive_set_error(a, ENOMEM, "Internal error initializing "
 167                     "compression library");
 168                 break;
 169         case Z_VERSION_ERROR:
 170                 archive_set_error(a, ARCHIVE_ERRNO_MISC,
 171                     "Internal error initializing "
 172                     "compression library: invalid library version");
 173                 break;
 174         }
 175
 176         return (ARCHIVE_FATAL);
 177 }
 178
 179 /*
 180  * Write data to the compressed stream.
 181  */
 182 static int
 183 archive_compressor_gzip_write(struct archive *a, const void *buff,
 184     size_t length)
 185 {
 186         struct private_data *state;
 187         int ret;
 188
 189         state = a->compression_data;
 190         if (a->client_writer == NULL) {
 191                 archive_set_error(a, ARCHIVE_ERRNO_PROGRAMMER,
 192                     "No write callback is registered?  "
 193                     "This is probably an internal programming error.");
 194                 return (ARCHIVE_FATAL);
 195         }
 196
 197         /* Update statistics */
 198         state->crc = crc32(state->crc, buff, length);
 199         state->total_in += length;
 200
 201         /* Compress input data to output buffer */
 202         SET_NEXT_IN(state, buff);
 203         state->stream.avail_in = length;
 204         if ((ret = drive_compressor(a, state, 0)) != ARCHIVE_OK)
 205                 return (ret);
 206
 207         a->file_position += length;
 208         return (ARCHIVE_OK);
 209 }
 210
 211
 212 /*
 213  * Finish the compression...
 214  */
 215 static int
 216 archive_compressor_gzip_finish(struct archive *a)
 217 {
 218         ssize_t block_length, target_block_length, bytes_written;
 219         int ret;
 220         struct private_data *state;
 221         unsigned tocopy;
 222         unsigned char trailer[8];
 223
 224         state = a->compression_data;
 225         ret = 0;
 226         if (a->client_writer == NULL) {
 227                 archive_set_error(a, ARCHIVE_ERRNO_PROGRAMMER,
 228                     "No write callback is registered?  "
 229                     "This is probably an internal programming error.");
 230                 ret = ARCHIVE_FATAL;
 231                 goto cleanup;
 232         }
 233
 234         /* By default, always pad the uncompressed data. */
 235         if (a->pad_uncompressed) {
 236                 tocopy = a->bytes_per_block -
 237                     (state->total_in % a->bytes_per_block);
 238                 while (tocopy > 0 && tocopy < (unsigned)a->bytes_per_block) {
 239                         SET_NEXT_IN(state, a->nulls);
 240                         state->stream.avail_in = tocopy < a->null_length ?
 241                             tocopy : a->null_length;
 242                         state->crc = crc32(state->crc, a->nulls,
 243                             state->stream.avail_in);
 244                         state->total_in += state->stream.avail_in;
 245                         tocopy -= state->stream.avail_in;
 246                         ret = drive_compressor(a, state, 0);
 247                         if (ret != ARCHIVE_OK)
 248                                 goto cleanup;
 249                 }
 250         }
 251
 252         /* Finish compression cycle */
 253         if (((ret = drive_compressor(a, state, 1))) != ARCHIVE_OK)
 254                 goto cleanup;
 255
 256         /* Build trailer: 4-byte CRC and 4-byte length. */
 257         trailer[0] = (state->crc)&0xff;
 258         trailer[1] = (state->crc >> 8)&0xff;
 259         trailer[2] = (state->crc >> 16)&0xff;
 260         trailer[3] = (state->crc >> 24)&0xff;
 261         trailer[4] = (state->total_in)&0xff;
 262         trailer[5] = (state->total_in >> 8)&0xff;
 263         trailer[6] = (state->total_in >> 16)&0xff;
 264         trailer[7] = (state->total_in >> 24)&0xff;
 265
 266         /* Add trailer to current block. */
 267         tocopy = 8;
 268         if (tocopy > state->stream.avail_out)
 269                 tocopy = state->stream.avail_out;
 270         memcpy(state->stream.next_out, trailer, tocopy);
 271         state->stream.next_out += tocopy;
 272         state->stream.avail_out -= tocopy;
 273
 274         /* If it overflowed, flush and start a new block. */
 275         if (tocopy < 8) {
 276                 bytes_written = (a->client_writer)(a, a->client_data,
 277                     state->compressed, state->compressed_buffer_size);
 278                 if (bytes_written <= 0) {
 279                         ret = ARCHIVE_FATAL;
 280                         goto cleanup;
 281                 }
 282                 a->raw_position += bytes_written;
 283                 state->stream.next_out = state->compressed;
 284                 state->stream.avail_out = state->compressed_buffer_size;
 285                 memcpy(state->stream.next_out, trailer + tocopy, 8-tocopy);
 286                 state->stream.next_out += 8-tocopy;
 287                 state->stream.avail_out -= 8-tocopy;
 288         }
 289
 290         /* Optionally, pad the final compressed block. */
 291         block_length = state->stream.next_out - state->compressed;
 292
 293
 294         /* Tricky calculation to determine size of last block. */
 295         target_block_length = block_length;
 296         if (a->bytes_in_last_block <= 0)
 297                 /* Default or Zero: pad to full block */
 298                 target_block_length = a->bytes_per_block;
 299         else
 300                 /* Round length to next multiple of bytes_in_last_block. */
 301                 target_block_length = a->bytes_in_last_block *
 302                     ( (block_length + a->bytes_in_last_block - 1) /
 303                         a->bytes_in_last_block);
 304         if (target_block_length > a->bytes_per_block)
 305                 target_block_length = a->bytes_per_block;
 306         if (block_length < target_block_length) {
 307                 memset(state->stream.next_out, 0,
 308                     target_block_length - block_length);
 309                 block_length = target_block_length;
 310         }
 311
 312         /* Write the last block */
 313         bytes_written = (a->client_writer)(a, a->client_data,
 314             state->compressed, block_length);
 315         if (bytes_written <= 0) {
 316                 ret = ARCHIVE_FATAL;
 317                 goto cleanup;
 318         }
 319         a->raw_position += bytes_written;
 320
 321         /* Cleanup: shut down compressor, release memory, etc. */
 322 cleanup:
 323         switch (deflateEnd(&(state->stream))) {
 324         case Z_OK:
 325                 break;
 326         default:
 327                 archive_set_error(a, ARCHIVE_ERRNO_MISC,
 328                     "Failed to clean up compressor");
 329                 ret = ARCHIVE_FATAL;
 330         }
 331         free(state->compressed);
 332         free(state);
 333
 334         /* Close the output */
 335         if (a->client_closer != NULL)
 336                 (a->client_closer)(a, a->client_data);
 337
 338         return (ret);
 339 }
 340
 341 /*
 342  * Utility function to push input data through compressor,
 343  * writing full output blocks as necessary.
 344  *
 345  * Note that this handles both the regular write case (finishing ==
 346  * false) and the end-of-archive case (finishing == true).
 347  */
 348 static int
 349 drive_compressor(struct archive *a, struct private_data *state, int finishing)
 350 {
 351         ssize_t bytes_written;
 352         int ret;
 353
 354         for (;;) {
 355                 if (state->stream.avail_out == 0) {
 356                         bytes_written = (a->client_writer)(a, a->client_data,
 357                             state->compressed, state->compressed_buffer_size);
 358                         if (bytes_written <= 0) {
 359                                 /* TODO: Handle this write failure */
 360                                 return (ARCHIVE_FATAL);
 361                         } else if ((size_t)bytes_written < state->compressed_buffer_size) {
 362                                 /* Short write: Move remaining to
 363                                  * front of block and keep filling */
 364                                 memmove(state->compressed,
 365                                     state->compressed + bytes_written,
 366                                     state->compressed_buffer_size - bytes_written);
 367                         }
 368                         a->raw_position += bytes_written;
 369                         state->stream.next_out
 370                             = state->compressed +
 371                             state->compressed_buffer_size - bytes_written;
 372                         state->stream.avail_out = bytes_written;
 373                 }
 374
 375                 ret = deflate(&(state->stream),
 376                     finishing ? Z_FINISH : Z_NO_FLUSH );
 377
 378                 switch (ret) {
 379                 case Z_OK:
 380                         /* In non-finishing case, check if compressor
 381                          * consumed everything */
 382                         if (!finishing && state->stream.avail_in == 0)
 383                                 return (ARCHIVE_OK);
 384                         /* In finishing case, this return always means
 385                          * there's more work */
 386                         break;
 387                 case Z_STREAM_END:
 388                         /* This return can only occur in finishing case. */
 389                         return (ARCHIVE_OK);
 390                 default:
 391                         /* Any other return value indicates an error. */
 392                         archive_set_error(a, ARCHIVE_ERRNO_MISC,
 393                             "GZip compression failed");
 394                         return (ARCHIVE_FATAL);
 395                 }
 396         }
 397 }
 398
 399 #endif /* HAVE_ZLIB_H */