Fossil

Check-in [0cd368f1]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:updated to miniz-1.16 beta r1
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | miniz-1.16br1
Files: files | file ages | folders
SHA1:0cd368f1bfde792e1827994d02a5b4f7fa86fa9e
User & Date: stephan 2016-02-02 15:58:03
Context
2016-02-02
15:58
updated to miniz-1.16 beta r1 Leaf check-in: 0cd368f1 user: stephan tags: miniz-1.16br1
2016-02-01
20:38
Add --https and --nossl options to the 'server' command. check-in: b8c7af5b user: mistachkin tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to src/miniz.c.

1
2
3
4
5
6
7
8
9















10
11
12
13
14
15
16
..
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
...
249
250
251
252
253
254
255

256
257
258
259
260
261
262
...
717
718
719
720
721
722
723






724


725


726





727




728





729

730
731
732
733
734
735
736
...
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
...
914
915
916
917
918
919
920








921
922
923
924
925
926
927
....
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
....
1406
1407
1408
1409
1410
1411
1412


1413
1414
1415
1416
1417
1418
1419
....
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
....
1664
1665
1666
1667
1668
1669
1670







1671
1672
1673
1674
1675

1676
1677
1678




1679
1680
1681
1682
1683
1684
1685
1686
....
2824
2825
2826
2827
2828
2829
2830

2831
2832
2833
2834
2835


























2836
2837
2838
2839
2840
2841
2842
....
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
/* miniz.c v1.15 - public domain deflate/inflate, zlib-subset, ZIP reading/writing/appending, PNG writing
   See "unlicense" statement at the end of this file.
   Rich Geldreich <richgel99@gmail.com>, last updated Oct. 13, 2013
   Implements RFC 1950: http://www.ietf.org/rfc/rfc1950.txt and RFC 1951: http://www.ietf.org/rfc/rfc1951.txt

   Most API's defined in miniz.c are optional. For example, to disable the archive related functions just define
   MINIZ_NO_ARCHIVE_APIS, or to get rid of all stdio usage define MINIZ_NO_STDIO (see the list below for more macros).

   * Change History















     10/13/13 v1.15 r4 - Interim bugfix release while I work on the next major release with Zip64 support (almost there!):
       - Critical fix for the MZ_ZIP_FLAG_DO_NOT_SORT_CENTRAL_DIRECTORY bug (thanks kahmyong.moon@hp.com) which could cause locate files to not find files. This bug
        would only have occured in earlier versions if you explicitly used this flag, OR if you used mz_zip_extract_archive_file_to_heap() or mz_zip_add_mem_to_archive_file_in_place()
        (which used this flag). If you can't switch to v1.15 but want to fix this bug, just remove the uses of this flag from both helper funcs (and of course don't use the flag).
       - Bugfix in mz_zip_reader_extract_to_mem_no_alloc() from kymoon when pUser_read_buf is not NULL and compressed size is > uncompressed size
       - Fixing mz_zip_reader_extract_*() funcs so they don't try to extract compressed data from directory entries, to account for weird zipfiles which contain zero-size compressed data on dir entries.
         Hopefully this fix won't cause any issues on weird zip archives, because it assumes the low 16-bits of zip external attributes are DOS attributes (which I believe they always are in practice).
................................................................................
       - Added tdefl_write_image_to_png_file_in_memory_ex(): supports Y flipping (super useful for OpenGL apps), and explicit control over the compression level (so you can
        set it to 1 for real-time compression).
       - Merged in some compiler fixes from paulharris's github repro.
       - Retested this build under Windows (VS 2010, including static analysis), tcc  0.9.26, gcc v4.6 and clang v3.3.
       - Added example6.c, which dumps an image of the mandelbrot set to a PNG file.
       - Modified example2 to help test the MZ_ZIP_FLAG_DO_NOT_SORT_CENTRAL_DIRECTORY flag more.
       - In r3: Bugfix to mz_zip_writer_add_file() found during merge: Fix possible src file fclose() leak if alignment bytes+local header file write faiiled
       - In r4: Minor bugfix to mz_zip_writer_add_from_zip_reader(): Was pushing the wrong central dir header offset, appears harmless in this release, but it became a problem in the zip64 branch
     5/20/12 v1.14 - MinGW32/64 GCC 4.6.1 compiler fixes: added MZ_FORCEINLINE, #include <time.h> (thanks fermtect).
     5/19/12 v1.13 - From jason@cornsyrup.org and kelwert@mtu.edu - Fix mz_crc32() so it doesn't compute the wrong CRC-32's when mz_ulong is 64-bit.
       - Temporarily/locally slammed in "typedef unsigned long mz_ulong" and re-ran a randomized regression test on ~500k files.
       - Eliminated a bunch of warnings when compiling with GCC 32-bit/64.
       - Ran all examples, miniz.c, and tinfl.c through MSVC 2008's /analyze (static analysis) option and fixed all warnings (except for the silly
        "Use of the comma-operator in a tested expression.." analysis warning, which I purposely use to work around a MSVC compiler warning).
       - Created 32-bit and 64-bit Codeblocks projects/workspace. Built and tested Linux executables. The codeblocks workspace is compatible with Linux+Win32/x64.
................................................................................

// Heap allocation callbacks.
// Note that mz_alloc_func parameter types purpsosely differ from zlib's: items/size is size_t, not unsigned long.
typedef void *(*mz_alloc_func)(void *opaque, size_t items, size_t size);
typedef void (*mz_free_func)(void *opaque, void *address);
typedef void *(*mz_realloc_func)(void *opaque, void *address, size_t items, size_t size);


#define MZ_VERSION          "9.1.15"
#define MZ_VERNUM           0x91F0
#define MZ_VER_MAJOR        9
#define MZ_VER_MINOR        1
#define MZ_VER_REVISION     15
#define MZ_VER_SUBREVISION  0

................................................................................

// Max size of LZ dictionary.
#define TINFL_LZ_DICT_SIZE 32768

// Return status.
typedef enum
{






  TINFL_STATUS_BAD_PARAM = -3,


  TINFL_STATUS_ADLER32_MISMATCH = -2,


  TINFL_STATUS_FAILED = -1,





  TINFL_STATUS_DONE = 0,




  TINFL_STATUS_NEEDS_MORE_INPUT = 1,





  TINFL_STATUS_HAS_MORE_OUTPUT = 2

} tinfl_status;

// Initializes the decompressor to its initial state.
#define tinfl_init(r) do { (r)->m_state = 0; } MZ_MACRO_END
#define tinfl_get_adler32(r) (r)->m_check_adler32

// Main low-level decompressor coroutine function. This is the only function actually needed for decompression. All the other functions are just high-level helpers for improved usability.
................................................................................

// tdefl_compress_mem_to_mem() compresses a block in memory to another block in memory.
// Returns 0 on failure.
size_t tdefl_compress_mem_to_mem(void *pOut_buf, size_t out_buf_len, const void *pSrc_buf, size_t src_buf_len, int flags);

// Compresses an image to a compressed PNG file in memory.
// On entry:
//  pImage, w, h, and num_chans describe the image to compress. num_chans may be 1, 2, 3, or 4.
//  The image pitch in bytes per scanline will be w*num_chans. The leftmost pixel on the top scanline is stored first in memory.
//  level may range from [0,10], use MZ_NO_COMPRESSION, MZ_BEST_SPEED, MZ_BEST_COMPRESSION, etc. or a decent default is MZ_DEFAULT_LEVEL
//  If flip is true, the image will be flipped on the Y axis (useful for OpenGL apps).
// On return:
//  Function returns a pointer to the compressed data, or NULL on failure.
//  *pLen_out will be set to the size of the PNG image file.
//  The caller must mz_free() the returned heap block (which will typically be larger than *pLen_out) when it's no longer needed.
................................................................................
#ifndef MINIZ_NO_ZLIB_APIS
// Create tdefl_compress() flags given zlib-style compression parameters.
// level may range from [0,10] (where 10 is absolute max compression, but may be much slower on some files)
// window_bits may be -15 (raw deflate) or 15 (zlib)
// strategy may be either MZ_DEFAULT_STRATEGY, MZ_FILTERED, MZ_HUFFMAN_ONLY, MZ_RLE, or MZ_FIXED
mz_uint tdefl_create_comp_flags_from_zip_params(int level, int window_bits, int strategy);
#endif // #ifndef MINIZ_NO_ZLIB_APIS









#ifdef __cplusplus
}
#endif

#endif // MINIZ_HEADER_INCLUDED

................................................................................
#define TINFL_MEMSET(p, c, l) memset(p, c, l)

#define TINFL_CR_BEGIN switch(r->m_state) { case 0:
#define TINFL_CR_RETURN(state_index, result) do { status = result; r->m_state = state_index; goto common_exit; case state_index:; } MZ_MACRO_END
#define TINFL_CR_RETURN_FOREVER(state_index, result) do { for ( ; ; ) { TINFL_CR_RETURN(state_index, result); } } MZ_MACRO_END
#define TINFL_CR_FINISH }

// TODO: If the caller has indicated that there's no more input, and we attempt to read beyond the input buf, then something is wrong with the input because the inflator never
// reads ahead more than it needs to. Currently TINFL_GET_BYTE() pads the end of the stream with 0's in this scenario.
#define TINFL_GET_BYTE(state_index, c) do { \
  if (pIn_buf_cur >= pIn_buf_end) { \
    for ( ; ; ) { \
      if (decomp_flags & TINFL_FLAG_HAS_MORE_INPUT) { \
        TINFL_CR_RETURN(state_index, TINFL_STATUS_NEEDS_MORE_INPUT); \
        if (pIn_buf_cur < pIn_buf_end) { \
          c = *pIn_buf_cur++; \
          break; \
        } \
      } else { \
        c = 0; \
        break; \
      } \
    } \
  } else c = *pIn_buf_cur++; } MZ_MACRO_END

#define TINFL_NEED_BITS(state_index, n) do { mz_uint c; TINFL_GET_BYTE(state_index, c); bit_buf |= (((tinfl_bit_buf_t)c) << num_bits); num_bits += 8; } while (num_bits < (mz_uint)(n))
#define TINFL_SKIP_BITS(state_index, n) do { if (num_bits < (mz_uint)(n)) { TINFL_NEED_BITS(state_index, n); } bit_buf >>= (n); num_bits -= (n); } MZ_MACRO_END
#define TINFL_GET_BITS(state_index, b, n) do { if (num_bits < (mz_uint)(n)) { TINFL_NEED_BITS(state_index, n); } b = bit_buf & ((1 << (n)) - 1); bit_buf >>= (n); num_bits -= (n); } MZ_MACRO_END

// TINFL_HUFF_BITBUF_FILL() is only used rarely, when the number of bytes remaining in the input buffer falls below 2.
// It reads just enough bytes from the input stream that are needed to decode the next Huffman code (and absolutely no more). It works by trying to fully decode a
................................................................................
    } TINFL_GET_BYTE(state_index, c); bit_buf |= (((tinfl_bit_buf_t)c) << num_bits); num_bits += 8; \
  } while (num_bits < 15);

// TINFL_HUFF_DECODE() decodes the next Huffman coded symbol. It's more complex than you would initially expect because the zlib API expects the decompressor to never read
// beyond the final byte of the deflate stream. (In other words, when this macro wants to read another byte from the input, it REALLY needs another byte in order to fully
// decode the next Huffman code.) Handling this properly is particularly important on raw deflate (non-zlib) streams, which aren't followed by a byte aligned adler-32.
// The slow path is only executed at the very end of the input buffer.


#define TINFL_HUFF_DECODE(state_index, sym, pHuff) do { \
  int temp; mz_uint code_len, c; \
  if (num_bits < 15) { \
    if ((pIn_buf_end - pIn_buf_cur) < 2) { \
       TINFL_HUFF_BITBUF_FILL(state_index, pHuff); \
    } else { \
       bit_buf |= (((tinfl_bit_buf_t)pIn_buf_cur[0]) << num_bits) | (((tinfl_bit_buf_t)pIn_buf_cur[1]) << (num_bits + 8)); pIn_buf_cur += 2; num_bits += 16; \
................................................................................
        counter--;
      }
      while (counter)
      {
        size_t n; while (pOut_buf_cur >= pOut_buf_end) { TINFL_CR_RETURN(9, TINFL_STATUS_HAS_MORE_OUTPUT); }
        while (pIn_buf_cur >= pIn_buf_end)
        {
          if (decomp_flags & TINFL_FLAG_HAS_MORE_INPUT)
          {
            TINFL_CR_RETURN(38, TINFL_STATUS_NEEDS_MORE_INPUT);
          }
          else
          {
            TINFL_CR_RETURN_FOREVER(40, TINFL_STATUS_FAILED);
          }
        }
        n = MZ_MIN(MZ_MIN((size_t)(pOut_buf_end - pOut_buf_cur), (size_t)(pIn_buf_end - pIn_buf_cur)), counter);
        TINFL_MEMCPY(pOut_buf_cur, pIn_buf_cur, n); pIn_buf_cur += n; pOut_buf_cur += n; counter -= (mz_uint)n;
      }
    }
    else if (r->m_type == 3)
    {
................................................................................
          if ((int)counter > 1)
            pOut_buf_cur[1] = pSrc[1];
          pOut_buf_cur += counter;
        }
      }
    }
  } while (!(r->m_final & 1));







  if (decomp_flags & TINFL_FLAG_PARSE_ZLIB_HEADER)
  {
    TINFL_SKIP_BITS(32, num_bits & 7); for (counter = 0; counter < 4; ++counter) { mz_uint s; if (num_bits) TINFL_GET_BITS(41, s, 8); else TINFL_GET_BYTE(42, s); r->m_z_adler32 = (r->m_z_adler32 << 8) | s; }
  }
  TINFL_CR_RETURN_FOREVER(34, TINFL_STATUS_DONE);

  TINFL_CR_FINISH

common_exit:




  r->m_num_bits = num_bits; r->m_bit_buf = bit_buf; r->m_dist = dist; r->m_counter = counter; r->m_num_extra = num_extra; r->m_dist_from_out_buf_start = dist_from_out_buf_start;
  *pIn_buf_size = pIn_buf_cur - pIn_buf_next; *pOut_buf_size = pOut_buf_cur - pOut_buf_next;
  if ((decomp_flags & (TINFL_FLAG_PARSE_ZLIB_HEADER | TINFL_FLAG_COMPUTE_ADLER32)) && (status >= 0))
  {
    const mz_uint8 *ptr = pOut_buf_next; size_t buf_len = *pOut_buf_size;
    mz_uint32 i, s1 = r->m_check_adler32 & 0xffff, s2 = r->m_check_adler32 >> 16; size_t block_len = buf_len % 5552;
    while (buf_len)
    {
................................................................................
  }
  // write footer (IDAT CRC-32, followed by IEND chunk)
  if (!tdefl_output_buffer_putter("\0\0\0\0\0\0\0\0\x49\x45\x4e\x44\xae\x42\x60\x82", 16, &out_buf)) { *pLen_out = 0; MZ_FREE(pComp); MZ_FREE(out_buf.m_pBuf); return NULL; }
  c = (mz_uint32)mz_crc32(MZ_CRC32_INIT,out_buf.m_pBuf+41-4, *pLen_out+4); for (i=0; i<4; ++i, c<<=8) (out_buf.m_pBuf+out_buf.m_size-16)[i] = (mz_uint8)(c >> 24);
  // compute final size of file, grab compressed data buffer and return
  *pLen_out += 57; MZ_FREE(pComp); return out_buf.m_pBuf;
}

void *tdefl_write_image_to_png_file_in_memory(const void *pImage, int w, int h, int num_chans, size_t *pLen_out)
{
  // Level 6 corresponds to TDEFL_DEFAULT_MAX_PROBES or MZ_DEFAULT_LEVEL (but we can't depend on MZ_DEFAULT_LEVEL being available in case the zlib API's where #defined out)
  return tdefl_write_image_to_png_file_in_memory_ex(pImage, w, h, num_chans, pLen_out, 6, MZ_FALSE);
}



























#ifdef _MSC_VER
#pragma warning (pop)
#endif

// ------------------- .ZIP archive reading

................................................................................

  // no zip64 support yet
  if ((pZip->m_total_files == 0xFFFF) || ((pZip->m_archive_size + num_alignment_padding_bytes + MZ_ZIP_LOCAL_DIR_HEADER_SIZE + MZ_ZIP_CENTRAL_DIR_HEADER_SIZE + comment_size + archive_name_size) > 0xFFFFFFFF))
    return MZ_FALSE;

  if (!mz_zip_get_file_modified_time(pSrc_filename, &dos_time, &dos_date))
    return MZ_FALSE;

  pSrc_file = MZ_FOPEN(pSrc_filename, "rb");
  if (!pSrc_file)
    return MZ_FALSE;
  MZ_FSEEK64(pSrc_file, 0, SEEK_END);
  uncomp_size = MZ_FTELL64(pSrc_file);
  MZ_FSEEK64(pSrc_file, 0, SEEK_SET);

|








>
>
>
>
>
>
>
>
>
>
>
>
>
>
>







 







|







 







>







 







>
>
>
>
>
>

>
>

>
>

>
>
>
>
>

>
>
>
>

>
>
>
>
>

>







 







|







 







>
>
>
>
>
>
>
>







 







<
<

|
<
|
<
<
<
<
<
<
<
<
<
<
|







 







>
>







 







|
<
<
<
<
<
<
<







 







>
>
>
>
>
>
>


|


>



>
>
>
>
|







 







>





>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>







 







|







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
..
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
...
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
...
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
...
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
...
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
....
1410
1411
1412
1413
1414
1415
1416


1417
1418

1419










1420
1421
1422
1423
1424
1425
1426
1427
....
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
....
1508
1509
1510
1511
1512
1513
1514
1515







1516
1517
1518
1519
1520
1521
1522
....
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
....
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
....
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
/* miniz.c v1.16 beta r1 - public domain deflate/inflate, zlib-subset, ZIP reading/writing/appending, PNG writing
   See "unlicense" statement at the end of this file.
   Rich Geldreich <richgel99@gmail.com>, last updated Oct. 13, 2013
   Implements RFC 1950: http://www.ietf.org/rfc/rfc1950.txt and RFC 1951: http://www.ietf.org/rfc/rfc1951.txt

   Most API's defined in miniz.c are optional. For example, to disable the archive related functions just define
   MINIZ_NO_ARCHIVE_APIS, or to get rid of all stdio usage define MINIZ_NO_STDIO (see the list below for more macros).

   * Change History
     10/19/13 v1.16 beta r1 - Two key inflator-only robustness and streaming related changes. Also merged in tdefl_compressor_alloc(), tdefl_compressor_free() helpers to make script bindings easier for rustyzip.
       - The inflator coroutine func. is subtle and complex so I'm being cautious about this release. I would greatly appreciate any help with testing or any feedback.
         I feel good about these changes, and they've been through several hours of automated testing, but they will probably not fix anything for the majority of prev. users so I'm
         going to mark this release as beta for a few weeks and continue testing it at work/home on various things.
       - The inflator in raw (non-zlib) mode is now usable on gzip or similiar data streams that have a bunch of bytes following the raw deflate data (problem discovered by rustyzip author williamw520).
         This version should *never* read beyond the last byte of the raw deflate data independent of how many bytes you pass into the input buffer. This issue was caused by the various Huffman bitbuffer lookahead optimizations, and
         would not be an issue if the caller knew and enforced the precise size of the raw compressed data *or* if the compressed data was in zlib format (i.e. always followed by the byte aligned zlib adler32).
         So in other words, you can now call the inflator on deflate streams that are followed by arbitrary amounts of data and it's guaranteed that decompression will stop exactly on the last byte.
       - The inflator now has a new failure status: TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS (-4). Previously, if the inflator was starved of bytes and could not make progress (because the input buffer was empty and the
         caller did not set the TINFL_FLAG_HAS_MORE_INPUT flag - say on truncated or corrupted compressed data stream) it would append all 0's to the input and try to soldier on.
         This is scary, because in the worst case, I believe it was possible for the prev. inflator to start outputting large amounts of literal data. If the caller didn't know when to stop accepting output
         (because it didn't know how much uncompressed data was expected, or didn't enforce a sane maximum) it could continue forever. v1.16 cannot fall into this failure mode, instead it'll return
         TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS immediately if it needs 1 or more bytes to make progress, the input buf is empty, and the caller has indicated that no more input is available. This is a "soft"
         failure, so you can call the inflator again with more input and it will try to continue, or you can give up and fail. This could be very useful in network streaming scenarios.
       - Added documentation to all the tinfl return status codes, fixed miniz_tester so it accepts double minus params for Linux, tweaked example1.c, added a simple "follower bytes" test to miniz_tester.cpp.
     10/13/13 v1.15 r4 - Interim bugfix release while I work on the next major release with Zip64 support (almost there!):
       - Critical fix for the MZ_ZIP_FLAG_DO_NOT_SORT_CENTRAL_DIRECTORY bug (thanks kahmyong.moon@hp.com) which could cause locate files to not find files. This bug
        would only have occured in earlier versions if you explicitly used this flag, OR if you used mz_zip_extract_archive_file_to_heap() or mz_zip_add_mem_to_archive_file_in_place()
        (which used this flag). If you can't switch to v1.15 but want to fix this bug, just remove the uses of this flag from both helper funcs (and of course don't use the flag).
       - Bugfix in mz_zip_reader_extract_to_mem_no_alloc() from kymoon when pUser_read_buf is not NULL and compressed size is > uncompressed size
       - Fixing mz_zip_reader_extract_*() funcs so they don't try to extract compressed data from directory entries, to account for weird zipfiles which contain zero-size compressed data on dir entries.
         Hopefully this fix won't cause any issues on weird zip archives, because it assumes the low 16-bits of zip external attributes are DOS attributes (which I believe they always are in practice).
................................................................................
       - Added tdefl_write_image_to_png_file_in_memory_ex(): supports Y flipping (super useful for OpenGL apps), and explicit control over the compression level (so you can
        set it to 1 for real-time compression).
       - Merged in some compiler fixes from paulharris's github repro.
       - Retested this build under Windows (VS 2010, including static analysis), tcc  0.9.26, gcc v4.6 and clang v3.3.
       - Added example6.c, which dumps an image of the mandelbrot set to a PNG file.
       - Modified example2 to help test the MZ_ZIP_FLAG_DO_NOT_SORT_CENTRAL_DIRECTORY flag more.
       - In r3: Bugfix to mz_zip_writer_add_file() found during merge: Fix possible src file fclose() leak if alignment bytes+local header file write faiiled
		 - In r4: Minor bugfix to mz_zip_writer_add_from_zip_reader(): Was pushing the wrong central dir header offset, appears harmless in this release, but it became a problem in the zip64 branch
     5/20/12 v1.14 - MinGW32/64 GCC 4.6.1 compiler fixes: added MZ_FORCEINLINE, #include <time.h> (thanks fermtect).
     5/19/12 v1.13 - From jason@cornsyrup.org and kelwert@mtu.edu - Fix mz_crc32() so it doesn't compute the wrong CRC-32's when mz_ulong is 64-bit.
       - Temporarily/locally slammed in "typedef unsigned long mz_ulong" and re-ran a randomized regression test on ~500k files.
       - Eliminated a bunch of warnings when compiling with GCC 32-bit/64.
       - Ran all examples, miniz.c, and tinfl.c through MSVC 2008's /analyze (static analysis) option and fixed all warnings (except for the silly
        "Use of the comma-operator in a tested expression.." analysis warning, which I purposely use to work around a MSVC compiler warning).
       - Created 32-bit and 64-bit Codeblocks projects/workspace. Built and tested Linux executables. The codeblocks workspace is compatible with Linux+Win32/x64.
................................................................................

// Heap allocation callbacks.
// Note that mz_alloc_func parameter types purpsosely differ from zlib's: items/size is size_t, not unsigned long.
typedef void *(*mz_alloc_func)(void *opaque, size_t items, size_t size);
typedef void (*mz_free_func)(void *opaque, void *address);
typedef void *(*mz_realloc_func)(void *opaque, void *address, size_t items, size_t size);

// TODO: I can't encode "1.16" here, argh
#define MZ_VERSION          "9.1.15"
#define MZ_VERNUM           0x91F0
#define MZ_VER_MAJOR        9
#define MZ_VER_MINOR        1
#define MZ_VER_REVISION     15
#define MZ_VER_SUBREVISION  0

................................................................................

// Max size of LZ dictionary.
#define TINFL_LZ_DICT_SIZE 32768

// Return status.
typedef enum
{
  // This flags indicates the inflator needs 1 or more input bytes to make forward progress, but the caller is indicating that no more are available. The compressed data
  // is probably corrupted. If you call the inflator again with more bytes it'll try to continue processing the input but this is a BAD sign (either the data is corrupted or you called it incorrectly).
  // If you call it again with no input you'll just get TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS again.
  TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS = -4,

  // This flag indicates that one or more of the input parameters was obviously bogus. (You can try calling it again, but if you get this error the calling code is wrong.)
  TINFL_STATUS_BAD_PARAM = -3,

  // This flags indicate the inflator is finished but the adler32 check of the uncompressed data didn't match. If you call it again it'll return TINFL_STATUS_DONE.
  TINFL_STATUS_ADLER32_MISMATCH = -2,

  // This flags indicate the inflator has somehow failed (bad code, corrupted input, etc.). If you call it again without resetting via tinfl_init() it it'll just keep on returning the same status failure code.
  TINFL_STATUS_FAILED = -1,

  // Any status code less than TINFL_STATUS_DONE must indicate a failure.

  // This flag indicates the inflator has returned every byte of uncompressed data that it can, has consumed every byte that it needed, has successfully reached the end of the deflate stream, and
  // if zlib headers and adler32 checking enabled that it has successfully checked the uncompressed data's adler32. If you call it again you'll just get TINFL_STATUS_DONE over and over again.
  TINFL_STATUS_DONE = 0,

  // This flag indicates the inflator MUST have more input data (even 1 byte) before it can make any more forward progress, or you need to clear the TINFL_FLAG_HAS_MORE_INPUT
  // flag on the next call if you don't have any more source data. If the source data was somehow corrupted it's also possible (but unlikely) for the inflator to keep on demanding input to
  // proceed, so be sure to properly set the TINFL_FLAG_HAS_MORE_INPUT flag.
  TINFL_STATUS_NEEDS_MORE_INPUT = 1,

  // This flag indicates the inflator definitely has 1 or more bytes of uncompressed data available, but it cannot write this data into the output buffer.
  // Note if the source compressed data was corrupted it's possible for the inflator to return a lot of uncompressed data to the caller. I've been assuming you know how much uncompressed data to expect
  // (either exact or worst case) and will stop calling the inflator and fail after receiving too much. In pure streaming scenarios where you have no idea how many bytes to expect this may not be possible
  // so I may need to add some code to address this.
  TINFL_STATUS_HAS_MORE_OUTPUT = 2

} tinfl_status;

// Initializes the decompressor to its initial state.
#define tinfl_init(r) do { (r)->m_state = 0; } MZ_MACRO_END
#define tinfl_get_adler32(r) (r)->m_check_adler32

// Main low-level decompressor coroutine function. This is the only function actually needed for decompression. All the other functions are just high-level helpers for improved usability.
................................................................................

// tdefl_compress_mem_to_mem() compresses a block in memory to another block in memory.
// Returns 0 on failure.
size_t tdefl_compress_mem_to_mem(void *pOut_buf, size_t out_buf_len, const void *pSrc_buf, size_t src_buf_len, int flags);

// Compresses an image to a compressed PNG file in memory.
// On entry:
//  pImage, w, h, and num_chans describe the image to compress. num_chans may be 1, 2, 3, or 4. 
//  The image pitch in bytes per scanline will be w*num_chans. The leftmost pixel on the top scanline is stored first in memory.
//  level may range from [0,10], use MZ_NO_COMPRESSION, MZ_BEST_SPEED, MZ_BEST_COMPRESSION, etc. or a decent default is MZ_DEFAULT_LEVEL
//  If flip is true, the image will be flipped on the Y axis (useful for OpenGL apps).
// On return:
//  Function returns a pointer to the compressed data, or NULL on failure.
//  *pLen_out will be set to the size of the PNG image file.
//  The caller must mz_free() the returned heap block (which will typically be larger than *pLen_out) when it's no longer needed.
................................................................................
#ifndef MINIZ_NO_ZLIB_APIS
// Create tdefl_compress() flags given zlib-style compression parameters.
// level may range from [0,10] (where 10 is absolute max compression, but may be much slower on some files)
// window_bits may be -15 (raw deflate) or 15 (zlib)
// strategy may be either MZ_DEFAULT_STRATEGY, MZ_FILTERED, MZ_HUFFMAN_ONLY, MZ_RLE, or MZ_FIXED
mz_uint tdefl_create_comp_flags_from_zip_params(int level, int window_bits, int strategy);
#endif // #ifndef MINIZ_NO_ZLIB_APIS

// Allocate the tdefl_compressor and tinfl_decompressor structures in C so that
// non-C language bindings to tdefl_ and tinfl_ API don't need to worry about
// structure size and allocation mechanism.
tdefl_compressor *tdefl_compressor_alloc();
void tdefl_compressor_free(tdefl_compressor *pComp);
tinfl_decompressor *tinfl_decompressor_alloc();
void tinfl_decompressor_free(tinfl_decompressor *pDecomp);

#ifdef __cplusplus
}
#endif

#endif // MINIZ_HEADER_INCLUDED

................................................................................
#define TINFL_MEMSET(p, c, l) memset(p, c, l)

#define TINFL_CR_BEGIN switch(r->m_state) { case 0:
#define TINFL_CR_RETURN(state_index, result) do { status = result; r->m_state = state_index; goto common_exit; case state_index:; } MZ_MACRO_END
#define TINFL_CR_RETURN_FOREVER(state_index, result) do { for ( ; ; ) { TINFL_CR_RETURN(state_index, result); } } MZ_MACRO_END
#define TINFL_CR_FINISH }



#define TINFL_GET_BYTE(state_index, c) do { \
  while (pIn_buf_cur >= pIn_buf_end) { \

    TINFL_CR_RETURN(state_index, (decomp_flags & TINFL_FLAG_HAS_MORE_INPUT) ? TINFL_STATUS_NEEDS_MORE_INPUT : TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS); \










  } c = *pIn_buf_cur++; } MZ_MACRO_END

#define TINFL_NEED_BITS(state_index, n) do { mz_uint c; TINFL_GET_BYTE(state_index, c); bit_buf |= (((tinfl_bit_buf_t)c) << num_bits); num_bits += 8; } while (num_bits < (mz_uint)(n))
#define TINFL_SKIP_BITS(state_index, n) do { if (num_bits < (mz_uint)(n)) { TINFL_NEED_BITS(state_index, n); } bit_buf >>= (n); num_bits -= (n); } MZ_MACRO_END
#define TINFL_GET_BITS(state_index, b, n) do { if (num_bits < (mz_uint)(n)) { TINFL_NEED_BITS(state_index, n); } b = bit_buf & ((1 << (n)) - 1); bit_buf >>= (n); num_bits -= (n); } MZ_MACRO_END

// TINFL_HUFF_BITBUF_FILL() is only used rarely, when the number of bytes remaining in the input buffer falls below 2.
// It reads just enough bytes from the input stream that are needed to decode the next Huffman code (and absolutely no more). It works by trying to fully decode a
................................................................................
    } TINFL_GET_BYTE(state_index, c); bit_buf |= (((tinfl_bit_buf_t)c) << num_bits); num_bits += 8; \
  } while (num_bits < 15);

// TINFL_HUFF_DECODE() decodes the next Huffman coded symbol. It's more complex than you would initially expect because the zlib API expects the decompressor to never read
// beyond the final byte of the deflate stream. (In other words, when this macro wants to read another byte from the input, it REALLY needs another byte in order to fully
// decode the next Huffman code.) Handling this properly is particularly important on raw deflate (non-zlib) streams, which aren't followed by a byte aligned adler-32.
// The slow path is only executed at the very end of the input buffer.
// v1.16: The original macro handled the case at the very end of the passed-in input buffer, but we also need to handle the case where the user passes in 1+zillion bytes
// following the deflate data and our non-conservative read-ahead path won't kick in here on this code. This is much trickier.
#define TINFL_HUFF_DECODE(state_index, sym, pHuff) do { \
  int temp; mz_uint code_len, c; \
  if (num_bits < 15) { \
    if ((pIn_buf_end - pIn_buf_cur) < 2) { \
       TINFL_HUFF_BITBUF_FILL(state_index, pHuff); \
    } else { \
       bit_buf |= (((tinfl_bit_buf_t)pIn_buf_cur[0]) << num_bits) | (((tinfl_bit_buf_t)pIn_buf_cur[1]) << (num_bits + 8)); pIn_buf_cur += 2; num_bits += 16; \
................................................................................
        counter--;
      }
      while (counter)
      {
        size_t n; while (pOut_buf_cur >= pOut_buf_end) { TINFL_CR_RETURN(9, TINFL_STATUS_HAS_MORE_OUTPUT); }
        while (pIn_buf_cur >= pIn_buf_end)
        {
          TINFL_CR_RETURN(38, (decomp_flags & TINFL_FLAG_HAS_MORE_INPUT) ? TINFL_STATUS_NEEDS_MORE_INPUT : TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS);







        }
        n = MZ_MIN(MZ_MIN((size_t)(pOut_buf_end - pOut_buf_cur), (size_t)(pIn_buf_end - pIn_buf_cur)), counter);
        TINFL_MEMCPY(pOut_buf_cur, pIn_buf_cur, n); pIn_buf_cur += n; pOut_buf_cur += n; counter -= (mz_uint)n;
      }
    }
    else if (r->m_type == 3)
    {
................................................................................
          if ((int)counter > 1)
            pOut_buf_cur[1] = pSrc[1];
          pOut_buf_cur += counter;
        }
      }
    }
  } while (!(r->m_final & 1));

  // Ensure byte alignment and put back any bytes from the bitbuf if we've looked ahead too far on gzip, or other Deflate streams followed by arbitrary data.
  // I'm being super conservative here. A number of simplifications can be made to the byte alignment part, and the Adler32 check shouldn't ever need to worry about reading from the bitbuf now.
  TINFL_SKIP_BITS(32, num_bits & 7);
  while ((pIn_buf_cur > pIn_buf_next) && (num_bits >= 8)) { --pIn_buf_cur; num_bits -= 8; } bit_buf &= (tinfl_bit_buf_t)((1ULL << num_bits) - 1ULL);
  MZ_ASSERT(!num_bits); // if this assert fires then we've read beyond the end of non-deflate/zlib streams with following data (such as gzip streams).

  if (decomp_flags & TINFL_FLAG_PARSE_ZLIB_HEADER)
  {
    for (counter = 0; counter < 4; ++counter) { mz_uint s; if (num_bits) TINFL_GET_BITS(41, s, 8); else TINFL_GET_BYTE(42, s); r->m_z_adler32 = (r->m_z_adler32 << 8) | s; }
  }
  TINFL_CR_RETURN_FOREVER(34, TINFL_STATUS_DONE);

  TINFL_CR_FINISH

common_exit:
  // As long as we aren't telling the caller that we NEED more input to make forward progress:
  // Put back any bytes from the bitbuf in case we've looked ahead too far on gzip, or other Deflate streams followed by arbitrary data.
  // We need to be very careful here to NOT push back any bytes we definitely know we need to make forward progress, though, or we'll lock the caller up into an inf loop.
  if ((status != TINFL_STATUS_NEEDS_MORE_INPUT) && (status != TINFL_STATUS_FAILED_CANNOT_MAKE_PROGRESS)) { while ((pIn_buf_cur > pIn_buf_next) && (num_bits >= 8)) { --pIn_buf_cur; num_bits -= 8; } }
  r->m_num_bits = num_bits; r->m_bit_buf = bit_buf & (tinfl_bit_buf_t)((1ULL << num_bits) - 1ULL); r->m_dist = dist; r->m_counter = counter; r->m_num_extra = num_extra; r->m_dist_from_out_buf_start = dist_from_out_buf_start;
  *pIn_buf_size = pIn_buf_cur - pIn_buf_next; *pOut_buf_size = pOut_buf_cur - pOut_buf_next;
  if ((decomp_flags & (TINFL_FLAG_PARSE_ZLIB_HEADER | TINFL_FLAG_COMPUTE_ADLER32)) && (status >= 0))
  {
    const mz_uint8 *ptr = pOut_buf_next; size_t buf_len = *pOut_buf_size;
    mz_uint32 i, s1 = r->m_check_adler32 & 0xffff, s2 = r->m_check_adler32 >> 16; size_t block_len = buf_len % 5552;
    while (buf_len)
    {
................................................................................
  }
  // write footer (IDAT CRC-32, followed by IEND chunk)
  if (!tdefl_output_buffer_putter("\0\0\0\0\0\0\0\0\x49\x45\x4e\x44\xae\x42\x60\x82", 16, &out_buf)) { *pLen_out = 0; MZ_FREE(pComp); MZ_FREE(out_buf.m_pBuf); return NULL; }
  c = (mz_uint32)mz_crc32(MZ_CRC32_INIT,out_buf.m_pBuf+41-4, *pLen_out+4); for (i=0; i<4; ++i, c<<=8) (out_buf.m_pBuf+out_buf.m_size-16)[i] = (mz_uint8)(c >> 24);
  // compute final size of file, grab compressed data buffer and return
  *pLen_out += 57; MZ_FREE(pComp); return out_buf.m_pBuf;
}

void *tdefl_write_image_to_png_file_in_memory(const void *pImage, int w, int h, int num_chans, size_t *pLen_out)
{
  // Level 6 corresponds to TDEFL_DEFAULT_MAX_PROBES or MZ_DEFAULT_LEVEL (but we can't depend on MZ_DEFAULT_LEVEL being available in case the zlib API's where #defined out)
  return tdefl_write_image_to_png_file_in_memory_ex(pImage, w, h, num_chans, pLen_out, 6, MZ_FALSE);
}

// Allocate the tdefl_compressor and tinfl_decompressor structures in C so that
// non-C language bindings to tdefL_ and tinfl_ API don't need to worry about
// structure size and allocation mechanism.
tdefl_compressor *tdefl_compressor_alloc()
{
  return (tdefl_compressor *)MZ_MALLOC(sizeof(tdefl_compressor));
}

void tdefl_compressor_free(tdefl_compressor* pComp)
{
  MZ_FREE(pComp);
}

tinfl_decompressor *tinfl_decompressor_alloc()
{
  tinfl_decompressor *pDecomp = (tinfl_decompressor *)MZ_MALLOC(sizeof(tinfl_decompressor));
  if (pDecomp)
    tinfl_init(pDecomp);
  return pDecomp;
}

void tinfl_decompressor_free(tinfl_decompressor *pDecomp)
{
  MZ_FREE(pDecomp);
}

#ifdef _MSC_VER
#pragma warning (pop)
#endif

// ------------------- .ZIP archive reading

................................................................................

  // no zip64 support yet
  if ((pZip->m_total_files == 0xFFFF) || ((pZip->m_archive_size + num_alignment_padding_bytes + MZ_ZIP_LOCAL_DIR_HEADER_SIZE + MZ_ZIP_CENTRAL_DIR_HEADER_SIZE + comment_size + archive_name_size) > 0xFFFFFFFF))
    return MZ_FALSE;

  if (!mz_zip_get_file_modified_time(pSrc_filename, &dos_time, &dos_date))
    return MZ_FALSE;
    
  pSrc_file = MZ_FOPEN(pSrc_filename, "rb");
  if (!pSrc_file)
    return MZ_FALSE;
  MZ_FSEEK64(pSrc_file, 0, SEEK_END);
  uncomp_size = MZ_FTELL64(pSrc_file);
  MZ_FSEEK64(pSrc_file, 0, SEEK_SET);