ChangeLog revision 33904
1
2		ChangeLog file for zlib
3
4Changes in 1.1.1 (27 Feb 98)
5- fix macros _tr_tally_* in deflate.h for debug mode  (Glenn Randers-Pehrson)
6- remove block truncation heuristic which had very marginal effect for zlib
7  (smaller lit_bufsize than in gzip 1.2.4) and degraded a little the
8  compression ratio on some files. This also allows inlining _tr_tally for
9  matches in deflate_slow.
10- added msdos/Makefile.w32 for WIN32 Microsoft Visual C++ (Bob Frazier)
11
12Changes in 1.1.0 (24 Feb 98)
13- do not return STREAM_END prematurely in inflate (John Bowler)
14- revert to the zlib 1.0.8 inflate to avoid the gcc 2.8.0 bug (Jeremy Buhler)
15- compile with -DFASTEST to get compression code optimized for speed only
16- in minigzip, try mmap'ing the input file first (Miguel Albrecht)
17- increase size of I/O buffers in minigzip.c and gzio.c (not a big gain
18  on Sun but significant on HP)
19
20- add a pointer to experimental unzip library in README (Gilles Vollant)
21- initialize variable gcc in configure (Chris Herborth)
22
23Changes in 1.0.9 (17 Feb 1998)
24- added gzputs and gzgets functions
25- do not clear eof flag in gzseek (Mark Diekhans)
26- fix gzseek for files in transparent mode (Mark Diekhans)
27- do not assume that vsprintf returns the number of bytes written (Jens Krinke)
28- replace EXPORT with ZEXPORT to avoid conflict with other programs
29- added compress2 in zconf.h, zlib.def, zlib.dnt
30- new asm code from Gilles Vollant in contrib/asm386
31- simplify the inflate code (Mark):
32 . Replace ZALLOC's in huft_build() with single ZALLOC in inflate_blocks_new()
33 . ZALLOC the length list in inflate_trees_fixed() instead of using stack
34 . ZALLOC the value area for huft_build() instead of using stack
35 . Simplify Z_FINISH check in inflate()
36
37- Avoid gcc 2.8.0 comparison bug a little differently than zlib 1.0.8
38- in inftrees.c, avoid cc -O bug on HP (Farshid Elahi)
39- in zconf.h move the ZLIB_DLL stuff earlier to avoid problems with
40  the declaration of FAR (Gilles VOllant)
41- install libz.so* with mode 755 (executable) instead of 644 (Marc Lehmann)
42- read_buf buf parameter of type Bytef* instead of charf*
43- zmemcpy parameters are of type Bytef*, not charf* (Joseph Strout)
44- do not redeclare unlink in minigzip.c for WIN32 (John Bowler)
45- fix check for presence of directories in "make install" (Ian Willis)
46
47Changes in 1.0.8 (27 Jan 1998)
48- fixed offsets in contrib/asm386/gvmat32.asm (Gilles Vollant)
49- fix gzgetc and gzputc for big endian systems (Markus Oberhumer)
50- added compress2() to allow setting the compression level
51- include sys/types.h to get off_t on some systems (Marc Lehmann & QingLong)
52- use constant arrays for the static trees in trees.c instead of computing
53  them at run time (thanks to Ken Raeburn for this suggestion). To create
54  trees.h, compile with GEN_TREES_H and run "make test".
55- check return code of example in "make test" and display result
56- pass minigzip command line options to file_compress
57- simplifying code of inflateSync to avoid gcc 2.8 bug
58
59- support CC="gcc -Wall" in configure -s (QingLong)
60- avoid a flush caused by ftell in gzopen for write mode (Ken Raeburn)
61- fix test for shared library support to avoid compiler warnings
62- zlib.lib -> zlib.dll in msdos/zlib.rc (Gilles Vollant)
63- check for TARGET_OS_MAC in addition to MACOS (Brad Pettit)
64- do not use fdopen for Metrowerks on Mac (Brad Pettit))
65- add checks for gzputc and gzputc in example.c
66- avoid warnings in gzio.c and deflate.c (Andreas Kleinert)
67- use const for the CRC table (Ken Raeburn)
68- fixed "make uninstall" for shared libraries
69- use Tracev instead of Trace in infblock.c
70- in example.c use correct compressed length for test_sync
71- suppress +vnocompatwarnings in configure for HPUX (not always supported)
72
73Changes in 1.0.7 (20 Jan 1998)
74- fix gzseek which was broken in write mode
75- return error for gzseek to negative absolute position
76- fix configure for Linux (Chun-Chung Chen)
77- increase stack space for MSC (Tim Wegner)
78- get_crc_table and inflateSyncPoint are EXPORTed (Gilles Vollant)
79- define EXPORTVA for gzprintf (Gilles Vollant)
80- added man page zlib.3 (Rick Rodgers)
81- for contrib/untgz, fix makedir() and improve Makefile
82
83- check gzseek in write mode in example.c
84- allocate extra buffer for seeks only if gzseek is actually called
85- avoid signed/unsigned comparisons (Tim Wegner, Gilles Vollant)
86- add inflateSyncPoint in zconf.h
87- fix list of exported functions in nt/zlib.dnt and mdsos/zlib.def
88
89Changes in 1.0.6 (19 Jan 1998)
90- add functions gzprintf, gzputc, gzgetc, gztell, gzeof, gzseek, gzrewind and
91  gzsetparams (thanks to Roland Giersig and Kevin Ruland for some of this code)
92- Fix a deflate bug occuring only with compression level 0 (thanks to
93  Andy Buckler for finding this one).
94- In minigzip, pass transparently also the first byte for .Z files.
95- return Z_BUF_ERROR instead of Z_OK if output buffer full in uncompress()
96- check Z_FINISH in inflate (thanks to Marc Schluper)
97- Implement deflateCopy (thanks to Adam Costello)
98- make static libraries by default in configure, add --shared option.
99- move MSDOS or Windows specific files to directory msdos
100- suppress the notion of partial flush to simplify the interface
101  (but the symbol Z_PARTIAL_FLUSH is kept for compatibility with 1.0.4)
102- suppress history buffer provided by application to simplify the interface
103  (this feature was not implemented anyway in 1.0.4)
104- next_in and avail_in must be initialized before calling inflateInit or
105  inflateInit2
106- add EXPORT in all exported functions (for Windows DLL)
107- added Makefile.nt (thanks to Stephen Williams)
108- added the unsupported "contrib" directory:
109   contrib/asm386/ by Gilles Vollant <info@winimage.com>
110	386 asm code replacing longest_match().
111   contrib/iostream/ by Kevin Ruland <kevin@rodin.wustl.edu>
112        A C++ I/O streams interface to the zlib gz* functions
113   contrib/iostream2/  by Tyge L�vset <Tyge.Lovset@cmr.no>
114	Another C++ I/O streams interface
115   contrib/untgz/  by "Pedro A. Aranda Guti\irrez" <paag@tid.es>
116	A very simple tar.gz file extractor using zlib
117   contrib/visual-basic.txt by Carlos Rios <c_rios@sonda.cl>
118        How to use compress(), uncompress() and the gz* functions from VB.
119- pass params -f (filtered data), -h (huffman only), -1 to -9 (compression
120  level) in minigzip (thanks to Tom Lane)
121
122- use const for rommable constants in deflate
123- added test for gzseek and gztell in example.c
124- add undocumented function inflateSyncPoint() (hack for Paul Mackerras)
125- add undocumented function zError to convert error code to string
126  (for Tim Smithers)
127- Allow compilation of gzio with -DNO_DEFLATE to avoid the compression code.
128- Use default memcpy for Symantec MSDOS compiler.
129- Add EXPORT keyword for check_func (needed for Windows DLL)
130- add current directory to LD_LIBRARY_PATH for "make test"
131- create also a link for libz.so.1
132- added support for FUJITSU UXP/DS (thanks to Toshiaki Nomura)
133- use $(SHAREDLIB) instead of libz.so in Makefile.in (for HPUX)
134- added -soname for Linux in configure (Chun-Chung Chen,
135- assign numbers to the exported functions in zlib.def (for Windows DLL)
136- add advice in zlib.h for best usage of deflateSetDictionary
137- work around compiler bug on Atari (cast Z_NULL in call of s->checkfn)
138- allow compilation with ANSI keywords only enabled for TurboC in large model
139- avoid "versionString"[0] (Borland bug)
140- add NEED_DUMMY_RETURN for Borland
141- use variable z_verbose for tracing in debug mode (Peter Deutsch).
142- allow compilation with CC
143- defined STDC for OS/2 (David Charlap)	
144- limit external names to 8 chars for MVS (Thomas Lund)
145- in minigzip.c, use static buffers only for 16-bit systems
146- fix suffix check for "minigzip -d foo.gz"
147- do not return an error for the 2nd of two consecutive gzflush() (Felix Lee)
148- use _fdopen instead of fdopen for MSC >= 6.0 (Thomas Fanslau)
149- added makelcc.bat for lcc-win32 (Tom St Denis)
150- in Makefile.dj2, use copy and del instead of install and rm (Frank Donahoe)
151- Avoid expanded $Id$. Use "rcs -kb" or "cvs admin -kb" to avoid Id expansion.
152- check for unistd.h in configure (for off_t)
153- remove useless check parameter in inflate_blocks_free
154- avoid useless assignment of s->check to itself in inflate_blocks_new
155- do not flush twice in gzclose (thanks to Ken Raeburn)
156- rename FOPEN as F_OPEN to avoid clash with /usr/include/sys/file.h
157- use NO_ERRNO_H instead of enumeration of operating systems with errno.h
158- work around buggy fclose on pipes for HP/UX
159- support zlib DLL with BORLAND C++ 5.0 (thanks to Glenn Randers-Pehrson)
160- fix configure if CC is already equal to gcc
161
162Changes in 1.0.5 (3 Jan 98)
163- Fix inflate to terminate gracefully when fed corrupted or invalid data
164- Use const for rommable constants in inflate
165- Eliminate memory leaks on error conditions in inflate
166- Removed some vestigial code in inflate
167- Update web address in README
168  
169Changes in 1.0.4 (24 Jul 96)
170- In very rare conditions, deflate(s, Z_FINISH) could fail to produce an EOF
171  bit, so the decompressor could decompress all the correct data but went
172  on to attempt decompressing extra garbage data. This affected minigzip too.
173- zlibVersion and gzerror return const char* (needed for DLL)
174- port to RISCOS (no fdopen, no multiple dots, no unlink, no fileno)
175- use z_error only for DEBUG (avoid problem with DLLs)
176
177Changes in 1.0.3 (2 Jul 96)
178- use z_streamp instead of z_stream *, which is now a far pointer in MSDOS
179  small and medium models; this makes the library incompatible with previous
180  versions for these models. (No effect in large model or on other systems.)
181- return OK instead of BUF_ERROR if previous deflate call returned with
182  avail_out as zero but there is nothing to do
183- added memcmp for non STDC compilers
184- define NO_DUMMY_DECL for more Mac compilers (.h files merged incorrectly)
185- define __32BIT__ if __386__ or i386 is defined (pb. with Watcom and SCO)
186- better check for 16-bit mode MSC (avoids problem with Symantec)
187
188Changes in 1.0.2 (23 May 96)
189- added Windows DLL support
190- added a function zlibVersion (for the DLL support)
191- fixed declarations using Bytef in infutil.c (pb with MSDOS medium model)
192- Bytef is define's instead of typedef'd only for Borland C
193- avoid reading uninitialized memory in example.c
194- mention in README that the zlib format is now RFC1950
195- updated Makefile.dj2
196- added algorithm.doc
197
198Changes in 1.0.1 (20 May 96) [1.0 skipped to avoid confusion]
199- fix array overlay in deflate.c which sometimes caused bad compressed data
200- fix inflate bug with empty stored block
201- fix MSDOS medium model which was broken in 0.99
202- fix deflateParams() which could generated bad compressed data.
203- Bytef is define'd instead of typedef'ed (work around Borland bug)
204- added an INDEX file
205- new makefiles for DJGPP (Makefile.dj2), 32-bit Borland (Makefile.b32),
206  Watcom (Makefile.wat), Amiga SAS/C (Makefile.sas)
207- speed up adler32 for modern machines without auto-increment
208- added -ansi for IRIX in configure
209- static_init_done in trees.c is an int
210- define unlink as delete for VMS
211- fix configure for QNX
212- add configure branch for SCO and HPUX
213- avoid many warnings (unused variables, dead assignments, etc...)
214- no fdopen for BeOS
215- fix the Watcom fix for 32 bit mode (define FAR as empty)
216- removed redefinition of Byte for MKWERKS
217- work around an MWKERKS bug (incorrect merge of all .h files)
218
219Changes in 0.99 (27 Jan 96)
220- allow preset dictionary shared between compressor and decompressor
221- allow compression level 0 (no compression)
222- add deflateParams in zlib.h: allow dynamic change of compression level
223  and compression strategy.
224- test large buffers and deflateParams in example.c
225- add optional "configure" to build zlib as a shared library
226- suppress Makefile.qnx, use configure instead
227- fixed deflate for 64-bit systems (detected on Cray)
228- fixed inflate_blocks for 64-bit systems (detected on Alpha)
229- declare Z_DEFLATED in zlib.h (possible parameter for deflateInit2)
230- always return Z_BUF_ERROR when deflate() has nothing to do
231- deflateInit and inflateInit are now macros to allow version checking
232- prefix all global functions and types with z_ with -DZ_PREFIX
233- make falloc completely reentrant (inftrees.c)
234- fixed very unlikely race condition in ct_static_init
235- free in reverse order of allocation to help memory manager
236- use zlib-1.0/* instead of zlib/* inside the tar.gz
237- make zlib warning-free with "gcc -O3 -Wall -Wwrite-strings -Wpointer-arith
238  -Wconversion -Wstrict-prototypes -Wmissing-prototypes"
239- allow gzread on concatenated .gz files
240- deflateEnd now returns Z_DATA_ERROR if it was premature
241- deflate is finally (?) fully deterministic (no matches beyond end of input)
242- Document Z_SYNC_FLUSH
243- add uninstall in Makefile
244- Check for __cpluplus in zlib.h
245- Better test in ct_align for partial flush
246- avoid harmless warnings for Borland C++
247- initialize hash_head in deflate.c
248- avoid warning on fdopen (gzio.c) for HP cc -Aa
249- include stdlib.h for STDC compilers
250- include errno.h for Cray
251- ignore error if ranlib doesn't exist
252- call ranlib twice for NeXTSTEP
253- use exec_prefix instead of prefix for libz.a
254- renamed ct_* as _tr_* to avoid conflict with applications
255- clear z->msg in inflateInit2 before any error return
256- initialize opaque in example.c, gzio.c, deflate.c and inflate.c
257- fixed typo in zconf.h (_GNUC__ => __GNUC__)
258- check for WIN32 in zconf.h and zutil.c (avoid farmalloc in 32-bit mode)
259- fix typo in Make_vms.com (f$trnlnm -> f$getsyi)
260- in fcalloc, normalize pointer if size > 65520 bytes
261- don't use special fcalloc for 32 bit Borland C++
262- use STDC instead of __GO32__ to avoid redeclaring exit, calloc, etc...
263- use Z_BINARY instead of BINARY
264- document that gzclose after gzdopen will close the file
265- allow "a" as mode in gzopen.
266- fix error checking in gzread
267- allow skipping .gz extra-field on pipes
268- added reference to Perl interface in README
269- put the crc table in FAR data (I dislike more and more the medium model :)
270- added get_crc_table
271- added a dimension to all arrays (Borland C can't count).
272- workaround Borland C bug in declaration of inflate_codes_new & inflate_fast
273- guard against multiple inclusion of *.h (for precompiled header on Mac)
274- Watcom C pretends to be Microsoft C small model even in 32 bit mode.
275- don't use unsized arrays to avoid silly warnings by Visual C++:
276     warning C4746: 'inflate_mask' : unsized array treated as  '__far'
277     (what's wrong with far data in far model?).
278- define enum out of inflate_blocks_state to allow compilation with C++
279
280Changes in 0.95 (16 Aug 95)
281- fix MSDOS small and medium model (now easier to adapt to any compiler)
282- inlined send_bits
283- fix the final (:-) bug for deflate with flush (output was correct but
284  not completely flushed in rare occasions).
285- default window size is same for compression and decompression
286  (it's now sufficient to set MAX_WBITS in zconf.h).
287- voidp -> voidpf and voidnp -> voidp (for consistency with other
288  typedefs and because voidnp was not near in large model).
289
290Changes in 0.94 (13 Aug 95)
291- support MSDOS medium model
292- fix deflate with flush (could sometimes generate bad output)
293- fix deflateReset (zlib header was incorrectly suppressed)
294- added support for VMS
295- allow a compression level in gzopen()
296- gzflush now calls fflush
297- For deflate with flush, flush even if no more input is provided.
298- rename libgz.a as libz.a
299- avoid complex expression in infcodes.c triggering Turbo C bug
300- work around a problem with gcc on Alpha (in INSERT_STRING)
301- don't use inline functions (problem with some gcc versions)
302- allow renaming of Byte, uInt, etc... with #define.
303- avoid warning about (unused) pointer before start of array in deflate.c
304- avoid various warnings in gzio.c, example.c, infblock.c, adler32.c, zutil.c
305- avoid reserved word 'new' in trees.c
306
307Changes in 0.93 (25 June 95)
308- temporarily disable inline functions
309- make deflate deterministic
310- give enough lookahead for PARTIAL_FLUSH
311- Set binary mode for stdin/stdout in minigzip.c for OS/2
312- don't even use signed char in inflate (not portable enough)
313- fix inflate memory leak for segmented architectures
314
315Changes in 0.92 (3 May 95)
316- don't assume that char is signed (problem on SGI)
317- Clear bit buffer when starting a stored block
318- no memcpy on Pyramid
319- suppressed inftest.c
320- optimized fill_window, put longest_match inline for gcc
321- optimized inflate on stored blocks.
322- untabify all sources to simplify patches
323
324Changes in 0.91 (2 May 95)
325- Default MEM_LEVEL is 8 (not 9 for Unix) as documented in zlib.h
326- Document the memory requirements in zconf.h
327- added "make install"
328- fix sync search logic in inflateSync
329- deflate(Z_FULL_FLUSH) now works even if output buffer too short
330- after inflateSync, don't scare people with just "lo world"
331- added support for DJGPP
332
333Changes in 0.9 (1 May 95)
334- don't assume that zalloc clears the allocated memory (the TurboC bug
335  was Mark's bug after all :)
336- let again gzread copy uncompressed data unchanged (was working in 0.71)
337- deflate(Z_FULL_FLUSH), inflateReset and inflateSync are now fully implemented
338- added a test of inflateSync in example.c
339- moved MAX_WBITS to zconf.h because users might want to change that.
340- document explicitly that zalloc(64K) on MSDOS must return a normalized
341  pointer (zero offset)
342- added Makefiles for Microsoft C, Turbo C, Borland C++
343- faster crc32()
344
345Changes in 0.8 (29 April 95)
346- added fast inflate (inffast.c)
347- deflate(Z_FINISH) now returns Z_STREAM_END when done. Warning: this
348  is incompatible with previous versions of zlib which returned Z_OK.
349- work around a TurboC compiler bug (bad code for b << 0, see infutil.h)
350  (actually that was not a compiler bug, see 0.81 above)
351- gzread no longer reads one extra byte in certain cases
352- In gzio destroy(), don't reference a freed structure
353- avoid many warnings for MSDOS
354- avoid the ERROR symbol which is used by MS Windows
355
356Changes in 0.71 (14 April 95)
357- Fixed more MSDOS compilation problems :( There is still a bug with
358  TurboC large model.
359
360Changes in 0.7 (14 April 95)
361- Added full inflate support.
362- Simplified the crc32() interface. The pre- and post-conditioning
363  (one's complement) is now done inside crc32(). WARNING: this is
364  incompatible with previous versions; see zlib.h for the new usage.
365
366Changes in 0.61 (12 April 95)
367- workaround for a bug in TurboC. example and minigzip now work on MSDOS.
368
369Changes in 0.6 (11 April 95)
370- added minigzip.c
371- added gzdopen to reopen a file descriptor as gzFile
372- added transparent reading of non-gziped files in gzread.
373- fixed bug in gzread (don't read crc as data)
374- fixed bug in destroy (gzio.c) (don't return Z_STREAM_END for gzclose).
375- don't allocate big arrays in the stack (for MSDOS)
376- fix some MSDOS compilation problems
377
378Changes in 0.5:
379- do real compression in deflate.c. Z_PARTIAL_FLUSH is supported but
380  not yet Z_FULL_FLUSH.
381- support decompression but only in a single step (forced Z_FINISH)
382- added opaque object for zalloc and zfree.
383- added deflateReset and inflateReset
384- added a variable zlib_version for consistency checking.
385- renamed the 'filter' parameter of deflateInit2 as 'strategy'.
386  Added Z_FILTERED and Z_HUFFMAN_ONLY constants.
387
388Changes in 0.4:
389- avoid "zip" everywhere, use zlib instead of ziplib.
390- suppress Z_BLOCK_FLUSH, interpret Z_PARTIAL_FLUSH as block flush
391  if compression method == 8.
392- added adler32 and crc32
393- renamed deflateOptions as deflateInit2, call one or the other but not both
394- added the method parameter for deflateInit2.
395- added inflateInit2
396- simplied considerably deflateInit and inflateInit by not supporting
397  user-provided history buffer. This is supported only in deflateInit2
398  and inflateInit2.
399
400Changes in 0.3:
401- prefix all macro names with Z_
402- use Z_FINISH instead of deflateEnd to finish compression.
403- added Z_HUFFMAN_ONLY
404- added gzerror()
405