summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* zlib 1.3v1.3Mark Adler2023-08-1831-74/+87
|
* Use original make and options when Makefile runs make.Mark Adler2023-08-171-2/+2
| | | | Also avoid the use of the -C option for non-GNU make.
* Avoid uninitialized and unused warnings in contrib/minizip.Mark Adler2023-08-172-2/+8
|
* Fix typo in preceding reversion commit.Mark Adler2023-08-171-1/+1
|
* Revert flipping of load flags in Makefile.in for z/OS.Mark Adler2023-08-171-6/+6
| | | | | It looked innocuous enough and worked on macOS, but failed on Ubuntu.
* Look for a cross-compile libtool first in configure.OldWorldOrdr2023-08-171-1/+3
| | | | Permit cross-compilation for Darwin.
* Avoid cmake deprecation warning.Bruno S Marques2023-08-171-1/+1
|
* Clarify requirement in zlib.h to avoid multiple flush markers.Mark Adler2023-08-171-2/+2
|
* Move load flags before object file in Makefile.in for Z/OS.Mark Adler2023-08-171-6/+6
|
* Add license to contrib/untgz.Bastian Germann2023-08-171-0/+16
| | | | A zlib license was agreed to by the authors.
* zlib now uses ANSI C function prototypes, so zlib2ansi not needed.Mark Adler2023-08-171-152/+0
|
* Fix bug when using gzflush() with a very small buffer.Mark Adler2023-08-141-2/+2
|
* Fix typos in contrib/ada.Dimitri Papadopoulos2023-08-142-3/+3
|
* Remove redundant includes in minizip.Mark Adler2023-08-132-4/+0
|
* Remove TRYFREE macro from minizip.Mark Adler2023-08-132-25/+19
|
* Replace gcc-9 with gcc-11 for macOS testing.Dimitri Papadopoulos2023-08-132-2/+2
| | | | | At some point, macos-latest stopped supporting gcc-9, and moved to gcc-11 and gcc-12.
* Suppress MSAN detections in deflate's slide_hash().Andrzej Hunt2023-08-121-0/+5
| | | | | | slide_hash() knowingly reads potentially uninitialized memory, see comment lower down about prev[n] potentially being garbage. In this case, the result is never used.
* Add memory sanitizer to configure (--memory).Mark Adler2023-08-121-5/+11
| | | | | | This also adds --address for the address sanitizer, in addition to the existing --sanitizer. -fno-omit-frame-pointer has been added for both sanitizers to improve the error reporting.
* Fix bug when gzungetc() is used immediately after gzopen().Mark Adler2023-08-111-0/+4
|
* Fix some spelling errors.Dimitri Papadopoulos2023-08-0313-17/+17
|
* Add minizip testing to Makefile.Mark Adler2023-08-031-1/+7
|
* Read multiple bytes instead of byte-by-byte in minizip unzip.c.Eugene Golushkov2023-08-031-96/+38
| | | | | Use a single ZREAD64 call in the unz64local_getShort/Long/Long64 implementation, rather than read it byte by byte.
* Support Haiku in minizip.Gilles Vollant2023-08-034-4/+4
|
* Correct dummy filetime() prototype in minizip.c.Xiang Xiao2023-08-031-1/+1
| | | | Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
* Match sign of printf directive to sign of argument in testzlib.Mark Adler2023-07-291-1/+1
|
* Match sign of printf directive to sign of argument in minizip.Mark Adler2023-07-291-1/+1
|
* Fix logic error in minizip argument processing.Mark Adler2023-07-291-1/+1
|
* Fix typos found by codespell in minizipDimitri Papadopoulos2023-07-298-36/+36
|
* Fix reading disk number start on zip64 files in minizip.Mark Adler2023-07-291-4/+2
|
* Remove duplicated code #806Gilles Vollant2023-07-291-1/+0
|
* minizip: Fix being unable to open empty zip fileRedworkDE2023-07-291-22/+26
|
* Avoid compiler complaints if _TIME_BITS defined when building zlib.Mark Adler2023-07-291-3/+2
| | | | | | zlib does not use time_t, so _TIME_BITS is irrelevant. However it may be defined anyway as part of a sledgehammer indiscriminately applied to all builds.
* Document in zlib.h the initialization of stream fields by the InitMark Adler2023-07-101-3/+5
| | | | and Reset functions.
* Correct comment in zlib.h on os setting in gzip header.Mark Adler2023-05-161-2/+3
| | | | | | The comment said that the os is set to 255, when in fact it has been set to the current os since zlib 1.2.3. Or at least our best guess at the os made at compile time.
* Remove duplicate "the" in zlib.h.Paul Marquess2023-05-011-1/+1
|
* Update Java and Perl links in README.Mark Adler2023-05-011-5/+4
|
* Fix test/example.c to work with FORCE_STORED.Mark Adler2023-04-181-5/+5
|
* Fix warnings on test/infcover.c.Mark Adler2023-04-171-2/+3
|
* Fix cast in minizip's ioapi.c for Windows.Mark Adler2023-04-171-1/+1
|
* No include file is needed for __int64 type on Windows.Mark Adler2023-04-173-3/+3
|
* Remove use of OF() from contrib/untgz and render it compilable.Mark Adler2023-04-151-35/+12
|
* Remove K&R function definitions from contrib/minizip.Mark Adler2023-04-1512-759/+514
|
* Remove K&R function definitions from infback9.Mark Adler2023-04-154-37/+20
|
* Remove K&R function definitions from zlib.Mark Adler2023-04-1524-1644/+931
| | | | | | | C2X has removed K&R definitions from the C function syntax. Though the standard has not yet been approved, some high-profile compilers are now issuing warnings when such definitions are encountered.
* Turn off C2X warning about deprecated K&R function syntax.Mark Adler2023-04-131-2/+2
| | | | This is a temporary workaround before excising the K&R prototypes.
* Add a NOPRIME #define to zran.c to not use inflatePrime().Mark Adler2023-04-131-2/+91
| | | | | This supports zlib versions earlier than 1.2.3 (July 2005), as well as incomplete clones of zlib that do not have inflatePrime().
* Rewrite of zran in examples. See version history in zran.c.Mark Adler2023-02-182-372/+348
|
* Assure that inflatePrime() can't shift a 32-bit integer by 32 bits.Mark Adler2023-02-171-0/+2
| | | | | | | | | The inflate() functions never leave state->bits greater than 24, so an inflatePrime() call could not cause this. The only way this could have happened would be by using inflatePrime() to fill the bit buffer with 32 bits, and then calling inflatePrime() a *second* time asking to insert zero bits, for some reason. This commit assures that a shift by 32 bits does not occur even in that case.
* Make z_size_t 64 bits when compiling on Windows with Z_SOLO.Mark Adler2023-01-253-3/+15
| | | | | | | Z_SOLO defines z_size_t as an unsigned long. However Windows and MinGW-w64 are LLP64, where a long is 32 bits, but a size_t is 64 bits. This makes z_size_t, used by adler32_z() and crc32_z(), 64 bits on those systems.
* Update zlib_how.html to match the web page, and add a license.Mark Adler2023-01-241-11/+15
|