7 - Added exUnixN option to allow creation of the "ux" extra field.
8 This allows 32-bit UID/GID to be stored.
9 - In one-shot mode use exUnixN rather than exUnix2 for the UID/GID.
11 * IO::Compress::Zlib::Extra::parseExtraField
12 - Fixed bad test for length of ID field
17 * t/105oneshot-zip-only.t
18 - CanonicalName test failure on Windows
22 - ExtAttr now populates MSDOS attributes
27 - Added CanonicalName option.
28 Note this option is set to true by default.
29 - Added FilterName option
32 - Fixed issue where setting $\ would corrupt the uncompressed data.
33 Thanks to Steffen Goeldner for reporting the issue.
36 - Handle case when external command contains a whitespace
42 - get globmapper tests working on VMS
45 * IO::Uncompress::Unzip
46 - Fixed limitation where Streamed Stored content was not supported.
50 * IO::Compress::Zip & IO::Uncompress::Unzip
51 - Added support for LZMA (method 14) compression/uncompresion.
54 - Fixed CRC issue when compression is Store or Bzip2 and Strict option
58 - Fixed Zip64 issue where the content size is exactly 0xFFFFFFFF
62 * RT #67931: Test failure on Windows
67 - Silence pod warnings.
70 - Removed duplicate words in pod.
74 - RT #56942: Testsuite fails when being run in parallel
76 - Reduce symbol import - patch from J. Nick Koston
78 - If the output buffer parameter passed to read has a value of
79 undef, and Append mode was specified when the file was opened,
80 and eof is reached, then the buffer paramer was left as undef.
81 This is different from when Append isn't specified - the buffer
82 parameter is set to an empty string.
84 - There area couple of issues with reading a file that contains an
85 empty file that is compressed.
86 Create with -- touch /tmp/empty; gzip /tmp/empty.
87 Issue 1 - eof is not true immediately. Have to read from the file
89 Issue 2 - readline incorrectly returns an empty string the first
90 time it is called, and (correctly) undef thereafter.
95 * Fixed typos & spelling errors.
100 * IO::Uncompress::Base
101 - An input file that had a valid header, and so would allow
102 creation of the uncompression object, but was then followed by
103 corrupt data would trigger an infinite loop when using the
108 - XFL default settings for max compression & fastest algorithm were
109 the wrong way around. Thanks to Andrey Zholos for spotting this.
111 * IO::Compress::Base::Common
112 - Fixed precedence problem in parameter parsing code.
117 - Updates to documentation.
118 - Changes default value for ExtAttr on Unix to 0100644
120 * IO::Uncompress::Unzip
121 Reworked the "Name" option and examples in the pod.
123 * IO::Uncompress::Base
124 Fixed problem with nextStream not returning 0 when there is no
125 next stream and Transparent is false.
130 Remove autoload code from Zlib.pm.
135 * IO::Uncompress::Zip
136 - Some updates to IO::Compress::Zip documentation.
137 - Fixed default setting for ExtAttr.
142 * IO::Uncompress::Unzip
143 The "Name" option wasn't documented.
145 * Allow zlib version check to be disabled by setting
146 TEST_SKIP_VERSION_CHECK environment variable.
152 Get memGunzip & memGzip to set $gzerrno
156 Export memGunzip, memGzip and zlib_version on demand
160 This sample was using IO::Uncompress::AnyInflate. Much better to
161 use IO::Uncompress::AnyUncompress.
163 2.023 9 November 2009
165 * IO::Compress::AnyUncompress
166 Added support for lzma_alone & xz.
170 * IO::Compress - Makefile.PL
175 * IO::Compress::Base.pm
176 - Less warnnings when reading from a closed filehandle.
178 - Fixed minor typo in an error message.
182 The PREREQ_PM dependency on Scalar::Util got dropped when
183 IO-Compress was created in 2.017.
186 * IO::Compress::Zip.pm
187 - Removed restriction that zip64 is only supported in streaming
189 - The "version made by" and "extract" fields in the zip64 end
190 central record were swapped.
191 - In the End Central Header record the "offset to the start of the
192 central directory" will now always be set to 0xFFFFFFFF when
194 - In the End Central Header record the "total entries in the
195 central directory" field will be set to 0xFFFF if zip64 is
196 enabled AND there are more than 0xFFFF entries present.
198 * IO::Uncompress::Unzip.pm
199 - Don't consume lots of memory when walking a zip file. This makes
200 life more bearable when dealing with zip64.
203 - documented that memGunzip cannot cope with concatenated gzip data
206 * Changed test harness so that it can cope with PERL5OPT=-MCarp=verbose
209 * IO::Compress::Gzip::Constants.pm
210 - GZIP_FEXTRA_MAX_SIZE was set to 0xFF. Should be 0xFFFF. This
211 issue came up when attempting to unzip a file created by MS
216 * IO::Uncompress::Base.pm
217 - Fixed problem with LimitOutput where a call to uncompress
218 created more uncompressed output, but didn't consume any of
219 the input buffer. The symptom is the underlying compression
220 library (zlib or bzip2) thinks the input stream is corrupt.
225 * IO::Uncompress::Adapter::Bunzip2
226 - Fixed problem with EOF check.
230 * IO::Uncompress::Bunzip2
231 - The interface to Compress-Raw-Bzip2 now uses the new LimitOutput
232 feature. This will make all of the bzip2-related IO-Compress modules
233 less greedy in their memory consumption.
236 - Fixed exTime & exUnix2
238 - Fixed 'Use of uninitialized value in pack' warning when using
243 * Merged IO-Compress-Base, IO-Compress-Bzip2, IO-Compress-Zlib &
244 Compress-Zlib into IO-Compress.
246 * The interface to Compress-Raw-Zlib now uses the new LimitOutput
247 feature. This will make all of the zlib-related IO-Compress modules
248 less greedy in their memory consumption.
250 * Removed MAN3PODS from Makefile.PL
252 * A few changes to get the test harness to work on VMS courtesy of
255 * IO::Compress::Base & IO::Uncompress::Base
256 Downgraded some croaks in the constructors to just set $! (by letting
257 the code attempt to open a file and fail).
258 This makes the behavior more consistent to a standard open.
261 * IO::Uncompress::Base
262 Doing a seek with MultiStream could drop some of the uncompressed
266 - Fixed problem with the uncompressed & uncompressed fields when
267 zip64 is enabled. They were set to 0x0000FFFF instead of
268 0xFFFFFFFF. Also the ZIP64 extra field was 4 bytes short.
269 Problem spotted by Dino Chiesa.
271 * IO::Uncompress::Unzip
272 - use POSIX::mktime instead of Time::Local::timelocal to convert
273 the zip DOS time field into Unix time.
276 - Documented Compress::Zlib::zlib_version()
279 2.015 3 September 2008
282 Backout changes made in 2.014
284 2.014 2 September 2008
287 Updated to check for indirect dependencies.
292 - Allow IO::Compress::Base::Parameters::parse to accept an
293 IO::Compress::Base::Parameters object.
298 - Silenced an uninitialised value warning when reading a line
299 at a time from a zip file where the content uses ZIP_CM_STORE.
300 [Problem spotted & fixed by Jeff Holt]
302 * IO::Compress::Base & IO::Uncompress::Base
303 - local-ise $!, $? et al in the DESTROY methods.
308 - Fixed problem that prevented the creation of a zip file that
309 contained more than one compression method.
311 * IO::Compress::Base::Common
312 - The private Validator class in this module clashes with another
313 CPAN module. Moved Validator into the IO::Compress::Base::Common
317 * IO::Uncompress::Unzip
318 - Print an error message if the zip file contains a
319 member compressed with bzip2 and IO::Uncompress::Bunzip2 is
321 - Could not cope with mixed compression zip files. For example a
322 zip file that contains both STORED and DEFLATED content.
327 * Fixed problem that meant Perl 5.10 could not upgrade this module.
332 * Removed the alpha status from File::GlobMapper
335 When writing output never output a zero length buffer.
336 Done to improve interoperability with other tied filenandle
339 * Changed IO::Uncompress::Base to not use the offset parameter of
340 the read method when reading from a filehandle.
342 The object returned from Net::FTP::retr implements a non-standard
343 read method. The third parameter is used for a timeout value
344 rather than an offset.
347 * Changed IO::Uncompress::Base to not use the eof method when
348 reading from a filehandle.
350 The object returned from Net::FTP::retr implements both the read
351 method and the eof method. Unfortunately the implementation of
352 the read method uses non-buffered IO (by using sysread) while
353 the eof method uses buffered IO. Mixing buffered and non-buffered
354 IO results in data corruption.
358 - Added exUnix2 option to allow storing of UID & GID.
359 - When running on a Unix derivative the ExtAttr option now defaults
360 to the equivalent of 0666. For all other systems the default
364 - Minor documentation issue with flush.
368 2.008 2 November 2007
370 * Minor documentation changes in README
372 * t/compress/truncate.pl
375 * IO::Compress::Gzip::Constants.pm
376 Tidied up the character classes used to defined invalid
377 FNAME & FCOMMENT fields for EBCDIC.
380 lib/Compress/Zlib.pm -- 1.x Backward Compatibility issues
381 gzclose - documented return value was wrong, should be 0 for ok.
382 gzflush - return value didn't match 1.x, should return 0 if ok.
383 [rt.cpan.org #29215] and Debian bug #440943 http://bugs.debian.org/440943
385 2.006 1 September 20007
388 Added INSTALLDIRS directive to install as a core module when built
391 * IO::Uncompress::RawDeflate
393 - Fixed export problem - "$RawDeflateError" and "rawdeflate" were
394 not being exported with ":all".
398 Fixed crc32 and adler32 tests in to remove ascii assumption.
400 - lib/Compress/Zlib.pm
401 Make gzreadline not support $/, just like in Compress::Zlib 1.x
402 Folk who want $/ with readline support can get it in
403 IO::Uncompress::Gunzip. [rt.cpan.org #28663] and
404 Debian bug #435656 http://bugs.debian.org/435656
409 * Stephen Turner reported a problem when using IO::Uncompress::Gunzip
410 with XML::Parser. Turns out there were two issues.
412 Firstly an IO::Uncompress object isn't an IO::Handle. It is now.
414 Secondly the implementation of "read" wasn't honouring this
416 SCALAR will be grown or shrunk to the length actually read.
418 In particular it didn't do the right thing on EOF.
421 * IO::Compress::Gzip & IO::Uncompress::Gunzip
423 - RFC1952 says that the FNAME & FCOMMENT header fields must be ISO
424 8859-1 (LATIN-1) characters. The code can optionally police this.
425 Added a fix for this logic when running on EBCDIC.
428 Added info about removing Compress::Zlib version 1, before
429 installing version 2.
433 * Made seek less wasteful of memory.
437 - Added Zip64 documentation.
439 - Fixed extended timestamp.
440 Creation time isn't available in Unix so only store the
441 modification time and the last access time in the extended field.
445 - Added ExtAttr option to control the value of the "external file
446 attributes" field in the central directory.
448 - Added Unix2 extended attribute ("Ux").
449 This stores the UID & GID.
453 - Fixed 050interop-gzip.t for Windows
455 * IO::Compress::Bzip2
457 - Fixed 050interop-bzip2.t for Windows
461 - rewrote memGzip using IO::Compress::Gzip::gzip
465 * Added explicit version checking
467 2.002 29 December 2006
469 * Documentation updates.
471 * Added IO::Handle to the ISA test in isaFilehandle
473 * Add an explicit use_ok test for Scalar::Util in the test harness.
474 The error message reported by 01misc implied the problem was
476 Also explicitly check that 'dualvar' is available.
479 - Fix append mode with gzopen.
482 - Allow gzopen to read from and write to a scalar reference.
484 2.001 1 November 2006
486 * Remove beta status.
488 2.000_14 26 October 2006
490 * IO::Uncompress::Base
491 Added support for $/ in record mode
493 * IO::Uncompress::Base
494 The readline interface was substantially slower than the 1.x
495 equivalent. This has now been sorted.
496 Thanks to Andreas J. Koenig for spotting the problem.
498 * IO::Uncompress::AnyUncompress
499 Added IO::Uncompress::Lzf to the list of supported uncompressors.
501 * IO::Uncompress::Base
502 Added TrailingData to one-shot interface.
504 * IO::Uncompress::AnyUncompress
505 Remove raw-deflate (RFC1951) from the default list of compressors
507 It can still be included if the new RawInflate parameter is
509 This change was made because the only way to tell if content is
510 raw-deflate is to attempt to uncompress it - a few false positives
511 have popped up recently, which suggests that auto-detecting raw
512 deflate is far from perfect.
513 The equivalent change has been made to IO::Uncompress::AnyInflate.
516 * Don't check that filehandles are writable. It would seem that
517 "-w *STDOUT" on windows returns false.
520 * IO::Uncompress::Deflate
521 Beefed up the magic signature check. Means less false positives
522 when auto-detecting the compression type.
524 * IO::Uncompress::UnZip
525 Tighten up the zip64 extra field processing to cope with the case
526 wheere only some of the local header fields are superseded.
528 * IO::Uncompress::AnyInflate
529 Remove raw-deflate (RFC 1951) from the default list of compressors
531 It can still be included if the new RawInflate parameter is
533 This change was made because the only way to tell if content is
534 raw-deflate is to attempt to uncompress it - a few false positives
535 have popped up recently, which suggests that auto-detecting raw
536 deflate is far from perfect.
537 The equivalent change has been made to IO::Uncompress::AnyUncompress.
540 2.000_13 20 June 2006
542 * Store compress & uncompressed sizes as 64-bit.
544 * For one-shot uncompression, like this
546 unzip "some.zip" => \@a, MultiStream => 1;
548 Push each uncompressed stream from "some.zip" onto @a.
550 * Added IO::Compress::Base::FilterEnvelope
552 * Added IO::Uncompress::Base::nextStream
554 * The '-' filehandle now maps to either *STDIN or *STDOUT.
555 This keeps mod_perl happier. Was using these before
560 * Preliminary support for reading zip files with zip64 members.
564 * Moved the code for creating and parsing the gzip extra field into
565 IO::Compress::Zlib::Extra.pm so that IO::Compress::Zip &
566 IO::Uncompress::Unzip can use it as well.
568 * Added ExtraFieldLocal & ExtraFieldCentral options to IO::Compress::Zip.
569 These allow the creation of user-defined extra fields in the local
570 and central headers, just like the ExtraField option in
573 * Moved the zip constants into IO::Compress::Zip::Constants
575 * Added exTime option to IO::Compress::Zip.
576 This allows creation of the extended timestamp extra field.
578 * Added Minimal option to IO::Compress::Zip.
579 This disables the creation of all extended fields.
581 * Added TextFlag option to IO::Compress::Zip.
583 * Documented Comment and ZipComment options in IO::Compress::Zip.
586 Fixed gzread to zap the output buffer to an empty string when zero
587 bytes are requested. This matches the behaviour of C::Z 1.x
589 2.000_11 10 April 2006
591 * Transparent + InputLength made more robust where input data is not
594 * Updated Documentation for zip modules.
596 * Changed IO::Compress::Zip 'Store' option to 'Method' and added
597 symbolic constants ZIP_CM_STORE, ZIP_CM_DEFLATE and ZIP_CM_BZIP2 to
598 allow the compression method to be picked by the user.
600 * Added support to allow bzip2 compressed data to be written/read
601 with IO::Compress::Zip and IO::Uncompress::Unzip.
603 * Beefed up 050interop-gzip.t to check that the external gzip command
604 works as expected before starting the tests. This means that
605 this test harness will just be skipped on problematic systems.
607 * Merged core patch 27565 from Steve Peters. This works around a
608 problem with gzip on OpenBSD where it doesn't seem to like
609 compressing files < 10 bytes long.
611 * Beefed up 050interop-bzip2.t to check that the external bzip2 command
612 works as expected before starting the tests. This means that
613 this test harness will just be skipped on problematic systems.
615 2.000_10 13 March 2006
617 * AnyUncompress doesn't assume that IO-Compress-Zlib is installed any
620 * Documentation updates.
623 Changed gzread so that its behaviour matches C::Z::gzread 1.x if it
624 is called after eof. In this case it will write an empty string
625 into the output parameter. This change is solely for backward
626 compatibility reasons.
628 2.000_09 3 March 2006
632 2.000_08 2 March 2006
634 * Split IO::Compress::Base into its own distribution.
636 * Split IO::Compress::Bzip2 into its own distribution.
638 * Added opened, autoflush and input_line_number.
640 * Beefed up support for $.
642 * Split IO::Compress::Zlib into its own distribution.
644 * Beefed up support for zip/unzip
646 * Breakout zlib specific code into separate modules.
648 * Limited support for reading/writing zip files
650 2.000_06 5 October 2005
652 * Added eof parameter to Compress::Zlib::inflate method.
654 * Fixed issue with 64-bit
656 2.000_05 4 October 2005
658 * Renamed IO::* to IO::Compress::* & IO::Uncompress::*
660 2.000_04 23 September 2005
662 * Fixed some more non-portable test that were failing on VMS.
664 * fixed problem where error messages in the oneshot interface were
667 2.000_03 12 September 2005
669 * Fixed some non-portable test that were failing on VMS.
671 * Fixed export of zlib constants from the IO::* classes
673 2.000_02 6 September 2005
675 * Split Append mode into Append and Merge
677 * Fixed typos in the documentation.
681 * Added libscan to Makefile.PL
683 * Added InputLength for IO::Gunzip et al
685 2.000_01 22 August 2005
687 * Fixed VERSION in Compress::Gzip::Constants
689 * Removed Compress::Gzip::Info from the distribution.
691 2.000_00 21 August 2005
693 * First Beta relase of Compress::zlib rewrite.
695 Compress-Zlib version 1 Changes
697 1.38 - 6 September 2005
699 * Integrate core change 25304 -- Symbian Update
701 * Added libscan to Makefile.PL
703 1.37 - 12 August 2005
705 * Change to t/03examples.t for VMS from Abe Timmerman
709 * Renamed zlib-src-1.2.3 to zlib-src to help VMS
711 * Fixed Makefile.PL for VMS
713 * Fixed t/03examples.t for VMS
715 * Added a couple of notes about incompatibility with Unix compress.
719 * Updated zlib source to 1.2.3
721 * Fixed problem with where two calls to gzclose would hang the debugger.
722 See https://rt.cpan.org/Ticket/Display.html?id=13789
724 * Added code from Alexey Tourbin to use XSLoader when available,
725 and DynaLoader otherwise.
727 * Documented that the compress & uncompress functions were not
728 the same as the Unix utilities of the same name.
730 * Fixed 05gzsetp -- it left a temp file hanging around.
732 * Integrate core change 24787 - SvUPGRADE returns void in blead
734 * Integrate core change 24788 - Makefile.PL adjustments for the core
737 1.34 - 30 January 2005
739 * Fixed typo in the README
741 * Fixed examples.t on Win32 where paths have embedded whitespace.
743 * Fix for Cygwin and core integration from Jos I. Boumans
745 * Upgrade zlib source to 1.2.2
747 1.33 - 14 January 2004
749 * Reworked Makefile.PL to avoid creating a private copy of zlib.
750 This both simplifies the build, plus it makes life easier for VMS.
752 * Patches for Makefile.PL to get it to work on VMS supplied by
755 * memGunzip has very slow on FreeBSD. Turns out to be down to
756 the way realloc works on FreeBSD. Changed both inflate & deflate
757 to use exponentially increasing buffer sizes when they need to
758 realloc. Thanks to Peter Jeremy for the lowdown on FreeBSD
761 1.32 - 26 November 2003
763 * Steve Hay reported a problem on rt.cpan.org with Windows and
764 MSCV++ 6.0 where the source from the zlib directory was getting
765 installed with the rest of the module.
766 https://rt.cpan.org/Ticket/Display.html?id=1741
768 This has been fixed by renaming the "zlib" directory to "zlib-src"
769 thus avoiding a conflict with the name of this Perl module.
771 * Fixed a bug in the inflate method where the input buffer is an
772 lvalue (via substr). Problem & solution reported by Salvador Fandiqo.
774 * Tightened up the logic in Makefile.PL when BUILD_ZLIB is
775 True. Issue spotted by Ralf S. Engelschall.
777 1.31 - 29 October 2003
779 * Reinstated the creation of .bak files - $^I seems to need a
780 backup file on Windows. For OpenVMS, the extension _bak is used.
782 1.30 - 28 October 2003
784 * Bundled a sub-set of the zlib source with the module and changed
785 the default make behaviour to build with the included zlib source.
786 The previous behaviour of using a pre-built zlib library is
787 still available for those that want it.
789 * Added prototypes to the subs in Zlib.pm that didn't already have
790 them. Patch from Ed Avis.
792 * No .bak files are created by Makefile.PL any more - this keep
793 distclean much happier. Patch suggested by Ed Avis.
794 This also fixes a similar problem reported by Dr. Martin Zinser
797 * Documentation for some of the gz functions updated.
799 * Format strings modified in DispStream to keep OpenVMS happy.
800 Problem reported by Dr. Martin Zinser.
805 * Makefile.PL now displays a warning about not installing
806 Compress::Zlib via the CPAN shell.
808 * Fix to allow intermingling of gzread & gzreadline - patch
809 supplied by Doug Perham.
811 * memGunzip will silently now work if the gzip trailer is
812 missing. Some HTTP Origin Servers seem to leave it out.
816 * Tests 148 & 150 from t/02zlib.t were failing on redhat 9.
818 * Added a few words about the problems with Mac OS X to the README file.
822 * Fixed bug in gzopen where $gzerrno wasn't being set correctly.
823 The symptom was $gzerrno was set to Z_MEM_ERROR although the file
824 was opened ok. If gzopen failed, $gzerrno was being set correctly.
825 This problem wasn't spotted before because the typical test
826 to determine whether gzopen passed or failed was to check its
829 1.19 - 31 October 2002
831 * fixed a problem with t/02zlib.t that was failing with redhat 8.
833 1.18 - 24 October 2002
835 * fixed a Win32 problem in t/02zlib.t by changing sysread to read.
837 * zlib 1.0.5 & older doesn't have gzsetparams & gzeof. Added a new
838 variable to config.in to flag an old version of zlib. Split
839 out the tests for gzsetparams into t/05gzsetp.t
843 * Moved the test to check the versions of libz & zlib.h into a separate
844 file and added troubleshooting notes to README.
846 * In gzopen, only attempt to call "tell" for normal files.
848 * Fixed to work in taint mode.
850 * Broke changes out of README into Changes file.
852 * Replaced internal use of Z_PARTIAL_FLUSH symbol with Z_SYNC_FLUSH.
853 zlib.h says /* will be removed, use Z_SYNC_FLUSH instead */
855 1.16 - 13 December 2001
857 * Fixed bug in Makefile.PL that stopped "perl Makefile.PL PREFIX=..."
860 1.15 - 4th December 2001
862 * Changes a few types to get the module to build on 64-bit Solaris
864 * Changed the up/downgrade logic to default to the older constructs, and
865 to only call a downgrade if specifically requested. Some older versions
866 of Perl were having problems with the in-place edit.
868 * added the new XS constant code.
870 1.14 - 27th August 2001
872 * Memory overwrite bug fixed in "inflate". Kudos to Rob Simons for
873 reporting the bug and to Anton Berezin for fixing it for me.
875 1.13 - 31st June 2001
877 * Make sure config.in is consistent when released.
879 1.12 - 28th April 2001
881 * Modified Makefile.PL to only enable the warnings pragma if
882 using perl 5.6.1 or better.
884 1.11 - 17th February 2001
886 * Added logic in Makefile.PL to toggle between using $^W and
887 the warnings pragma in the module.
889 * The module, the examples & the test harness are now all strict
892 1.10 - 5th February 2001
894 * fixed a bug in memGunzip. Used Z_ERR instead of Z_DATA_ERROR.
896 1.09 - 15th January 2001
898 * Silenced a few compiler warnings.
900 * Updated zlib home site in README & Zlib.pm to www.info-zip.org
902 * Minor typo in Zlib.pm - a link used AUTHORS instead of AUTHOR
903 -- spotted by Frank Martini.
905 * Mention Archive::Zip
907 * added memGunzip. This is largely based on code provided by Jim Leonard.
909 * $deflate->flush can now take an optional parameter. Valid
910 values are Z_NO_FLUSH, Z_PARTIAL_FLUSH, Z_SYNC_FLUSH, Z_FULL_FLUSH
911 and Z_FINISH. The default is Z_FINISH.
915 * uncompress was clobbering its input parameter. Now it doesn't.
916 This bug was spotted by Deven T. Corzine.
918 * If a variable that only ever contained a number was given
919 to compress or deflate, it would not be compressed properly. Now
920 it will be coerced to a string and then compressed. This bug
921 was spotted by Deven T. Corzine.
925 * ANSI-ified the static functions in Zlib.xs
927 * Added the ability to build zlib along with the module.
928 This feature is 90% based on a Makefile provided by Gurusamy
933 * Fixed a nasty problem where inflate could truncate the data
934 returned. Thanks to Douglas Thomson <dougt@mugc.cc.monash.edu.au>
935 for both spotting the problem and fixing the bug.
937 * Added a note about the undocumented features in zlib that are
938 required when accessing zip files.
940 * gzclose will now get called automatically when the gzip object is
945 * Previous release used newSVpvn, which doesn't exist in 5.004_04
946 or earlier. Changed to use newSVpv instead.
948 * The module needs Perl 5.004 or better, so updated the version
949 checking in Zlib.pm and Makefile.PL
953 * Bug 19990527.001: compress(undef) core dumps -- Fixed.
957 * Updated to use the new PL_ symbols.
958 Means the module can be built with Perl 5.005_5*
962 * The return codes for gzread, gzreadline and gzwrite were
963 documented incorrectly as returning a status code.
965 * The test harness was missing a "gzclose". This caused problem
966 showed up on an amiga. Thanks to Erik van Roode for reporting
969 * Patched zlib.t for OS/2. Thanks to Ilya Zakharevich for the patch.
973 * A number of fixes to the test suite and the example scripts to
974 allow them to work under win32. All courtesy of Gurusamy
979 * Fixed crc32 & adler32. They were very broken.
981 * The following functions can now take a scalar reference in
982 place of a scalar for their buffer parameters:
991 This should mean applications that make use of the module don't
992 have to copy large buffers around.
995 * Normally the inflate method consumes I<all> of the input buffer
996 before returning. The exception to this is when inflate detects
997 the end of the stream (Z_STREAM_END). In this case the input
998 buffer need not be completely consumed. To allow processing of
999 file formats that embed a deflation stream (e.g. zip, gzip),
1000 the inflate method now sets the buffer parameter to be what
1001 remains after inflation.
1003 When the return status is Z_STREAM_END, it will be what remains
1004 of the buffer (if any) after deflation. When the status is Z_OK
1005 it will be an empty string.
1007 This change means that the buffer parameter must be a lvalue.
1009 * Fixed crc32 and adler32. They were both very broken.
1011 * Added the Compress::Zlib::memGzip function.
1013 0.5 - Confirmed that no changes were necessary for zlib 1.0.3, or 1.0.4.
1015 The optional parameters for deflateInit and inflateInit can now
1016 be specified as an associative array in addition to a reference
1017 to an associative array. They can also accept the -Name
1020 gzopen can now optionally take a reference to an open
1021 filehandle in place of a filename. In this case it will call
1024 Added gzstream example script.
1026 0.4 - Upgrade to support zlib 0.99
1028 Added dictionary interface.
1030 Fixed bug in gzreadline - previously it would keep returning
1031 the same buffer. This bug was reported by Helmut Jarausch
1033 Removed dependency to zutil.h and so dropped support for
1035 DEF_MEM_LEVEL (use MAX_MEM_LEVEL instead)
1036 DEF_WBITS (use MAX_WBITS instead)
1038 0.3 - Added prototype specification.
1040 0.2 - Fixed a minor allocation problem in Zlib.xs
1042 0.1 - first alpha release. 2nd October 1995