view release on metacpan or  search on metacpan

Changes  view on Meta::CPAN

1.25 Sat 11 Oct 2008 - Adam Kennedy
    - Removing "use warnings" instances that somehow slipped in
    - Skip test if Digest::MD5 is not available

1.24 Sun 23 Aug 2008 - Adam Kennedy
    - Blatantly pander to CPANTS by adding use strict to a deprecated module
    - Add an explicit load of FileHandle since in some circumstances,
      calling GLOB->print() failed.
    - Fixed http://rt.cpan.org/Public/Bug/Display.html?id=25925 :
    - Archive-Zip wrote faulty .zip files when $\ was set (such as when running
      using perl -l).
    - Incorporated a heavily modified version of ECARROLL's test file.
    - Thanks for ECARROLL for reporting it, and helping with the investigation.
    - The fix was to convert all $fh->print(@data) to $self->_print($fh, @data)
      where the _print() method localizes $\ to undef.
    - Fixed http://rt.cpan.org/Ticket/Display.html?id=14132 :
    - Incorrect file permissions after extraction.
    - Archive-Zip did not set the file permissions correctly in
      extractToFileNamed().
    - Added t/10_chmod.t and t/data/chmod.zip. Changed
      lib/Archive/Zip/Member.pm.
    - Reported by ak2 and jlv (Thanks!)
    - SHLOMIF wrote the test script.
    - (SHLOMIF)
    - Removed a double "required module" from the Archive::Zip POD.
    - Fixed http://rt.cpan.org/Ticket/Display.html?id=24557 ("documentation
      improvement"):
    - mentioned Archive::Zip::MemberRead in a few places.
    - TODO:
        - 1. Add a method to Archive::Zip to get a ::MemberRead from an
          archive member using -> notation. (?)
        - 2. In the POD of ::MemberRead - replace the indirect object
          call.
        - Changed the POD of ::MemberRead:
        - replaced the indirect object construction with $PKG->new().
        - Fixed http://rt.cpan.org/Public/Bug/Display.html?id=34103 :
        - changed the example to read unless ( .. == AZ_OK) instead of
          unless ( != AZ_OK), which was incorrect.

1.23 Thu  8 Nov 2007 - Adam Kennedy
    - Temporarily skilling some failing tests on Win32 in the
      interests of toolchain sanity. (until we work out the
      real problem here)

1.22 Fri  2 Nov 2007 - Adam Kennedy
    - Fixing platform compatibility bugs in the new regression tests
      from 1.21.

1.21 Thu  1 Nov 2007 - Adam Kennedy
    - Tidying up copyright formatting a bit.
    - Disable the GPBF_HAS_DATA_DESCRIPTOR_MASK bit when auto-switching
      directory storage to STORED because of a WinZip workaround because
      the read code in Java JAR which was... ok, I really don't understand,
      but Roland from Verisign says this one extra line unbreaks JAR files,
      so I just applied it :)
    - fixed http://rt.cpan.org/Public/Bug/Display.html?id=27463 with a
      regression test - cannot add files whose entire filenames are "0".
      (SHLOMIF).
    - fixed http://rt.cpan.org/Public/Bug/Display.html?id=26384 with a
      regression test - Archive::Zip::MemberRead::getline ignores
      $INPUT_RECORD_SEPARATOR . The modified file in the bug had it to be
      reworked a bit and tests were added in the file
      08_readmember_record_sep.t.
    - Thanks to kovesp [...] sympatico.ca
    - (SHLOMIF)

1.20 Tue  5 Jun 2007 - Adam Kennedy
    - Removing dependency on File::Which due to public outburst of flaming
      on cpanra(n)tings by H.Merijn Brand. Try a simple email next time. :(
    - Embedding an entire copy of File::Which inside the tests instead as
      an alternative to compensating for the lack of build_requires.
    - Removing the docs directory.
      It only had out of date files and non-free copyrighted materials.
      The tarball was probably illegal to distribute as a result.
      (reported by Debian devs)

1.19 Internal use, public release skipped

1.18 Wed 25 Oct 2006 - Adam Kennedy
    - Changing to a production version for final release
    - No other changes of any kind

1.17_05 Tue 19 Sep 2006 - Adam Kennedy
    - Seperated the classes from the main file into seperate packages.
    - Merged the Zip.pod into the main Zip.pm file.
    - Applied default Perl::Tidy to all of the source files, to improve
      the readability and maintainability of the files.
    - Added license in Makefile.PL
    - Added some additional entries to the realclean files

1.17_03 Sat 16 Sep 2006 - Adam Kennedy
    - Adding dependency on File::Which to deal with problems on systems
      that lack zip and unzip programs. This really should be a build-time
      dependency only, but ExtUtils::MakeMaker lacks that capability.
    - Builds and tests cleanly on Win32 now.

1.17_02 Sun  7 May 2006 - Adam Kennedy
    - Renamed the test scripts to the more conventional 01_name.t style
    - Upgraded all test scripts from Test.pm to Test::More (removing Test.pm dependency)
    - Various other miscellaneous cleanups of the test scripts
    - Removed MANIFEST and pod.t from repository (will be auto-generated)
    - Some cleaning up of the POD documentation for readability
    - Added SUPPORT section to docs
    - Merged external TODO file into the POD as a more-common TO DO section
    - Added a BUGS section to the docs

1.17_01 Sun 30 Apr 2006 - Adam Kennedy
    - Imported Archive::Zip into http://svn.ali.as/cpan/ orphanage.
      If you have a CPAN login and have released a module, ask ADAMK about an
      account and you can repair your bug directly in the repository.
    - Removed the revision comments from the old CVS repository
    - DOS DateTime Format doesn't support dates before 1980 and goes crazy when
      decoding back to unix time. If we don't get passed a time at all
      (0 or undef) we now throw an error.
    - DOS DateTime Format doesn't support dates before 1980, so if we find any
      we warn and use Jan 1 12:01pm 1980 if we encounter any
    - Win32 doesn't support directory modification times.
      Tentatively use the current time as the mod-time to prevent sending
      null times to the unix2dos converter (and the resulting error)
    - Reformat the expected empty zip warning in the output to add a note that
      the warning is entirely normal. Would be nice if some time later we can