Log message:
Bump the PKGREVISION for all packages which depend directly on perl,
to trigger/signal a rebuild for the transition 5.10.1 -> 5.12.1.

The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=..."), minus the packages updated after
the perl package update.

sno@ was right after all, obache@ kindly asked and he@ led the
way. Thanks!

Log message:
Bump the PKGREVISION for all packages which depend directly on perl,
to trigger/signal a rebuild for the transition 5.8.8 -> 5.10.0.

The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=...").

Log message:
Add DESTDIR support.

Log message:
Remove empty PLISTs from pkgsrc since revision 1.33 of plist/plist.mk
can handle packages having no PLIST files.

Log message:
Upgrade Update p5-Apache-Session-Wrapper from version 0.31 to 0.33.

Changes:

0.33   2006-09-26

- When setting up tests with Apache::Test, if the user has apache2 we
  need to make sure that they have the mod_apreq2.so library
  available. If they don't, the live tests will be skipped. Based on a
  patch from Derek Price. RT #21552.

0.32   2006-09-18

- Work around a (maybe) bug in Apache2::Cookie. Set -value to undef to
  in Apach2::Cookie->new causes it to return undef, rather than a
  cookie with no value. This meant that delete_session() caused an
  "attempt to call method of undefined value" error. Reported by Derek
  Price. RT #20992.

Log message:
Whitespace cleanup, courtesy of pkglint.
Patch provided by Sergey Svishchev in private mail.

Log message:
Update p5-Apache-Session-Wrapper to 0.31.
Patch provided by Martin Wilke via PR 34339.

0.31   2006-07-11

- Exactly 0.30 but now the Build.PL creates a "traditional"
  Makefile.PL. This fixes the problem that the generated "passthrough"
  Makefile.PL was requiring Apache::TestMB, because it's available on
  my system.

0.30   2006-07-10

- Exactly like 0.29 but does not declare a "build_requires" dependency
  on Apache::TestMB. Declaring this dependency causes CPAN.pm (and
  probably CPANPLUS) to try and install Apache::Test, which may not
  always install cleanly. If you do have it already installed, then it
  will be used to run the live tests.

0.29   2006-06-09

- When not running under mod_perl, Apache::Session::Wrapper simply did
  not even try to load CGI::Cookie to generate cookies. Patch by Derek
  Poon.

- When running under mod_perl 1 without Apache::Request, an exception
  was generated when trying to send a cookie.

- Apache::Session::Wrapper now only checks for err_headers_out and
  headers_out on the header_object, since this works for both mod_perl
  1 and 2, as well as Mason's FakeApache object.

- As of version CGI.pm 3.16, CGI::Cookie has a bake() method, but now
  Apache::Session::Wrapper explicitly will not call it, because if
  you're not running under mod_perl, it ends up calling print() to
  send the cookie header when it is called, which may not be what a
  user of this module wants. Patch by Derek Poon.

0.28   2006-05-23

- Allow a session_id parameter to the constructor, to force the use of
  a particular existing session.

- Fixed a bug where failing to create a new session was ignored if
  allow_invalid_id was true.

- Detect very weird case where tying to tie to Apache::Session
  succeeds but the tied hash has no _session_id key. This now throws
  an exception.

- Fixed bug when the header_object ->can('header_out') but this method
  does not return an object, as with HTML::Mason::FakeApache. Reported
  by Hans Dieter.

- The logic to load an appropriate cookie class if the class was
  already loaded and an exception had previously been thrown by an
  unrelated class.

- Added a way to register session subclass via RegisterClass() and
  RegisterFlexClass(), to allow people to use any session subclass
  they want without needing explicit support in
  Apache::Session::Wrapper.

0.27   not released

- This was a version used internally at Socialtext, but never
  released.

0.26   2005-09-20

- Finally should work properly under both mp1 and mp2.  Patches and
  tests by Daisuke Maki.

- No longer depends on Apache::Cookie or Apache2::Cookie, but it will
  use them if it can instead of CGI::Cookie.

0.25   2005-08-14

- Silenced a warning that could occur when using session length
  cookies under mod_perl.

0.24   2005-08-01

- Optional parameters for store modules were not handled properly when
  using Flex as the session class.  Added some tests.  Reported by
  Gavin Carr.

0.23   2005-07-26

- If two sets of valid parameters for a class shared a parameter, an
  error would be thrown if you used the second set, even if it was
  complete.  Reported by Victor Felix.

0.22   2005-07-18

- Another fix for mod_perl2, this one from Jonas.

0.21   2005-06-20

- The fix for mod_perl2 caused a new cookie to be sent for each
  request under mod_perl 1.  I love it!

0.20   2005-06-08

- Fix to make getting cookies work under mod_perl2.  Reported by
  Baldvin Kovacs.

Log message:
Point MAINTAINER to pkgsrc-users@NetBSD.org in the case where no
developer is officially maintaining the package.

The rationale for changing this from "tech-pkg" to \ 
"pkgsrc-users" is
that it implies that any user can try to maintain the package (by
submitting patches to the mailing list).  Since the folks most likely
to care about the package are the folks that want to use it or are
already using it, this would leverage the energy of users who aren't
developers.

Log message:
Bump the PKGREVISIONs of all (638) packages that hardcode the locations
of Perl files to deal with the perl-5.8.7 update that moved all
pkgsrc-installed Perl files into the "vendor" directories.

Log message:
Turn PERL5_PACKLIST into a relative path instead of an absolute path.
These paths are now relative to PERL5_PACKLIST_DIR, which currently
defaults to ${PERL5_SITEARCH}.  There is no change to the binary
packages.