use -fno-reorder-blocks for sparc64, mips, and vax and GCC 4.5*.

something in op.c (as miniop.c) is mis-compiled with this option which
is enabled by -O2, when using GCC 4.5.  i didn't try to figure out
exactly what as op.c is 419,359 bytes long and the assembler output
is almost 100% different and approximiately 1.5MB either way (the
diff of the asm output is larger than the combined inputs), so for now
we have this hack.  this problem doesn't appear to occur in newer GCC.

XXX: pullup to 2015Q2.

(mrg)

Updated mail/p5-Email-Send to 2.201

(wen)

Update to 2.201

Upstream changes:
2.201    2015-03-05
          ** PLEASE CONSIDER USING EMAIL-SENDER INSTEAD **

          add x_deprecated metadata

(wen)

Updated mail/p5-Mail-POP3Client to 2.19

(wen)

Update to 2.19
Add LICENSE

Upstream changes:
2.19 2013/19/03
        + RT82713 - version number contains a space.

(wen)

Updated mail/p5-Email-Address to 1.907

(wen)

Update to 1.907

Upstream changes:
1.907    2015-02-03 17:48:10-05:00 America/New_York
        - restore 5.8.x compatibility

1.906    2015-02-03 16:48:11-05:00 America/New_York
        - the test suite tests parses more directly, which eliminates a number
          of misleading (and possibly wrong) test cases
        - escaped characters in a quoted string are now properly unescaped
          for the ->phrase method, which means things should round trip better
          from their decomposed form

(wen)

Recursive revbump associated with lang/ocaml update.

(jaapb)

Updated lang/camlp5 to 6.14

(jaapb)

Updated package to latest version, 6.14. No changes to this release,
except that camlp5 is now compatible with ocaml 4.02.3.

(jaapb)

Updated lang/ocaml to 4.02.3

(jaapb)

This is an update of lang/ocaml to its latest version, 4.02.3. Changes
include:

Bug fixes:
- Top-level custom printing for GADTs: interface change in 4.02.2
  (Gr辿goire Henry, report by Jeremy Yallop)
- corrupted final_table (ygrek)
- R#6926: Regression: ocamldoc lost unattached comment
  (Damien Doligez, report by Fran巽ois Bobot)
- Aliased result type of GADT constructor results in assertion failure
  (Jacques Garrigue)

Feature wishes:
- PR#6691: install .cmt[i] files for stdlib and compiler-libs
  (David Sheets, request by Gabriel Radanne)
- New primitive: caml_alloc_dummy_function
  (Hugo Heuzard)

(jaapb)

Remove PKGREVISION setting from Makefile.common. This should never be here.

Noted by Matthias Ferdinand on pkgsrc-users.

(wiz)

netatalk requires libevent 2.x, the builtin one on NetBSD<7 is older.
Fixes pkg/50084.

(fhajny)

Updated math/R-DBI to 0.3.1

(wen)

Update to 0.3.1

Upstream changes:
# Version 0.3.1

* Actually export `dbIsValid()` :/

* `dbGetQuery()` uses `dbFetch()` in the default implementation.

# Version 0.3.0

## New and enhanced generics

* `dbIsValid()` returns a logical value describing whether a connection or
  result set (or other object) is still valid. (#12).

* `dbQuoteString()` and `dbQuoteIdentifier()` to implement database specific
  quoting mechanisms.

* `dbFetch()` added as alias to `fetch()` to provide consistent name.
  Implementers should define methods for both `fetch()` and `dbFetch()` until
  `fetch()` is deprecated in 2015. For now, the default method for `dbFetch()`
  calls `fetch()`.

* `dbBegin()` begins a transaction (#17). If not supported, DB specific
  methods should throw an error (as should `dbCommit()` and `dbRollback()`).

## New default methods

* `dbGetStatement()`, `dbGetRowsAffected()`, `dbHasCompleted()`, and
  `dbGetRowCount()` gain default methods that extract the appropriate elements
  from `dbGetInfo()`. This means that most drivers should no longer need to
  implement these methods (#13).

* `dbGetQuery()` gains a default method for `DBIConnection` which uses
  `dbSendQuery()`, `fetch()` and `dbClearResult()`.

## Deprecated features

* The following functions are soft-deprecated. They are going away,
  and developers who use the DBI should begin preparing. The formal deprecation
  process will begin in July 2015, where these function will emit warnings
  on use.

    * `fetch()` is replaced by `dbFetch()`.

    * `make.db.names()`, `isSQLKeyword()` and `SQLKeywords()`: a black list
      based approach is fundamentally flawed; instead quote strings and
      identifiers with `dbQuoteIdentifier()` and `dbQuoteString()`.

* `dbGetDBIVersion()` is deprecated since it's now just a thin wrapper
  around `packageVersion("DBI")`.

* `dbSetDataMappings()` (#9) and `dbCallProc()` (#7) are deprecated as no
  implementations were ever provided.

## Other improvements

* `dbiCheckCompliance()` makes it easier for implementors to check that their
  package is in compliance with the DBI specification.

* All examples now use the RSQLite package so that you can easily try out
  the code samples (#4).

* `dbDriver()` gains a more effective search mechanism that doesn't rely on
  packages being loaded (#1).

* DBI has been converted to use roxygen2 for documentation, and now most
  functions have their own documentation files. I would love your feedback
  on how we could make the documentation better!

(wen)

Updated mail/p5-Mail-Sender to 0.8.23

(wen)

Update to 0.8.23
Add LICENSE

Upstream changes:
Version 0.8.21
- fixed the case in the header names (Content-type -> Content-Type)
- fixed the encoding of non-ascii stuff in headers (RFC 2047)
- fixed using the ctype even for body with MailFile

Version 0.8.23
- fixed problems with long subjects containing accentuated characters
- fixed problems with IO::Socket::SSL
- trims addresses passed in an array (per request)
- changed licence per request from Fedora
- fix to allow very long passwords
- fix error in TestServer()

(wen)

Updated textproc/p5-Pod-Tree to 1.25

(wen)

Update to 1.25

Upstream changes:
1.25 2015.05.25
    - Rename test files to have number prefix.
    - Move test reqirements to TEST_REQUIRES or BUILD_REQUIRES for older EUMM
    - Older versions of EU::MM require quotes around 2-dot versions (CHORNY)

1.24  2015.05.22
    - Include the rc files in the distribution to use the proper
      Perl::Critic configuration.
    - use Test::Version to make sure we have the same version number
      in every module.
    - Configure Perl::Critic to be level 4.
    - Lots of other refactorings.

(wen)

Updated devel/p5-Convert-Binary-C to 0.77

(wen)

Update to 0.77

Upstream changes:
Version 0.77 (2015-07-22)
-------------------------

* ensure consistent order of enums when generating code; this was
  a real bug and caused the configuration part of C::B::C to report
  wrong (even invalid) configurations back to the user when built
  using a recent version of Perl that randomises hash keys

* fix cpan #85264: the POD test was reporting a missing =back tag
  in one of the support modules; the ultimate fix was to ensure only
  the provided modules were POD checked

* fix cpan #83167: an artificially low refcnt for immortal variables
  introduced recently for debug Perl builds was causing the refcnt
  tests to fail for SvNULLs; thanks to Reini Urban for his patch;
  also thanks to Nicholas Clark for pointing me to the Perl commit

* fix cpan #78165: spelling errors in the documentation; thanks to
  gregor herrmann for providing a patch

* fix deprecated use of unescaped braces in regex

* fix some gcc and Perl warnings

* switch from SnapshotCM to git repo: github.com/mhx/Convert-Binary-C/

(wen)

Restore i386 distfile.
Fix PR pkg/50130.

(ryoon)

Updated devel/p5-Module-Pluggable to 5.2

(wen)

Update to 5.2

Upstream changes:
2015-08-05 - 5.2
    Use Module::Runtime if it's installed (thanks to Olivier Mengu茅)
    Switch to using ExtUtils::MakeMaker (thanks to Karen Etheridge and Tim Orling)

(wen)

Add support for CFLAGS

No functional change intended.

(khorben)

Add support for CFLAGS

No functional change intended.

(khorben)

Add support for CFLAGS

No functional change intended.

(khorben)

Add support for CFLAGS

No functional change intended.

(khorben)

Add support for CFLAGS

No functional change intended.

(khorben)

Add support for CFLAGS

No functional change intended.

(khorben)

Updated chat/finch to 2.10.11nb2

(wiz)

Disable gstreamer option by default.
Suggested by Michael McConville to get rid of the x11 dependency
in finch, which you really don't want. Use chat/pidgin for X11.

Bump PKGREVISION.

(wiz)

Added databases/py-firebase version 1.2.

(kleink)

+ py-firebase

(kleink)

Import python-firebase-1.2 as databases/py-firebase.

Python interface to Firebase's REST API.

(kleink)

Updated devel/libgsf to 1.14.34

(wiz)

Update to 1.14.34, so that gnumeric112 has a chance of building.

libgsf 1.14.34

Morten:
* Fix OLE2 property parsing problem.  [#748528]
* Fuzzed file fixes.  [#749120] [#749169] [#749183] [#750809]
  [#751273]
* xlsx: fix problem with absolute relation targets.  [#751120]
* Fix text line problem with very long lines.

--------------------------------------------------------------------------
libgsf 1.14.33

Morten:
* Ensure floats roundtrip through our xml functions.
* Improve handling of unknown xml tags.
* xml parser improvements.
* xml pretty-printing improvements.

--------------------------------------------------------------------------
libgsf 1.14.32

Morten:
* Fix obscure problem with xml parser.
* Add vml to openpkg utils.
* Allow NULL content type for openpkg.
* Plug leak.

--------------------------------------------------------------------------
libgsf 1.14.31

Andreas:
* Add two new localc name spaces to gsf-opendoc-utils

Allin Cottrell:
* Add support for non-default zip compression level.  [#722470]

Morten:
* Plug leaks.
* Dead kittens.
* Reduce number of CRITICALs for corrupted files.
* Read zip files with 64k+ members.  [Part of #732209]
* Read zip files members larger than 4G.  [Part of #732209]
* Speed up zip file reading for lots of members.
* Speed up stdio directory handling with large number of files.
* Fix various issues with files larger than 4G.
* Fix minor zip file issues.
* Write zip archives with more than 64k+ members.
* Store unix modtime in zip.  (Until that overflows.)
* Fix seekability checks in GsfOutputIOChannel.
* Avoid implementation defined behaviour of shifts.
* Start a test suite.

(wiz)

Updated converters/psiconv to 0.9.9

(wiz)

Update to 0.9.9:

0.9.9  20141019  Bugfix in HTML4 target. Updated toolchain.
                Support for GraphicsMagick. Updated manual page psiconv(1).

0.9.8  20051115  Small bug fixes

(wiz)

Updated sysutils/spice-protocol to 0.12.9

(wiz)

Update to 0.12.9:

Major changes in 0.12.9
=======================
* rename newly introduced SpiceImageCompress enum to SpiceImageCompression
  as otherwise it was clashing with the definition used by older spice-server
  releases, breaking QEMU build

Major changes in 0.12.8
=======================
* add LZ4 support
* add audio volume synchronization
* deprecate unused vdi-dev
* add 'preferred-compression' message/capability
* add a new Windows driver escape code to send
  monitors capability from guest to client

(wiz)

Updated net/p5-Net to 3.07

(wiz)

Update to 3.07:

3.07 2015-07-17

    - Net::FTP::rmdir() has been made more robust by making use of the MLSD
      command in addition to the NLST command since the latter is known not to
      be processed correctly by some FTP servers.

      [Chris Lindee, CPAN RT#100694]

    - Net::FTP, Net::NNTP, Net::POP3 and Net::SMTP can now restrict domain to
      IPv4 even if IPv6 is available by using the new Domain or Family argument.

      Net::NNTP now supports the LocalPort argument in addition to LocalAddr.

      Net::POP3 now supports the LocalAddr and LocalPort arguments in addition
      to ResvPort (which is retained for backwards compatibility).

      [Steffen Ullrich, PR#18]

    - Fixed a bug in Net::Cmd::datasend() which caused octets in [\x80-\xFF]
      stored in a "binary string" to be replaced with their UTF-8 encodings if
      the string happened to be stored internally in an "upgraded" state (i.e.
      with the UTF-8 flag on).  (As noted below, strings passed to datasend()
      should always be encoded first, and therefore not stored in such a state
      anyway, but it is all too easy for perl to change this internal state
      unless the encodeing is done at the very last minute before calling
      datasend(), so it helps if datasend() plays more nicely in this case.  In
      particular, it was wrong of datasend() to treat upgraded and downgraded
      strings differently when their contents were identical at the Perl level.)

      This bugfix results in a breaking change to the case of a "text string"
      with characters in U+0080..U+00FF stored internally in an upgraded state
      since those characters are likewise no longer encoded to UTF-8 by
      datasend(), but callers of datasend() should not have been relying on this
      behaviour anyway: In general, datasend() has no idea what encoding is
      required for output so callers should always encode the data to be output
      to whatever encoding is required first.  This has now been clarified in
      the documentation.

      Finally, a text string with characters >= U+0100 will now cause a "Wide
      character in print" warning from datasend() since such characters cannot
      be output as bytes and datasend() no longer encodes to UTF-8.  In this
      case, UTF-8 bytes will still be output as before since that happens to be
      the internal representation of such characters, but the warning is new.
      Callers should heed this warning and encode such strings to whatever
      encoding is required before calling datasend(), as noted above.

      [Ricardo Signes, CPAN RT#104433]

(wiz)

Updated misc/p5-Locale-libintl to 1.24

(wiz)

Update to 1.24:

Version 1.24 - 08 Apr 2013

* Better portability for tests.

* Updated documentation.

* License changed to GPLv3.

* Bugs fixed:

  https://rt.cpan.org/Ticket/Display.html?id=98109
  https://rt.cpan.org/Ticket/Display.html?id=47968
  https://rt.cpan.org/Ticket/Display.html?id=83980
  https://rt.cpan.org/Ticket/Display.html?id=102288

(wiz)

Updated devel/p5-Module-Build-XSUtil to 0.16

(wiz)

Update to 0.16:

0.16 2015-06-29T15:55:13Z
    - Fix regexp of _gcc_version for Ubuntu and Debian
        (syohex)
    - Set -std=c99 explicitly for older GCC
        (syohex)

(wiz)

Updated devel/p5-Module-Build to 0.42140

(wiz)

Update to 0.4214:

0.4214 - Fri Jun 12 00:25:00 CEST 2015

  - Released 0.42_13 as 0.4214

0.42_13 - Sat Jun  6 21:18:24 CEST 2015

  [BUG FIXES]

  - Handle failure to guess license from key better

  [ENHANCEMENTS]

  - Output data in a stable order [J辿r辿my Bobbio]

(wiz)

Added devel/p5-inc-latest version 0.500

(wiz)

+ p5-inc-latest.

(wiz)

Import p5-inc-latest-0.500 as devel/p5-inc-latest.

The inc::latest module helps bootstrap configure-time dependencies
for CPAN distributions. These dependencies get bundled into the
inc directory within a distribution and are used by Makefile.PL or
Build.PL.

Arguments to inc::latest are module names that are checked against
both the current @INC array and against specially-named directories
in inc. If the bundled version is newer than the installed one (or
the module isn't installed, then, the bundled directory is added
to the start of @INC and the module is loaded from there.

(wiz)

Updated geography/py-geojson to 1.2.2.

(kleink)

Update py-geojson to 1.2.2.

1.2.2 (2015-07-13)
------------------

- Fix tests by including test file into build

  - https://github.com/frewsxcv/python-geojson/issues/61

- Build universal wheels

  - https://packaging.python.org/en/latest/distributing.html#universal-wheels

1.2.1 (2015-06-25)
------------------

- Encode long types correctly with Python 2.x

  - https://github.com/frewsxcv/python-geojson/pull/57

1.2.0 (2015-06-19)
------------------

- Utility function to validate GeoJSON objects

  - https://github.com/frewsxcv/python-geojson/pull/56

1.1.0 (2015-06-08)
------------------

- Stop outputting invalid GeoJSON value id=null on Features

  - https://github.com/frewsxcv/python-geojson/pull/53

(kleink)

add and enable go14

(jnemeth)

Added devel/p5-inc version 0.06

(wiz)

+ p5-inc.

(wiz)

Import p5-inc-0.06 as devel/p5-inc.

The inc module redefines @INC to a list of predefined smart objects.
These objects are really just code refs for handy lookup techniques.
For example, only finding modules that were core in Perl 5.8.1, or
only finding non-core modules that are declared prerequisites of
some module.

(wiz)

Updated converters/p5-JSON-MaybeXS to 1.003005

(wiz)

Update to 1.003005:

1.003005 - 2015-03-22
- fix x_contributors metadata that was killing metacpan (see
  https://github.com/CPAN-API/cpan-api/issues/401)

(wiz)

Updated devel/p5-SUPER to 1.20141117

(wiz)

Update to 1.20141117:

1.20141117 2014-11-17 09:12:01-08:00 America/Los_Angeles
    - improved export mechanism
    - removed useless Exporter tests
    - improved docs (RT #79681)

1.20141116 2014-11-16 14:07:18-08:00 America/Los_Angeles
    - resolved Test::More changes (RT #97939)

(wiz)

Updated devel/p5-Test-DependentModules to 0.21

(wiz)

Update to 0.21. Improve DESCR.

0.21  2015-07-17

- If PERL_TEST_DM_CPAN_VERBOSE was true and Test::More had not been loaded in
  the process you would get an error. GH #9.

(wiz)

Updated devel/p5-Test-Class to 0.50

(wiz)

Update to 0.50:

0.50  [2015-06-07]
    - add links to Ovid's tutorial series on Test::Class (Tim Vroom, PR#19)
    - add links to Test::Class::Most, Test::Class::Moose (Tim Vroom, PR#20)
    - list some distributions that use Test::Class in their test suite (Tim
      Vroom, PR#21)
    - update documentation about running individual tests (Matthew Horsfall,
      PR#22)
    - fix some tests to work with the new Test::Stream (Chad Granum, PR#27)
    - switch packaging to ExtUtils::MakeMaker

(wiz)

Updated devel/p5-Inline-C to 0.76

(wiz)

Update to 0.76. Fixes test failure noted in Makefile.

0.76 Wed Jul  1 08:09:50 PDT 2015
- Support for cppflags

(wiz)

Updated devel/p5-IO-All to 0.86

(wiz)

Update to 0.86:

0.86 Thu Jan  1 18:44:25 PST 2015
- Revamp docs

(wiz)

Updated www/p5-Hijk to 0.24

(wiz)

Update to 0.24.

0.24: # 2015-07-05T13:40:00+0200

- Minor copyediting and formatting changes to the documentation. No
  code changes at all.

0.23: # 2015-07-03T17:00:00+0200

- The "Host" header can now be overriden by supplying a new
  `no_default_host_header` option along with a `Host` header in `head
  => []` to request().

  Before this we'd always send "Host: $host" over, where $host was the
  host we were connecting to, now you can customize this.

- Fixed a bug where if passed passed `head => []` to request() we'd
  emit a ":" header, i.e. just an empty header name with an empty
  value.

  You could have just not passed the `head => ` value if the array was
  empty, but no we won't screw up and emit a single line consisting of
  ":" if given an empty array.

0.22: # 2015-05-27T07:54:17+0200

- No feature change. Re-package due to a missing file in the tarball:
  https://rt.cpan.org/Ticket/Display.html?id=104624

0.21: # 2015-05-22T15:26:23+0200

- Fix "Too many CRLF" issue. Hijk has been always generating HTTP
  request with an extra CRLF at the end. While many HTTP servers are
  ignoring those, some treat it as errors. We now eliminate the extra
  CRLF at the end of every request.
  See also http://www.w3.org/Protocols/rfc2616/rfc2616-sec4.html

- Handle better when select() is interrupted by signals.

(wiz)

Updated www/p5-HTTP-Server-Simple to 0.50

(wiz)

Update to 0.50, add missing p5-CGI dependency:

0.50    2015-02-02
  - All collected changes since 0.44

0.45_08 2015-01-29
- IPv6 skipping code always turned off tests, rather than just
  skipping on bad hosts.

0.45_07 2015-01-27T22:24Z
- Skip IPv6 tests in 01live.t on machines where IPv6 isn't configured

0.45_06 2015-01-23T17:55Z
- Further test iteration and diagnostics for smoke failures

0.45_05 2015-01-22T16:23Z
- Test diagnostics for cpantester failures
- Hopefully better localhost handling

0.45_04 2015-01-20T17:40Z
- Fix compatibility for the IPv6 support in .45_02
- Hopefully improve the live tests

0.45_03 2015-01-14T19:48Z
- Remove useless VERSION from HTTP::Server::Simple::CGI and
  HTTP::Server::Simple::CGI::Environment

0.45_02 2012-05-20T18:15Z
- Support for IPv6 - Daniel Kahn Gillmor [rt.cpan.org #61200]

0.45_01 2011-09-22T10:10:41Z
- Support full URIs as required by RFC2616 - penfold [rt.cpan.org #69445]

(wiz)

Updated devel/p5-File-DesktopEntry to 0.12

(wiz)

Update to 0.12:

0.12 2015-06-20
- Remove spurious Module::Build from test. Reported by gregor herrmann,
  Debian team, https://bugs.debian.org/789210

0.11 2015-06-11
- Explicitly state minimum perl version as 5.8 in Makefile.PL.

0.10 2015-06-09
- Include new test in MANIFEST
    RT 105112 - SREZIC

0.09 2015-06-09
- Exclude newlines from whitespace on either side of '='
    RT 65394, fix by sdme.

(wiz)

Updated security/p5-Digest-JHash to 0.09

(wiz)

Update to 0.09:

0.09    2015-05-30
    - Convert to Dist-Zilla.
        - This implies several enhancements including links to the VCS
        repositories in META.yml/META.json.

(wiz)

Updated devel/p5-Devel-CheckCompiler to 0.06

(wiz)

Update to 0.06:

0.06 2015-06-30T05:22:51Z
    - More strict C99 check code
      (tokuhirom)
    - Fix for older GCC(< 5.0)
      (Syohei YOSHIDA)

(wiz)

Updated devel/p5-Devel-CheckBin to 0.04

(wiz)

Update to 0.04:

0.04 2015-07-22T07:35:08Z

    - Test case fix.
      * win32 fix.
      * added quotes for perl installed into a path with spaces in it
      (mohawk2++)

(wiz)

Updated devel/p5-Class-Load to 0.23

(wiz)

Update to 0.23:

0.23    2015-06-25
    - remove use of namespace::clean

(wiz)

Updated time/p5-DateTime-Format-Strptime to 1.5600

(wiz)

Update to 1.56. Changes not found, but fixes test failure noted in pkgsrc
Makefile.

(wiz)

Updated devel/p5-Data-Dump to 1.23

(wiz)

Update to 1.23:

2015-06-09  Gisle Aas <gisle@ActiveState.com>

  Release 1.23

  Avoid "Negative repeat count does nothing"-warnings from perl-5.22

  Typo fixes by David Steinbrunner

(wiz)

Updated databases/p5-DBM-Deep to 2.0012

(wiz)

Update to 2.0012:

2.0012 Jun 17 20:30:00 2015 PDT
    - Improved transaction validation and warnings (GH#12)

(wiz)

Updated devel/p5-Test-Exception to 0.40

(wiz)

Update to 0.40:

0.40  [2015-06-05]
    -  Updated for changes in Test::Stream (Use Test::Stream::Sync)

0.39  [2015-06-04]
    -  Updated for changes in Test::Stream

(wiz)

Updated devel/p5-Test-Requires to 0.10

(wiz)

Update to 0.10:

0.10 2015-07-21T12:37:03Z

    - Support 5.6 again.

0.09 2015-07-21T04:01:50Z

    - requires 5.8.1

(wiz)

Updated net/p5-Test-RequiresInternet to 0.05

(wiz)

Update to 0.05:

0.05        2015-05-25
            Fix test error when NO_NETWORK_TESTING is set (Karen Etheridge,
            RT#101996, GH#3)

(wiz)

Updated time/p5-Time-Duration to 1.20

(wiz)

Update to 1.20:

1.20 2015-04-25 NEILB
    - Switched to Dist::Zilla, in the process fixing RT#90701, RT#101548,
      and RT#101434
    - Reformatted this file as per CPAN::Changes::Spec

1.1_01 2015-04-18 NEILB
    - This release had the changes in 1.20 above, done first as a developer
      release to see what CPAN Testers said.

(wiz)

Updated time/p5-Time-Duration-Parse to 0.12

(wiz)

Update to 0.12:

0.12 2015-07-02 NEILB
    - Added [MetaJSON] to dist.ini, so the dist will get a META.json file.
      RT#105627 from ETHER.

(wiz)

Updated devel/p5-Variable-Magic to 0.58

(wiz)

Update to 0.58:

0.58    2015-07-21 16:00 UTC
        + Add : If a non-len magic callback returns a reference, it will now
                only be freed at the end of the statement that caused the
                magic to trigger. This allows the user to attach free magic
                (or a plain destructor) to a token returned from the callbacks
                in order to defer an action after the magic is processed by
                perl.
        + Fix : Test failures of threads tests on systems with harsh resource
                constraints causing the threads to exit() during run.

(wiz)

Updated www/p5-WWW-Mechanize to 1.75

(wiz)

Update to 1.75, add missing p5-CGI dependency.

1.75
========================================
[OTHER CHANGES]

- WWW::Mechanize::Image and WWW::Mechanize::Link now have a defined $VERSION
- fixed warning about the use of the encoding pragma (new in 5.22) (RT#91971)
- fixed warning about the use of CGI::param in list context (RT#103096)

(wiz)

Updated textproc/p5-YAML-LibYAML to 0.59

(wiz)

Update to 0.59:

0.59 Mon Jan 26 15:04:10 PST 2015
- PR/23 Better scalar dump heuristics
- More closely match YAML.pm
- Thanks Matthias Bethke

0.58 Tue Jan 20 21:01:49 PST 2015
- Add a VERSION statement to YAML::LibYAML (issue#8)

0.57 Thu Jan 15 23:05:15 EST 2015
- Applied fix for PR/21. nawglan++

0.56 Thu Jan 15 22:21:47 EST 2015
- Update copyright year
- Use Swim cpan-tail block functions in doc

0.55 Mon Dec 22 17:26:27 PST 2014
- Get YAML::XS using latest libyaml

(wiz)

Updated devel/p5-namespace-autoclean to 0.26

(wiz)