Update Zope modules, add py-ZopeEvent

(fhajny)

Update devel/py-ZopeComponent to 4.2.2.
Fix missing dependencies.

4.2.2 (2015-06-04)
- Fix test cases for PyPy and PyPy3.

(fhajny)

Update devel/py-ZopeInterface to 4.1.3.

- Fix installation without a C compiler on Python 3.5.

(fhajny)

Import zope.event 4.1.0 as devel/py-ZopeEvent.

The zope.event package provides a simple event system, including:
- An event publishing API, intended for use by applications which
  are unaware of any subscribers to their events.
- A very simple event-dispatching system on which more sophisticated
  event dispatching systems can be built. For example, a type-based
  event dispatching system that builds on zope.event can be found
  in zope.component.

(fhajny)

Revert. Committer was explicitly asked to not commit that change as it
is incompletely and incorrect.

(joerg)

pbulk-base-0.51: Allow symbolic host names in distributed case.

(asau)

Doesn't like nbpax for extraction, so use bsdtar.

(joerg)

Updated net/dhcpcd to 6.9.4

(roy)

Update to dhcpcd-6.9.4 with the following changes:
  *  dhcpcd will now configure chrony if installed and ntp isn't
  *  dhcpcd no longer attempts temporary address management on Linux
  *  replace the SixRD decode function with a generic definition
  *  try harder to ensure only 1 lladdr exists per interface on BSD
  *  kFreeBSD compiles once more, thanks to JS Junior
  *  change IPv6 routes on MTU change
  *  -p works with -x on an already running process started without -p
  *  fix TEST for IPv4LL
  *  Correct size allocation for prefix delegation, thanks to Jade
  *  Add an option to enable DHCPv6 Information Request without the
    need for dhcpcd to recieve an IPv6 Router Advertisement with the
    Other Configuration bit set.
  *  Introduce the optional option type, which allows embedded options
    to be optional
  *  Mark our logger function as sysloglike because we enjoy using %m
  *  Don't check link state if not instruted to before working out if
    we can fork early or not.
  *  Add a -N --renew option to renew any existing address early
  *  Obey the hostname_short option even for configured FQDN hostnames
  *  -U, --dumplease now works with standard input.
    It no longer works with a filename.
  *  If dumping leases, skip authentication and address expiry checks
  *  Fix adding host routes via a gateway on Linux
  *  Fix adding static routes via a gateway on BSD
  *  Always send LOG_DEBUG to syslog(3) even if we are in quiet mode.
    It's upto syslog to filter it.
  *  If testing or dumping leases, don't send to syslog only
    stdout/stderr.

(roy)

Add Following line for make test
+BUILD_DEPENDS+=        p5-Test-NoTabs-[0-9]*:../../devel/p5-Test-NoTabs

(mef)

Try to fix breakage, optionalize x11 on Darwin. Bump.

(tnn)

Updated devel/libsigc++ to 2.6.2

(adam)

Changes 2.6.2:
* slot: Handle auto-disconnection when a slot contains a slot
* slot and signal: Correct move constructors and move assignments.
* trackable, slot_base, signal_base, connection: Add some noexcept specs
* trackable, slot, signal: Remove noexcept from move operations.
* slot and signal: Add missing move constructors and move assignments
* signal_impl: =delete copy and move operations.
* typed_slot_rep: =delete unimplemented copy and move operations.
* slot_rep: =delete copy and move operations.

(adam)

Added devel/p5-Test-EOL version 1.6

(mef)

Import p5-Test-EOL-1.6 as devel/p5-Test-EOL.

This module scans your project/distribution for any perl files (scripts,
modules, etc) for the presence of windows line endings.

(mef)

Added devel/p5-CPAN-Meta version 2.150005

(mef)

Import p5-Test-CPAN-Meta-0.25 as devel/p5-Test-CPAN-Meta.

This distribution was written to ensure that a META.yml file, provided with a
standard distribution uploaded to CPAN, meets the specifications that are
slowly being introduced to module uploads, via the use of package makers
and installers such as ExtUtils::MakeMaker, Module::Build and
Module::Install.

(mef)

Add following line for make test
  +BUILD_DEPENDS+=        p5-Text-Diff-[0-9]*:../../textproc/p5-Text-Diff

(mef)

Updated converters/help2man to 1.47.3

(adam)

Changes 1.47.3:
* Filter more untranslatable strings from help2man-texi.pot, and add a
  note to translators about translation of cross-references.

(adam)

Fix build with latest gstreamer-1.

(markd)

Updated databases/hiredis to 0.13.3

(wiedi)

Update hiredis to 0.13.3

Changes:
* Revert "Clear `REDIS_CONNECTED` flag when connection is closed".
* Make tests pass on FreeBSD (Thanks, Giacomo Olgeni)

(wiedi)

Note update of devel/ruby-tins package to 1.7.0.

(taca)

Update ruby-tins to 1.7.0.

2015-11-09 Release 1.7.0

* Officially require ruby >= 2.0 and use new hash syntax.
* New feature attempt: sleep:-30 sleeps upto 30 seconds while retrying
  with an exponential waiting pattern.
* Remove rotate and shuffle, recent rubies all support them.
* Add an object oriented view on method signatures and use it to create
  method descriptions.

(taca)

Note update of devel/ruby-rspec-rails package to 3.4.0.

(taca)

Update ruby-rspec-rails to 3.4.0.

### 3.4.0 / 2015-11-11
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.3.3...v3.4.0)

Enhancements:

* Improved the failure message for `have_rendered` matcher on a redirect
  response. (Alex Egan, #1440)
* Add configuration option to filter out Rails gems from backtraces.
  (Bradley Schaefer, #1458)
* Enable resolver cache for view specs for a large speed improvement
  (Chris Zetter, #1452)
* Add `have_enqueued_job` matcher for checking if a block has queued jobs.
  (Wojciech Wn��trzak, #1464)

Bug Fixes:

* Fix another load order issued which causes an undefined method `fixture_path` error
  when loading rspec-rails after a spec has been created. (Nikki Murray, #1430)
* Removed incorrect surrounding whitespace in the rspec-rails backtrace
  exclusion pattern for its own `lib` code. (Jam Black, #1439)

(taca)

Note update of ruby-rspec packages:

devel/ruby-rspec-support 3.4.1
devel/ruby-rspec-core 3.4.1
devel/ruby-rspec-mocks 3.4.0
devel/ruby-rspec 3.4.0

(taca)

Update ruby-rspec to 3.4.0.

This is a meta ruby gem package for rspec-core, rspec-expectations
and rspec-mocks.

(taca)

Update ruby-rspec-mocks to 3.4.0.

### 3.4.0 / 2015-11-11
[Full Changelog](http://github.com/rspec/rspec-mocks/compare/v3.3.2...v3.4.0)

Enhancements:

* Make `expect(...).to have_received` work without relying upon
  rspec-expectations. (Myron Marston, #978)
* Add option for failing tests when expectations are set on `nil`.
  (Liz Rush, #983)

Bug Fixes:

* Fix `have_received { ... }` so that any block passed when the message
  was received is forwarded to the `have_received` block. (Myron Marston, #1006)
* Fix infinite loop in error generator when stubbing `respond_to?`.
  (Alex Dowad, #1022)
* Fix issue with using `receive` on subclasses (at a class level) with 1.8.7.
  (Alex Dowad, #1026)

(taca)

Update ruby-rspec-expectations to 3.4.0.

### 3.4.0 / 2015-11-11
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.3.1...v3.4.0)

Enhancements:

* Warn when `RSpec::Matchers` is included in a superclass after it has
  already been included in a subclass on MRI 1.9, since that situation
  can cause uses of `super` to trigger infinite recursion. (Myron Marston, #816)
* Stop rescuing `NoMemoryError`, `SignalExcepetion`, `Interrupt` and
  `SystemExit`. It is dangerous to interfere with these. (Myron Marston, #845)
* Add `#with_captures` to the
  [match matcher](https://www.relishapp.com/rspec/rspec-expectations/docs/built-in-matchers/match-matcher)
  which allows a user to specify expected captures when matching a regex
  against a string. (Sam Phippen, #848)
* Always print compound failure messages in the multi-line form. Trying
  to print it all on a single line didn't read very well. (Myron Marston, #859)

Bug Fixes:

* Fix failure message from dynamic predicate matchers when the object
  does not respond to the predicate so that it is inspected rather
  than relying upon it's `to_s` -- that way for `nil`, `"nil"` is
  printed rather than an empty string. (Myron Marston, #841)
* Fix SystemStackError raised when diffing an Enumerable object
  whose `#each` includes the object itself. (Yuji Nakayama, #857)

(taca)

Update ruby-rspec-core to 3.4.1.

### 3.4.1 / 2015-11-18
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.4.0...v3.4.1)

Bug Fixes:

* Fix backtrace formatter to handle backtraces that are `nil`.
  (Myron Marston, #2118)

### 3.4.0 / 2015-11-11
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.3.2...v3.4.0)

Enhancements:

* Combine multiple `--pattern` arguments making them equivalent to
  `--pattern=1,2,...,n`. (Jon Rowe, #2002)
* Improve `inspect` and `to_s` output for `RSpec::Core::Example`
  objects, replacing Ruby's excessively verbose output. (Gavin Miller, #1922)
* Add `silence_filter_announcements` configuration option.
  (David Raffensperger, #2007)
* Add optional `example_finished` notification to the reporter protocol for
  when you don't care about the example outcome. (Jon Rowe, #2013)
* Switch `--bisect` to a recursion-based bisection algorithm rather than
  a permutation-based one. This better handles cases where an example
  depends upon multiple other examples instead of just one and minimizes
  the number of runs necessary to determine that an example set cannot be
  minimized further. (Simon Coffey, #1997)
* Allow simple filters (e.g. `:symbol` key only) to be triggered by truthey
  values. (Tim Mertens, #2035)
* Remove unneeded warning about need for `ansicon` on Windows when using
  RSpec's `--color` option. (Ashley Engelund, #2038)
* Add option to configure RSpec to raise errors when issuing warnings.
  (Jon Rowe, #2052)
* Append the root `cause` of a failure or error to the printed failure
  output when a `cause` is available. (Adam Magan)
* Stop rescuing `NoMemoryError`, `SignalExcepetion`, `Interrupt` and
  `SystemExit`. It is dangerous to interfere with these. (Myron Marston, #2063)
* Add `config.project_source_dirs` setting which RSpec uses to determine
  if a backtrace line comes from your project source or from some
  external library. It defaults to `spec`, `lib` and `app` but can be
  configured differently. (Myron Marston, #2088)
* Improve failure line detection so that it looks for the failure line
  in any project source directory instead of just in the spec file.
  In addition, if no backtrace lines can be found from a project source
  file, we fall back to displaying the source of the first backtrace
  line. This should virtually eliminate the "Unable to find matching
  line from backtrace" messages. (Myron Marston, #2088)
* Add support for `:extra_failure_lines` example metadata that will
  be appended to the failure output. (bootstraponline, #2092).
* Add `RSpec::Core::Example#duplicate_with` to produce new examples
  with cloned metadata. (bootstraponline, #2098)
* Add `RSpec::Core::Configuration#on_example_group_definition` to register
  hooks to be invoked when example groups are created. (bootstraponline, #2094)
* Add `add_example` and `remove_example` to `RSpec::Core::ExampleGroup` to
  allow  manipulating an example groups examples. (bootstraponline, #2095)
* Display multiline failure source lines in failure output when Ripper is
  available (MRI >= 1.9.2, and JRuby >= 1.7.5 && < 9.0.0.0.rc1).
  (Yuji Nakayama, #2083)
* Add `max_displayed_failure_line_count` configuration option
  (defaults to 10). (Yuji Nakayama, #2083)
* Enhance `fail_fast` option so it can take a number (e.g. `--fail-fast=3`)
  to force the run to abort after the specified number of failures.
  (Jack Scotti, #2065)
* Syntax highlight the failure snippets in text formatters when `color`
  is enabled and the `coderay` gem is installed on a POSIX system.
  (Myron Marston, #2109)

Bug Fixes:

* Lock `example_status_persistence_file` when reading from and writing
  to it to prevent race conditions when multiple processes try to use
  it. (Ben Woosley, #2029)
* Fix regression in 3.3 that caused spec file names with square brackets in
  them (such as `1[]_spec.rb`) to not be loaded properly. (Myron Marston, #2041)
* Fix output encoding issue caused by ASCII literal on 1.9.3 (Jon Rowe, #2072)
* Fix requires in `rspec/core/rake_task.rb` to avoid double requires
  seen by some users. (Myron Marston, #2101)

(taca)

Update ruby-rspec-support to 3.4.1.

### 3.4.1 / 2015-11-20
[Full Changelog](http://github.com/rspec/rspec-support/compare/v3.4.0...v3.4.1)

Bug Fixes:

* Fix `RSpec::Support::RubyFeature.ripper_supported?` so it returns
  `false` on Rubinius since the Rubinius team has no plans to support
  it. This prevents rspec-core from trying to load and use ripper to
  extract failure snippets. (Aaron Stone, #251)

### 3.4.0 / 2015-11-11
[Full Changelog](http://github.com/rspec/rspec-support/compare/v3.3.0...v3.4.0)

Enhancements:

* Improve formatting of `Delegator` based objects (e.g. `SimpleDelgator`) in
  failure messages and diffs. (Andrew Horner, #215)
* Add `ComparableVersion`. (Yuji Nakayama, #245)
* Add `Ripper` support detection. (Yuji Nakayama, #245)

Bug Fixes:

* Work around bug in JRuby that reports that `attr_writer` methods
  have no parameters, causing RSpec's verifying doubles to wrongly
  fail when mocking or stubbing a writer method on JRuby. (Myron Marston, #225)

(taca)

Note update of devel/ruby-racc package to 1.4.14.

(taca)

Update ruby-racc to 1.4.14.

Changes are not available, please refer:
<https://github.com/tenderlove/racc/commits/master>.

(taca)

Note update of devel/ruby-power_assert package to 0.2.6.

(taca)

Update ruby-power_assert to 0.2.6.

* Fix escape.
* Add workaround for https://bugs.ruby-lang.org/issues/11182.

(taca)

Note update of devel/ruby-parser package to 3.7.2.

(taca)

Update ruby-parser to 3.7.2.

=== 3.7.2 / 2015-10-26

* 1 bug fix:

  * I hate regexen. Fixed a lexing bug confusing strings vs labels. (phiggins)

(taca)

Note update of devel/ruby-msgpack package to 0.7.1.

(taca)

Update ruby-msgpack to 0.7.1.

2015-11-20 version 0.7.1:

* Fixed bug to pack/unpack ext type objects larger than 256bytes incorrectly.

2015-10-24 version 0.7.0:

* Add extention types support.
* Fix to share almost all test cases between CRuby and JRuby implementations.
* Fixed JRuby implementation to raise UnknownExtTypeError for unregistered ext type ids
  instead to generate MessagePack::ExtensionValue instances.
  (Specify `allow_unknown_ext: true` as unpacker option for v0.6.x behavior.)

(taca)

Note update of devel/ruby-mixlib-shellout package to 2.2.3.

(taca)

Update ruby-mixlib-shellout to 2.2.3.

Release 2.2.3

* Kill all child processes on Windows when a command times out.

Release 2.2.2

* Ship gemspec and Gemfiles to facilate testing.
* Fix #111 by pulling in an updated version of win-32/process and correctly
  patching Process::create.
* Kill all child processes on Windows when a command times out.

(taca)

Note update of devel/ruby-minitest package to 5.8.3.

(taca)

Update ruby-minitest to 5.8.3.

=== 5.8.3 / 2015-11-17

* 1 minor enhancement:

  * Added extra note about mocks and threads to readme. (zamith)

* 1 bug fix:

  * Fixed bug in Mock#verify. (pithub/zamith)

=== 5.8.2 / 2015-10-26

* 1 bug fix:

  * Fixed using parallelize_me! and capture_io (or any locking io). (arlt/tenderlove)

(taca)

Note update of devel/ruby-gettext_i18n_rails package to 1.3.1.

(taca)

Update ruby-gettext_i18n_rails to 1.3.1.

* Reimplement extraction using SexpProcessor.
* Fix ruby_parser and sexp_processor requirements.
* Move require sexp_processor to extractor.
* Add gem hints to make upgrading easy.
* Properly pass file name to RubyParser.
* Unify interface of "parse" with gettext.
* Spec to test correct functioning of HAML parser with parse options.

(taca)

Note update of devel/ruby-flexmock package to 2.0.2.

(taca)

Update ruby-flexmock to 2.0.2.

* Avoid infinite recursion when a mocked method is called within #inspect.
* Remove duplicate method SpyDescribers.call_description / FlexMock.format_call.

(taca)

Note update of devel/ruby-fast_gettext package to 1.0.0.

(taca)

Update ruby-fast_gettext to 1.0.0.

1.0.0 -- do not enforce attr_accessible unless ProtectedAttributes are loaded

(taca)

Updated mail/msmtp to 1.6.3

(leot)

Update mail/msmtp to 1.6.3.

pkgsrc changes:
- Remove patches/patch-src_Makefile.in that seems no more needed

Changes:
Version 1.6.3:
- A bug in SOCKS support was fixed.
- Handling non-fatal errors in TLS handshakes was fixed.

(leot)

Updated print/cups-filters to 1.2.0

(leot)

Update print/cups-filters to 1.2.0.

Changes:
CHANGES IN V1.2.0
- cups-browsed: When using IP-address-based device URIs via
  the "IPBasedDeviceURIs" directive in cups-browsed.conf, add
  two additional settings to restrict the used IP addresses to
  either only IPv4 addresses or only IPv6 addresses.
- foomatic-rip: SECURITY FIX: Also consider the back tick
  ('`') as an illegal shell escape character. Thanks to Michal
  Kowalczyk from the Google Security Team for the hint
  (CVE-2015-8327).

(leot)

Note update of devel/ruby-backports package to 3.6.7.

(taca)

Update ruby-backports to 3.6.7.

* Fix pledgie image markup.
* Fix embarassing bug with bsearch for floats.

(taca)

Note update of databases/ruby-sequel package to 4.28.0.

(taca)

Update sequel to 4.28.0.

=== 4.28.0 (2015-11-02)

* Add boolean_subsets plugin, which adds a subset for each boolean column (jeremyevans)

* Add subset_conditions plugin, which adds a method for each subset returning the filter conditions for the subset (jeremyevans)

* Make the list plugin work better with the auto_validations plugin when there is a validation on the position column (jeremyevans)

* Make to_csv for model datasets call instance methods, just like Model#to_csv, in the csv_serializer plugin (skrobul) (#1088)

* Raise Sequel::NoExistingObject instead of generic error if Model#refresh can't find the related row (jeremyevans)

(taca)

Note update of databases/ruby-pg package to 0.18.4.

(taca)

Update ruby-pg to 0.18.4.

== v0.18.4 [2015-11-13] Michael Granger <ged@FaerieMUD.org>

Enhancements:
- Fixing compilation problems with Microsoft Visual Studio 2008. GH #10
- Avoid name clash with xcode and jemalloc. PR#22, PR#23

Bugfixes:
- Avoid segfault, when quote_ident or TextEncoder::Identifier
  is called with Array containing non-strings. #226

(taca)

Do not "force" dvdread option via "--enable-dvdread" (but let the MPlayer's
configure script to autorecognise it).

In that way pkg-config will be used to set the CFLAGS and LDFLAGS (while this
does not happen via --enable-* for various options).

Pointed out by Joyent SmartOS bulk builds.

(leot)

Allow work on PHP 5.5/5.6.  Patch is based on patch
on https://bugzilla.redhat.com/show_bug.cgi?id=974928.

(taca)

Note update of databases/ruby-mysql2 package to 0.4.2.

(taca)

Update ruby-mysql2 to 0.4.2.

* Free the MYSQL client handle after mysql_close to avoid a memory leak.
* Avoid crashing when Statement#close is called before a Result is
  garbage collected.

(taca)

Reset PKGREVISION along with php 5.6.16 update.

(taca)

Reset MAINTAINER.

(taca)

Mention where the command tools can be obtained from to save time, as the Apple developer
portal search doesn't offer it as a result.

(sevan)

Latest valgrind is also for plattform SunOS-*-i386 and SunOS-*-x86_64.
Adjust PLIST

(wiedi)

Updated textproc/p5-PDF-Create to 1.22

(wen)

Update to 1.22

Upstream changes:
1.22  2015-11-25 10:35:00 MANWAR
      - Fixed the debug level handling.
      - Updated pod document about the same.

1.21  2015-11-24 14:00:00 MANWAR
      - Fixed issue RT# 109562.

1.20  2015-11-24 MANWAR
      - Updated method PDF::Create::Page::string() to handle character/word spacing. (Thanks David Kitcher-Jones).
      - Updated method PDF::Create::Page::text() to handle font correctly. (Thanks David Kitcher-Jones).
      - Added new method PDF::Create::Page::block_text() to manage block of text.  (Thanks David Kitcher-Jones).
      - Updated method PDF::Create::Page::block_text() to handle multiple paragraphs. (MANWAR)
      - Documented method text() and block_text(). (MANWAR)

(wen)

Updated textproc/p5-List-Compare to 0.53

(wen)

Updated multimedia/ffmpeg2 to 2.8.3

(adam)

Update to 0.53

Upstream changes:
0.50 Sat May  9 09:06:41 EDT 2015
    - Improve performance of is_LsubsetR via patches submitted by Mich Rawson
      at New York Perl Hackathon (May 2 2015).

0.51 Thu May 14 21:49:00 EDT 2015
    - Changes to List::Compare::Functional::is_LsubsetR introduced in v0.50
      found to be buggy.  See: https://rt.cpan.org/Ticket/Display.html?id=104452.
      Pending further investigation, reverted to v0.49 of
      List::Compare::Base::_Auxiliary; added two test files.

0.52 Thu May 21 21:57:18 EDT 2015
    - v0.51 changes were not applied to master branch; apply them.
      Remove two hash-to-hash assignments in List::Compare::Base::_Auxiliary::_subset_subengine.

0.53 Sun Jun  7 10:49:24 MDT 2015
    - Accept patch submitted by Paulo Custodio reworking, and correcting,
      changes originally published in 0.50.

(wen)

Updated devel/p5-Log-Report to 1.09

(wen)

version 2.8.3
- avcodec/cabac: Check initial cabac decoder state
- avcodec/cabac_functions: Fix "left shift of negative value -31767"
- avcodec/h264_slice: Limit max_contexts when slice_context_count is initialized
- rtmpcrypt: Do the xtea decryption in little endian mode
- avformat/matroskadec: Check subtitle stream before dereferencing
- avcodec/pngdec: Replace assert by request for sample for unsupported TRNS cases
- avformat/utils: Do not init parser if probing is unfinished
- avcodec/jpeg2000dec: Fix potential integer overflow with tile dimensions
- avcodec/jpeg2000: Use av_image_check_size() in ff_jpeg2000_init_component()
- avcodec/wmaprodec: Check for overread in decode_packet()
- avcodec/smacker: Check that the data size is a multiple of a sample vector
- avcodec/takdec: Skip last p2 sample (which is unused)
- avcodec/dxtory: Fix input size check in dxtory_decode_v1_410()
- avcodec/dxtory: Fix input size check in dxtory_decode_v1_420()
- avcodec/error_resilience: avoid accessing previous or next frames tables beyond height
- avcodec/dpx: Move need_align to act per line
- avcodec/flashsv: Check size before updating it
- avcodec/ivi: Check image dimensions
- avcodec/utils: Better check for channels in av_get_audio_frame_duration()
- avcodec/jpeg2000dec: Check for duplicate SIZ marker
- aacsbr: don't call sbr_dequant twice without intermediate read_sbr_data
- hqx: correct type and size check of info_offset
- mxfdec: check edit_rate also for physical_track
- avcodec/jpeg2000: Change coord to 32bit to support larger than 32k width or height
- avcodec/jpeg2000dec: Check SIZ dimensions to be within the supported range
- avcodec/jpeg2000: Check comp coords to be within the supported size
- mpegvideo: clear overread in clear_context
- avcodec/avrndec: Use the AVFrame format instead of the context
- dds: disable palette flag for compressed images
- dds: validate compressed source buffer size
- dds: validate source buffer size before copying
- dvdsubdec: validate offset2 similar to offset1
- brstm: reject negative sample rate
- aacps: avoid division by zero in stereo_processing
- softfloat: assert when the argument of av_sqrt_sf is negative

(adam)

Update to 1.09

Upstream changes:
version 1.09: Tue 20 Oct 09:26:00 CEST 2015

Fixes:
- try: do not ignore is_fatal parameter

Improvements:
- dispatcher() new action 'active-try'
- many, many improvements to Dancer2::*  [Andrew Beverley]

version 1.08: Thu  8 Oct 17:55:39 CEST 2015

Fixes:
- tests on Windows [cpantesters]

Improvements:
- avoid use of 'package Dancer::Logger' to circumvent complaints
  of Pause.
- Log::Report configure message_class [Andrew Beverley]
- Dancer2 plugin improved a lot [Andrew Beverley]

(wen)

Update devel/valgrind to 3.11.0.

(alnsn)

Update devel/valgrind to version 3.11.0.

  Release 3.11.0 (22 September 2015)
  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  3.11.0 is a feature release with many improvements and the usual
  collection of bug fixes.
  This release supports X86/Linux, AMD64/Linux, ARM32/Linux,
  ARM64/Linux, PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux,
  MIPS32/Linux, MIPS64/Linux, ARM/Android, ARM64/Android,
  MIPS32/Android, X86/Android, X86/Solaris, AMD64/Solaris, X86/MacOSX
  10.10 and AMD64/MacOSX 10.10.  There is also preliminary support for
  X86/MacOSX 10.11, AMD64/MacOSX 10.11 and TILEGX/Linux.
  * ================== PLATFORM CHANGES =================
  * Support for Solaris/x86 and Solaris/amd64 has been added.
  * Preliminary support for Mac OS X 10.11 (El Capitan) has been added.
  * Preliminary support for the Tilera TileGX architecture has been added
  .
  * s390x: It is now required for the host to have the "long displacement
  "
    facility.  The oldest supported machine model is z990.
  * x86: on an SSE2 only host, Valgrind in 32 bit mode now claims to be a
    Pentium 4.  3.10.1 wrongly claimed to be a Core 2, which is SSSE3.
  * The JIT's register allocator is significantly faster, making the JIT
    as a whole somewhat faster, so JIT-intensive activities, for example
    program startup, are modestly faster, around 5%.
  * There have been changes to the default settings of several command
    line flags, as detailed below.
  * Intel AVX2 support is more complete (64 bit targets only).  On AVX2
    capable hosts, the simulated CPUID will now indicate AVX2 support.
  * ==================== TOOL CHANGES ====================
  * Memcheck:
    - The default value for --leak-check-heuristics has been changed from
      "none" to "all". This helps to reduce the number of possibly
      lost blocks, in particular for C++ applications.
    - The default value for --keep-stacktraces has been changed from
      "malloc-then-free" to "malloc-and-free".  This has a small cost in
      memory (one word per malloc-ed block) but allows Memcheck to show t
  he
      3 stacktraces of a dangling reference: where the block was allocate
  d,
      where it was freed, and where it is acccessed after being freed.
    - The default value for --partial-loads-ok has been changed from "no"
    to
      "yes", so as to avoid false positive errors resulting from some kin
  ds
      of vectorised loops.
    - A new monitor command 'xb <addr> <len>' shows the validity bits of
      <len> bytes at <addr>.  The monitor command 'xb' is easier to use
      than get_vbits when you need to associate byte data value with
      their corresponding validity bits.
    - The 'block_list' monitor command has been enhanced:
        o it can print a range of loss records
        o it now accepts an optional argument 'limited <max_blocks>'
          to control the number of blocks printed.
        o if a block has been found using a heuristic, then
          'block_list' now shows the heuristic after the block size.
        o the loss records/blocks to print can be limited to the blocks
          found via specified heuristics.
    - The C helper functions used to instrument loads on
      x86-{linux,solaris} and arm-linux (both 32-bit only) have been
      replaced by handwritten assembly sequences.  This gives speedups
      in the region of 0% to 7% for those targets only.
    - A new command line option, --expensive-definedness-checks=yes|no,
      has been added.  This is useful for avoiding occasional invalid
      uninitialised-value errors in optimised code.  Watch out for
      runtime degradation, as this can be up to 25%.  As always, though,
      the slowdown is highly application specific.  The default setting
      is "no".
  * Massif:
    - A new monitor command 'all_snapshots <filename>' dumps all
      snapshots taken so far.
  * Helgrind:
    - Significant memory reduction and moderate speedups for
      --history-level=full for applications accessing a lot of memory
      with many different stacktraces.
    - The default value for --conflict-cache-size=N has been doubled to
      2000000.  Users that were not using the default value should
      preferably also double the value they give.
      The default was changed due to the changes in the "full history"
      implementation.  Doubling the value gives on average a slightly mor
  e
      complete history and uses similar memory (or significantly less mem
  ory
      in the worst case) than the previous implementation.

    - The Helgrind monitor command 'info locks' now accepts an optional
      argument 'lock_addr', which shows information about the lock at the
      given address only.
    - When using --history-level=full, the new Helgrind monitor command
      'accesshistory <addr> [<len>]' will show the recorded accesses for
      <len> (or 1) bytes at <addr>.
  * ==================== OTHER CHANGES ====================
  * The default value for the --smc-check option has been changed from
    "stack" to "all-non-file" on targets that provide automatic D-I
    cache coherence (x86, amd64 and s390x).  The result is to provide,
    by default, transparent support for JIT generated and self-modifying
    code on all targets.
  * Mac OS X only: the default value for the --dsymutil option has been
    changed from "no" to "yes", since any serious usage on Mac OS X
    always required it to be "yes".
  * The command line options --db-attach and --db-command have been remov
  ed.
    They were deprecated in 3.10.0.
  * When a process dies due to a signal, Valgrind now shows the signal
    and the stacktrace at default verbosity (i.e. verbosity 1).
  * The address description logic used by Memcheck and Helgrind now
    describes addresses in anonymous segments, file mmap-ed segments,
    shared memory segments and the brk data segment.
  * The new option --error-markers=<begin>,<end> can be used to mark the
    begin/end of errors in textual output mode, to facilitate
    searching/extracting errors in output files that mix valgrind errors
    with program output.
  * The new option --max-threads=<number> can be used to change the numbe
  r
    of threads valgrind can handle.  The default is 500 threads which
    should be more than enough for most applications.
  * The new option --valgrind-stacksize=<number> can be used to change th
  e
    size of the private thread stacks used by Valgrind.  This is useful
    for reducing memory use or increasing the stack size if Valgrind
    segfaults due to stack overflow.
  * The new option --avg-transtab-entry-size=<number> can be used to spec
  ify
    the expected instrumented block size, either to reduce memory use or
    to avoid excessive retranslation.
  * Valgrind can be built with Intel's ICC compiler, version 14.0 or late
  r.
  * New and modified GDB server monitor features:
    - When a signal is reported in GDB, you can now use the GDB convenien
  ce
      variable $_siginfo to examine detailed signal information.

    - Valgrind's gdbserver now allows the user to change the signal
      to deliver to the process.  So, use 'signal SIGNAL' to continue exe
  cution
      with SIGNAL instead of the signal reported to GDB. Use 'signal 0' t
  o
      continue without passing the signal to the process.
    - With GDB >= 7.10, the command 'target remote'
      will automatically load the executable file of the process running
      under Valgrind. This means you do not need to specify the executabl
  e
      file yourself, GDB will discover it itself.  See GDB documentation
  about
      'qXfer:exec-file:read' packet for more info.
  * ==================== FIXED BUGS ====================
  The following bugs have been fixed or resolved.  Note that "n-i-bz"
  stands for "not in bugzilla" -- that is, a bug that was reported to us
  but never got a bugzilla entry.  We encourage you to file bugs in
  bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
  than mailing the developers (or mailing lists) directly -- bugs that
  are not entered into bugzilla tend to get forgotten about or ignored.
  To see details of a given bug, visit
    https://bugs.kde.org/show_bug.cgi?id=XXXXXX
  where XXXXXX is the bug number as listed below.
  116002  VG_(printf): Problems with justification of strings and integer
  s
  155125  avoid cutting away file:lineno after long function name
  197259  Unsupported arch_prtctl PR_SET_GS option
  201152  ppc64: Assertion in ppc32g_dirtyhelper_MFSPR_268_269
  201216  Fix Valgrind does not support pthread_sigmask() on OS X
  201435  Fix Darwin: -v does not show kernel version
  208217  "Warning: noted but unhandled ioctl 0x2000747b" on Mac OS X
  211256  Fixed an outdated comment regarding the default platform.
  211529  Incomplete call stacks for code compiled by newer versions of M
  SVC
  211926  Avoid compilation warnings in valgrind.h with -pedantic
  212291  Fix unhandled syscall: unix:132 (mkfifo) on OS X
          == 263119
  226609  Crediting upstream authors in man page
  231257  Valgrind omits path when executing script from shebang line
  254164  OS X task_info: UNKNOWN task message [id 3405, to mach_task_sel
  f() [..]
  269360  s390x: Fix addressing mode selection for compare-and-swap
  302630  Memcheck: Assertion failed: 'sizeof(UWord) == sizeof(UInt)'
          == 326797
  312989  ioctl handling needs to do POST handling on generic ioctls and
  [..]
  319274  Fix unhandled syscall: unix:410 (sigsuspend_nocancel) on OS X
  324181  mmap does not handle MAP_32BIT (handle it now, rather than fail
    it)
  327745  Fix valgrind 3.9.0 build fails on Mac OS X 10.6.8
  330147  libmpiwrap PMPI_Get_count returns undefined value
  333051  mmap of huge pages fails due to incorrect alignment
          == 339163
  334802  valgrind does not always explain why a given option is bad
  335618  mov.w rN, pc/sp (ARM32)
  335785  amd64->IR 0xC4 0xE2 0x75 0x2F (vmaskmovpd)
          == 307399
          == 343175
          == 342740
          == 346912
  335907  segfault when running wine's ddrawex/tests/surface.c under valg
  rind
  338602  AVX2 bit in CPUID missing
  338606  Strange message for scripts with invalid interpreter
  338731  ppc: Fix testuite build for toolchains not supporting -maltivec
  338995  shmat with hugepages (SHM_HUGETLB) fails with EINVAL
  339045  Getting valgrind to compile and run on OS X Yosemite (10.10)
          == 340252
  339156  gdbsrv not called for fatal signal
  339215  Valgrind 3.10.0 contain 2013 in copyrights notice
  339288  support Cavium Octeon MIPS specific BBIT*32 instructions
  339636  Use fxsave64 and fxrstor64 mnemonics instead of old-school rex6
  4 prefix
  339442  Fix testsuite build failure on OS X 10.9
  339542  Enable compilation with Intel's ICC compiler
  339563  The DVB demux DMX_STOP ioctl doesn't have a wrapper
  339688  Mac-specific ASM does not support .version directive (cpuid,
          tronical and pushfpopf tests)
  339745  Valgrind crash when check Marmalade app (partial fix)
  339755  Fix known deliberate memory leak in setenv() on Mac OS X 10.9
  339778  Linux/TileGx platform support for Valgrind
  339780  Fix known uninitialised read in pthread_rwlock_init() on Mac OS
    X 10.9
  339789  Fix none/tests/execve test on Mac OS X 10.9
  339808  Fix none/tests/rlimit64_nofile test on Mac OS X 10.9
  339820  vex amd64->IR: 0x66 0xF 0x3A 0x63 0xA 0x42 0x74 0x9 (pcmpistri
  $0x42)
  340115  Fix none/tests/cmdline[1|2] tests on systems which define TMPDI
  R
  340392  Allow user to select more accurate definedness checking in memc
  heck
          to avoid invalid complaints on optimised code
  340430  Fix some grammatical weirdness in the manual.
  341238  Recognize GCC5/DWARFv5 DW_LANG constants (Go, C11, C++11, C++14
  )
  341419  Signal handler ucontext_t not filled out correctly on OS X
  341539  VG_(describe_addr) should not describe address as belonging to
  client
          segment if it is past the heap end
  341613  Enable building of manythreads and thread-exits tests on Mac OS
    X
  341615  Fix none/tests/darwin/access_extended test on Mac OS X
  341698  Valgrind's AESKEYGENASSIST gives wrong result in words 0 and 2
  [..]
  341789  aarch64: shmat fails with valgrind on ARMv8
  341997  MIPS64: Cavium OCTEON insns - immediate operand handled incorre
  ctly
  342008  valgrind.h needs type cast [..] for clang/llvm in 64-bit mode
  342038  Unhandled syscalls on aarch64 (mbind/get/set_mempolicy)
  342063  wrong format specifier for test mcblocklistsearch in gdbserver_
  tests
  342117  Hang when loading PDB file for MSVC compiled Firefox under Wine
  342221  socket connect false positive uninit memory for unknown af fami
  ly
  342353  Allow dumping full massif output while valgrind is still runnin
  g
  342571  Valgrind chokes on AVX compare intrinsic with _CMP_GE_QS
          == 346476
          == 348387
          == 350593
  342603  Add I2C_SMBUS ioctl support
  342635  OS X 10.10 (Yosemite) - missing system calls and fcntl code
  342683  Mark memory past the initial brk limit as unaddressable
  342783  arm: unhandled instruction 0xEEFE1ACA = "vcvt.s32.f32 s3, s3, #
  12"
  342795  Internal glibc __GI_mempcpy call should be intercepted
  342841  s390x: Support instructions fiebr(a) and fidbr(a)
  343012  Unhandled syscall 319 (memfd_create)
  343069  Patch updating v4l2 API support
  343173  helgrind crash during stack unwind
  343219  fix GET_STARTREGS for arm
  343303  Fix known deliberate memory leak in setenv() on Mac OS X 10.10
  343306  OS X 10.10: UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option
  343332  Unhandled instruction 0x9E310021 (fcvtmu) on aarch64
  343335  unhandled instruction 0x1E638400 (fccmp) aarch64
  343523  OS X mach_ports_register: UNKNOWN task message [id 3403, to [..
  ]
  343525  OS X host_get_special_port: UNKNOWN host message [id 412, to [.
  .]
  343597  ppc64le: incorrect use of offseof macro
  343649  OS X host_create_mach_voucher: UNKNOWN host message [id 222, to
    [..]
  343663  OS X 10.10  Memchecj always reports a leak regardless of [..]
  343732  Unhandled syscall 144 (setgid) on aarch64
  343733  Unhandled syscall 187 (msgctl and related) on aarch64
  343802  s390x: False positive "conditional jump or move depends on [..]
  343902  --vgdb=yes doesn't break when --xml=yes is used
  343967  Don't warn about setuid/setgid/setcap executable for directorie
  s
  343978  Recognize DWARF5/GCC5 DW_LANG_Fortran 2003 and 2008 constants
  344007  accept4 syscall unhandled on arm64 (242) and ppc64 (344)
  344033  Helgrind on ARM32 loses track of mutex state in pthread_cond_wa
  it
  344054  www - update info for Solaris/illumos
  344416  'make regtest' does not work cleanly on OS X
  344235  Remove duplicate include of pub_core_aspacemgr.h
  344279  syscall sendmmsg on arm64 (269) and ppc32/64 (349) unhandled
  344295  syscall recvmmsg on arm64 (243) and ppc32/64 (343) unhandled
  344307  2 unhandled syscalls on aarch64/arm64: umount2(39), mount (40)
  344314  callgrind_annotate ... warnings about commands containing newli
  nes
  344318  socketcall should wrap recvmmsg and sendmmsg
  344337  Fix unhandled syscall: mach:41 (_kernelrpc_mach_port_guard_trap
  )
  344416  Fix 'make regtest' does not work cleanly on OS X
  344499  Fix compilation for Linux kernel >= 4.0.0
  344512  OS X: unhandled syscall: unix:348 (__pthread_chdir),
          unix:349 (__pthread_fchdir)
  344559  Garbage collection of unused segment names in address space man
  ager
  344560  Fix stack traces missing penultimate frame on OS X
  344621  Fix memcheck/tests/err_disable4 test on OS X
  344686  Fix suppression for pthread_rwlock_init on OS X 10.10
  344702  Fix missing libobjc suppressions on OS X 10.10
          == 344543
  344936  Fix unhandled syscall: unix:473 (readlinkat) on OS X 10.10
  344939  Fix memcheck/tests/xml1 on OS X 10.10
  345016  helgrind/tests/locked_vs_unlocked2 is failing sometimes
  345079  Fix build problems in VEX/useful/test_main.c
  345126  Incorrect handling of VIDIOC_G_AUDIO and G_AUDOUT
  345177  arm64: prfm (reg) not implemented
  345215  Performance improvements for the register allocator
  345248  add support for Solaris OS in valgrind
  345338  TIOCGSERIAL and TIOCSSERIAL ioctl support on Linux
  345394  Fix memcheck/tests/strchr on OS X
  345637  Fix memcheck/tests/sendmsg on OS X
  345695  Add POWERPC support for AT_DCACHESIZE and HWCAP2
  345824  Fix aspacem segment mismatch: seen with none/tests/bigcode
  345887  Fix an assertion in the address space manager
  345928  amd64: callstack only contains current function for small stack
  s
  345984  disInstr(arm): unhandled instruction: 0xEE193F1E
  345987  MIPS64: Implement cavium LHX instruction
  346031  MIPS: Implement support for the CvmCount register (rhwr %0, 31)
  346185  Fix typo saving altivec register v24
  346267  Compiler warnings for PPC64 code on call to LibVEX_GuestPPC64_g
  et_XER()
          and LibVEX_GuestPPC64_get_CR()
  346270  Regression tests none/tests/jm_vec/isa_2_07 and
          none/tests/test_isa_2_07_part2 have failures on PPC64 little en
  dian
  346307  fuse filesystem syscall deadlocks
  346324  PPC64 missing support for lbarx, lharx, stbcx and sthcx instruc
  tions
  346411  MIPS: SysRes::_valEx handling is incorrect
  346416  Add support for LL_IOC_PATH2FID and LL_IOC_GETPARENT Lustre ioc
  tls
  346474  PPC64 Power 8, spr TEXASRU register not supported
  346487  Compiler generates "note" about a future ABI change for PPC64
  346562  MIPS64: lwl/lwr instructions are performing 64bit loads
          and causing spurious "invalid read of size 8" warnings
  346801  Fix link error on OS X: _vgModuleLocal_sf_maybe_extend_stack
  347151  Fix suppression for pthread_rwlock_init on OS X 10.8
  347233  Fix memcheck/tests/strchr on OS X 10.10 (Haswell)
  347322  Power PC regression test cleanup
  347379  valgrind --leak-check=full leak errors from system libs on OS X
    10.8
          == 217236
  347389  unhandled syscall: 373 (Linux ARM syncfs)
  347686  Patch set to cleanup PPC64 regtests
  347978  Remove bash dependencies where not needed
  347982  OS X: undefined symbols for architecture x86_64: "_global" [..]
  347988  Memcheck: the 'impossible' happened: unexpected size for Addr (
  OSX/wine)
          == 345929
  348102  Patch updating v4l2 API support
  348247  amd64 front end: jno jumps wrongly when overflow is not set
  348269  Improve mmap MAP_HUGETLB support.
  348334  (ppc) valgrind does not simulate dcbfl - then my program termin
  ates
  348345  Assertion fails for negative lineno
  348377  Unsupported ARM instruction: yield
  348565  Fix detection of command line option availability for clang
  348574  vex amd64->IR pcmpistri SSE4.2 unsupported (pcmpistri $0x18)
  348728  Fix broken check for VIDIOC_G_ENC_INDEX
  348748  Fix redundant condition
  348890  Fix clang warning about unsupported --param inline-unit-growth=
  900
  348949  Bogus "ERROR: --ignore-ranges: suspiciously large range"
  349034  Add Lustre ioctls LL_IOC_GROUP_LOCK and LL_IOC_GROUP_UNLOCK
  349086  Fix UNKNOWN task message [id 3406, to mach_task_self(), [..]
  349087  Fix UNKNOWN task message [id 3410, to mach_task_self(), [..]
  349626  Implemented additional Xen hypercalls
  349769  Clang/osx: ld: warning: -read_only_relocs cannot be used with x
  86_64
  349790  Clean up of the hardware capability checking utilities.
  349828  memcpy intercepts memmove causing src/dst overlap error (ppc64
  ld.so)
  349874  Fix typos in source code
  349879  memcheck: add handwritten assembly for helperc_LOADV*
  349941  di_notify_mmap might create wrong start/size DebugInfoMapping
  350062  vex x86->IR: 0x66 0xF 0x3A 0xB (ROUNDSD) on OS X
  350202  Add limited param to 'monitor block_list'
  350290  s390x: Support instructions fixbr(a)
  350359  memcheck/tests/x86/fxsave hangs indefinetely on OS X
  350809  Fix none/tests/async-sigs for Solaris
  350811  Remove reference to --db-attach which has been removed.
  350813  Memcheck/x86: enable handwritten assembly helpers for x86/Solar
  is too
  350854  hard-to-understand code in VG_(load_ELF)()
  351140  arm64 syscalls setuid (146) and setresgid (149) not implemented
  351386  Solaris: Cannot run ld.so.1 under Valgrind
  351474  Fix VG_(iseqsigset) as obvious
  351534  Fix incorrect header guard
  351632  Fix UNKNOWN fcntl 97 on OS X 10.11
  351756  Intercept platform_memchr$VARIANT$Haswell on OS X
  351858  ldsoexec support on Solaris
  351873  Newer gcc doesn't allow __builtin_tabortdc[i] in ppc32 mode
  352130  helgrind reports false races for printfs using mempcpy on FILE*
    state
  352284  s390: Conditional jump depends on uninitialised value(s) in vfp
  rintf
  352320  arm64 crash on none/tests/nestedfs
  352765  Vbit test fails on Power 6
  352768  The mbar instruction is missing from the Power PC support
  352769  Power PC program priority register (PPR) is not supported
  n-i-bz  Provide implementations of certain compiler builtins to support
          compilers that may not provide those
  n-i-bz  Old STABS code is still being compiled, but never used. Remove
  it.
  n-i-bz  Fix compilation on distros with glibc < 2.5
  n-i-bz  (vex 3098) Avoid generation of Neon insns on non-Neon hosts
  n-i-bz  Enable rt_sigpending syscall on ppc64 linux.
  n-i-bz  mremap did not work properly on shared memory
  n-i-bz  Fix incorrect sizeof expression in syswrap-xen.c reported by Co
  verity
  n-i-bz  In VALGRIND_PRINTF write out thread name, if any, to xml
  (3.11.0.TEST1:  8 September 2015, vex r3187, valgrind r15646)
  (3.11.0.TEST2: 21 September 2015, vex r3193, valgrind r15667)
  (3.11.0:      22 September 2015, vex r3195, valgrind r15674)

(alnsn)

Added math/nlopt version 2.4.2

(wen)

Add nlopt

(wen)

Import nlopt-2.4.2 as math/nlopt.

NLopt is a free/open-source library for nonlinear optimization,
providing a common interface for a number of different free optimization
outines available online as well as original implementations of various
other algorithms.

Its features include:
- Callable from C, C++, Fortran, Matlab or GNU Octave, Python,
  GNU Guile, Julia, GNU R, Lua, and OCaml.
- A common interface for many different algorithms -- try a different
  algorithm just by changing one parameter.
- Support for large-scale optimization (some algorithms scalable to
  millions of parameters and thousands of constraints).
- Both global and local optimization algorithms.
- Algorithms using function values only (derivative-free) and also
  algorithms exploiting user-supplied gradients.
- Algorithms for unconstrained optimization, bound-constrained
  optimization, and general nonlinear inequality/equality constraints.

Reviewed by: wiz@

(wen)

Note update of lang/php56 package to 5.6.16.

(taca)

Update php56 to 5.6.16 (PHP 5.6.16).

26 Nov 2015, PHP 5.6.16

- Core:
  . Fixed bug #70828 (php-fpm 5.6 with opcache crashes when referencing a
    non-existent constant). (Laruence)
  . Fixed bug #70748 (Segfault in ini_lex () at Zend/zend_ini_scanner.l).
    (Laruence)

- Mysqlnd:
  . Fixed bug #68344 (MySQLi does not provide way to disable peer certificate
    validation) by introducing MYSQLI_CLIENT_SSL_DONT_VERIFY_SERVER_CERT
        connection flag. (Andrey)

- OCI8:
  . Fixed bug #68298 (OCI int overflow). (Senthil)

- PDO_DBlib:
  . Fixed bug #69757 (Segmentation fault on nextRowset).
    (miracle at rpz dot name)

- SOAP:
  . Fixed bug #70875 (Segmentation fault if wsdl has no targetNamespace
    attribute). (Matteo)

- SPL:
  . Fixed bug #70852 (Segfault getting NULL offset of an ArrayObject).
    (Reeze Xia)

(taca)

Update HOMEPAGE.

(taca)

Clean up www/contao*:

o Do not set version in contao/Makefile.common but in each Makefile of
  contao?? package.
o Utilize mk/fetch/github.mk.
o Remove unused CT_FILES.

(taca)

Note update of www/contao35 package to 3.5.6.

(taca)

Update contao35 to 3.5.6.

Version 3.5.6 (2015-11-27)
--------------------------

### Fixed
Correctly determin the protocol delimiter in `Idna::encodeUrl()`.

### Fixed
Handle relative URLs when following redirects in the Request class (see #7799).

### Fixed
Correctly handle empty UUIDs when comparing versions (see #7971).

### Fixed
Remove the "required" attribute when setting up TinyMCE (see #8131).

Version 3.5.5 (2015-11-25)
--------------------------

### Fixed
Fix the domain when forwarding in the page controllers (see #8123).

### Fixed
Use the feed URL instead of the base URL for enclosures (see #8116).

### Fixed
Fix the `<time>` tags and standardize the event templates (see #8012).

### Fixed
Handle empty `href` attributes in the book navigation (see #8104).

### Fixed
Do not store e-mail addresses in the newsletter (un)subscription log.

### Fixed
Correctly encrypt fields upon registration (see #8110).

### Fixed
Correctly render required single checkboxes in the back end (see #7731).

### Fixed
Correctly store multi select menus if no value is selected (see #7760).

### Fixed
Prevent recursion when rendering 403/404 pages (see #8060).

### Fixed
Map the `FileTree` widget to `FormFileUpload` in the front end (see #8091).

### Fixed
Preserve the user input when loading image meta data (see #8108).

### Fixed
Show the "toggle all" buttons in "edit multiple" mode (see #5622).

### Fixed
Disable the gallery pagination if the images are sorted randomly (see #8033).

### Fixed
Set the correct empty value when copying elements (see #8064).

### Fixed
Correctly hide forward pages with no public subpages (see #8054).

### Fixed
Correctly render the page picker if the value starts with `#` (see #8055).

### Fixed
Correctly render the "group" option in the radio button and checkbox widgets.

### Fixed
Correctly set the ID when toggling fields via Ajax (see #8043).

### Fixed
Support call, sms and app hyperlinks when converting relative URLs (see #8102).

### Fixed
Correctly check if a folder is protected when loading subfolders.

### Fixed
Correctly check the synchronization status when copying or moving files.

### Fixed
Adjust the code to be compatible with PHP7 (see #8018).

### Fixed
Correctly show the UUID in the back end file manager popup (see #8058).

(taca)

Updated www/p5-Catalyst-Runtime to 5.90103

(mef)

Update 5.90053 to 5.90102
----------------------------------------------------------------
ChangeLog reaches some 500 lines for this update. Following is a
just Release numbers and their date for that.
----------------------------------------------------------------
5.90103 - 2015-11-12
    introduced in 5.90102.  This is a recommended upgrade (tsibley++, mst++)
5.90102 - 2015-10-29
5.90101 - 2015-09-04
5.90100 - 2015-08-24
5.90097 - 2015-07-28
5.90096 - 2015-07-27
5.90095 - 2015-07-27
5.90094 - 2015-07-24
5.90093 - 2015-05-29
5.90092 - 2015-05-19
  - Refactor the change introduced in 5.90091 to solve reported issues (for
5.90091 - 2015-05-08
5.90090 - 2015-04-29
  - Merged all changes from 5.90089_XXX development cycle.
5.90089_004 - 2015-04-28
5.90089_003 - 2015-04-27
5.90089_002 - 2015-04-17
5.90089_001 - 2015-03-26
  - New development branch synched with 5.90085.
5.90085 - 2015-03-25
5.90084 - 2015-02-23
5.90083 - 2015-02-16
5.90082 - 2015-01-10
5.90081 - 2015-01-10
5.90080 - 2015-01-09
5.90079_008  - 2015-01-07
5.90079_007  - 2015-01-07
  - Merged from Stable (5.90079)
    changes introduced in v5.90080.
5.90079_006  - 2015-01-02
5.90079_005 - 2014-12-31
  - Merged changes from 5.90078
5.90079_004 - 2014-12-26
5.90079_003 - 2014-12-03
5.90079_002 - 2014-12-02
5.90079_001 - 2014-12-02
5.90079 - 2015-01-02
5.90078 - 2014-12-30
    introduced in 5.90060.  You can use this if that feature is causing you
5.90077 - 2014-11-18
5.90076 - 2014-11-13
5.90075 - 2014-10-06
5.90074 - 2014-10-01
5.90073 - 2014-09-23
5.90072 - 2014-09-15
5.90071 - 2014-08-10
    in the wild (Catalyst-Plugin-Authentication-0.10023) - (removed in 5.90070)
  - Reverted changes to debug log/handling (5.90069_003) to fix
5.90070 - 2014-08-07
5.90069_004
5.90069_003
5.90069_002
5.90069_001
5.90065 - 2014-06-04
5.90064 - 2014-05-05
5.90063 - 2014-05-01
5.90062 - 2014-04-14
5.90061 - 2014-03-10
5.90060 - 2014-02-07
  - Same as 5.90059_006, just marking it as stable, no functional changes.
5.90059_006 - 2014-02-06
5.90059_005 - 2014-01-28
5.90059_004 - 2014-01-27
5.90059_003 - 2013-12-24
5.90059_002 - 2013-12-21
5.90059_001 - 2013-12-19
5.90053 - 2013-12-21

(mef)

Added www/p5-Plack-Middleware-MethodOverride version 0.15

(mef)

Import p5-Plack-Middleware-MethodOverride-0.15 as www/p5-Plack-Middleware-MethodOverride.

Writing RESTful apps is a good thing, but if you're also trying to support
web browsers, it would be nice not to be reduced to GET and POST for
everything.

This middleware allows for POST requests that pretend to be something else:
by adding either a header named X-HTTP-Method-Override to the request, or a
query parameter named x-tunneled-method to the URI, the client can say what
method it actually meant. That is, as long as it meant one of these:
    GET
    POST
    HEAD
    PUT
    DELETE
    OPTIONS
    TRACE
    CONNECT
    PATCH
If so, then the REQUEST_METHOD in the PSGI environment will be replaced
with the client's desired value. The original request method is always
stored under the plack.original_request_method key.

(mef)

distfile fixes

(nils)