Link [ pkgsrc | NetBSD | pkgsrc git mirror | PR fulltext-search | netbsd commit viewer ]


   
        usage: [branch:branch] [user:user] [path[@revision]] keyword [... [-excludekeyword [...]]] (e.g. branch:MAIN pkgtools/pkg)




switch to index mode

recent branches: MAIN (38m)  pkgsrc-2024Q1 (15d)  pkgsrc-2023Q4 (42d)  pkgsrc-2023Q2 (75d)  pkgsrc-2023Q3 (154d) 

2024-05-13 10:28:50 UTC Now

2017-06-08 09:08:51 UTC MAIN commitmail json YAML

2017-06-08 08:47:39 UTC MAIN commitmail json YAML

Updated graphics/asymptote to 2.41

(wiz)

2017-06-08 08:47:26 UTC MAIN commitmail json YAML

Updated asymptote to 2.41.

Release Notes for Version 2.41

The rendering of Bezier patches was improved. Minor improvements were made
to perpendicular marks in the geometry package. A perl issue was addressed.
Two unwanted temporary files are now removed after TeX processing.
EPS output is now supported with all TeX engines. A workaround was
implemented for the ImageMagick jpeg black background bug.

Release Notes for Version 2.40

A partial workaround for the OpenGL transparency bug was implemented, by
presorting transparent triangles of the same material. The examples were
updated and a segmentation fault was fixed. Multisample detection, surface
rendering, and crack filling algorithms were fixed. The default compilation
flags now specify -std=c++11.

Release Notes for Version 2.39

A workaround was implemented for the backwards incompatibility in the
TeXLive 2016 graphicx package. Empty tick labels in graphs are now avoided.
A paletteticks NoTicks option was added. Support for lualatex was improved.
Renderers for Bezier patches and curves more efficient than those
in the deprecated GLU library were implemented.

(wiz)

2017-06-08 08:38:35 UTC MAIN commitmail json YAML

Restrict to python-2.7 due to py-scp.

(wiz)

2017-06-08 08:33:46 UTC MAIN commitmail json YAML

2017-06-08 08:29:30 UTC MAIN commitmail json YAML

I was too hopeful. This does not really support gnupg2 yet, see
https://rt.cpan.org/Public/Bug/Display.html?id=102651

Switch back to gnupg1. Bump PKGREVISION.

(wiz)

2017-06-08 04:42:08 UTC MAIN commitmail json YAML

Updated security/py-oauth2client to 4.1.1

(adam)

2017-06-08 04:41:45 UTC MAIN commitmail json YAML

v4.1.1
New features:
Allow passing prompt='consent' via the flow_from_clientsecrets

(adam)

2017-06-08 04:39:23 UTC MAIN commitmail json YAML

2017-06-08 04:32:20 UTC MAIN commitmail json YAML

Updated graphics/libepoxy to 1.4.3

(adam)

2017-06-08 04:31:34 UTC MAIN commitmail json YAML

Changes 1.4.3:
Require Meson 0.39.1; the previous release already had an implicit requirement on this version
Allow disabling EGL support when building Epoxy
Add support for building on Haiku
Update EGL registry from Khronos

(adam)

2017-06-08 01:28:10 UTC MAIN commitmail json YAML

2017-06-08 00:17:24 UTC MAIN commitmail json YAML

2017-06-07 23:53:58 UTC MAIN commitmail json YAML

2017-06-07 23:49:25 UTC MAIN commitmail json YAML

2017-06-07 23:19:05 UTC MAIN commitmail json YAML

Mark gnatmake as NOT_PAX_MPROTECT_SAFE

Bumps PKGREVISION.

(khorben)

2017-06-07 22:57:02 UTC MAIN commitmail json YAML

Fix "make install" in devel/kyua without "tests" package option.

* Add a space after the "@comment" PLIST keyword inserted in the
  case where "tests" option is disabled.

* Only create the "${PREFIX}/tests" directory and install the
  Kyuafile there if the "tests" option is enabled.

(jlam)

2017-06-07 22:29:41 UTC MAIN commitmail json YAML

2017-06-07 20:46:34 UTC MAIN commitmail json YAML

2017-06-07 20:28:02 UTC MAIN commitmail json YAML

Updated security/py-yara to 3.6.1

(khorben)

2017-06-07 20:27:51 UTC MAIN commitmail json YAML

Updated security/yara to 3.6.1

(khorben)

2017-06-07 20:27:37 UTC MAIN commitmail json YAML

Package yara 3.6.1

In the release notes:

* BUGFIX: Stack overflow caused by uncontrolled recursiveness (CVE-2017-9304)
* BUGFIX: pe.overlay.size was undefined if the PE didn't have an overlay. Now it's set to 0 in those cases.
* BUGFIX: Fix initalization issue that could cause a crash if rules compiled with a 32bit yarac is used with a 64bit yara.

(khorben)

2017-06-07 20:24:27 UTC MAIN commitmail json YAML

Fix the value being substituted for TTMKFDIR in install scripts.

@TYPE1INST@ should be replaced with ${TOOLS_PATH.type1inst}.
Before this change, the sed(1) script being generated was not
replaced @TYPE1INST@ at all, and instead generating a no-op
substitution pattern for TTMKFDIR.

(jlam)

2017-06-07 20:24:19 UTC MAIN commitmail json YAML

Substitute for LDCONFIG_ADD_CMD and LDCONFIG_REMOVE_CMD in scripts.

LDCONFIG_ADD_CMD and LDCONFIG_REMOVE_CMD are set by emulator/*.mk
in order to update the library search paths in the emulated system
directories.  Substitute for those values, but default them to the
empty string so that the default action taken by the "shlibs" task
can be used.

(jlam)

2017-06-07 20:24:10 UTC MAIN commitmail json YAML

Fix subsitutions for @PKG_ADMIN@ and @PKG_INFO@.

The substituted values need to refer to the _CMD variables that
hold the path to the actual commands.

(jlam)

2017-06-07 20:24:01 UTC MAIN commitmail json YAML

Subsitute for @CROSSBASE@ and @VARBASE@ in install scripts.

${CROSSBASE} and ${VARBASE} both point to important directories
for installed packages, so always substitute those values into
deinstall/install scripts.

(jlam)

2017-06-07 20:12:35 UTC MAIN commitmail json YAML

Updated security/py-yara to 3.6.0

(khorben)

2017-06-07 20:12:21 UTC MAIN commitmail json YAML

Updated security/yara to 3.6.0

(khorben)

2017-06-07 20:11:42 UTC MAIN commitmail json YAML

Package yara 3.6.0

In the release notes:
* .NET module (Wesley Shields)
* New features for ELF module (Jacob Baines)
* Fix endianness issues (Hilko Bengen)
* Function yr_compiler_add_fd added to libyara
* MAX_THREADS limit can be arbitrarily increased (Emerson R. Wiley)
* Added --fail-on-warnings command-line option
* Multiple bug fixes

(khorben)

2017-06-07 19:17:29 UTC MAIN commitmail json YAML

Remove new sentence, new line warning. mandoc learned it.

>From Ingo Schwarze.

Ride bump.

(wiz)

2017-06-07 19:10:05 UTC MAIN commitmail json YAML

Updated textproc/mdoclint to 1.55

(wiz)

2017-06-07 19:09:55 UTC MAIN commitmail json YAML

Remove -n, mandoc now has a warning for this case.

>From Ingo Schwarze.

Bump version.

(wiz)

2017-06-07 18:07:50 UTC MAIN commitmail json YAML

fix build with perl 5.26, copied from p5-Data-ICal's version of this.

(mrg)

2017-06-07 14:56:58 UTC MAIN commitmail json YAML

2017-06-07 14:53:52 UTC MAIN commitmail json YAML

2017-06-07 14:51:18 UTC MAIN commitmail json YAML

2017-06-07 14:46:59 UTC MAIN commitmail json YAML

2017-06-07 14:45:31 UTC MAIN commitmail json YAML

2017-06-07 14:42:26 UTC MAIN commitmail json YAML

2017-06-07 14:34:47 UTC MAIN commitmail json YAML

2017-06-07 14:33:24 UTC MAIN commitmail json YAML

2017-06-07 14:31:37 UTC MAIN commitmail json YAML

2017-06-07 14:29:59 UTC MAIN commitmail json YAML

2017-06-07 14:28:46 UTC MAIN commitmail json YAML

Add a missing patch to fix build with Perl 5.26.0

(ryoon)

2017-06-07 14:27:12 UTC MAIN commitmail json YAML

Fix build with Perl 5.26.0

(ryoon)

2017-06-07 11:17:14 UTC MAIN commitmail json YAML

Updated lang/nodejs6 to 6.11.0

(fhajny)

2017-06-07 11:09:44 UTC MAIN commitmail json YAML

Update lang/nodejs6 to 6.11.0

Notable Changes

build:
- support for building mips64el
cluster:
- disconnect() now returns a reference to the disconnected worker.
crypto:
- ability to select cert store at runtime
- Use system CAs instead of using bundled ones
- The Decipher methods setAuthTag() and setAAD now return this.
- adding support for OPENSSL_CONF again
- make LazyTransform compabile with Streams1
deps:
- upgrade libuv to 1.11.0
dns:
- Implemented {ttl: true} for resolve4() and resolve6().
process:
- add NODE_NO_WARNINGS environment variable
readline:
- add option to stop duplicates in history
src:
- support "--" after "-e" as end-of-options
tls:
- new tls.TLSSocket() supports sec ctx options
- Allow obvious key/passphrase combinations.

(fhajny)

2017-06-07 10:29:13 UTC MAIN commitmail json YAML

Note update of net/py-junos-eznc to 2.1.3.

(he)

2017-06-07 10:25:30 UTC MAIN commitmail json YAML

Update py-junos-eznc to version 2.1.3.

Pkgsrc changes:
===============
* Update dependencies to match requirements.txt
* Adapt to PLIST changes

Upstream changes:
=================

Release 2.1.3
-------------

Features Added:
* Ephemeral config support #707
* Add a srx_cluster_redundancy_group fact. #711

Bugs Fixed:
* ignore_warning fails when single that is first child of . #712
* mode='telnet' did not logout non-cli user #713
* JSONLoadError was thrown when load valid JSON config #717/#718
* Fix XML normalization feature when using NETCONF over console. #719/#720
* Handle differences in |display xml rpc #722

Release 2.1.2
-------------

Bugs Fixed:
* Doc badge was pointing to older version #694
* Fix new-style fact gathering for SRX clusters. #697/#698
* Properly handle SW upgrades on multi-RE and/or multi-chassis
  systems when using new-style fact gathering. #700
* Raise JSONLoadError if json config is malformed #706
* Handle ConnectClosedError exception for lock() and unlock() #708

Release 2.1.1
-------------

Bugs Fixed:
* Fix regressions caused by ignore_warning. #691

Release 2.1.0
-------------

Features Added:
* Enhanced fact gathering. Facts are now gathered "on demand."
  Additional facts are added.
* The definition of facts and return values are properly documented. #638
* Support for YANG get RPCs. #672
* Add an ignore_warning argument to suppress RpcError exceptions
  for warnings. #672/#685
* Enhanced the sw.install() method with basic ISSU and NSSU
  support using the issu and nssu
* boolean arguments. #606/#630/#632
* ** NSSU support has not yet been tested and should currently
  be considered experimental.
* Provide a master property and a re_name property for Device. #682
* Enhanced reboot() method to take an all_re boolean parameter
  which controls if only the connected
* Routing Engine, or all Routing Engines, are rebooted. #613
* Enhanced the warning message produced by the cli() method to
  recommend the corresponding
* dev.rpc.<method>() call. #603
* Add support for update parameter to configuration load() method. #681
* Added directory_usage to utils #629/#631/#636
* Adding support for NFX/JDM fact gathering. #652/#659
* Connected property. #664

Bugs Fixed:
* Updated the interface-name glob pattern to correctly match
  et-<x>/<y>/<z> interfaces
* in several tables and views. #609
* Take care of special chars on screen during console connection. #610
* Address issue with fact gathering stopping when it encounters
  a problem. #615
* Minor typos fixed in RuntimeError exception message and in comments. #621
* Added console_has_banner parameter. #622
* Add CentOS Support to install instructions #623
* Key value is needed in _IsisAdjacencyLogTable #627
* Improved functionality and documentation of Docker build.
  #637/#673/#674/#677
* added remote port ID to lldp.yml (OP) #645
* Fix documentation for rollback() #647
* Fix for fact gathering pprint. #660/#661
* update ospf view, add bgp/inventory #665
* Updated doc string for close function #686
* Add Travis builds for Python 3.5 and 3.6 #687
* StartShell.run to take this as None for non returning commands #680
* Modify ignore_warning return value to mimic normal RPC return value. #688

Release 2.0.1
-------------
* StartShell to take timeout (30 second by default) as paramter
* Proper exception handling in case of Console connection #595
* Fix: Config.lock() return exception when normalize is on
* Added microbadge badge for the Docker image #593
* Fix: print dev for Console conn was printing object not Device(....) #591
* Fix: To take care of special chars with StartShell->run function call #589
* Fix: ssh private key file to be considered for scp util #586
* Added Dockerfile to enable automated image builds on project commits #585

(he)

2017-06-07 10:23:17 UTC MAIN commitmail json YAML

2017-06-07 10:22:19 UTC MAIN commitmail json YAML

Note update of net/py-ncclient to 0.5.3.

(he)

2017-06-07 10:21:35 UTC MAIN commitmail json YAML

Update py-ncclient to version 0.5.3.

Pkgsrc changes:
* Adapt PLIST and patch to README.rst.

Upstream changes:

v0.5.3
- Add notifications support
- Add support for ecdsa keys
- Various bug fixes

(he)

2017-06-07 09:54:59 UTC MAIN commitmail json YAML

Note addition of textproc/py-jxmlease version 1.0.1.

(he)

2017-06-07 09:54:17 UTC MAIN commitmail json YAML

Add py-jxmlease.

(he)

2017-06-07 09:53:42 UTC MAIN commitmail json YAML

Add py-jxmlease version 1.0.1, imported from pkgsrc-wip.

This is a Python module for converting XML to intelligent Python
data structures, and converting Python data structures to XML.

(he)

2017-06-07 09:45:43 UTC MAIN commitmail json YAML

Note addition of net/py-scp version 0.10.2.

(he)

2017-06-07 09:44:05 UTC MAIN commitmail json YAML

Add py-scp.

(he)

2017-06-07 09:43:15 UTC MAIN commitmail json YAML

Add py-scp version 0.10.2 from pkgsrc-wip.

Py-scp is a pure python scp module.

The scp.py module uses a paramiko transport to send and recieve
files via the scp1 protocol. This is the protocol as referenced
from the openssh scp program, and has only been tested with this
implementation.

(he)

2017-06-07 09:14:39 UTC MAIN commitmail json YAML

Added math/py-bottleneck version 1.2.1; Updated math/py-pandas to 0.20.2, multimedia/ffmpeg3 to 3.3.2, mail/dovecot2 to 2.2.30.2

(adam)

2017-06-07 09:12:41 UTC MAIN commitmail json YAML

Hopefully the last 2.2.30.x..
- auth: Multiple failed authentications within short time caused
  crashes
- push-notification: OX driver crashed at deinit

(adam)

2017-06-07 08:48:23 UTC MAIN commitmail json YAML

version 3.3.2:
- avcodec/mpeg4videodec: Fix runtime error: signed integer overflow: 53098 * 40448 cannot be represented in type 'int'
- avcodec/pafvideo: Fix assertion failure
- avcodec/takdec: Fix multiple runtime error: signed integer overflow: 637072 * 4096 cannot be represented in type 'int'
- avcodec/mjpegdec: Check that reference frame matches the current frame
- avcodec/tiff: Avoid loosing allocated geotag values
- avcodec/cavs: Fix runtime error: signed integer overflow: -12648062 * 256 cannot be represented in type 'int'
- avformat/hls: Check local file extensions
- avcodec/qdrw: Fix null pointer dereference
- avutil/softfloat: Fix sign error in and improve documentation of av_int2sf()
- avcodec/hevc_ps: Fix runtime error: index 32 out of bounds for type 'uint8_t [32]'
- avcodec/dxv: Check remaining bytes in dxv_decompress_raw()
- avcodec/pafvideo: Check packet size and frame code before ff_reget_buffer()
- avcodec/ac3dec_fixed: Fix runtime error: left shift of 419 by 23 places cannot be represented in type 'int'
- avformat/options: log filename on open
- avcodec/aacps: Fix runtime error: left shift of 1073741824 by 1 places cannot be represented in type 'INTFLOAT' (aka 'int')
- avcodec/wavpack: Fix runtime error: shift exponent 32 is too large for 32-bit type 'int'
- avcodec/cfhd: Fix runtime error: signed integer overflow: 65280 * 65288 cannot be represented in type 'int'
- avcodec/wavpack: Fix runtime error: signed integer overflow: 2013265955 - -134217694 cannot be represented in type 'int'
- avcodec/cinepak: Check input packet size before frame reallocation
- avcodec/hevc_ps: Fix runtime error: signed integer overflow: 2147483628 + 256 cannot be represented in type 'int'
- avcodec/ra144: Fixes runtime error: signed integer overflow: 7160 * 327138 cannot be represented in type 'int'
- avcodec/pnm: Use ff_set_dimensions()
- avcodec/cavsdec: Fix runtime error: signed integer overflow: 59 + 2147483600 cannot be represented in type 'int'
- avcodec/nvenc: fix hw accelerated transcode with bframes
- libavformat/hls: Observe Set-Cookie headers
- libavformat/http: Ignore expired cookies
- avformat/avidec: Limit formats in gab2 to srt and ass/ssa
- avcodec/acelp_pitch_delay: Fix runtime error: value 4.83233e+39 is outside the range of representable values of type 'float'
- avcodec/wavpack: Check float_shift
- avcodec/wavpack: Fix runtime error: signed integer overflow: 24 * -2147483648 cannot be represented in type 'int'
- avcodec/ansi: Fix frame memleak
- avcodec/dds: Fix runtime error: left shift of 145 by 24 places cannot be represented in type 'int'
- avcodec/jpeg2000dec: Use ff_set_dimensions()
- avcodec/truemotion2: Fix passing null pointer to memset()
- avcodec/truemotion2: Fix runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
- avcodec/ra144: Fix runtime error: signed integer overflow: -2449 * 1398101 cannot be represented in type 'int'
- avcodec/ra144: Fix runtime error: signed integer overflow: 11184810 * 404 cannot be represented in type 'int'
- avcodec/aac_defines: Add missing () to AAC_HALF_SUM() macro
- avcodec/webp: Fixes null pointer dereference
- avcodec/aacdec_fixed: Fix runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
- avcodec/ylc: Check count in build_vlc()
- avcodec/snow: Fix runtime error: signed integer overflow: 1086573993 + 1086573994 cannot be represented in type 'int'
- avcodec/jpeg2000: Fix runtime error: signed integer overflow: 4185 + 2147483394 cannot be represented in type 'int'
- avcodec/jpeg2000dec: Check tile offsets more completely
- avcodec/sheervideo: Check input buffer size before allocating and decoding
- avcodec/aacdec_fixed: Fix multiple runtime error: shift exponent 127 is too large for 32-bit type 'int'
- avcodec/wnv1: More strict buffer size check
- avcodec/libfdk-aacdec: Correct buffer_size parameter
- avcodec/sbrdsp_template: Fix: runtime error: signed integer overflow: 849815297 + 1315389781 cannot be represented in type 'int'
- avcodec/ivi_dsp: Fix runtime error: left shift of negative value -2
- doc/filters: Clarify scale2ref example
- avcodec/mlpdec: Do not leave invalid values in matrix_out_ch[] on error
- avcodec/ra144dec: Fix runtime error: left shift of negative value -17
- avcodec/pixlet: Fix runtime error: signed integer overflow: 2147483647 + 32 cannot be represented in type 'int'
- avformat/mux: Fix copy an paste typo
- avutil/internal: Do not enable CHECKED with DEBUG
- avcodec/clearvideo: Check buf_size before decoding frame
- avcodec/aacdec_fixed: Fix runtime error: signed integer overflow: -2147483648 * -1 cannot be represented in type 'int'
- avcodec/smc: Check remaining input
- avcodec/diracdec: Fix off by 1 error in quant check
- avcodec/jpeg2000dec: Fix copy and paste error
- avcodec/jpeg2000dec: Check tile offsets
- avcodec/sanm: Fix uninitialized reference frames
- avcodec/jpeglsdec: Check get_bits_left() before decoding a picture
- avcodec/fmvc: Fix use of uninitialized memory when the first frame is not a keyframe
- avcodec/ivi_dsp: Fix multiple runtime error: left shift of negative value -71
- avcodec/mjpegdec: Fix runtime error: signed integer overflow: -32767 * 130560 cannot be represented in type 'int'
- avcodec/aacdec_fixed: Fix runtime error: shift exponent 34 is too large for 32-bit type 'int'
- avcodec/mpeg4videodec: Check for multiple VOL headers
- avcodec/vp9block: fix runtime error: signed integer overflow: 196675 * 20670 cannot be represented in type 'int'
- avcodec/vmnc: Check location before use
- avcodec/takdec: Fix runtime error: signed integer overflow: 8192 * 524308 cannot be represented in type 'int'
- avcodec/aac_defines: Fix: runtime error: left shift of negative value -2
- avcodec/takdec: Fix runtime error: left shift of negative value -63
- avcodec/mlpdsp: Fix runtime error: signed integer overflow: -24419392 * 128 cannot be represented in type 'int'
- avcodec/sbrdsp_fixed: fix runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
- avcodec/aacsbr_fixed: Fix multiple runtime error: shift exponent 170 is too large for 32-bit type 'int'
- avcodec/mlpdec: Do not leave a invalid num_primitive_matrices in the context
- avcodec/aacsbr_fixed: Fix multiple runtime error: shift exponent 150 is too large for 32-bit type 'int'
- avcodec/mimic: Use ff_set_dimensions() to set the dimensions
- avcodec/fic: Fix multiple runtime error: signed integer overflow: 5793 * 419752 cannot be represented in type 'int'
- avcodec/pixlet: Fix reading invalid numbers of bits
- avcodec/mlpdec: Fix: runtime error: left shift of negative value -8
- avcodec/dfa: Fix: runtime error: signed integer overflow: -14202 * 196877 cannot be represented in type 'int'
- avcodec/aacdec: Fix runtime error: signed integer overflow: 2147483520 + 255 cannot be represented in type 'int'
- avcodec/aacdec_template: Fix fixed point scale in decode_cce()
- avcodec/fmvc: Fix off by 1 error
- avcodec/flicvideo: Check frame_size before decrementing
- avcodec/mlpdec: Fix runtime error: left shift of negative value -1
- avcodec/takdec: Fix  runtime error: left shift of negative value -42
- avcodec/hq_hqa: Fix: runtime error: signed integer overflow: -255 * 10180917 cannot be represented in type 'int'
- avcodec/scpr: mask bits to prevent out of array read
- avcodec/truemotion1: Fix multiple runtime error: signed integer overflow: 1246906962 * 2 cannot be represented in type 'int'
- avcodec/svq3: Fix runtime error: left shift of negative value -6
- avcodec/tiff: reset sampling[] if its invalid
- configure: Fix the msvcrt version check for mingw32
- lavf/mov: make invalid m{d,v}hd time_scale default to 1 instead of erroring out
- lavc/ffjni: add missing '\n'
- lavc/mediacodec_wrapper: do not declare JNIAMedia{Codec,CodecList,Format}Fields on the stack
- lavc/mediacodec_wrapper: fix local reference leaks
- avcodec/nvenc: remove unnecessary alignment
- Use AVOnce as a static variable consistently
- avfilter: take_samples: do not directly return frame when samples are skipped
- avutil/hwcontext_dxva2: Don't improperly free IDirect3DSurface9 objects

(adam)

2017-06-07 08:13:56 UTC MAIN commitmail json YAML

v0.20.2:
This is a minor bug-fix release in the 0.20.x series and includes some small regression fixes, bug fixes and performance improvements. We recommend that all users upgrade to this version.

(adam)

2017-06-07 08:12:47 UTC MAIN commitmail json YAML

2017-06-07 05:41:53 UTC MAIN commitmail json YAML

update to version 1.8.20p2

upstream changelog:
2017-05-31  Todd C. Miller  <Todd.Miller@courtesan.com>

        * NEWS, configure, configure.ac:
        Sudo 1.8.20p2
        [47836f4c9834]

        * src/ttyname.c:
        A command name may also contain newline characters so read
        /proc/self/stat until EOF. It is not legal for /proc/self/stat to
        contain embedded NUL bytes so treat the file as corrupt if we see
        any. With help from Qualys.

        This is not exploitable due to the /dev traversal changes in sudo
        1.8.20p1 (thanks Solar!).
        [15a46f4007dd]

2017-05-30  Todd C. Miller  <Todd.Miller@courtesan.com>

        * src/ttyname.c:
        Use /proc/self consistently on Linux. As far as I know, only AIX
        doesn't support /proc/self.
        [6f3d9816541b]

(spz)

2017-06-07 04:41:41 UTC MAIN commitmail json YAML

dump PKGREVISION

(snj)

2017-06-07 04:41:22 UTC MAIN commitmail json YAML

Updated chat/irssi to 1.0.3

(snj)

2017-06-07 04:38:55 UTC MAIN commitmail json YAML

Update irssi to 1.0.3.  Changes:
- Fix out of bounds read when scanning expandos (GL!11).
- Fix invalid memory access with quoted filenames in DCC
  (GL#8, GL!12).
- Fix null-pointer dereference on DCC without address (GL#9, GL!13).
- Improve integer overflow handling. Originally reported by
  oss-fuzz#525 (#706).
- Improve nicklist performance from O(N^2) to O(N) (#705).
- Fix initial screen redraw delay. By Stephen Oberholtzer
  (#680, bdo#856201).
- Fix incorrect reset of true colours when resetting background. (#711).
- Fix missing -notls option in /SERVER. By Jari Matilainen (#117, #702).
- Fix minor history glitch on overcounter (#462, #685).
- Improved OpenSSL detection at compile time. By Rodrigo Rebello (#677).
- Improved NetBSD Terminfo detection. By Maya Rashish (#694, #698).
- Add missing syntax info for COMPLETION (#687, #688).
- Minor typo correction in help. By Michael Hansen (#707).

>From maya@

(snj)

2017-06-07 00:36:24 UTC MAIN commitmail json YAML

2017-06-07 00:34:15 UTC MAIN commitmail json YAML

2017-06-07 00:32:18 UTC MAIN commitmail json YAML

2017-06-07 00:28:24 UTC MAIN commitmail json YAML

2017-06-07 00:25:40 UTC MAIN commitmail json YAML

2017-06-07 00:23:02 UTC MAIN commitmail json YAML

2017-06-07 00:12:53 UTC MAIN commitmail json YAML

2017-06-07 00:10:55 UTC MAIN commitmail json YAML

2017-06-07 00:08:37 UTC MAIN commitmail json YAML

2017-06-07 00:05:51 UTC MAIN commitmail json YAML

2017-06-07 00:01:33 UTC MAIN commitmail json YAML

2017-06-07 00:00:24 UTC MAIN commitmail json YAML

2017-06-06 23:58:52 UTC MAIN commitmail json YAML

2017-06-06 20:41:06 UTC MAIN commitmail json YAML

Updated security/p5-GnuPG-Interface to 0.52nb4

(wiz)

2017-06-06 20:40:56 UTC MAIN commitmail json YAML

Allow selecting gnupg version. Switch default to gnupg2.
Bump PKGREVISION.

(wiz)

2017-06-06 20:37:24 UTC MAIN commitmail json YAML

Updated security/caff to 2.5nb1

(wiz)

2017-06-06 20:37:15 UTC MAIN commitmail json YAML

Allow choosing gnupg1, gnupg2, or gnupg21. Default to gnupg2. Bump PKGREVISION.

(wiz)

2017-06-06 17:34:33 UTC MAIN commitmail json YAML

Needs py-setuptools_scm.

(wiz)

2017-06-06 15:19:11 UTC MAIN commitmail json YAML

2017-06-06 15:16:58 UTC MAIN commitmail json YAML

2017-06-06 15:14:28 UTC MAIN commitmail json YAML

2017-06-06 15:12:05 UTC MAIN commitmail json YAML

2017-06-06 15:09:42 UTC MAIN commitmail json YAML

2017-06-06 15:03:01 UTC MAIN commitmail json YAML

2017-06-06 14:58:09 UTC MAIN commitmail json YAML

2017-06-06 14:54:15 UTC MAIN commitmail json YAML

2017-06-06 14:51:04 UTC MAIN commitmail json YAML

2017-06-06 14:48:44 UTC MAIN commitmail json YAML

2017-06-06 14:45:48 UTC MAIN commitmail json YAML

2017-06-06 14:38:05 UTC MAIN commitmail json YAML

2017-06-06 14:30:31 UTC MAIN commitmail json YAML

2017-06-06 14:27:23 UTC MAIN commitmail json YAML

2017-06-06 14:22:36 UTC MAIN commitmail json YAML

2017-06-06 12:37:50 UTC MAIN commitmail json YAML

2017-06-06 11:12:05 UTC MAIN commitmail json YAML

2017-06-06 09:41:29 UTC MAIN commitmail json YAML

2017-06-06 09:41:18 UTC MAIN commitmail json YAML

Update misc/tmux to 2.5.

CHANGES FROM 2.4 to 2.5 09 May 2017

- Reset updated flag when restarting #() command so that new output is
  properly
  recognised. GitHub issue 922.
- Fix ECH with a background colour.
- Do not rely on the terminal not moving the cursor after DL or EL.
- Fix send-keys and send-prefix in copy-mode (so C-b C-b works).
- Set the current pane for rotate-window so it works in command
  sequences.
- Add pane_mode format.
- Differentiate M-Up from Escape+Up when possible (that is, in
  terminals with xterm(1) style function keys). GitHub issue 907.
- Add session_stack and window_stack_index formats.
- Some new control mode notifications and corresponding hooks:
  pane-mode-changed, window-pane-changed, client-session-changed,
  session-window-changed.
- Format pane_search_string for last search term while in copy mode
  (useful with command-prompt -I).
- Fix a problem with high CPU usage and multiple clients with #().
- Fix UTF-8 combining characters in column 0.
- Fix reference counting so that panes are properly destroyed and
  their processes killed.
- Clamp SU (CSI S) parameter to work around a bug in Konsole.
- Tweak line wrapping in full width panes to play more nicely with
  terminal copy and paste.
- Fix when we emit SGR 0 in capture-pane -e.
- Do not change TERM until after config file parsing has finished, so
  that commands run inside the config file can use it to make decisions
  (typically about default-terminal).
- Make the initial client wait until config file parsing has finished
  to avoid racing with commands.
- Fix core when if-shell fails.
- Only use ED to clear screen if the pane is at the bottom.
- Fix multibyte UTF-8 output.
- Code improvements around target (-t) resolution.
- Change how the default target (for commands without -t) is managed
  across command sequences: now it is set up at the start and commands
  are required to update it if needed. Fixes binding command sequences
  to mouse keys.
- Make if-shell from the config file work correctly.
- Change to always check the root key table if no binding is found in
  the current table (prefix table or copy-mode table or whatever). This
  means that root key bindings will take effect even in copy mode, if not
  overridden by a copy mode key binding.
- Fix so that the history file works again.
- Run config file without a client rather than using the first client,
  restores previous behaviour.
- If a #() command doesn't exit, continue to read from it and use its
  last full line of output.
- Handle slow terminals and fast output better: when the amount of
  data outstanding gets too large, discard output until it is drained and
  we are able to do a full redraw. Prevents tmux sitting on a huge buffer
  that the terminal will take forever to consume.
- Do not redraw a client unless we realistically think it can accept
  the data - defer redraws until the client has nothing else waiting to write.

(fhajny)

2017-06-06 09:12:59 UTC MAIN commitmail json YAML

2017-06-06 08:22:45 UTC MAIN commitmail json YAML

2017-06-06 08:02:55 UTC MAIN commitmail json YAML

2017-06-06 07:34:50 UTC MAIN commitmail json YAML

+ ImageMagick-7.0.5.10, afl-2.42b, alsa-lib-1.1.4.1, colordiff-1.0.18,
  libgcrypt-1.7.7, libuv-1.12.0, mcookie-2.30, mtr-0.92, notmuch-0.24.2,
  npth-1.5, p5-Date-Manip-6.59, p5-DateTime-TimeZone-2.13,
  p5-List-SomeUtils-0.54, p5-List-SomeUtils-XS-0.53,
  p5-Regexp-Common-2017060201, p5-Type-Tiny-1.002000,
  p5-WebService-MusicBrainz-1.0.2, par2-0.7.2, py-cryptography-1.9,
  wireshark-2.2.7, x264-devel-20170603, xterm-328.

(wiz)

2017-06-06 07:30:28 UTC MAIN commitmail json YAML

+ gnuplot-5.2, libmicrohttpd-0.9.55, nodejs-8.0.0, p5-DateTime-1.4300,
  p5-JSON-2.94, p5-Mojolicious-7.32, p5-Net-HTTP-6.16, syncthing-0.14.29,
  tmux-2.5, wine-devel-2.9.

(wiz)

2017-06-06 03:36:34 UTC MAIN commitmail json YAML

Enable bootsrapping on NetBSD 8.99.1

(ryoon)

2017-06-05 23:19:54 UTC MAIN commitmail json YAML

2017-06-05 21:14:26 UTC MAIN commitmail json YAML

Updated lang/nodejs to 8.0.0

(fhajny)

2017-06-05 21:14:04 UTC MAIN commitmail json YAML

Update lang/nodejs to 8.0.0

Node.js 8.0.0 is a major new release that includes a significant number of
semver-major and semver-minor changes. Notable changes are listed below.

The Node.js 8.x release branch is scheduled to become the next actively
maintained Long Term Support (LTS) release line in October, 2017 under the
LTS codename 'Carbon'.

### Notable Changes

* Async Hooks
  * The `async_hooks` module has landed in core

* Buffer
  * Using the `--pending-deprecation` flag will cause Node.js to emit a
    deprecation warning when using `new Buffer(num)` or `Buffer(num)`.
  * `new Buffer(num)` and `Buffer(num)` will zero-fill new `Buffer` instances
  * Many `Buffer` methods now accept `Uint8Array` as input

* Child Process
  * Argument and kill signal validations have been improved
  * Child Process methods accept `Uint8Array` as input

* Console
  * Error events emitted when using `console` methods are now supressed.

* Dependencies
  * The npm client has been updated to 5.0.0
  * V8 has been updated to 5.8 with forward ABI stability to 6.0

* Domains
  * Native `Promise` instances are now `Domain` aware

* Errors
  * We have started assigning static error codes to errors generated by Node.js.
    This has been done through multiple commits and is still a work in
    progress.

* File System
  * The utility class `fs.SyncWriteStream` has been deprecated
  * The deprecated `fs.read()` string interface has been removed

* HTTP
  * Improved support for userland implemented Agents
  * Outgoing Cookie headers are concatenated into a single string
  * The `httpResponse.writeHeader()` method has been deprecated
  * New methods for accessing HTTP headers have been added to `OutgoingMessage`

* Lib
  * All deprecation messages have been assigned static identifiers
  * The legacy `linkedlist` module has been removed

* N-API
  * Experimental support for the new N-API API has been added

* Process
  * Process warning output can be redirected to a file using the
    `--redirect-warnings` command-line argument
  * Process warnings may now include additional detail

* REPL
  * REPL magic mode has been deprecated

* Src
  * `NODE_MODULE_VERSION` has been updated to 57
  * Add `--pending-deprecation` command-line argument and
    `NODE_PENDING_DEPRECATION` environment variable
  * The `--debug` command-line argument has been deprecated. Note that
    using `--debug` will enable the *new* Inspector-based debug protocol
    as the legacy Debugger protocol previously used by Node.js has been
    removed.
  * Throw when the `-c` and `-e` command-line arguments are used at the same
    time
  * Throw when the `--use-bundled-ca` and `--use-openssl-ca` command-line
    arguments are used at the same time.

* Stream
  * `Stream` now supports `destroy()` and `_destroy()` APIs
  * `Stream` now supports the `_final()` API

* TLS
  * The `rejectUnauthorized` option now defaults to `true`
  * The `tls.createSecurePair()` API now emits a runtime deprecation
  * A runtime deprecation will now be emitted when `dhparam` is less than
    2048 bits

* URL
  * The WHATWG URL implementation is now a fully-supported Node.js API

* Util
  * `Symbol` keys are now displayed by default when using `util.inspect()`
  * `toJSON` errors will be thrown when formatting `%j`
  * Convert `inspect.styles` and `inspect.colors` to prototype-less objects
  * The new `util.promisify()` API has been added

* Zlib
  * Support `Uint8Array` in Zlib convenience methods
  * Zlib errors now use `RangeError` and `TypeError` consistently

(fhajny)

2017-06-05 20:37:34 UTC MAIN commitmail json YAML

Updated net/youtube-dl to 20170605

(leot)

2017-06-05 20:37:21 UTC MAIN commitmail json YAML

Update net/youtube-dl to 20170605

Changes:
version 2017.06.05

Core
* [YoutubeDL] Don't emit ANSI escape codes on Windows (#13270)

Extractors
+ [bandcamp:weekly] Add support for bandcamp weekly (#12758)
* [pornhub:playlist] Fix extraction (#13281)
- [godtv] Remove extractor (#13175)
* [safari] Fix typo (#13252)
* [youtube] Improve chapters extraction (#13247)
* [1tv] Lower preference for HTTP formats (#13246)
* [francetv] Relax URL regular expression
* [drbonanza] Fix extraction (#13231)
* [packtpub] Fix authentication (#13240)

version 2017.05.29

Extractors
* [youtube] Fix DASH MPD extraction for videos with non-encrypted format URLs
  (#13211)
* [xhamster] Fix uploader and like/dislike count extraction (#13216))
+ [xhamster] Extract categories (#11728)
+ [abcnews] Add support for embed URLs (#12851)
* [gaskrank] Fix extraction (#12493)
* [medialaan] Fix videos with missing videoUrl (#12774)
* [dvtv] Fix playlist support
+ [dvtv] Add support for DASH and HLS formats (#3063)
+ [beam:vod] Add support for beam.pro/mixer.com VODs (#13032))
* [cbsinteractive] Relax URL regular expression (#13213)
* [adn] Fix formats extraction
+ [youku] Extract more metadata (#10433)
* [cbsnews] Fix extraction (#13205)

version 2017.05.26

Core
+ [utils] strip_jsonp() can recognize more patterns
* [postprocessor/ffmpeg] Fix metadata filename handling on Python 2 (#13182)

Extractors
+ [youtube] DASH MPDs with cipher signatures are recognized now (#11381)
+ [bbc] Add support for authentication
* [tudou] Merge into youku extractor (#12214)
* [youku:show] Fix extraction
* [youku] Fix extraction (#13191)
* [udemy] Fix extraction for outputs' format entries without URL (#13192)
* [vimeo] Fix formats' sorting (#13189)
* [cbsnews] Fix extraction for 60 Minutes videos (#12861)

version 2017.05.23

Core
+ [downloader/external] Pass -loglevel to ffmpeg downloader (#13183)
+ [adobepass] Add support for Bright House Networks (#13149)

Extractors
+ [streamcz] Add support for subtitles (#13174)
* [youtube] Fix DASH manifest signature decryption (#8944, #13156)
* [toggle] Relax URL regular expression (#13172)
* [toypics] Fix extraction (#13077)
* [njpwworld] Fix extraction (#13162, #13169)
+ [hitbox] Add support for smashcast.tv (#13154)
* [mitele] Update app key regular expression (#13158)

version 2017.05.18.1

Core
* [jsinterp] Fix typo and cleanup regular expressions (#13134)

version 2017.05.18

Core
+ [jsinterp] Add support for quoted names and indexers (#13123, #13124, #13125,
  #13126, #13128, #13129, #13130, #13131, #13132)
+ [extractor/common] Add support for schemeless URLs in _extract_wowza_formats
  (#13088, #13092)
+ [utils] Recognize more audio codecs (#13081)

Extractors
+ [vier] Extract more metadata (#12539)
* [vier] Improve extraction (#12801)
    + Add support for authentication
    * Bypass authentication when no credentials provided
    * Improve extraction robustness
* [dailymail] Fix sources extraction (#13057)
* [dailymotion] Extend URL regular expression (#13079)

(leot)

2017-06-05 20:22:37 UTC MAIN commitmail json YAML

Updated textproc/the_silver_searcher to 2.0.0

(leot)

2017-06-05 20:22:24 UTC MAIN commitmail json YAML

Update textproc/the_silver_searcher to 2.0.0

Changes:
2.0.0
-----
- Remove support for `.agignore' file, please use `.ignore' instead
- Add --print-all-files option.
  This option causes ag to print headings for all files that it
  searches, not just those that contain matches. This will be useful
  in the case where ag is being called by another program that needs
  to show the user a progress indicator of how many paths have been
  searched.
- Add Haxe support
- Add support for jsp tag and fragment files

(leot)

2017-06-05 19:04:20 UTC MAIN commitmail json YAML

2017-06-05 18:58:44 UTC MAIN commitmail json YAML

2017-06-05 18:01:06 UTC MAIN commitmail json YAML

Security update for File::Path to 2.13.

2.13 2017-05-31
    - Document security vulnerability reported as CVE-2017-6512.

2.12_008 2017-05-07
    - Patch from John Lightsey.

2.12_007 2017-04-22
    - Skip tests where filesystem doesn't support permissions (RT 121248).
    - Add AppVeyor configuration; thanks to Charlie Gonzalez and Hayo Baan.

2.12_006 2017-04-21
    - Modernize README, Makefile.PL, updating of version number
      and release date in documentation.

2.12_005 2017-04-21
    - Recommend use of 'safe => 1' in remove_tree() and rmtree().
    - Warn if mkpath() or make_path() is passed implausible options on
      Windows.
    - Corrections to errors in previous release.

2.12_004 2017-04-18
    - Certain functions used in tests are not available on Windows;
      skip them.
    - Move certain functions used in testing to t/FilePathTest.pm.

2.12_003 2017-04-07
    - Add tests to improve coverage ratios as measured by Devel::Cover
    - No functional changes.

2.12_002 2017-03-12
    - GH#41 RT 117019 Fixed File::Path::remove_tree option hash is auto
                      populated and cannot be reused
    - GH#40 Unskip in path root t
    - GH#39 Remove superfluous assignment to $arg{perm}
    - GH#38 Minor grammatical doc fixes.
    - GH#37 Minor grammatical doc fixes.

2.12_001 2016-09-18
    - RT 94209  document that the thread safety issue will not change and
      communicate alternative.
    - RT 85878  be more generous with error check regex given we could be
      dealing with a pre-1.25 Carp.
    - GH #33    Be more precise in documentation example for make_path
      error checking.
    - GH #34 Skip Windows 2000 and earlier unit tests (test change).
    - GH #36 Do not hardcode ENOENT (test change).

(bsiegert)

2017-06-05 16:06:57 UTC MAIN commitmail json YAML

* Drop task for release from Rakefile which depends on files in prior
  version 12.0 of rake.

(taca)

2017-06-05 16:02:21 UTC MAIN commitmail json YAML

Note update of devel/php-libawl package to 0.57.

(taca)

2017-06-05 16:01:56 UTC MAIN commitmail json YAML

Update php-libawl to 0.57.

2017-01-23 Florian Schlichting <fsfs@debian.org>
* release awl 0.57
* Update AUTHORS and ChangeLog

2017-01-10 Florian Schlichting <fsfs@debian.org>
* Browser: add ExtraRowFormat() to format additional rows differently

2017-01-08 Florian Schlichting <fsfs@debian.org>
* delete session after logout, clean up old sessions (fixes davical#65, Debian #643907)
* always use full the ID as username, similar to davical's
  HTTPAuthSession (fix #1, fix #2, debian #703139)

2017-01-07 Florian Schlichting <fsfs@debian.org>
* Session::Log and ::Dbg are deprecated
* vComponent: add missing strtoupper, so that BEGIN:TYPE...END:TYPE
  are really case-insensitive

2017-01-02 Florian Schlichting <fsfs@debian.org>
* vProperty: always escape backslash, and escape semicolon for "other"
  properties (fixes Debian bug #837154)

2016-12-31 Florian Schlichting <fsfs@debian.org>
* improve debug logging: ignore components, filter by remote IP or username

2016-12-28 Florian Schlichting <fsfs@debian.org>
* eliminate trailing whitespace, expand tabs
* doc typos

2016-12-02 Florian Schlichting <fsfs@debian.org>
* Remove deprecated and unused class iCalendar along with remaining
  $c->local_tzid references

2016-12-01 Paul Kallnbach <p.kallnbach@gorilla-computing.de>
* Update vProperty.php

2016-06-13 Florian Schlichting <fsfs@debian.org>
* Bump AWL version to 0.57 (API change)
* fix if-else logic, so that "Unable to Reset Password" does not get
  overwritten by "Temporary Password" form

2016-06-13 Basti <basti122303@gitlab>
* Fix link after password reset in case DAViCal is not installed in
  the webserver root. Closes:davical-project/davical#94

2016-05-08 Andrew McMillan <andrew@morphoss.com>
* A semi-colon might be escaped within the text value.
* Rename get_fields() function.
* Improve access to the slow query threshold.

2016-05-10 Florian Schlichting <fsfs@debian.org>
* always remake the apidoc, and remove the stale copy from git
* update Makefile to work with both php5 and php7

2016-02-22 ClemensN <c.nuebel@gorilla-computing.de>
* fixed issue with escaped ',' in CATEGORIES Property that are seperators in this case

(taca)

2017-06-05 16:00:34 UTC MAIN commitmail json YAML

Note update of databases/ruby-dm-serializer package to 1.2.2nb2.

(taca)

2017-06-05 16:00:00 UTC MAIN commitmail json YAML

Tweak gemspec dependency.

Bump PKGREVISION.

(taca)

2017-06-05 15:40:22 UTC MAIN commitmail json YAML

Note update of www/ruby-sass package to 3.4.24.

(taca)

2017-06-05 15:39:59 UTC MAIN commitmail json YAML

Update ruby-sass to 3.4.24.

3.4.24 (18 May 2017)

* Elements without a namespace (such as div) are no longer unified with
  elements with the empty namespace (such as |div). This unification didn't
  match the results returned by is-superselector(), and was not guaranteed to
  be valid.

(taca)

2017-06-05 15:35:52 UTC MAIN commitmail json YAML

Note update of www/ruby-rest-client package to 2.0.2.

(taca)

2017-06-05 15:35:29 UTC MAIN commitmail json YAML

Update ruby-rest-client to 2.0.2.

# 2.0.2

- Suppress the header override warning introduced in 2.0.1 if the value is the
  same. There's no conflict if the value is unchanged. (#578)

(taca)

2017-06-05 15:34:54 UTC MAIN commitmail json YAML

Note update of www/ruby-raindrops package to 0.18.0.

(taca)

2017-06-05 15:34:33 UTC MAIN commitmail json YAML

Update ruby-raindrops to 0.18.0.

=== raindrops 0.18.0 / 2017-03-23 02:44 UTC

  The most notable feature of this release is the addition of
  FreeBSD and OpenBSD TCP_INFO support.  This includes the
  Raindrops::TCP for portably mapping TCP state names to
  platform-dependent numeric values:

    https://bogomips.org/raindrops/Raindrops.html#TCP

  Thanks to Jeremy Evans and Simon Eskildsen on the
  unicorn-public@bogomips.org mailing list for inspiring
  these changes to raindrops.

  There's also a few internal cleanups, and documentation
  improvements, including some fixes to the largely-forgotten
  Raindrops::Aggreage::PMQ class:

    https://bogomips.org/raindrops/Raindrops/Aggregate/PMQ.html

  20 changes since 0.17.0:

        test_inet_diag_socket: fix Fixnum deprecation warning
        TODO: add item for IPv6 breakage
        ext: fix documentation for C ext-defined classes
        TCP_Info: custom documentation for #get!
        TypedData C-API conversion
        test_watcher: disable test correctly when aggregate is missing
        tcp_info: support this struct under FreeBSD
        define Raindrops::TCP hash for TCP states
        linux_inet_diag: reduce stack usage and simplify
        avoid reading errno repeatedly
        aggregate/pmq: avoid false sharing of lock buffers
        aggregate/pmq: remove io-extra requirement
        aggregate/pmq: avoid File#stat allocation
        Merge remote-tracking branch 'origin/freebsd'
        Merge remote-tracking branch 'origin/aggregate-pmq'
        doc: remove private email support address
        doc: update location of TCP_INFO-related stuff
        build: avoid olddoc for building the RubyGem
        doc: document Raindrops::TCP hash
        aggregate/pmq: update version numbers for Ruby and Linux

(taca)

2017-06-05 15:33:50 UTC MAIN commitmail json YAML

Note update of www/ruby-http-form_data package to 1.0.3.

(taca)

2017-06-05 15:33:26 UTC MAIN commitmail json YAML

Update ruby-http-form_data to 1.0.3.

## 1.0.3 (2017-05-18)

* [#16](https://github.com/httprb/form_data/issues/16)
  Fix ruby < 2.0.0 support.
  [@ixti][]

## 1.0.2 (2017-05-08)

* [#5](https://github.com/httprb/form_data.rb/issues/5)
  Allow setting Content-Type non-file parts
  [@abotalov]

* [#6](https://github.com/httprb/form_data.rb/issues/6)
  Creation of file parts without filename
  [@abotalov]

* [#11](https://github.com/httprb/form_data.rb/pull/11)
  Deprecate `HTTP::FormData::File#mime_type`. Use `#content_type` instead.
  [@ixti]

(taca)

2017-06-05 15:32:52 UTC MAIN commitmail json YAML

Note update of www/ruby-http package to 2.2.2.

(taca)

2017-06-05 15:32:15 UTC MAIN commitmail json YAML

Update ruby-http to 2.2.2.

## 2.2.2 (2017-04-27)

* [#404](https://github.com/httprb/http/issues/404),
  [#405](https://github.com/httprb/http/pull/405)
  Make keepalive timeout configurable.
  ([@nestegg])

(taca)

2017-06-05 15:31:16 UTC MAIN commitmail json YAML

Note addition of time/ruby-iso8601 package version 0.9.1.

(taca)

2017-06-05 15:30:47 UTC MAIN commitmail json YAML

Add and enable ruby-iso8601.

(taca)

2017-06-05 15:30:17 UTC MAIN commitmail json YAML

Add ruby-iso8601 package version 0.9.1.

ISO8601 is a simple implementation in Ruby of the ISO 8601 (Data elements and
interchange formats - Information interchange - Representation of dates
and times) standard.

(taca)

2017-06-05 15:28:52 UTC MAIN commitmail json YAML

Note update of time/ruby-hitimes package to 1.2.5.

(taca)

2017-06-05 15:28:26 UTC MAIN commitmail json YAML

Update ruby-hitimes to 1.2.5.

## Version 1.2.5 2017-05-25

* Update dependencies
* Add ruby 2.4 to windows fatbinary
* Update docs to indicate windows ruby before 2.0 is no longer supported

(taca)

2017-06-05 15:27:51 UTC MAIN commitmail json YAML

Note update of textproc/ruby-yard package to 0.9.9.

(taca)

2017-06-05 15:27:28 UTC MAIN commitmail json YAML

Update ruby-yard to 0.9.9.

Release v0.9.9 2017/04/23

* Added gem uninstall hooks to remove YARD documentation files. (#1083)
* Added support for C++ namespaces. (#809)
* Fixed issue where loading a .html page via an anchor would not scroll to
  the anchor section. (#1082)
* Hide some Ruby warnings.
* Improve progress indicator icons in terminal.

(taca)

2017-06-05 15:27:13 UTC MAIN commitmail json YAML

Added www/py-django-rest-framework-proxy version 1.6.0

(adam)

2017-06-05 15:24:25 UTC MAIN commitmail json YAML

Note update of textproc/ruby-plist package to 3.3.0.

(taca)

2017-06-05 15:24:02 UTC MAIN commitmail json YAML

Update ruby-plist to 3.3.0.

=== 3.3.0 (2017-04-28)

https://github.com/patsplat/plist/compare/dece870...v3.3.0

* Fix ASCII/UTF-8 error (https://github.com/patsplat/plist/pull/38).
* Fix Fixnum, Bignum deprecations in Ruby 2.4
* Fix unused variable `e` warning

(taca)

2017-06-05 15:23:21 UTC MAIN commitmail json YAML

Note update of textproc/ruby-nokogiri package to 1.8.0.

(taca)

2017-06-05 15:23:01 UTC MAIN commitmail json YAML

Provides views to redirect incoming request to another API server.

Features:
* Masquerade paths
* HTTP Basic Auth (between your API and backend API)
* Token Auth
* Supported methods: GET/POST/PUT/PATCH
* File uploads

(adam)

2017-06-05 15:22:56 UTC MAIN commitmail json YAML

Update ruby-nokogiri to 1.8.0.

# 1.8.0 / 2017-06-04

## Backwards incompatibilities

This release ends support for Ruby 2.1 on Windows in the `x86-mingw32` and `x64-mingw32` platform gems (containing pre-compiled DLLs). Official support ended for Ruby 2.1 on 2017-04-01.

Please note that this deprecation note only applies to the precompiled Windows gems. Ruby 2.1 continues to be supported (for now) in the default gem when compiled on installation.

## Dependencies

* [Windows] Upgrade iconv from 1.14 to 1.15 (unless --use-system-libraries)
* [Windows] Upgrade zlib from 1.2.8 to 1.2.11 (unless --use-system-libraries)
* [MRI] Upgrade rake-compiler dependency from 0.9.2 to 1.0.3
* [MRI] Upgrade mini-portile2 dependency from `~> 2.1.0` to `~> 2.2.0`

## Compatibility notes

* [JRuby] Removed support for `jruby --1.8` code paths. [#1607] (Thanks, @kares!)
* [MRI Windows] Retrieve zlib source from http://zlib.net/fossils to avoid deprecation issues going forward. See #1632 for details around this problem.

## Features

* NodeSet#clone is not an alias for NodeSet#dup [#1503] (Thanks, @stephankaag!)
* Allow Processing Instructions and Comments as children of a document root. [#1033] (Thanks, @windwiny!)
* [MRI] PushParser#replace_entities and #replace_entities= will control whether entities are replaced or not. [#1017] (Thanks, @spraints!)
* [MRI] SyntaxError#to_s now includes line number, column number, and log level if made available by the parser. [#1304, #1637] (Thanks, @spk and @ccarruitero!)
* [MRI] Cross-built Windows gems now support Ruby 2.4
* [MRI] Support for frozen string literals. [#1413]
* [MRI] Support for installing Nokogiri on a machine in FIPS-enabled mode [#1544]
* [MRI] Vendored libraries are verified with SHA-256 hashes (formerly some MD5 hashes were used) [#1544]
* [JRuby] (performance) remove unnecessary synchronization of class-cache [#1563] (Thanks, @kares!)
* [JRuby] (performance) remove unnecessary cloning of objects in XPath searches [#1563] (Thanks, @kares!)
* [JRuby] (performance) more performance improvements, particularly in XPath, Reader, XmlNode, and XmlNodeSet [#1597] (Thanks, @kares!)

## Bugs

* HTML::SAX::Parser#parse_io now correctly parses HTML and not XML [#1577] (Thanks for the test case, @gregors!)
* Support installation on systems with a `lib64` site config. [#1562]
* [MRI] on OpenBSD, do not require gcc if using system libraries [#1515] (Thanks, @jeremyevans!)
* [MRI] XML::Attr.new checks type of Document arg to prevent segfaults. [#1477]
* [MRI] Prefer xmlCharStrdup (and friends) to strdup (and friends), which can cause problems on some platforms. [#1517] (Thanks, @jeremy!)
* [JRuby] correctly append a text node before another text node [#1318] (Thanks, @jkraemer!)
* [JRuby] custom xpath functions returning an integer now work correctly [#1595] (Thanks, @kares!)
* [JRuby] serializing (`#to_html`, `#to_s`, et al) a document with explicit encoding now works correctly. [#1281, #1440] (Thanks, @kares!)
* [JRuby] XML::Reader now returns parse errors [#1586] (Thanks, @kares!)
* [JRuby] Empty NodeSets are now decorated properly. [#1319] (Thanks, @kares!)
* [JRuby] Merged nodes no longer results in Java exceptions during XPath queries. [#1320] (Thanks, @kares!)

# 1.7.2 / 2017-05-09

## Security Notes

[MRI] Upstream libxslt patches are applied to the vendored libxslt 1.1.29 which address CVE-2017-5029 and CVE-2016-4738.

For more information:

* https://github.com/sparklemotion/nokogiri/issues/1634
* http://people.canonical.com/~ubuntu-security/cve/2017/CVE-2017-5029.html
* http://people.canonical.com/~ubuntu-security/cve/2016/CVE-2016-4738.html

(taca)

2017-06-05 15:21:56 UTC MAIN commitmail json YAML

Note update of textproc/ruby-ferret package to 0.11.8.7.

(taca)

2017-06-05 15:21:20 UTC MAIN commitmail json YAML

2017-06-05 15:15:41 UTC MAIN commitmail json YAML

Note update of security/ruby-oauth package to 0.5.3.

(taca)

2017-06-05 15:15:01 UTC MAIN commitmail json YAML

Update ruby-oauth to 0.5.3

=== 0.5.3 2017-05-24

* Removing legacy scripts (James Pinto)
* Fix #145 - broken CLI required loading active_support (James Pinto)

=== 0.5.2 2017-05-17

* Adding a development dependency that had not been mentioned (James Pinto)
* Use assert_nil so as to silence a Minitest 6 deprecation warning (James Pinto)
* Stop bundling tests files in the gem (Michal Papis)
* Minor cleanup on tests (James Pinto)
* TravisCI no longer needs libcurl-dev (James Pinto)
* Nokogiri 1.7 does not accept Ruby 2.0 (James Pinto)
* Upgrading to CodeClimate 1.0 (James Pinto)
* Adding support to Ruby 2.4 and head (James Pinto)
* Locking gemspec to Rails 4 so as to allow our next version for Rails 5 (James Pinto)
* Fix #113 adding paths when a full URL has been specified  (James Pinto)
* moving development dependency to gemspec (James Pinto)
* Silencing 'Net::HTTPResponse#header is obsolete' (James Pinto)
* Silencing some test warnings (James Pinto)
* Silencing 'loading in progress, circular require considered harmful' (James Pinto)
* Silence 'URI.escape obsolete' (James Pinto)
* Refactored CLI (James Pinto)
* Bug Fix, webmock 2.0 has introduced a new bug (James Pinto)
* Moving test files into test/units/ (James Pinto)
* Adding CodeClimate (James Pinto)
* Reimplementing #82 - Debug Output Option (James Pinto)
* Making a test/support dir (James Pinto)
* Fix #177 - Adjusting to webmock latest recommended implementation for minitest (James Pinto)
* Adding support to Ruby 2.4 and head (James Pinto)
* Upgrading to CodeClimate 1.0 (James Pinto)
* Nokogiri 1.7 does not accept Ruby 2.0 (James Pinto)

(taca)

2017-06-05 14:53:24 UTC MAIN commitmail json YAML

Add and enable ruby-proxifier.

(taca)

2017-06-05 14:52:55 UTC MAIN commitmail json YAML

Add ruby-proxifier package version 1.0.3.

Proxifier is a gem to force ruby to use a proxy.

This gem was created for 2 purposes.

First is to enable ruby programmers to use HTTP or SOCKS proxies
interchangeably when using TCPSockets. Either manually with
Proxifier::Proxy#open or by require "proxifier/env".

The second purpose is to use ruby code that doesn't user proxies for users
that have to use proxies.

The pruby and pirb executables are simple wrappers for their respective ruby
executables that support proxies from environment variables.

(taca)

2017-06-05 14:50:54 UTC MAIN commitmail json YAML

Note update of net/ruby-domain_name package to 0.5.20170404.

(taca)

2017-06-05 14:50:32 UTC MAIN commitmail json YAML

Update ruby-domain_name to 0.5.20170404.

# Change Log

## [v0.5.20170404](https://github.com/knu/ruby-domain_name/tree/v0.5.20170404) (2017-04-04)
[Full Changelog](https://github.com/knu/ruby-domain_name/compare/v0.5.20170223...v0.5.20170404)

- Update the eTLD database to 2017-04-04T20:20:25Z

(taca)

2017-06-05 14:49:58 UTC MAIN commitmail json YAML

Note update of net/ruby-dnsruby package to 1.60.1.

(taca)

2017-06-05 14:49:26 UTC MAIN commitmail json YAML

Update ruby-dnsruby to 1.60.1.

## v1.60.1

* DNSSEC validation switched OFF by default (but can still be switched on)
* Add APL RR support (thanks Manabu Sonoda)
* Various test fixes (thanks Keith Bennett)
* 'include' issues fixed (thanks Keith Bennett!)
* Fixnum replacement (thanks Keith Bennett)
* Zone transfer fixes (thanks Manabu Sonoda)
* Name decoding fix
* MX record passing error now raised
* CAA RR support (thanks Richard Luther)
* TLSA RR support (thanks Manabu Sonoda)

(taca)

2017-06-05 14:48:52 UTC MAIN commitmail json YAML

Note update of net/ruby-amq-protocol package to 2.2.0.

(taca)

2017-06-05 14:48:30 UTC MAIN commitmail json YAML

Update ruby-amq-protocol to 2.2.0.

## Changes between 2.1.0 and 2.2.0 (unreleased)

### Timestamps are Encoded as 64-bit Unsigned Integers

This is a potentially **breaking change**. It is recommended that
all applications that use this gem and pass date/time values in message
properties or headers are upgraded at the same time.

GitHub issue: [#64](https://github.com/ruby-amqp/amq-protocol/issues/64).

Contributed by Carl Hoerberg.

(taca)

2017-06-05 14:45:54 UTC MAIN commitmail json YAML

Increment version number of perl5

(ryoon)

2017-06-05 14:43:40 UTC MAIN commitmail json YAML

Note update of misc/ruby-ohai package to 13.1.0.

(taca)

2017-06-05 14:43:01 UTC MAIN commitmail json YAML

Update ruby-ohai to 13.1.0.

v13.1.0 (2017-05-09)

* Improvements to EC2 metadata handling to reuse connections #995 (tas50)
* EC2: Poll EC2 metadata from the new 2016 metadata API versions #992 (tas50)
* Inject sane paths into shell_out #991 (akitada)
* mdadm: Add members devices array #989 (jaymzh)

v13.0.1 (2017-04-12)

Fixed bugs:

* Fix Eucalyptus plugin to use the HttpHelper mixin #987 (akitada)
* Fix OpenStack plugin to use the HttpHelper mixin #986 (akitada)

v13.0.0 (2017-04-06)

Implemented enhancements:

* Load all the plugin paths for ohai plugin dependency resolution #984
  (thommay)
* Require Ruby 2.3+ to match Chef itself #981 (tas50)
* Allow ohai to load a plugin path #980 (thommay)
* Fully rename the cloud_v2 plugin to cloud #978 (tas50)
* Replace the existing filesystem plugin on Linux and Darwin with
  filesystem_v2 #974 (tas50)
* Replace the cloud plugin with the cloud_v2 plugin #973 (tas50)
* Freeze all string values coming out of Ohai. #972 (coderanger)
* Update amazon to use the platform_family of amazon not RHEL #971
  (lamont-granquist)
* Add DMI type 40,41, and 42 from the latest man page #969 (tas50)
* [ec2] Add additional data from identity document #964 (webframp)
* Move duplicate http logic into a helper #951 (tas50)
* Remove deprecated config logic #939 (tas50)
* Require Ruby 2.2+ #938 (tas50)
* Remove the deprecated run_command and popen4 methods #933 (tas50)
* Remove usage of the Sigar gem #930 (tas50)

Fixed bugs:

* Remove sbt version detect as it's not possible in the current sbt #982 (tas50)
* Fix scala detection when version output contains a warning #959 (tas50)
* Fix lua detection on new versions of lua #958 (tas50)
* Fix logger issues #955 (lamont-granquist)
* Rescue exception in DMI plugin #952 (tas50)
* Use name for Windows CPU model_name #918 (tduffield)

8.24.0 (2017-05-08)

* base: Load additional ohai plugins from /etc/chef/ohai/plugins or
  C:\chef\ohai\plugins\
* ec2: Poll EC2 metadata from the new 2016 metadata API versions #992 (tas50)
  mdadm: Add a new 'members' attribute for member devices in the array #989
  (jaymzh)
* dmi: Add DMI type 40,41, and 42 from the latest man page #969 (tas50)
* ec2: Gather availability_zone and region data #964 (webframp)
* scala: Fix scala detection when version output contains a warning #959 (tas50)
* lua: Fix lua detection on new versions of lua #958 (tas50)
* dmi: Rescue exception in DMI plugin to improve debug logs #952 (tas50)

(taca)

2017-06-05 14:33:54 UTC MAIN commitmail json YAML

Note update of misc/ruby-mini_portile2 package to 2.2.0.

(taca)

2017-06-05 14:33:28 UTC MAIN commitmail json YAML

Update ruby-mini_portile2 to 2.2.0.

### 2.2.0 / 2017-06-04

#### Enhancements

* Remove MD5 hashing of configure options, not avialbale in FIPS mode. (#78)
* Add experimental support for cmake-based projects.
* Retry on HTTP failures during downloads. [#63] (Thanks, @jtarchie and @jvshahid!)
* Support Ruby 2.4 frozen string literals.
* Support applying patches for users with misconfigured git worktree. [#69]
* Support gpg signature verification of download resources.

(taca)

2017-06-05 14:32:47 UTC MAIN commitmail json YAML

Note update of misc/ruby-bundler package to 1.15.1.

(taca)

2017-06-05 14:32:24 UTC MAIN commitmail json YAML

Update ruby-bundler to 1.15.1.

## 1.15.1 (2017-06-02)

Bugfixes:

  - `bundle lock --update GEM` will fail gracefully when the gem is not in the lockfile (#5693, @segiddins)
  - `bundle init --gemspec` will fail gracefully when the gemspec is invalid (@colby-swandale)
  - `bundle install --force` works when the gemfile contains git gems (#5678, @segiddins)
  - `bundle env` will print well-formed markdown when there are no settings (#5677, @segiddins)

## 1.15.0 (2017-05-19)

This space intentionally left blank.

## 1.15.0.pre.4 (2017-05-10)

Bugfixes:

  - avoid conflicts when `Gem.finish_resolve` is called after the bundle has been set up (@segiddins)
  - ensure that `Gem::Specification.find_by_name` always returns an object that can have `#to_spec` called on it (#5592, @jules2689)

## 1.15.0.pre.3 (2017-04-30)

Bugfixes:

  - avoid redundant blank lines in the readme generated by `bundle gem` (@koic)
  - ensure that `open-uri` is not loaded after `bundle exec` (@segiddins)
  - print a helpful error message when an activated default gem conflicts with
    a gem in the gemfile (@segiddins)
  - only shorten `ref` option for git gems when it is a SHA (#5620, @segiddins)

## 1.15.0.pre.2 (2017-04-23)

Bugfixes:

  - ensure pre-existing fit caches are updated from remote sources (#5423, @alextaylor000)
  - avoid duplicating specs in the lockfile after updating with the gem uninstalled (#5599, @segiddins)
  - ensure git gems have their extensions available at runtime (#5594, @jules2689, @segiddins)

## 1.15.0.pre.1 (2017-04-16)

Features:

  - print a notification when a newer version of bundler is available (#4683, @segiddins)
  - add man pages for all bundler commands (#4988, @feministy)
  - add the `bundle info` command (@fredrb, @colby-swandale)
  - all files created with `bundle gem` comply with the bundler style guide (@zachahn)
  - if installing a gem fails, print out the reason the gem needed to be installed (#5078, @segiddins)
  - allow setting `gem.push_key` to set the key used when running `rake release` (@DTrierweiler)
  - print gem versions that are regressing during `bundle update` in yellow (#5506, @brchristian)
  - avoid printing extraneous dependencies when the resolver encounters a conflict (@segiddins)
  - add the `bundle issue` command that prints instructions for reporting issues (#4871, @jonathanpike)
  - add `--source` and `--group` options to the `bundle inject` command (#5452, @Shekharrajak)
  - add the `bundle add` command to add a gem to the gemfile (@denniss)
  - add the `bundle pristine` command to re-install gems from cached `.gem` files (#4509, @denniss)
  - add a `--parseable` option for `bundle config` (@JuanitoFatas, @colby-swandale)

Performance:

  - speed up gemfile initialization by storing locked dependencies as a hash (@jules2689)
  - speed up gemfile initialization by making locked dependency comparison lazy, avoiding object allocation (@jules2689)
  - only validate git gems when they are downloaded, instead of every time `Bundler.setup` is run (@segiddins)
  - avoid regenerating the lockfile when nothing has changed (@segiddins)
  - avoid diffing large arrays when no sources in the gemfile have changed (@segiddins)
  - avoid evaluating full gemspecs when running with RubyGems 2.5+ (@segiddins)

Bugfixes:

  - fix cases where `bundle update` would print a resolver conflict instead of updating the selected gems (#5031, #5095, @segiddins)
  - print out a stack trace after an interrupt when running in debug mode (@segiddins)
  - print out when bundler starts fetching a gem from a remote server (@segiddins)
  - fix `bundle gem` failing when `git` is unavailable (#5458, @Shekharrajak, @colby-swandale)
  - suggest the appropriate command to unfreeze a bundle (#5009, @denniss)
  - ensure nested calls to `bundle exec` resolve default gems correctly (#5500, @segiddins)
  - ensure that a plugin failing to install doesn't uninstall other plugins (@kerrizor, @roseaboveit)
  - ensure `socket` is required before being referenced (#5533, @rafaelfranca)
  - allow running `bundle outdated` when gems aren't installed locally (#5553, @segiddins)
  - print a helpful error when `bundle exec`ing to a gem that isn't included in the bundle (#5487, @segiddins)
  - print an error message when a non-git gem is given a `branch` option (#5530, @colby-swandale)
  - allow interrupts to exit the process after gems have been installed (@segiddins)
  - print the underlying error when downloading gem metadata fails (#5579, @segiddins)
  - avoid deadlocking when installing with a lockfile that is missing dependencies (#5378, #5480, #5519, #5526, #5529, #5549, #5572, @segiddins)

(taca)

2017-06-05 14:31:29 UTC MAIN commitmail json YAML

Note update of math/ruby-gsl package to 2.1.0.3.

(taca)

2017-06-05 14:30:52 UTC MAIN commitmail json YAML

Update ruby-gsl to 2.1.0.3.

* Suppress warnings in Ruby 2.4 by suppressing use of Fixnum.

(taca)

2017-06-05 14:27:13 UTC MAIN commitmail json YAML

Note update of mail/ruby-mail package to 2.6.5.

(taca)

2017-06-05 14:26:51 UTC MAIN commitmail json YAML

Update ruby-mail to 2.6.5.

== Version 2.6.5 - 2017-04-26 Jeremy Daer <jeremydaer@gmail.com>

Features:
* #1053 - Ruby 2.4.0 compatibility. Fixnum+Bignum unified as Integer. (peterkovacs)

Bugs:
* #605 - Fix Mail::Address#name for nil addresses (peterkovacs)
* #1003 - Fix decoding some b encoded headers on specific rubies that don't account for lack of base64 padding (kjg)
* #1023 - Fix double-quoting in display names. (garethrees)

(taca)

2017-06-05 14:26:14 UTC MAIN commitmail json YAML

Note update of graphics/ruby-cairo package to 1.15.9.

(taca)

2017-06-05 14:25:47 UTC MAIN commitmail json YAML

Update ruby-cairo to 1.15.9.

Release 1.15.9 (2017-06-03) Kouhei Sutou <kou@cozmixng.org>
===========================================================

Improvements
------------

  * Supported RubyInstaller2.
    [GitHub#47][Reported by PLURIX]

Thanks
------

  * PLURIX

Release 1.15.8 (2017-05-29) Kouhei Sutou <kou@cozmixng.org>
===========================================================

Improvements
------------

  * Added Cairo::FreeTypeFotnFace.
    [GitHub#45][Patch by Stafford Brunk]

  * Windows: Supported Ruby 2.4.
    [ruby-list:50525][Reported by 5.5]

  * Windows: Supported MSYS2.

  * Windows: Updated bundled libraries:
    * zlib: 1.2.8 -> 1.2.11

Thanks
------

  * Stafford Brunk

  * 5.5

Release 1.15.7 (2017-05-01) Kouhei Sutou <kou@cozmixng.org>
===========================================================

Fixes
-----

  * Windows: Fixed zlib version. We should use the same version
    as Ruby Installer. Ruby Installer uses zlib 1.2.8.
    [GitHub#47][Reported by jjgh]

Thanks
------

  * jigh

Release 1.15.6 (2017-04-27) Kouhei Sutou <kou@cozmixng.org>
===========================================================

Improvements
------------

  * Added Cairo::Context#refrence_count.
  * Windows: Updated bundled libraries:
    * zlib: 1.2.10 -> 1.2.11
    * libpng: 1.6.27 -> 1.6.29
    * FreeType: 2.7 -> 2.7.1
  * Used native-package-installer gem.

(taca)

2017-06-05 14:25:36 UTC MAIN commitmail json YAML

2017-06-05 14:24:48 UTC MAIN commitmail json YAML

Note update of fonts/ruby-ttfunk package to 1.5.1.

(taca)

2017-06-05 14:24:20 UTC MAIN commitmail json YAML

Update ruby-ttfunkto 1.5.1.

## [1.5.1]

### Fixed

* loca table corruption during subsetting. The loca table serialization code
  didn't properly detect suitable table format.

* Fixed checksum calculation for empty tables.

(taca)

2017-06-05 14:23:46 UTC MAIN commitmail json YAML

Note update of devel/ruby-tins package to 1.14.0.

(taca)

2017-06-05 14:23:21 UTC MAIN commitmail json YAML

Update ruby-tinsto 1.14.0.

2017-05-16 Release 1.14.0

* Add visit proc to determine if file should be visited.

(taca)

2017-06-05 14:18:26 UTC MAIN commitmail json YAML

Note update of devel/ruby-test-unit package to 3.2.4.

(taca)

2017-06-05 14:17:43 UTC MAIN commitmail json YAML

Update ruby-test-unit to 3.2.4.

## 3.2.4 - 2017-05-23 {#version-3-2-4}

### Improvements

  * Updated tests for Ruby 2.4. [GitHUb#136][Patch by Kazuki Tsujimoto]

  * Supported power\_assert 1.0.0. [GitHub#137][Patch by Kazuki Tsujimoto]

  * Added the getting started document.
    [GitHub#139][GitHub#141][Patch by Hiroyuki Sato]

  * Added the document for `attribute`.
    [GitHub#143][Patch by Fumiaki MATSUSHIMA]

  * Improved a link for GitHub. [GitHub#144][Patch by rochefort]

  * Updated `.travis.yml`. [GitHub#145][Patch by Jun Aruga]

### Fixes

  * Fixed a contributor name. [GitHub#131][Patch by Akira Matsuda]

  * Fixed typos in document. [GitHub#132][Patch by Akira Matsuda]

  * Fixed typos in document. [GitHub#134][Patch by Yuji Yaginuma]

  * Fixed a bug that data label with "(" isn't supported.
    [GitHub#135][Reported by Kazuki Tsujimoto]

  * Fixed assertion message in English.
    [GitHub#133][Reported by Khalil Fazal]

  * Fixed a typo in typo fix. [GitHub#138][Patch by kami]

  * Fixed a bug that target location finder may return wrong
    location. [GitHub#146][Patch by Yuki Ito]

  * Fixed a bug that `--no-show-detail-immediately` raises an error.
    [GitHub#147][Reported by MSP-Greg]

### Thanks

  * Akira Matsuda

  * Yuji Yaginuma

  * Kazuki Tsujimoto

  * Khalil Fazal

  * kami

  * Hiroyuki Sato

  * Fumiaki MATSUSHIMA

  * rochefort

  * Jun Aruga

  * Yuki Ito

  * MSP-Greg

(taca)

2017-06-05 14:16:26 UTC MAIN commitmail json YAML

Note update of devel/ruby-simplecov-html package to 0.10.1.

(taca)

2017-06-05 14:15:59 UTC MAIN commitmail json YAML

Update ruby-simplecov-html to 0.10.1.

0.10.1 2017-05-17
========

## Bugfixes

* circumvent a regression that happens in the new JRuby 9.1.9.0 release. See [#53](https://github.com/colszowka/simplecov-html/pull/53) thanks @koic

(taca)

2017-06-05 14:11:20 UTC MAIN commitmail json YAML

Note update of devel/ruby-native-package-installer package to 1.0.4.

(taca)

2017-06-05 14:10:53 UTC MAIN commitmail json YAML

Update ruby-native-package-installer to 1.0.4.

## 1.0.4 - 2017-06-04

### Fixes

  * Fixed MSYS2 detection.

(taca)

2017-06-05 14:09:19 UTC MAIN commitmail json YAML

Note update of Ruby RSpec package family.

devel/ruby-rspec-support
devel/ruby-rspec-core
devel/ruby-rspec-expectations
devel/ruby-rspec-mocks
devel/ruby-rspec
devel/ruby-rspec-rails

(taca)

2017-06-05 14:08:17 UTC MAIN commitmail json YAML

Update ruby-rspec-railsto 3.6.0.

### 3.6.0 / 2017-05-04
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.6.0.beta2...v3.6.0)

Enhancements:

* Add compatibility for Rails 5.1. (Sam Phippen, Yuichiro Kaneko, #1790)

Bug Fixes:

* Fix scaffold generator so that it does not generate broken controller specs
  on Rails 3.x and 4.x. (Yuji Nakayama, #1710)

### 3.6.0.beta2 / 2016-12-12
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.6.0.beta1...v3.6.0.beta2)

Enhancements:

* Improve failure output of ActiveJob matchers by listing queued jobs.
  (Wojciech Wn��trzak, #1722)
* Load `spec_helper.rb` earlier in `rails_helper.rb` by default.
  (Kevin Glowacz, #1795)

### 3.6.0.beta1 / 2016-10-09
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.2...v3.6.0.beta1)

Enhancements:

* Add support for `rake notes` in Rails `>= 5.1`. (John Meehan, #1661)
* Remove `assigns` and `assert_template` from scaffold spec generators (Josh
  Justice, #1689)
* Add support for generating scaffolds for api app specs. (Krzysztof Zych, #1685)

(taca)

2017-06-05 14:07:47 UTC MAIN commitmail json YAML

Update ruby-rspec to 3.6.0.

This is a kind of meta package, contains ruby-rspec-core,
ruby-rspec-expectations and ruby-rspec-mocks.

(taca)

2017-06-05 14:06:11 UTC MAIN commitmail json YAML

Update ruby-rspec-mocks to 3.6.0.

### 3.6.0 / 2017-05-04
[Full Changelog](http://github.com/rspec/rspec-mocks/compare/v3.6.0.beta2...v3.6.0)

Bug Fixes:

* Fix "instance variable @color not initialized" warning when using
  rspec-mocks without rspec-core. (Myron Marston, #1142)
* Restore aliased module methods properly when stubbing on 1.8.7.
  (Samuel Giddins, #1144)
* Allow a message chain expectation to be constrained by argument(s).
  (Jon Rowe, #1156)

### 3.6.0.beta2 / 2016-12-12
[Full Changelog](http://github.com/rspec/rspec-mocks/compare/v3.6.0.beta1...v3.6.0.beta2)

Enhancements:

* Add new `without_partial_double_verification { }` API that lets you
  temporarily turn off partial double verification for an example.
  (Jon Rowe, #1104)

### 3.6.0.beta1 / 2016-10-09
[Full Changelog](http://github.com/rspec/rspec-mocks/compare/v3.5.0...v3.6.0.beta1)

Bug Fixes:

* Return the test double instance form `#freeze` (Alessandro Berardi, #1109)
* Allow the special logic for stubbing `new` to work when `<Class>.method` has
  been redefined. (Proby, #1119)

(taca)

2017-06-05 14:05:45 UTC MAIN commitmail json YAML

Update ruby-rspec-expectations to 3.6.0.

### 3.6.0 / 2017-05-04
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.6.0.beta2...v3.6.0)

Enhancements:

* Treat NoMethodError as a failure for comparison matchers. (Jon Rowe, #972)
* Allow for scoped aliased and negated matchers--just call
  `alias_matcher` or `define_negated_matcher` from within an example
  group. (Markus Reiter, #974)
* Improve failure message of `change` matcher with block and `satisfy` matcher
  by including the block snippet instead of just describing it as `result` or
  `block` when Ripper is available. (Yuji Nakayama, #987)

Bug Fixes:

* Fix `yield_with_args` and `yield_successive_args` matchers so that
  they compare expected to actual args at the time the args are yielded
  instead of at the end, in case the method that is yielding mutates the
  arguments after yielding. (Alyssa Ross, #965)

### 3.6.0.beta2 / 2016-12-12
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.6.0.beta1...v3.6.0.beta2)

Bug Fixes:

* Using the exist matcher on `File` no longer produces a deprecation warning.
  (Jon Rowe, #954)

### 3.6.0.beta1 / 2016-10-09
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.5.0...v3.6.0.beta1)

Bug Fixes:

* Fix `contain_exactly` to work correctly with ranges. (Myron Marston, #940)
* Fix `change` to work correctly with sets. (Marcin Gajewski, #939)

(taca)

2017-06-05 14:05:16 UTC MAIN commitmail json YAML

Update ruby-rspec-core to 3.6.0.

### 3.6.0 / 2017-05-04
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.6.0.beta2...v3.6.0)

Enhancements:

* Add seed information to JSON formatter output. (#2388, Mitsutaka Mimura)
* Include example id in the JSON formatter output. (#2369, Xavier Shay)
* Respect changes to `config.output_stream` after formatters have been
  setup. (#2401, #2419, Ilya Lavrov)

Bug Fixes:

* Delay formatter loading until the last minute to allow accessing the reporter
  without triggering formatter setup. (Jon Rowe, #2243)
* Ensure context hook failures running before an example can access the
  reporter. (Jon Jensen, #2387)
* Multiple fixes to allow using the runner multiple times within the same
  process: `RSpec.clear_examples` resets the formatter and no longer clears
  shared examples, and streams can be used across multiple runs rather than
  being closed after the first. (#2368, Xavier Shay)
* Prevent unexpected `example_group_finished` notifications causing an error.
  (#2396, VTJamie)
* Fix bugs where `config.when_first_matching_example_defined` hooks would fire
  multiple times in some cases. (Yuji Nakayama, #2400)
* Default `last_run_status` to "unknown" when the `status` field in the
  persistence file contains an unrecognized value. (#2360, matrinox)
* Prevent `let` from defining an `initialize` method. (#2414, Jon Rowe)

### 3.6.0.beta2 / 2016-12-12
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.6.0.beta1...v3.6.0.beta2)

Enhancements:

* Include count of errors occurring outside examples in default summaries.
  (#2351, Jon Rowe)
* Warn when including shared example groups recursively. (#2356, Jon Rowe)
* Improve failure snippet syntax highlighting with CodeRay to highlight
  RSpec "keywords" like `expect`. (#2358, Myron Marston)

### 3.6.0.beta1 / 2016-10-09
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.4...v3.6.0.beta1)

Enhancements:

* Warn when duplicate shared examples definitions are loaded due to being
  defined in files matching the spec pattern (e.g. `_spec.rb`) (#2278, Devon Estes)
* Improve metadata filtering so that it can match against any object
  that implements `===` instead of treating regular expressions as
  special. (Myron Marston, #2294)
* Improve `rspec -v` so that it prints out the versions of each part of
  RSpec to prevent confusion. (Myron Marston, #2304)
* Add `config.fail_if_no_examples` option which causes RSpec to fail if
  no examples are found. (Ewa Czechowska, #2302)
* Nicely format errors encountered while loading spec files.
  (Myron Marston, #2323)
* Improve the API for enabling and disabling color output (Josh
  Justice, #2321):
  * Automatically enable color if the output is a TTY, since color is
    nearly always desirable if the output can handle it.
  * Introduce new CLI flag to force color on (`--force-color`), even
    if the output is not a TTY. `--no-color` continues to work as well.
  * Introduce `config.color_mode` for configuring the color from Ruby.
    `:automatic` is the default and will produce color if the output is
    a TTY. `:on` forces it on and `:off` forces it off.

(taca)

2017-06-05 14:04:42 UTC MAIN commitmail json YAML

Update ruby-rspec-support to 3.6.0.

### 3.6.0 / 2017-05-04
[Full Changelog](http://github.com/rspec/rspec-support/compare/v3.6.0.beta2...3.6.0)

Enhancements:

* Import `Source` classes from rspec-core. (Yuji Nakayama, #315)

### 3.6.0.beta2 / 2016-12-12
[Full Changelog](http://github.com/rspec/rspec-support/compare/v3.6.0.beta1...v3.6.0.beta2)

No user-facing changes.

### 3.6.0.beta1 / 2016-10-09
[Full Changelog](http://github.com/rspec/rspec-support/compare/v3.5.0...v3.6.0.beta1)

Bug Fixes:

* Prevent truncated formatted object output from mangling console codes. (#294, Anson Kelly)

(taca)

2017-06-05 13:43:44 UTC MAIN commitmail json YAML

Updated lang/perl5 to 5.26.0

(ryoon)

2017-06-05 13:41:22 UTC MAIN commitmail json YAML

Update to 5.26.0

Changelog:
NAME
      perldelta - what is new for perl v5.26.0

DESCRIPTION
      This document describes the differences between the 5.24.0 release and
      the 5.26.0 release.

Notice
      This release includes three updates with widespread effects:

      o  "." no longer in @INC

          For security reasons, the current directory (".") is no longer
          included by default at the end of the module search path (@INC).
          This may have widespread implications for the building, testing and
          installing of modules, and for the execution of scripts.  See the
          section "Removal of the current directory (".") from @INC" for the
          full details.

      o  "do" may now warn

          "do" now gives a deprecation warning when it fails to load a file
          which it would have loaded had "." been in @INC.

      o  In regular expression patterns, a literal left brace "{" should be
          escaped

          See "Unescaped literal "{" characters in regular expression
          patterns are no longer permissible".

Core Enhancements
  Lexical subroutines are no longer experimental
      Using the "lexical_subs" feature introduced in v5.18 no longer emits a
      warning.  Existing code that disables the "experimental::lexical_subs"
      warning category that the feature previously used will continue to
      work.  The "lexical_subs" feature has no effect; all Perl code can use
      lexical subroutines, regardless of what feature declarations are in
      scope.

  Indented Here-documents
      This adds a new modifier "~" to here-docs that tells the parser that it
      should look for "/^\s*$DELIM\n/" as the closing delimiter.

      These syntaxes are all supported:

          <<~EOF;
          <<~\EOF;
          <<~'EOF';
          <<~"EOF";
          <<~`EOF`;
          <<~ 'EOF';
          <<~ "EOF";
          <<~ `EOF`;

      The "~" modifier will strip, from each line in the here-doc, the same
      whitespace that appears before the delimiter.

      Newlines will be copied as-is, and lines that don't include the proper
      beginning whitespace will cause perl to croak.

      For example:

          if (1) {
            print <<~EOF;
              Hello there
              EOF
          }

      prints "Hello there\n" with no leading whitespace.

  New regular expression modifier "/xx"
      Specifying two "x" characters to modify a regular expression pattern
      does everything that a single one does, but additionally TAB and SPACE
      characters within a bracketed character class are generally ignored and
      can be added to improve readability, like "/[ ^ A-Z d-f p-x ]/xx".
      Details are at "/x and /xx" in perlre.

  "@{^CAPTURE}", "%{^CAPTURE}", and "%{^CAPTURE_ALL}"
      "@{^CAPTURE}" exposes the capture buffers of the last match as an
      array.  So $1 is "${^CAPTURE}[0]".  This is a more efficient equivalent
      to code like "substr($matched_string,$-[0],$+[0]-$-[0])", and you don't
      have to keep track of the $matched_string either.  This variable has no
      single character equivalent.  Note that, like the other regex magic
      variables, the contents of this variable is dynamic; if you wish to
      store it beyond the lifetime of the match you must copy it to another
      array.

      "%{^CAPTURE}" is equivalent to "%+" (i.e., named captures).  Other than
      being more self-documenting there is no difference between the two
      forms.

      "%{^CAPTURE_ALL}" is equivalent to "%-" (i.e., all named captures).
      Other than being more self-documenting there is no difference between
      the two forms.

  Declaring a reference to a variable
      As an experimental feature, Perl now allows the referencing operator to
      come after "my()", "state()", "our()", or "local()".  This syntax must
      be enabled with "use feature 'declared_refs'".  It is experimental, and
      will warn by default unless "no warnings 'experimental::refaliasing'"
      is in effect.  It is intended mainly for use in assignments to
      references.  For example:

          use experimental 'refaliasing', 'declared_refs';
          my \$a = \$b;

      See "Assigning to References" in perlref for more details.

  Unicode 9.0 is now supported
      A list of changes is at
      <http://www.unicode.org/versions/Unicode9.0.0/>.  Modules that are
      shipped with core Perl but not maintained by p5p do not necessarily
      support Unicode 9.0.  Unicode::Normalize does work on 9.0.

  Use of "\p{script}" uses the improved Script_Extensions property
      Unicode 6.0 introduced an improved form of the Script ("sc") property,
      and called it Script_Extensions ("scx").  Perl now uses this improved
      version when a property is specified as just "\p{script}".  This should
      make programs more accurate when determining if a character is used in
      a given script, but there is a slight chance of breakage for programs
      that very specifically needed the old behavior.  The meaning of
      compound forms, like "\p{sc=script}" are unchanged.  See "Scripts" in
      perlunicode.

  Perl can now do default collation in UTF-8 locales on platforms that
      support it
      Some platforms natively do a reasonable job of collating and sorting in
      UTF-8 locales.  Perl now works with those.  For portability and full
      control, Unicode::Collate is still recommended, but now you may not
      need to do anything special to get good-enough results, depending on
      your application.  See "Category "LC_COLLATE": Collation: Text
      Comparisons and Sorting" in perllocale.

  Better locale collation of strings containing embedded "NUL" characters
      In locales that have multi-level character weights, "NUL"s are now
      ignored at the higher priority ones.  There are still some gotchas in
      some strings, though.  See "Collation of strings containing embedded
      "NUL" characters" in perllocale.

  "CORE" subroutines for hash and array functions callable via reference
      The hash and array functions in the "CORE" namespace ("keys", "each",
      "values", "push", "pop", "shift", "unshift" and "splice") can now be
      called with ampersand syntax ("&CORE::keys(\%hash") and via reference
      ("my $k = \&CORE::keys; $k->(\%hash)").  Previously they could only be
      used when inlined.

  New Hash Function For 64-bit Builds
      We have switched to a hybrid hash function to better balance
      performance for short and long keys.

      For short keys, 16 bytes and under, we use an optimised variant of One
      At A Time Hard, and for longer keys we use Siphash 1-3.  For very long
      keys this is a big improvement in performance.  For shorter keys there
      is a modest improvement.

Security
  Removal of the current directory (".") from @INC
      The perl binary includes a default set of paths in @INC.  Historically
      it has also included the current directory (".") as the final entry,
      unless run with taint mode enabled ("perl -T").  While convenient, this
      has security implications: for example, where a script attempts to load
      an optional module when its current directory is untrusted (such as
      /tmp), it could load and execute code from under that directory.

      Starting with v5.26, "." is always removed by default, not just under
      tainting.  This has major implications for installing modules and
      executing scripts.

      The following new features have been added to help ameliorate these
      issues.

      o  Configure -Udefault_inc_excludes_dot

          There is a new Configure option, "default_inc_excludes_dot"
          (enabled by default) which builds a perl executable without ".";
          unsetting this option using "-U" reverts perl to the old behaviour.
          This may fix your path issues but will reintroduce all the security
          concerns, so don't build a perl executable like this unless you're
          really confident that such issues are not a concern in your
          environment.

      o  "PERL_USE_UNSAFE_INC"

          There is a new environment variable recognised by the perl
          interpreter.  If this variable has the value 1 when the perl
          interpreter starts up, then "." will be automatically appended to
          @INC (except under tainting).

          This allows you restore the old perl interpreter behaviour on a
          case-by-case basis.  But note that this is intended to be a
          temporary crutch, and this feature will likely be removed in some
          future perl version.  It is currently set by the "cpan" utility and
          "Test::Harness" to ease installation of CPAN modules which have not
          been updated to handle the lack of dot.  Once again, don't use this
          unless you are sure that this will not reintroduce any security
          concerns.

      o  A new deprecation warning issued by "do".

          While it is well-known that "use" and "require" use @INC to search
          for the file to load, many people don't realise that "do "file""
          also searches @INC if the file is a relative path.  With the
          removal of ".", a simple "do "file.pl"" will fail to read in and
          execute "file.pl" from the current directory.  Since this is
          commonly expected behaviour, a new deprecation warning is now
          issued whenever "do" fails to load a file which it otherwise would
          have found if a dot had been in @INC.

      Here are some things script and module authors may need to do to make
      their software work in the new regime.

      o  Script authors

          If the issue is within your own code (rather than within included
          modules), then you have two main options.  Firstly, if you are
          confident that your script will only be run within a trusted
          directory (under which you expect to find trusted files and
          modules), then add "." back into the path; e.g.:

              BEGIN {
                  my $dir = "/some/trusted/directory";
                  chdir $dir or die "Can't chdir to $dir: $!\n";
                  # safe now
                  push @INC, '.';
              }

              use "Foo::Bar"; # may load /some/trusted/directory/Foo/Bar.pm
              do "config.pl"; # may load /some/trusted/directory/config.pl

          On the other hand, if your script is intended to be run from within
          untrusted directories (such as /tmp), then your script suddenly
          failing to load files may be indicative of a security issue.  You
          most likely want to replace any relative paths with full paths; for
          example,

              do "foo_config.pl"

          might become

              do "$ENV{HOME}/foo_config.pl"

          If you are absolutely certain that you want your script to load and
          execute a file from the current directory, then use a "./" prefix;
          for example:

              do "./foo_config.pl"

      o  Installing and using CPAN modules

          If you install a CPAN module using an automatic tool like "cpan",
          then this tool will itself set the "PERL_USE_UNSAFE_INC"
          environment variable while building and testing the module, which
          may be sufficient to install a distribution which hasn't been
          updated to be dot-aware.  If you want to install such a module
          manually, then you'll need to replace the traditional invocation:

              perl Makefile.PL && make && make test && make install

          with something like

              (export PERL_USE_UNSAFE_INC=1; \
                perl Makefile.PL && make && make test && make install)

          Note that this only helps build and install an unfixed module.
          It's possible for the tests to pass (since they were run under
          "PERL_USE_UNSAFE_INC=1"), but for the module itself to fail to
          perform correctly in production.  In this case, you may have to
          temporarily modify your script until a fixed version of the module
          is released.  For example:

              use Foo::Bar;
              {
                  local @INC = (@INC, '.');
                  # assuming read_config() needs '.' in @INC
                  $config = Foo::Bar->read_config();
              }

          This is only rarely expected to be necessary.  Again, if doing
          this, assess the resultant risks first.

      o  Module Authors

          If you maintain a CPAN distribution, it may need updating to run in
          a dotless environment.  Although "cpan" and other such tools will
          currently set the "PERL_USE_UNSAFE_INC" during module build, this
          is a temporary workaround for the set of modules which rely on "."
          being in @INC for installation and testing, and this may mask
          deeper issues.  It could result in a module which passes tests and
          installs, but which fails at run time.

          During build, test, and install, it will normally be the case that
          any perl processes will be executing directly within the root
          directory of the untarred distribution, or a known subdirectory of
          that, such as t/.  It may well be that Makefile.PL or t/foo.t will
          attempt to include local modules and configuration files using
          their direct relative filenames, which will now fail.

          However, as described above, automatic tools like cpan will (for
          now) set the "PERL_USE_UNSAFE_INC" environment variable, which
          introduces dot during a build.

          This makes it likely that your existing build and test code will
          work, but this may mask issues with your code which only manifest
          when used after install.  It is prudent to try and run your build
          process with that variable explicitly disabled:

              (export PERL_USE_UNSAFE_INC=0; \
                perl Makefile.PL && make && make test && make install)

          This is more likely to show up any potential problems with your
          module's build process, or even with the module itself.  Fixing
          such issues will ensure both that your module can again be
          installed manually, and that it will still build once the
          "PERL_USE_UNSAFE_INC" crutch goes away.

          When fixing issues in tests due to the removal of dot from @INC,
          reinsertion of dot into @INC should be performed with caution, for
          this too may suppress real errors in your runtime code.  You are
          encouraged wherever possible to apply the aforementioned approaches
          with explicit absolute/relative paths, or to relocate your needed
          files into a subdirectory and insert that subdirectory into @INC
          instead.

          If your runtime code has problems under the dotless @INC, then the
          comments above on how to fix for script authors will mostly apply
          here too.  Bear in mind though that it is considered bad form for a
          module to globally add a dot to @INC, since it introduces both a
          security risk and hides issues of accidentally requiring dot in
          @INC, as explained above.

  Escaped colons and relative paths in PATH
      On Unix systems, Perl treats any relative paths in the "PATH"
      environment variable as tainted when starting a new process.
      Previously, it was allowing a backslash to escape a colon (unlike the
      OS), consequently allowing relative paths to be considered safe if the
      PATH was set to something like "/\:.".  The check has been fixed to
      treat "." as tainted in that example.

  New "-Di" switch is now required for PerlIO debugging output
      This is used for debugging of code within PerlIO to avoid recursive
      calls.  Previously this output would be sent to the file specified by
      the "PERLIO_DEBUG" environment variable if perl wasn't running setuid
      and the "-T" or "-t" switches hadn't been parsed yet.

      If perl performed output at a point where it hadn't yet parsed its
      switches this could result in perl creating or overwriting the file
      named by "PERLIO_DEBUG" even when the "-T" switch had been supplied.

      Perl now requires the "-Di" switch to be present before it will produce
      PerlIO debugging output.  By default this is written to "stderr", but
      can optionally be redirected to a file by setting the "PERLIO_DEBUG"
      environment variable.

      If perl is running setuid or the "-T" switch was supplied,
      "PERLIO_DEBUG" is ignored and the debugging output is sent to "stderr"
      as for any other "-D" switch.

Incompatible Changes
  Unescaped literal "{" characters in regular expression patterns are no
      longer permissible
      You have to now say something like "\{" or "[{]" to specify to match a
      LEFT CURLY BRACKET; otherwise, it is a fatal pattern compilation error.
      This change will allow future extensions to the language.

      These have been deprecated since v5.16, with a deprecation message
      raised for some uses starting in v5.22.  Unfortunately, the code added
      to raise the message was buggy and failed to warn in some cases where
      it should have.  Therefore, enforcement of this ban for these cases is
      deferred until Perl 5.30, but the code has been fixed to raise a
      default-on deprecation message for them in the meantime.

      Some uses of literal "{" occur in contexts where we do not foresee the
      meaning ever being anything but the literal, such as the very first
      character in the pattern, or after a "|" meaning alternation.  Thus

        qr/{fee|{fie/

      matches either of the strings "{fee" or "{fie".  To avoid forcing
      unnecessary code changes, these uses do not need to be escaped, and no
      warning is raised about them, and there are no current plans to change
      this.

      But it is always correct to escape "{", and the simple rule to remember
      is to always do so.

      See Unescaped left brace in regex is illegal here.

  "scalar(%hash)" return signature changed
      The value returned for "scalar(%hash)" will no longer show information
      about the buckets allocated in the hash.  It will simply return the
      count of used keys.  It is thus equivalent to "0+keys(%hash)".

      A form of backward compatibility is provided via
      "Hash::Util::bucket_ratio()" which provides the same behavior as
      "scalar(%hash)" provided in Perl 5.24 and earlier.

  "keys" returned from an lvalue subroutine
      "keys" returned from an lvalue subroutine can no longer be assigned to
      in list context.

          sub foo : lvalue { keys(%INC) }
          (foo) = 3; # death
          sub bar : lvalue { keys(@_) }
          (bar) = 3; # also an error

      This makes the lvalue sub case consistent with "(keys %hash) = ..." and
      "(keys @_) = ...", which are also errors.  [perl #128187]
      <https://rt.perl.org/Public/Bug/Display.html?id=128187>

  The "${^ENCODING}" facility has been removed
      The special behaviour associated with assigning a value to this
      variable has been removed.  As a consequence, the encoding pragma's
      default mode is no longer supported.  If you still need to write your
      source code in encodings other than UTF-8, use a source filter such as
      Filter::Encoding on CPAN or encoding's "Filter" option.

  "POSIX::tmpnam()" has been removed
      The fundamentally unsafe "tmpnam()" interface was deprecated in Perl
      5.22 and has now been removed.  In its place, you can use, for example,
      the File::Temp interfaces.

  require ::Foo::Bar is now illegal.
      Formerly, "require ::Foo::Bar" would try to read /Foo/Bar.pm.  Now any
      bareword require which starts with a double colon dies instead.

  Literal control character variable names are no longer permissible
      A variable name may no longer contain a literal control character under
      any circumstances.  These previously were allowed in single-character
      names on ASCII platforms, but have been deprecated there since Perl
      5.20.  This affects things like "$\cT", where \cT is a literal control
      (such as a "NAK" or "NEGATIVE ACKNOWLEDGE" character) in the source
      code.

  "NBSP" is no longer permissible in "\N{...}"
      The name of a character may no longer contain non-breaking spaces.  It
      has been deprecated to do so since Perl 5.22.

Deprecations
  String delimiters that aren't stand-alone graphemes are now deprecated
      For Perl to eventually allow string delimiters to be Unicode grapheme
      clusters (which look like a single character, but may be a sequence of
      several ones), we have to stop allowing a single character delimiter
      that isn't a grapheme by itself.  These are unlikely to exist in actual
      code, as they would typically display as attached to the character in
      front of them.

  "\cX" that maps to a printable is no longer deprecated
      This means we have no plans to remove this feature.  It still raises a
      warning, but only if syntax warnings are enabled.  The feature was
      originally intended to be a way to express non-printable characters
      that don't have a mnemonic ("\t" and "\n" are mnemonics for two non-
      printable characters, but most non-printables don't have a mnemonic.)
      But the feature can be used to specify a few printable characters,
      though those are more clearly expressed as the printable itself.  See
      <http://www.nntp.perl.org/group/perl.perl5.porters/2017/02/msg242944.html>.

Performance Enhancements
      o  A hash in boolean context is now sometimes faster, e.g.

              if (!%h) { ... }

          This was already special-cased, but some cases were missed (such as
          "grep %$_, @AoH"), and even the ones which weren't have been
          improved.

      o  New Faster Hash Function on 64 bit builds

          We use a different hash function for short and long keys.  This
          should improve performance and security, especially for long keys.

      o  readline is faster

          Reading from a file line-by-line with "readline()" or "<>" should
          now typically be faster due to a better implementation of the code
          that searches for the next newline character.

      o  Assigning one reference to another, e.g. "$ref1 = $ref2" has been
          optimized in some cases.

      o  Remove some exceptions to creating Copy-on-Write strings. The
          string buffer growth algorithm has been slightly altered so that
          you're less likely to encounter a string which can't be COWed.

      o  Better optimise array and hash assignment: where an array or hash
          appears in the LHS of a list assignment, such as "(..., @a) =
          (...);", it's likely to be considerably faster, especially if it
          involves emptying the array/hash. For example, this code runs about
          a third faster compared to Perl 5.24.0:

              my @a;
              for my $i (1..10_000_000) {
                  @a = (1,2,3);
                  @a = ();
              }

      o  Converting a single-digit string to a number is now substantially
          faster.

      o  The "split" builtin is now slightly faster in many cases: in
          particular for the two specially-handled forms

              my    @a = split ...;
              local @a = split ...;

      o  The rather slow implementation for the experimental subroutine
          signatures feature has been made much faster; it is now comparable
          in speed with the traditional "my ($a, $b, @c) = @_".

      o  Bareword constant strings are now permitted to take part in
          constant folding.  They were originally exempted from constant
          folding in August 1999, during the development of Perl 5.6, to
          ensure that "use strict "subs"" would still apply to bareword
          constants.  That has now been accomplished a different way, so
          barewords, like other constants, now gain the performance benefits
          of constant folding.

          This also means that void-context warnings on constant expressions
          of barewords now report the folded constant operand, rather than
          the operation; this matches the behaviour for non-bareword
          constants.

Modules and Pragmata
  Updated Modules and Pragmata
      o  IO::Compress has been upgraded from version 2.069 to 2.074.

      o  Archive::Tar has been upgraded from version 2.04 to 2.24.

      o  arybase has been upgraded from version 0.11 to 0.12.

      o  attributes has been upgraded from version 0.27 to 0.29.

          The deprecation message for the ":unique" and ":locked" attributes
          now mention that they will disappear in Perl 5.28.

      o  B has been upgraded from version 1.62 to 1.68.

      o  B::Concise has been upgraded from version 0.996 to 0.999.

          Its output is now more descriptive for "op_private" flags.

      o  B::Debug has been upgraded from version 1.23 to 1.24.

      o  B::Deparse has been upgraded from version 1.37 to 1.40.

      o  B::Xref has been upgraded from version 1.05 to 1.06.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  base has been upgraded from version 2.23 to 2.25.

      o  bignum has been upgraded from version 0.42 to 0.47.

      o  Carp has been upgraded from version 1.40 to 1.42.

      o  charnames has been upgraded from version 1.43 to 1.44.

      o  Compress::Raw::Bzip2 has been upgraded from version 2.069 to 2.074.

      o  Compress::Raw::Zlib has been upgraded from version 2.069 to 2.074.

      o  Config::Perl::V has been upgraded from version 0.25 to 0.28.

      o  CPAN has been upgraded from version 2.11 to 2.18.

      o  CPAN::Meta has been upgraded from version 2.150005 to 2.150010.

      o  Data::Dumper has been upgraded from version 2.160 to 2.167.

          The XS implementation now supports Deparse.

      o  DB_File has been upgraded from version 1.835 to 1.840.

      o  Devel::Peek has been upgraded from version 1.23 to 1.26.

      o  Devel::PPPort has been upgraded from version 3.32 to 3.35.

      o  Devel::SelfStubber has been upgraded from version 1.05 to 1.06.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  diagnostics has been upgraded from version 1.34 to 1.36.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  Digest has been upgraded from version 1.17 to 1.17_01.

      o  Digest::MD5 has been upgraded from version 2.54 to 2.55.

      o  Digest::SHA has been upgraded from version 5.95 to 5.96.

      o  DynaLoader has been upgraded from version 1.38 to 1.42.

      o  Encode has been upgraded from version 2.80 to 2.88.

      o  encoding has been upgraded from version 2.17 to 2.19.

          This module's default mode is no longer supported.  It now dies
          when imported, unless the "Filter" option is being used.

      o  encoding::warnings has been upgraded from version 0.12 to 0.13.

          This module is no longer supported.  It emits a warning to that
          effect and then does nothing.

      o  Errno has been upgraded from version 1.25 to 1.28.

          It now documents that using "%!" automatically loads Errno for you.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  ExtUtils::Embed has been upgraded from version 1.33 to 1.34.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  ExtUtils::MakeMaker has been upgraded from version 7.10_01 to 7.24.

      o  ExtUtils::Miniperl has been upgraded from version 1.05 to 1.06.

      o  ExtUtils::ParseXS has been upgraded from version 3.31 to 3.34.

      o  ExtUtils::Typemaps has been upgraded from version 3.31 to 3.34.

      o  feature has been upgraded from version 1.42 to 1.47.

      o  File::Copy has been upgraded from version 2.31 to 2.32.

      o  File::Fetch has been upgraded from version 0.48 to 0.52.

      o  File::Glob has been upgraded from version 1.26 to 1.28.

          It now Issues a deprecation message for "File::Glob::glob()".

      o  File::Spec has been upgraded from version 3.63 to 3.67.

      o  FileHandle has been upgraded from version 2.02 to 2.03.

      o  Filter::Simple has been upgraded from version 0.92 to 0.93.

          It no longer treats "no MyFilter" immediately following "use
          MyFilter" as end-of-file.  [perl #107726]
          <https://rt.perl.org/Public/Bug/Display.html?id=107726>

      o  Getopt::Long has been upgraded from version 2.48 to 2.49.

      o  Getopt::Std has been upgraded from version 1.11 to 1.12.

      o  Hash::Util has been upgraded from version 0.19 to 0.22.

      o  HTTP::Tiny has been upgraded from version 0.056 to 0.070.

          Internal 599-series errors now include the redirect history.

      o  I18N::LangTags has been upgraded from version 0.40 to 0.42.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  IO has been upgraded from version 1.36 to 1.38.

      o  IO::Socket::IP has been upgraded from version 0.37 to 0.38.

      o  IPC::Cmd has been upgraded from version 0.92 to 0.96.

      o  IPC::SysV has been upgraded from version 2.06_01 to 2.07.

      o  JSON::PP has been upgraded from version 2.27300 to 2.27400_02.

      o  lib has been upgraded from version 0.63 to 0.64.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  List::Util has been upgraded from version 1.42_02 to 1.46_02.

      o  Locale::Codes has been upgraded from version 3.37 to 3.42.

      o  Locale::Maketext has been upgraded from version 1.26 to 1.28.

      o  Locale::Maketext::Simple has been upgraded from version 0.21 to
          0.21_01.

      o  Math::BigInt has been upgraded from version 1.999715 to 1.999806.

      o  Math::BigInt::FastCalc has been upgraded from version 0.40 to
          0.5005.

      o  Math::BigRat has been upgraded from version 0.260802 to 0.2611.

      o  Math::Complex has been upgraded from version 1.59 to 1.5901.

      o  Memoize has been upgraded from version 1.03 to 1.03_01.

      o  Module::CoreList has been upgraded from version 5.20170420 to
          5.20170530.

      o  Module::Load::Conditional has been upgraded from version 0.64 to
          0.68.

      o  Module::Metadata has been upgraded from version 1.000031 to
          1.000033.

      o  mro has been upgraded from version 1.18 to 1.20.

      o  Net::Ping has been upgraded from version 2.43 to 2.55.

          IPv6 addresses and "AF_INET6" sockets are now supported, along with
          several other enhancements.

      o  NEXT has been upgraded from version 0.65 to 0.67.

      o  Opcode has been upgraded from version 1.34 to 1.39.

      o  open has been upgraded from version 1.10 to 1.11.

      o  OS2::Process has been upgraded from version 1.11 to 1.12.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  overload has been upgraded from version 1.26 to 1.28.

          Its compilation speed has been improved slightly.

      o  parent has been upgraded from version 0.234 to 0.236.

      o  perl5db.pl has been upgraded from version 1.50 to 1.51.

          It now ignores /dev/tty on non-Unix systems.  [perl #113960]
          <https://rt.perl.org/Public/Bug/Display.html?id=113960>

      o  Perl::OSType has been upgraded from version 1.009 to 1.010.

      o  perlfaq has been upgraded from version 5.021010 to 5.021011.

      o  PerlIO has been upgraded from version 1.09 to 1.10.

      o  PerlIO::encoding has been upgraded from version 0.24 to 0.25.

      o  PerlIO::scalar has been upgraded from version 0.24 to 0.26.

      o  Pod::Checker has been upgraded from version 1.60 to 1.73.

      o  Pod::Functions has been upgraded from version 1.10 to 1.11.

      o  Pod::Html has been upgraded from version 1.22 to 1.2202.

      o  Pod::Perldoc has been upgraded from version 3.25_02 to 3.28.

      o  Pod::Simple has been upgraded from version 3.32 to 3.35.

      o  Pod::Usage has been upgraded from version 1.68 to 1.69.

      o  POSIX has been upgraded from version 1.65 to 1.76.

          This remedies several defects in making its symbols exportable.
          [perl #127821]
          <https://rt.perl.org/Public/Bug/Display.html?id=127821>

          The "POSIX::tmpnam()" interface has been removed, see
          "POSIX::tmpnam() has been removed".

          The following deprecated functions have been removed:

              POSIX::isalnum
              POSIX::isalpha
              POSIX::iscntrl
              POSIX::isdigit
              POSIX::isgraph
              POSIX::islower
              POSIX::isprint
              POSIX::ispunct
              POSIX::isspace
              POSIX::isupper
              POSIX::isxdigit
              POSIX::tolower
              POSIX::toupper

          Trying to import POSIX subs that have no real implementations (like
          "POSIX::atend()") now fails at import time, instead of waiting
          until runtime.

      o  re has been upgraded from version 0.32 to 0.34

          This adds support for the new "/xx" regular expression pattern
          modifier, and a change to the "use re 'strict'" experimental
          feature.  When "re 'strict'" is enabled, a warning now will be
          generated for all unescaped uses of the two characters "}" and "]"
          in regular expression patterns (outside bracketed character
          classes) that are taken literally.  This brings them more in line
          with the ")" character which is always a metacharacter unless
          escaped.  Being a metacharacter only sometimes, depending on an
          action at a distance, can lead to silently having the pattern mean
          something quite different than was intended, which the
          "re 'strict'" mode is intended to minimize.

      o  Safe has been upgraded from version 2.39 to 2.40.

      o  Scalar::Util has been upgraded from version 1.42_02 to 1.46_02.

      o  Storable has been upgraded from version 2.56 to 2.62.

          Fixes [perl #130098]
          <https://rt.perl.org/Public/Bug/Display.html?id=130098>.

      o  Symbol has been upgraded from version 1.07 to 1.08.

      o  Sys::Syslog has been upgraded from version 0.33 to 0.35.

      o  Term::ANSIColor has been upgraded from version 4.04 to 4.06.

      o  Term::ReadLine has been upgraded from version 1.15 to 1.16.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  Test has been upgraded from version 1.28 to 1.30.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  Test::Harness has been upgraded from version 3.36 to 3.38.

      o  Test::Simple has been upgraded from version 1.001014 to 1.302073.

      o  Thread::Queue has been upgraded from version 3.09 to 3.12.

      o  Thread::Semaphore has been upgraded from 2.12 to 2.13.

          Added the "down_timed" method.

      o  threads has been upgraded from version 2.07 to 2.15.

      o  threads::shared has been upgraded from version 1.51 to 1.56.

      o  Tie::Hash::NamedCapture has been upgraded from version 0.09 to
          0.10.

      o  Time::HiRes has been upgraded from version 1.9733 to 1.9741.

          It now builds on systems with C++11 compilers (such as G++ 6 and
          Clang++ 3.9).

          Now uses "clockid_t".

      o  Time::Local has been upgraded from version 1.2300 to 1.25.

      o  Unicode::Collate has been upgraded from version 1.14 to 1.19.

      o  Unicode::UCD has been upgraded from version 0.64 to 0.68.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  version has been upgraded from version 0.9916 to 0.9917.

      o  VMS::DCLsym has been upgraded from version 1.06 to 1.08.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

      o  warnings has been upgraded from version 1.36 to 1.37.

      o  XS::Typemap has been upgraded from version 0.14 to 0.15.

      o  XSLoader has been upgraded from version 0.21 to 0.27.

          Fixed a security hole in which binary files could be loaded from a
          path outside of @INC.

          It now uses 3-arg "open()" instead of 2-arg "open()".  [perl
          #130122] <https://rt.perl.org/Public/Bug/Display.html?id=130122>

Documentation
  New Documentation
      perldeprecation

      This file documents all upcoming deprecations, and some of the
      deprecations which already have been removed.  The purpose of this
      documentation is two-fold: document what will disappear, and by which
      version, and serve as a guide for people dealing with code which has
      features that no longer work after an upgrade of their perl.

  Changes to Existing Documentation
      We have attempted to update the documentation to reflect the changes
      listed in this document.  If you find any we have missed, send email to
      perlbug@perl.org <mailto:perlbug@perl.org>.

      Additionally, all references to Usenet have been removed, and the
      following selected changes have been made:

      perlfunc

      o  Removed obsolete text about "defined()" on aggregates that should
          have been deleted earlier, when the feature was removed.

      o  Corrected documentation of "eval()", and "evalbytes()".

      o  Clarified documentation of "seek()", "tell()" and "sysseek()"
          emphasizing that positions are in bytes and not characters.  [perl
          #128607] <https://rt.perl.org/Public/Bug/Display.html?id=128607>

      o  Clarified documentation of "sort()" concerning the variables $a and
          $b.

      o  In "split()" noted that certain pattern modifiers are legal, and
          added a caution about its use in Perls before v5.11.

      o  Removed obsolete documentation of "study()", noting that it is now
          a no-op.

      o  Noted that "vec()" doesn't work well when the string contains
          characters whose code points are above 255.

      perlguts

      o  Added advice on formatted printing of operands of "Size_t" and
          "SSize_t"

      perlhack

      o  Clarify what editor tab stop rules to use, and note that we are
          migrating away from using tabs, replacing them with sequences of
          SPACE characters.

      perlhacktips

      o  Give another reason to use "cBOOL" to cast an expression to
          boolean.

      o  Note that the macros "TRUE" and "FALSE" are available to express
          boolean values.

      perlinterp

      o  perlinterp has been expanded to give a more detailed example of how
          to hunt around in the parser for how a given operator is handled.

      perllocale

      o  Some locales aren't compatible with Perl.  Note that these can
          cause core dumps.

      perlmod

      o  Various clarifications have been added.

      perlmodlib

      o  Updated the site mirror list.

      perlobj

      o  Added a section on calling methods using their fully qualified
          names.

      o  Do not discourage manual @ISA.

      perlootut

      o  Mention "Moo" more.

      perlop

      o  Note that white space must be used for quoting operators if the
          delimiter is a word character (i.e., matches "\w").

      o  Clarify that in regular expression patterns delimited by single
          quotes, no variable interpolation is done.

      perlre

      o  The first part was extensively rewritten to incorporate various
          basic points, that in earlier versions were mentioned in sort of an
          appendix on Version 8 regular expressions.

      o  Note that it is common to have the "/x" modifier and forget that
          this means that "#" has to be escaped.

      perlretut

      o  Add introductory material.

      o  Note that a metacharacter occurring in a context where it can't
          mean that, silently loses its meta-ness and matches literally.
          "use re 'strict'" can catch some of these.

      perlunicode

      o  Corrected the text about Unicode BYTE ORDER MARK handling.

      o  Updated the text to correspond with changes in Unicode UTS#18,
          concerning regular expressions, and Perl compatibility with what it
          says.

      perlvar

      o  Document @ISA.  It was documented in other places, but not in
          perlvar.

Diagnostics
  New Diagnostics
      New Errors

      o  A signature parameter must start with '$', '@' or '%'

      o  Bareword in require contains "%s"

      o  Bareword in require maps to empty filename

      o  Bareword in require maps to disallowed filename "%s"

      o  Bareword in require must not start with a double-colon: "%s"

      o  %s: command not found

          (A) You've accidentally run your script through bash or another
          shell instead of Perl.  Check the "#!" line, or manually feed your
          script into Perl yourself.  The "#!" line at the top of your file
          could look like:

            #!/usr/bin/perl

      o  %s: command not found: %s

          (A) You've accidentally run your script through zsh or another
          shell instead of Perl.  Check the "#!" line, or manually feed your
          script into Perl yourself.  The "#!" line at the top of your file
          could look like:

            #!/usr/bin/perl

      o  The experimental declared_refs feature is not enabled

          (F) To declare references to variables, as in "my \%x", you must
          first enable the feature:

              no warnings "experimental::declared_refs";
              use feature "declared_refs";

          See "Declaring a reference to a variable".

      o  Illegal character following sigil in a subroutine signature

      o  Indentation on line %d of here-doc doesn't match delimiter

      o  Infinite recursion via empty pattern.

          Using the empty pattern (which re-executes the last successfully-
          matched pattern) inside a code block in another regex, as in "/(?{
          s!!new! })/", has always previously yielded a segfault.  It now
          produces this error.

      o  Malformed UTF-8 string in "%s"

      o  Multiple slurpy parameters not allowed

      o  '#' not allowed immediately following a sigil in a subroutine
          signature

      o  panic: unknown OA_*: %x

      o  Unescaped left brace in regex is illegal here

          Unescaped left braces are now illegal in some contexts in regular
          expression patterns.  In other contexts, they are still just
          deprecated; they will be illegal in Perl 5.30.

      o  Version control conflict marker

          (F) The parser found a line starting with "<<<<<<<", ">>>>>>>", or
          "=======".  These may be left by a version control system to mark
          conflicts after a failed merge operation.

      New Warnings

      o  Can't determine class of operator %s, assuming "BASEOP"

      o  Declaring references is experimental

          (S experimental::declared_refs) This warning is emitted if you use
          a reference constructor on the right-hand side of "my()",
          "state()", "our()", or "local()".  Simply suppress the warning if
          you want to use the feature, but know that in doing so you are
          taking the risk of using an experimental feature which may change
          or be removed in a future Perl version:

              no warnings "experimental::declared_refs";
              use feature "declared_refs";
              $fooref = my \$foo;

          See "Declaring a reference to a variable".

      o  do "%s" failed, '.' is no longer in @INC

          Since "." is now removed from @INC by default, "do" will now
          trigger a warning recommending to fix the "do" statement.

      o  "File::Glob::glob()" will disappear in perl 5.30. Use
          "File::Glob::bsd_glob()" instead.

      o  Unescaped literal '%c' in regex; marked by <-- HERE in m/%s/

      o  Use of unassigned code point or non-standalone grapheme for a
          delimiter will be a fatal error starting in Perl 5.30

          See "Deprecations"

  Changes to Existing Diagnostics
      o  When a "require" fails, we now do not provide @INC when the
          "require" is for a file instead of a module.

      o  When @INC is not scanned for a "require" call, we no longer display
          @INC to avoid confusion.

      o  Attribute "locked" is deprecated, and will disappear in Perl 5.28

          This existing warning has had the and will disappear text added in
          this release.

      o  Attribute "unique" is deprecated, and will disappear in Perl 5.28

          This existing warning has had the and will disappear text added in
          this release.

      o  Calling POSIX::%s() is deprecated

          This warning has been removed, as the deprecated functions have
          been removed from POSIX.

      o  Constants from lexical variables potentially modified elsewhere are
          deprecated. This will not be allowed in Perl 5.32

          This existing warning has had the this will not be allowed text
          added in this release.

      o  Deprecated use of "my()" in false conditional. This will be a fatal
          error in Perl 5.30

          This existing warning has had the this will be a fatal error text
          added in this release.

      o  "dump()" better written as "CORE::dump()". "dump()" will no longer
          be available in Perl 5.30

          This existing warning has had the no longer be available text added
          in this release.

      o  Experimental %s on scalar is now forbidden

          This message is now followed by more helpful text.  [perl #127976]
          <https://rt.perl.org/Public/Bug/Display.html?id=127976>

      o  Experimental "%s" subs not enabled

          This warning was been removed, as lexical subs are no longer
          experimental.

      o  Having more than one /%c regexp modifier is deprecated

          This deprecation warning has been removed, since "/xx" now has a
          new meaning.

      o  %s() is deprecated on ":utf8" handles. This will be a fatal error
          in Perl 5.30 .

          where "%s" is one of "sysread", "recv", "syswrite", or "send".

          This existing warning has had the this will be a fatal error text
          added in this release.

          This warning is now enabled by default, as all "deprecated"
          category warnings should be.

      o  $* is no longer supported. Its use will be fatal in Perl 5.30

          This existing warning has had the its use will be fatal text added
          in this release.

      o  $# is no longer supported. Its use will be fatal in Perl 5.30

          This existing warning has had the its use will be fatal text added
          in this release.

      o  Malformed UTF-8 character%s

          Details as to the exact problem have been added at the end of this
          message

      o  Missing or undefined argument to %s

          This warning used to warn about "require", even if it was actually
          "do" which being executed. It now gets the operation name right.

      o  NO-BREAK SPACE in a charnames alias definition is deprecated

          This warning has been removed as the behavior is now an error.

      o  Odd name/value argument for subroutine '%s'

          This warning now includes the name of the offending subroutine.

      o  Opening dirhandle %s also as a file. This will be a fatal error in
          Perl 5.28

          This existing warning has had the this will be a fatal error text
          added in this release.

      o  Opening filehandle %s also as a directory. This will be a fatal
          error in Perl 5.28

          This existing warning has had the this will be a fatal error text
          added in this release.

      o  panic: ck_split, type=%u

          panic: pp_split, pm=%p, s=%p

          These panic errors have been removed.

      o  Passing malformed UTF-8 to "%s" is deprecated

          This warning has been changed to the fatal Malformed UTF-8 string
          in "%s"

      o  Setting $/ to a reference to %s as a form of slurp is deprecated,
          treating as undef. This will be fatal in Perl 5.28

          This existing warning has had the this will be fatal text added in
          this release.

      o  "${^ENCODING}" is no longer supported. Its use will be fatal in
          Perl 5.28

          This warning used to be: "Setting "${^ENCODING}" is deprecated".

          The special action of the variable "${^ENCODING}" was formerly used
          to implement the "encoding" pragma. As of Perl 5.26, rather than
          being deprecated, assigning to this variable now has no effect
          except to issue the warning.

      o  Too few arguments for subroutine '%s'

          This warning now includes the name of the offending subroutine.

      o  Too many arguments for subroutine '%s'

          This warning now includes the name of the offending subroutine.

      o  Unescaped left brace in regex is deprecated here (and will be fatal
          in Perl 5.30), passed through in regex; marked by <-- HERE in m/%s/

          This existing warning has had the here (and will be fatal...) text
          added in this release.

      o  Unknown charname '' is deprecated. Its use will be fatal in Perl
          5.28

          This existing warning has had the its use will be fatal text added
          in this release.

      o  Use of bare << to mean <<"" is deprecated. Its use will be fatal in
          Perl 5.28

          This existing warning has had the its use will be fatal text added
          in this release.

      o  Use of code point 0x%s is deprecated; the permissible max is 0x%s.
          This will be fatal in Perl 5.28

          This existing warning has had the this will be fatal text added in
          this release.

      o  Use of comma-less variable list is deprecated. Its use will be
          fatal in Perl 5.28

          This existing warning has had the its use will be fatal text added
          in this release.

      o  Use of inherited "AUTOLOAD" for non-method %s() is deprecated. This
          will be fatal in Perl 5.28

          This existing warning has had the this will be fatal text added in
          this release.

      o  Use of strings with code points over 0xFF as arguments to %s
          operator is deprecated. This will be a fatal error in Perl 5.28

          This existing warning has had the this will be a fatal error text
          added in this release.

Utility Changes
  c2ph and pstruct
      o  These old utilities have long since superceded by h2xs, and are now
          gone from the distribution.

  Porting/pod_lib.pl
      o  Removed spurious executable bit.

      o  Account for the possibility of DOS file endings.

  Porting/sync-with-cpan
      o  Many improvements.

  perf/benchmarks
      o  Tidy file, rename some symbols.

  Porting/checkAUTHORS.pl
      o  Replace obscure character range with "\w".

  t/porting/regen.t
      o  Try to be more helpful when tests fail.

  utils/h2xs.PL
      o  Avoid infinite loop for enums.

  perlbug
      o  Long lines in the message body are now wrapped at 900 characters,
          to stay well within the 1000-character limit imposed by SMTP mail
          transfer agents.  This is particularly likely to be important for
          the list of arguments to Configure, which can readily exceed the
          limit if, for example, it names several non-default installation
          paths.  This change also adds the first unit tests for perlbug.
          [perl #128020]
          <https://rt.perl.org/Public/Bug/Display.html?id=128020>

Configuration and Compilation
      o  "-Ddefault_inc_excludes_dot" has added, and enabled by default.

      o  The "dtrace" build process has further changes [perl #130108]
          <https://rt.perl.org/Public/Bug/Display.html?id=130108>:

          o  If the "-xnolibs" is available, use that so a dtrace perl can
              be built within a FreeBSD jail.

          o  On systems that build a dtrace object file (FreeBSD, Solaris,
              and SystemTap's dtrace emulation), copy the input objects to a
              separate directory and process them there, and use those
              objects in the link, since "dtrace -G" also modifies these
              objects.

          o  Add libelf to the build on FreeBSD 10.x, since dtrace adds
              references to libelf symbols.

          o  Generate a dummy dtrace_main.o if "dtrace -G" fails to build
              it.  A default build on Solaris generates probes from the
              unused inline functions, while they don't on FreeBSD, which
              causes "dtrace -G" to fail.

      o  You can now disable perl's use of the "PERL_HASH_SEED" and
          "PERL_PERTURB_KEYS" environment variables by configuring perl with
          "-Accflags=NO_PERL_HASH_ENV".

      o  You can now disable perl's use of the "PERL_HASH_SEED_DEBUG"
          environment variable by configuring perl with
          "-Accflags=-DNO_PERL_HASH_SEED_DEBUG".

      o  Configure now zeroes out the alignment bytes when calculating the
          bytes for 80-bit "NaN" and "Inf" to make builds more reproducible.
          [perl #130133]
          <https://rt.perl.org/Public/Bug/Display.html?id=130133>

      o  Since v5.18, for testing purposes we have included support for
          building perl with a variety of non-standard, and non-recommended
          hash functions.  Since we do not recommend the use of these
          functions, we have removed them and their corresponding build
          options.  Specifically this includes the following build options:

              PERL_HASH_FUNC_SDBM
              PERL_HASH_FUNC_DJB2
              PERL_HASH_FUNC_SUPERFAST
              PERL_HASH_FUNC_MURMUR3
              PERL_HASH_FUNC_ONE_AT_A_TIME
              PERL_HASH_FUNC_ONE_AT_A_TIME_OLD
              PERL_HASH_FUNC_MURMUR_HASH_64A
              PERL_HASH_FUNC_MURMUR_HASH_64B

      o  Remove "Warning: perl appears in your path"

          This install warning is more or less obsolete, since most platforms
          already will have a /usr/bin/perl or similar provided by the OS.

      o  Reduce verbosity of "make install.man"

          Previously, two progress messages were emitted for each manpage:
          one by installman itself, and one by the function in install_lib.pl
          that it calls to actually install the file.  Disabling the second
          of those in each case saves over 750 lines of unhelpful output.

      o  Cleanup for "clang -Weverything" support.  [perl #129961]
          <https://rt.perl.org/Public/Bug/Display.html?id=129961>

      o  Configure: signbit scan was assuming too much, stop assuming
          negative 0.

      o  Various compiler warnings have been silenced.

      o  Several smaller changes have been made to remove impediments to
          compiling under C++11.

      o  Builds using "USE_PAD_RESET" now work again; this configuration had
          bit-rotted.

      o  A probe for "gai_strerror" was added to Configure that checks if
          the "gai_strerror()" routine is available and can be used to
          translate error codes returned by "getaddrinfo()" into human
          readable strings.

      o  Configure now aborts if both "-Duselongdouble" and "-Dusequadmath"
          are requested.  [perl #126203]
          <https://rt.perl.org/Public/Bug/Display.html?id=126203>

      o  Fixed a bug in which Configure could append "-quadmath" to the
          archname even if it was already present.  [perl #128538]
          <https://rt.perl.org/Public/Bug/Display.html?id=128538>

      o  Clang builds with "-DPERL_GLOBAL_STRUCT" or
          "-DPERL_GLOBAL_STRUCT_PRIVATE" have been fixed (by disabling Thread
          Safety Analysis for these configurations).

      o  make_ext.pl no longer updates a module's pm_to_blib file when no
          files require updates.  This could cause dependencies, perlmain.c
          in particular, to be rebuilt unnecessarily.  [perl #126710]
          <https://rt.perl.org/Public/Bug/Display.html?id=126710>

      o  The output of "perl -V" has been reformatted so that each
          configuration and compile-time option is now listed one per line,
          to improve readability.

      o  Configure now builds "miniperl" and "generate_uudmap" if you invoke
          it with "-Dusecrosscompiler" but not "-Dtargethost=somehost".  This
          means you can supply your target platform "config.sh", generate the
          headers and proceed to build your cross-target perl.  [perl
          #127234] <https://rt.perl.org/Public/Bug/Display.html?id=127234>

      o  Perl built with "-Accflags=-DPERL_TRACE_OPS" now only dumps the
          operator counts when the environment variable "PERL_TRACE_OPS" is
          set to a non-zero integer.  This allows "make test" to pass on such
          a build.

      o  When building with GCC 6 and link-time optimization (the "-flto"
          option to "gcc"), Configure was treating all probed symbols as
          present on the system, regardless of whether they actually exist.
          This has been fixed.  [perl #128131]
          <https://rt.perl.org/Public/Bug/Display.html?id=128131>

      o  The t/test.pl library is used for internal testing of Perl itself,
          and also copied by several CPAN modules.  Some of those modules
          must work on older versions of Perl, so t/test.pl must in turn
          avoid newer Perl features.  Compatibility with Perl 5.8 was
          inadvertently removed some time ago; it has now been restored.
          [perl #128052]
          <https://rt.perl.org/Public/Bug/Display.html?id=128052>

      o  The build process no longer emits an extra blank line before
          building each "simple" extension (those with only *.pm and *.pod
          files).

Testing
      Tests were added and changed to reflect the other additions and changes
      in this release.  Furthermore, these substantive changes were made:

      o  A new test script, comp/parser_run.t, has been added that is like
          comp/parser.t but with test.pl included so that "runperl()" and the
          like are available for use.

      o  Tests for locales were erroneously using locales incompatible with
          Perl.

      o  Some parts of the test suite that try to exhaustively test edge
          cases in the regex implementation have been restricted to running
          for a maximum of five minutes.  On slow systems they could
          otherwise take several hours, without significantly improving our
          understanding of the correctness of the code under test.

      o  A new internal facility allows analysing the time taken by the
          individual tests in Perl's own test suite; see
          Porting/harness-timer-report.pl.

      o  t/re/regexp_nonull.t has been added to test that the regular
          expression engine can handle scalars that do not have a null byte
          just past the end of the string.

      o  A new test script, t/op/decl-refs.t, has been added to test the new
          feature "Declaring a reference to a variable".

      o  A new test script, t/re/keep_tabs.t has been added to contain tests
          where "\t" characters should not be expanded into spaces.

      o  A new test script, t/re/anyof.t, has been added to test that the
          ANYOF nodes generated by bracketed character classes are as
          expected.

      o  There is now more extensive testing of the Unicode-related API
          macros and functions.

      o  Several of the longer running API test files have been split into
          multiple test files so that they can be run in parallel.

      o  t/harness now tries really hard not to run tests which are located
          outside of the Perl source tree.  [perl #124050]
          <https://rt.perl.org/Public/Bug/Display.html?id=124050>

      o  Prevent debugger tests (lib/perl5db.t) from failing due to the
          contents of $ENV{PERLDB_OPTS}.  [perl #130445]
          <https://rt.perl.org/Public/Bug/Display.html?id=130445>

Platform Support
  New Platforms
      NetBSD/VAX
          Perl now compiles under NetBSD on VAX machines.  However, it's not
          possible for that platform to implement floating-point infinities
          and NaNs compatible with most modern systems, which implement the
          IEEE-754 floating point standard.  The hexadecimal floating point
          ("0x...p[+-]n" literals, "printf %a") is not implemented, either.
          The "make test" passes 98% of tests.

          o  Test fixes and minor updates.

          o  Account for lack of "inf", "nan", and "-0.0" support.

  Platform-Specific Notes
      Darwin
          o  Don't treat "-Dprefix=/usr" as special: instead require an
              extra option "-Ddarwin_distribution" to produce the same
              results.

          o  OS X El Capitan doesn't implement the "clock_gettime()" or
              "clock_getres()" APIs; emulate them as necessary.

          o  Deprecated syscall(2) on macOS 10.12.

      EBCDIC
          Several tests have been updated to work (or be skipped) on EBCDIC
          platforms.

      HP-UX
          The Net::Ping UDP test is now skipped on HP-UX.

      Hurd
          The hints for Hurd have been improved, enabling malloc wrap and
          reporting the GNU libc used (previously it was an empty string when
          reported).

      VAX VAX floating point formats are now supported on NetBSD.

      VMS
          o  The path separator for the "PERL5LIB" and "PERLLIB" environment
              entries is now a colon (":") when running under a Unix shell.
              There is no change when running under DCL (it's still "|").

          o  configure.com now recognizes the VSI-branded C compiler and no
              longer recognizes the "DEC"-branded C compiler (as there hasn't
              been such a thing for 15 or more years).

      Windows
          o  Support for compiling perl on Windows using Microsoft Visual
              Studio 2015 (containing Visual C++ 14.0) has been added.

              This version of VC++ includes a completely rewritten C run-time
              library, some of the changes in which mean that work done to
              resolve a socket "close()" bug in perl #120091 and perl #118059
              is not workable in its current state with this version of VC++.
              Therefore, we have effectively reverted that bug fix for VS2015
              onwards on the basis that being able to build with VS2015
              onwards is more important than keeping the bug fix.  We may
              revisit this in the future to attempt to fix the bug again in a
              way that is compatible with VS2015.

              These changes do not affect compilation with GCC or with Visual
              Studio versions up to and including VS2013, i.e., the bug fix
              is retained (unchanged) for those compilers.

              Note that you may experience compatibility problems if you mix
              a perl built with GCC or VS <= VS2013 with XS modules built
              with VS2015, or if you mix a perl built with VS2015 with XS
              modules built with GCC or VS <= VS2013. Some incompatibility
              may arise because of the bug fix that has been reverted for
              VS2015 builds of perl, but there may well be incompatibility
              anyway because of the rewritten CRT in VS2015 (e.g., see
              discussion at <http://stackoverflow.com/questions/30412951>).

          o  It now automatically detects GCC versus Visual C and sets the
              VC version number on Win32.

      Linux
          Drop support for Linux a.out executable format. Linux has used ELF
          for over twenty years.

      OpenBSD 6
          OpenBSD 6 still does not support returning "pid", "gid", or "uid"
          with "SA_SIGINFO".  Make sure to account for it.

      FreeBSD
          t/uni/overload.t: Skip hanging test on FreeBSD.

      DragonFly BSD
          DragonFly BSD now has support for "setproctitle()".  [perl #130068]
          <https://rt.perl.org/Public/Bug/Display.html?id=130068>.

Internal Changes
      o  A new API function "sv_setpv_bufsize()" allows simultaneously
          setting the length and the allocated size of the buffer in an "SV",
          growing the buffer if necessary.

      o  A new API macro "SvPVCLEAR()" sets its "SV" argument to an empty
          string, like Perl-space "$x = ''", but with several optimisations.

      o  Several new macros and functions for dealing with Unicode and
          UTF-8-encoded strings have been added to the API, as well as some
          changes in the functionality of existing functions (see "Unicode
          Support" in perlapi for more details):

          o  New versions of the API macros like "isALPHA_utf8" and
              "toLOWER_utf8" have been added, each with the suffix "_safe",
              like "isSPACE_utf8_safe".  These take an extra parameter,
              giving an upper limit of how far into the string it is safe to
              read.  Using the old versions could cause attempts to read
              beyond the end of the input buffer if the UTF-8 is not well-
              formed, and their use now raises a deprecation warning.
              Details are at "Character classification" in perlapi.

          o  Macros like "isALPHA_utf8" and "toLOWER_utf8" now die if they
              detect that their input UTF-8 is malformed.  A deprecation
              warning had been issued since Perl 5.18.

          o  Several new macros for analysing the validity of utf8
              sequences. These are:

              "UTF8_GOT_ABOVE_31_BIT" "UTF8_GOT_CONTINUATION"
              "UTF8_GOT_EMPTY" "UTF8_GOT_LONG" "UTF8_GOT_NONCHAR"
              "UTF8_GOT_NON_CONTINUATION" "UTF8_GOT_OVERFLOW"
              "UTF8_GOT_SHORT" "UTF8_GOT_SUPER" "UTF8_GOT_SURROGATE"
              "UTF8_IS_INVARIANT" "UTF8_IS_NONCHAR" "UTF8_IS_SUPER"
              "UTF8_IS_SURROGATE" "UVCHR_IS_INVARIANT" "isUTF8_CHAR_flags"
              "isSTRICT_UTF8_CHAR" "isC9_STRICT_UTF8_CHAR"

          o  Functions that are all extensions of the "is_utf8_string_*()"
              functions, that apply various restrictions to the UTF-8
              recognized as valid:

              "is_strict_utf8_string", "is_strict_utf8_string_loc",
              "is_strict_utf8_string_loclen",

              "is_c9strict_utf8_string", "is_c9strict_utf8_string_loc",
              "is_c9strict_utf8_string_loclen",

              "is_utf8_string_flags", "is_utf8_string_loc_flags",
              "is_utf8_string_loclen_flags",

              "is_utf8_fixed_width_buf_flags",
              "is_utf8_fixed_width_buf_loc_flags",
              "is_utf8_fixed_width_buf_loclen_flags".

              "is_utf8_invariant_string".  "is_utf8_valid_partial_char".
              "is_utf8_valid_partial_char_flags".

          o  The functions "utf8n_to_uvchr" and its derivatives have had
              several changes of behaviour.

              Calling them, while passing a string length of 0 is now
              asserted against in DEBUGGING builds, and otherwise, returns
              the Unicode REPLACEMENT CHARACTER.  If you have nothing to
              decode, you shouldn't call the decode function.

              They now return the Unicode REPLACEMENT CHARACTER if called
              with UTF-8 that has the overlong malformation and that
              malformation is allowed by the input parameters.  This
              malformation is where the UTF-8 looks valid syntactically, but
              there is a shorter sequence that yields the same code point.
              This has been forbidden since Unicode version 3.1.

              They now accept an input flag to allow the overflow
              malformation.  This malformation is when the UTF-8 may be
              syntactically valid, but the code point it represents is not
              capable of being represented in the word length on the
              platform.  What "allowed" means, in this case, is that the
              function doesn't return an error, and it advances the parse
              pointer to beyond the UTF-8 in question, but it returns the
              Unicode REPLACEMENT CHARACTER as the value of the code point
              (since the real value is not representable).

              They no longer abandon searching for other malformations when
              the first one is encountered.  A call to one of these functions
              thus can generate multiple diagnostics, instead of just one.

          o  "valid_utf8_to_uvchr()" has been added to the API (although it
              was present in core earlier). Like "utf8_to_uvchr_buf()", but
              assumes that the next character is well-formed.  Use with
              caution.

          o  A new function, "utf8n_to_uvchr_error", has been added for use
              by modules that need to know the details of UTF-8 malformations
              beyond pass/fail.  Previously, the only ways to know why a
              sequence was ill-formed was to capture and parse the generated
              diagnostics or to do your own analysis.

          o  There is now a safer version of utf8_hop(), called
              "utf8_hop_safe()".  Unlike utf8_hop(), utf8_hop_safe() won't
              navigate before the beginning or after the end of the supplied
              buffer.

          o  Two new functions, "utf8_hop_forward()" and "utf8_hop_back()"
              are similar to "utf8_hop_safe()" but are for when you know
              which direction you wish to travel.

          o  Two new macros which return useful utf8 byte sequences:

              "BOM_UTF8"

              "REPLACEMENT_CHARACTER_UTF8"

      o  Perl is now built with the "PERL_OP_PARENT" compiler define enabled
          by default.  To disable it, use the "PERL_NO_OP_PARENT" compiler
          define.  This flag alters how the "op_sibling" field is used in
          "OP" structures, and has been available optionally since perl 5.22.

          See "Internal Changes" in perl5220delta for more details of what
          this build option does.

      o  Three new ops, "OP_ARGELEM", "OP_ARGDEFELEM", and "OP_ARGCHECK"
          have been added.  These are intended principally to implement the
          individual elements of a subroutine signature, plus any overall
          checking required.

      o  The "OP_PUSHRE" op has been eliminated and the "OP_SPLIT" op has
          been changed from class "LISTOP" to "PMOP".

          Formerly the first child of a split would be a "pushre", which
          would have the "split"'s regex attached to it. Now the regex is
          attached directly to the "split" op, and the "pushre" has been
          eliminated.

      o  The "op_class()" API function has been added.  This is like the
          existing "OP_CLASS()" macro, but can more accurately determine what
          struct an op has been allocated as.  For example "OP_CLASS()" might
          return "OA_BASEOP_OR_UNOP" indicating that ops of this type are
          usually allocated as an "OP" or "UNOP"; while "op_class()" will
          return "OPclass_BASEOP" or "OPclass_UNOP" as appropriate.

      o  All parts of the internals now agree that the "sassign" op is a
          "BINOP"; previously it was listed as a "BASEOP" in regen/opcodes,
          which meant that several parts of the internals had to be special-
          cased to accommodate it.  This oddity's original motivation was to
          handle code like "$x ||= 1"; that is now handled in a simpler way.

      o  The output format of the "op_dump()" function (as used by "perl
          -Dx") has changed: it now displays an "ASCII-art" tree structure,
          and shows more low-level details about each op, such as its address
          and class.

      o  The "PADOFFSET" type has changed from being unsigned to signed, and
          several pad-related variables such as "PL_padix" have changed from
          being of type "I32" to type "PADOFFSET".

      o  The "DEBUGGING"-mode output for regex compilation and execution has
          been enhanced.

      o  Several obscure SV flags have been eliminated, sometimes along with
          the macros which manipulate them: "SVpbm_VALID", "SVpbm_TAIL",
          "SvTAIL_on", "SvTAIL_off", "SVrepl_EVAL", "SvEVALED".

      o  An OP "op_private" flag has been eliminated: "OPpRUNTIME". This
          used to often get set on "PMOP" ops, but had become meaningless
          over time.

Selected Bug Fixes
      o  Perl no longer panics when switching into some locales on machines
          with buggy "strxfrm()" implementations in their libc.  [perl
          #121734] <https://rt.perl.org/Public/Bug/Display.html?id=121734>

      o  " $-{$name} " would leak an "AV" on each access if the regular
          expression had no named captures.  The same applies to access to
          any hash tied with Tie::Hash::NamedCapture and "all => 1".  [perl
          #130822] <https://rt.perl.org/Public/Bug/Display.html?id=130822>

      o  Attempting to use the deprecated variable $# as the object in an
          indirect object method call could cause a heap use after free or
          buffer overflow.  [perl #129274]
          <https://rt.perl.org/Public/Bug/Display.html?id=129274>

      o  When checking for an indirect object method call, in some rare
          cases the parser could reallocate the line buffer but then continue
          to use pointers to the old buffer.  [perl #129190]
          <https://rt.perl.org/Public/Bug/Display.html?id=129190>

      o  Supplying a glob as the format argument to "formline" would cause
          an assertion failure.  [perl #130722]
          <https://rt.perl.org/Public/Bug/Display.html?id=130722>

      o  Code like " $value1 =~ qr/.../ ~~ $value2 " would have the match
          converted into a "qr//" operator, leaving extra elements on the
          stack to confuse any surrounding expression.  [perl #130705]
          <https://rt.perl.org/Public/Bug/Display.html?id=130705>

      o  Since v5.24 in some obscure cases, a regex which included code
          blocks from multiple sources (e.g., via embedded via "qr//"
          objects) could end up with the wrong current pad and crash or give
          weird results.  [perl #129881]
          <https://rt.perl.org/Public/Bug/Display.html?id=129881>

      o  Occasionally "local()"s in a code block within a patterns weren't
          being undone when the pattern matching backtracked over the code
          block.  [perl #126697]
          <https://rt.perl.org/Public/Bug/Display.html?id=126697>

      o  Using "substr()" to modify a magic variable could access freed
          memory in some cases.  [perl #129340]
          <https://rt.perl.org/Public/Bug/Display.html?id=129340>

      o  Under "use utf8", the entire source code is now checked for being
          UTF-8 well formed, not just quoted strings as before.  [perl
          #126310] <https://rt.perl.org/Public/Bug/Display.html?id=126310>.

      o  The range operator ".." on strings now handles its arguments
          correctly when in the scope of the "unicode_strings" feature.  The
          previous behaviour was sufficiently unexpected that we believe no
          correct program could have made use of it.

      o  The "split" operator did not ensure enough space was allocated for
          its return value in scalar context.  It could then write a single
          pointer immediately beyond the end of the memory block allocated
          for the stack.  [perl #130262]
          <https://rt.perl.org/Public/Bug/Display.html?id=130262>

      o  Using a large code point with the "W" pack template character with
          the current output position aligned at just the right point could
          cause a write of a single zero byte immediately beyond the end of
          an allocated buffer.  [perl #129149]
          <https://rt.perl.org/Public/Bug/Display.html?id=129149>

      o  Supplying a format's picture argument as part of the format
          argument list where the picture specifies modifying the argument
          could cause an access to the new freed compiled form.at.  [perl
          #129125] <https://rt.perl.org/Public/Bug/Display.html?id=129125>

      o  The sort() operator's built-in numeric comparison function didn't
          handle large integers that weren't exactly representable by a
          double.  This now uses the same code used to implement the "<=>"
          operator.  [perl #130335]
          <https://rt.perl.org/Public/Bug/Display.html?id=130335>

      o  Fix issues with "/(?{ ... <<EOF })/" that broke Method::Signatures.
          [perl #130398]
          <https://rt.perl.org/Public/Bug/Display.html?id=130398>

      o  Fixed an assertion failure with "chop" and "chomp", which could be
          triggered by "chop(@x =~ tr/1/1/)".  [perl #130198]
          <https://rt.perl.org/Public/Bug/Display.html?id=130198>.

      o  Fixed a comment skipping error in patterns under "/x"; it could
          stop skipping a byte early, which could be in the middle of a UTF-8
          character.  [perl #130495]
          <https://rt.perl.org/Public/Bug/Display.html?id=130495>.

      o  perldb now ignores /dev/tty on non-Unix systems.  [perl #113960]
          <https://rt.perl.org/Public/Bug/Display.html?id=113960>;

      o  Fix assertion failure for "{}->$x" when $x isn't defined.  [perl
          #130496] <https://rt.perl.org/Public/Bug/Display.html?id=130496>.

      o  Fix an assertion error which could be triggered when a lookahead
          string in patterns exceeded a minimum length.  [perl #130522]
          <https://rt.perl.org/Public/Bug/Display.html?id=130522>.

      o  Only warn once per literal number about a misplaced "_".  [perl
          #70878] <https://rt.perl.org/Public/Bug/Display.html?id=70878>.

      o  The "tr///" parse code could be looking at uninitialized data after
          a perse error.  [perl #129342]
          <https://rt.perl.org/Public/Bug/Display.html?id=129342>.

      o  In a pattern match, a back-reference ("\1") to an unmatched capture
          could read back beyond the start of the string being matched.
          [perl #129377]
          <https://rt.perl.org/Public/Bug/Display.html?id=129377>.

      o  "use re 'strict'" is supposed to warn if you use a range (such as
          "/(?[ [ X-Y ] ])/") whose start and end digit aren't from the same
          group of 10.  It didn't do that for five groups of mathematical
          digits starting at "U+1D7E".

      o  A sub containing a "forward" declaration with the same name (e.g.,
          "sub c { sub c; }") could sometimes crash or loop infinitely.
          [perl #129090]
          <https://rt.perl.org/Public/Bug/Display.html?id=129090>

      o  A crash in executing a regex with a non-anchored UTF-8 substring
          against a target string that also used UTF-8 has been fixed.  [perl
          #129350] <https://rt.perl.org/Public/Bug/Display.html?id=129350>

      o  Previously, a shebang line like "#!perl -i u" could be erroneously
          interpreted as requesting the "-u" option.  This has been fixed.
          [perl #129336]
          <https://rt.perl.org/Public/Bug/Display.html?id=129336>

      o  The regex engine was previously producing incorrect results in some
          rare situations when backtracking past an alternation that matches
          only one thing; this showed up as capture buffers ($1, $2, etc.)
          erroneously containing data from regex execution paths that weren't
          actually executed for the final match.  [perl #129897]
          <https://rt.perl.org/Public/Bug/Display.html?id=129897>

      o  Certain regexes making use of the experimental "regex_sets" feature
          could trigger an assertion failure.  This has been fixed.  [perl
          #129322] <https://rt.perl.org/Public/Bug/Display.html?id=129322>

      o  Invalid assignments to a reference constructor (e.g., "\eval=time")
          could sometimes crash in addition to giving a syntax error.  [perl
          #125679] <https://rt.perl.org/Public/Bug/Display.html?id=125679>

      o  The parser could sometimes crash if a bareword came after
          "evalbytes".  [perl #129196]
          <https://rt.perl.org/Public/Bug/Display.html?id=129196>

      o  Autoloading via a method call would warn erroneously ("Use of
          inherited AUTOLOAD for non-method") if there was a stub present in
          the package into which the invocant had been blessed.  The warning
          is no longer emitted in such circumstances.  [perl #47047]
          <https://rt.perl.org/Public/Bug/Display.html?id=47047>

      o  The use of "splice" on arrays with non-existent elements could
          cause other operators to crash.  [perl #129164]
          <https://rt.perl.org/Public/Bug/Display.html?id=129164>

      o  A possible buffer overrun when a pattern contains a fixed utf8
          substring.  [perl #129012]
          <https://rt.perl.org/Public/Bug/Display.html?id=129012>

      o  Fixed two possible use-after-free bugs in perl's lexer.  [perl
          #129069] <https://rt.perl.org/Public/Bug/Display.html?id=129069>

      o  Fixed a crash with "s///l" where it thought it was dealing with
          UTF-8 when it wasn't.  [perl #129038]
          <https://rt.perl.org/Public/Bug/Display.html?id=129038>

      o  Fixed a place where the regex parser was not setting the syntax
          error correctly on a syntactically incorrect pattern.  [perl
          #129122] <https://rt.perl.org/Public/Bug/Display.html?id=129122>

      o  The "&." operator (and the "&" operator, when it treats its
          arguments as strings) were failing to append a trailing null byte
          if at least one string was marked as utf8 internally.  Many code
          paths (system calls, regexp compilation) still expect there to be a
          null byte in the string buffer just past the end of the logical
          string.  An assertion failure was the result.  [perl #129287]
          <https://rt.perl.org/Public/Bug/Display.html?id=129287>

      o  Avoid a heap-after-use error in the parser when creating an error
          messge for a syntactically invalid heredoc.  [perl #128988]
          <https://rt.perl.org/Public/Bug/Display.html?id=128988>

      o  Fix a segfault when run with "-DC" options on DEBUGGING builds.
          [perl #129106]
          <https://rt.perl.org/Public/Bug/Display.html?id=129106>

      o  Fixed the parser error handling in subroutine attributes for an
          '":attr(foo"' that does not have an ending '")"'.

      o  Fix the perl lexer to correctly handle a backslash as the last char
          in quoted-string context. This actually fixed two bugs, [perl
          #129064] <https://rt.perl.org/Public/Bug/Display.html?id=129064>
          and [perl #129176]
          <https://rt.perl.org/Public/Bug/Display.html?id=129176>.

      o  In the API function "gv_fetchmethod_pvn_flags", rework separator
          parsing to prevent possible string overrun with an invalid "len"
          argument.  [perl #129267]
          <https://rt.perl.org/Public/Bug/Display.html?id=129267>

      o  Problems with in-place array sorts: code like "@a = sort { ... }
          @a", where the source and destination of the sort are the same
          plain array, are optimised to do less copying around.  Two side-
          effects of this optimisation were that the contents of @a as seen
          by sort routines were partially sorted; and under some
          circumstances accessing @a during the sort could crash the
          interpreter.  Both these issues have been fixed, and Sort functions
          see the original value of @a.  [perl #128340]
          <https://rt.perl.org/Public/Bug/Display.html?id=128340>

      o  Non-ASCII string delimiters are now reported correctly in error
          messages for unterminated strings.  [perl #128701]
          <https://rt.perl.org/Public/Bug/Display.html?id=128701>

      o  "pack("p", ...)" used to emit its warning ("Attempt to pack pointer
          to temporary value") erroneously in some cases, but has been fixed.

      o  @DB::args is now exempt from "used once" warnings.  The warnings
          only occurred under -w, because warnings.pm itself uses @DB::args
          multiple times.

      o  The use of built-in arrays or hash slices in a double-quoted string
          no longer issues a warning ("Possible unintended interpolation...")
          if the variable has not been mentioned before.  This affected code
          like "qq|@DB::args|" and "qq|@SIG{'CHLD', 'HUP'}|".  (The special
          variables "@-" and "@+" were already exempt from the warning.)

      o  "gethostent" and similar functions now perform a null check
          internally, to avoid crashing with the torsocks library.  This was
          a regression from v5.22.  [perl #128740]
          <https://rt.perl.org/Public/Bug/Display.html?id=128740>

      o  "defined *{'!'}", "defined *{'['}", and "defined *{'-'}" no longer
          leak memory if the typeglob in question has never been accessed
          before.

      o  Mentioning the same constant twice in a row (which is a syntax
          error) no longer fails an assertion under debugging builds.  This
          was a regression from v5.20.  [perl #126482]
          <https://rt.perl.org/Public/Bug/Display.html?id=126482>

      o  Many issues relating to "printf "%a"" of hexadecimal floating point
          were fixed.  In addition, the "subnormals" (formerly known as
          "denormals") floating point numbers are now supported both with the
          plain IEEE 754 floating point numbers (64-bit or 128-bit) and the
          x86 80-bit "extended precision".  Note that subnormal hexadecimal
          floating point literals will give a warning about "exponent
          underflow".  [perl #128843]
          <https://rt.perl.org/Public/Bug/Display.html?id=128843> [perl
          #128889] <https://rt.perl.org/Public/Bug/Display.html?id=128889>
          [perl #128890]
          <https://rt.perl.org/Public/Bug/Display.html?id=128890> [perl
          #128893] <https://rt.perl.org/Public/Bug/Display.html?id=128893>
          [perl #128909]
          <https://rt.perl.org/Public/Bug/Display.html?id=128909> [perl
          #128919] <https://rt.perl.org/Public/Bug/Display.html?id=128919>

      o  A regression in v5.24 with "tr/\N{U+...}/foo/" when the code point
          was between 128 and 255 has been fixed.  [perl #128734]
          <https://rt.perl.org/Public/Bug/Display.html?id=128734>.

      o  Use of a string delimiter whose code point is above 2**31 now works
          correctly on platforms that allow this.  Previously, certain
          characters, due to truncation, would be confused with other
          delimiter characters with special meaning (such as "?" in
          "m?...?"), resulting in inconsistent behaviour.  Note that this is
          non-portable, and is based on Perl's extension to UTF-8, and is
          probably not displayable nor enterable by any editor.  [perl
          #128738] <https://rt.perl.org/Public/Bug/Display.html?id=128738>

      o  "@{x" followed by a newline where "x" represents a control or non-
          ASCII character no longer produces a garbled syntax error message
          or a crash.  [perl #128951]
          <https://rt.perl.org/Public/Bug/Display.html?id=128951>

      o  An assertion failure with "%: = 0" has been fixed.  [perl #128238]
          <https://rt.perl.org/Public/Bug/Display.html?id=128238>

      o  In Perl 5.18, the parsing of "$foo::$bar" was accidentally changed,
          such that it would be treated as "$foo."::".$bar".  The previous
          behavior, which was to parse it as "$foo:: . $bar", has been
          restored.  [perl #128478]
          <https://rt.perl.org/Public/Bug/Display.html?id=128478>

      o  Since Perl 5.20, line numbers have been off by one when perl is
          invoked with the -x switch.  This has been fixed.  [perl #128508]
          <https://rt.perl.org/Public/Bug/Display.html?id=128508>

      o  Vivifying a subroutine stub in a deleted stash (e.g., "delete
          $My::{"Foo::"}; \&My::Foo::foo") no longer crashes.  It had begun
          crashing in Perl 5.18.  [perl #128532]
          <https://rt.perl.org/Public/Bug/Display.html?id=128532>

      o  Some obscure cases of subroutines and file handles being freed at
          the same time could result in crashes, but have been fixed.  The
          crash was introduced in Perl 5.22.  [perl #128597]
          <https://rt.perl.org/Public/Bug/Display.html?id=128597>

      o  Code that looks for a variable name associated with an
          uninitialized value could cause an assertion failure in cases where
          magic is involved, such as $ISA[0][0].  This has now been fixed.
          [perl #128253]
          <https://rt.perl.org/Public/Bug/Display.html?id=128253>

      o  A crash caused by code generating the warning "Subroutine
          STASH::NAME redefined" in cases such as "sub P::f{} undef *P::;
          *P::f =sub{};" has been fixed.  In these cases, where the STASH is
          missing, the warning will now appear as "Subroutine NAME
          redefined".  [perl #128257]
          <https://rt.perl.org/Public/Bug/Display.html?id=128257>

      o  Fixed an assertion triggered by some code that handles deprecated
          behavior in formats, e.g., in cases like this:

              format STDOUT =
              @
              0"$x"

          [perl #128255]
          <https://rt.perl.org/Public/Bug/Display.html?id=128255>

      o  A possible divide by zero in string transformation code on Windows
          has been avoided, fixing a crash when collating an empty string.
          [perl #128618]
          <https://rt.perl.org/Public/Bug/Display.html?id=128618>

      o  Some regular expression parsing glitches could lead to assertion
          failures with regular expressions such as "/(?<=/" and "/(?<!/".
          This has now been fixed.  [perl #128170]
          <https://rt.perl.org/Public/Bug/Display.html?id=128170>

      o  " until ($x = 1) { ... } " and " ... until $x = 1 " now properly
          warn when syntax warnings are enabled.  [perl #127333]
          <https://rt.perl.org/Public/Bug/Display.html?id=127333>

      o  socket() now leaves the error code returned by the system in $! on
          failure.  [perl #128316]
          <https://rt.perl.org/Public/Bug/Display.html?id=128316>

      o  Assignment variants of any bitwise ops under the "bitwise" feature
          would crash if the left-hand side was an array or hash.  [perl
          #128204] <https://rt.perl.org/Public/Bug/Display.html?id=128204>

      o  "require" followed by a single colon (as in "foo() ? require : ..."
          is now parsed correctly as "require" with implicit $_, rather than
          "require """.  [perl #128307]
          <https://rt.perl.org/Public/Bug/Display.html?id=128307>

      o  Scalar "keys %hash" can now be assigned to consistently in all
          scalar lvalue contexts.  Previously it worked for some contexts but
          not others.

      o  List assignment to "vec" or "substr" with an array or hash for its
          first argument used to result in crashes or "Can't coerce" error
          messages at run time, unlike scalar assignment, which would give an
          error at compile time.  List assignment now gives a compile-time
          error, too.  [perl #128260]
          <https://rt.perl.org/Public/Bug/Display.html?id=128260>

      o  Expressions containing an "&&" or "||" operator (or their synonyms
          "and" and "or") were being compiled incorrectly in some cases.  If
          the left-hand side consisted of either a negated bareword constant
          or a negated "do {}" block containing a constant expression, and
          the right-hand side consisted of a negated non-foldable expression,
          one of the negations was effectively ignored.  The same was true of
          "if" and "unless" statement modifiers, though with the left-hand
          and right-hand sides swapped.  This long-standing bug has now been
          fixed.  [perl #127952]
          <https://rt.perl.org/Public/Bug/Display.html?id=127952>

      o  "reset" with an argument no longer crashes when encountering stash
          entries other than globs.  [perl #128106]
          <https://rt.perl.org/Public/Bug/Display.html?id=128106>

      o  Assignment of hashes to, and deletion of, typeglobs named *::::::
          no longer causes crashes.  [perl #128086]
          <https://rt.perl.org/Public/Bug/Display.html?id=128086>

      o  Perl wasn't correctly handling true/false values in the LHS of a
          list assign; specifically the truth values returned by boolean
          operators.  This could trigger an assertion failure in something
          like the following:

              for ($x > $y) {
                  ($_, ...) = (...); # here $_ is aliased to a truth value
              }

          This was a regression from v5.24.  [perl #129991]
          <https://rt.perl.org/Public/Bug/Display.html?id=129991>

      o  Assertion failure with user-defined Unicode-like properties.  [perl
          #130010] <https://rt.perl.org/Public/Bug/Display.html?id=130010>

      o  Fix error message for unclosed "\N{" in a regex.  An unclosed "\N{"
          could give the wrong error message: "\N{NAME} must be resolved by
          the lexer".

      o  List assignment in list context where the LHS contained aggregates
          and where there were not enough RHS elements, used to skip scalar
          lvalues.  Previously, "(($a,$b,@c,$d) = (1))" in list context
          returned "($a)"; now it returns "($a,$b,$d)".  "(($a,$b,$c) = (1))"
          is unchanged: it still returns "($a,$b,$c)".  This can be seen in
          the following:

              sub inc { $_++ for @_ }
              inc(($a,$b,@c,$d) = (10))

          Formerly, the values of "($a,$b,$d)" would be left as
          "(11,undef,undef)"; now they are "(11,1,1)".

      o  Code like this: "/(?{ s!!! })/" could trigger infinite recursion on
          the C stack (not the normal perl stack) when the last successful
          pattern in scope is itself.  We avoid the segfault by simply
          forbidding the use of the empty pattern when it would resolve to
          the currently executing pattern.  [perl #129903]
          <https://rt.perl.org/Public/Bug/Display.html?id=129903>

      o  Avoid reading beyond the end of the line buffer in perl's lexer
          when there's a short UTF-8 character at the end.  [perl #128997]
          <https://rt.perl.org/Public/Bug/Display.html?id=128997>

      o  Alternations in regular expressions were sometimes failing to match
          a utf8 string against a utf8 alternate.  [perl #129950]
          <https://rt.perl.org/Public/Bug/Display.html?id=129950>

      o  Make "do "a\0b"" fail silently (and return "undef" and set $!)
          instead of throwing an error.  [perl #129928]
          <https://rt.perl.org/Public/Bug/Display.html?id=129928>

      o  "chdir" with no argument didn't ensure that there was stack space
          available for returning its result.  [perl #129130]
          <https://rt.perl.org/Public/Bug/Display.html?id=129130>

      o  All error messages related to "do" now refer to "do"; some formerly
          claimed to be from "require" instead.

      o  Executing "undef $x" where $x is tied or magical no longer
          incorrectly blames the variable for an uninitialized-value warning
          encountered by the tied/magical code.

      o  Code like "$x = $x . "a"" was incorrectly failing to yield a use of
          uninitialized value warning when $x was a lexical variable with an
          undefined value. That has now been fixed.  [perl #127877]
          <https://rt.perl.org/Public/Bug/Display.html?id=127877>

      o  "undef *_; shift" or "undef *_; pop" inside a subroutine, with no
          argument to "shift" or "pop", began crashing in Perl 5.14, but has
          now been fixed.

      o  "string$scalar->$*" now correctly prefers concatenation overloading
          to string overloading if "$scalar->$*" returns an overloaded
          object, bringing it into consistency with $$scalar.

      o  "/@0{0*->@*/*0" and similar contortions used to crash, but no
          longer do, but merely produce a syntax error.  [perl #128171]
          <https://rt.perl.org/Public/Bug/Display.html?id=128171>

      o  "do" or "require" with an argument which is a reference or typeglob
          which, when stringified, contains a null character, started
          crashing in Perl 5.20, but has now been fixed.  [perl #128182]
          <https://rt.perl.org/Public/Bug/Display.html?id=128182>

      o  Improve the error message for a missing "tie()" package/method.
          This brings the error messages in line with the ones used for
          normal method calls.

      o  Parsing bad POSIX charclasses no longer leaks memory.  [perl
          #128313] <https://rt.perl.org/Public/Bug/Display.html?id=128313>

Known Problems
      o  G++ 6 handles subnormal (denormal) floating point values
          differently than gcc 6 or g++ 5 resulting in "flush-to-zero". The
          end result is that if you specify very small values using the
          hexadecimal floating point format, like "0x1.fffffffffffffp-1022",
          they become zeros.  [perl #131388]
          <https://rt.perl.org/Ticket/Display.html?id=131388>

Errata From Previous Releases
      o  Fixed issues with recursive regexes.  The behavior was fixed in
          Perl 5.24.  [perl #126182]
          <https://rt.perl.org/Public/Bug/Display.html?id=126182>

Obituary
      Jon Portnoy (AVENJ), a prolific Perl author and admired Gentoo
      community member, has passed away on August 10, 2016.  He will be
      remembered and missed by all those who he came in contact with, and
      enriched with his intellect, wit, and spirit.

      It is with great sadness that we also note Kip Hampton's passing.
      Probably best known as the author of the Perl & XML column on XML.com,
      he was a core contributor to AxKit, an XML server platform that became
      an Apache Foundation project.  He was a frequent speaker in the early
      days at OSCON, and most recently at YAPC::NA in Madison.  He was
      frequently on irc.perl.org as ubu, generally in the #axkit-dahut
      community, the group responsible for YAPC::NA Asheville in 2011.

      Kip and his constant contributions to the community will be greatly
      missed.

Acknowledgements
      Perl 5.26.0 represents approximately 13 months of development since
      Perl 5.24.0 and contains approximately 360,000 lines of changes across
      2,600 files from 86 authors.

      Excluding auto-generated files, documentation and release tools, there
      were approximately 230,000 lines of changes to 1,800 .pm, .t, .c and .h
      files.

      Perl continues to flourish into its third decade thanks to a vibrant
      community of users and developers.  The following people are known to
      have contributed the improvements that became Perl 5.26.0:

      Aaron Crane, Abigail, AEvar Arnfjoerd` Bjarmason, Alex Vandiver,
      Andreas Koenig, Andreas Voegele, Andrew Fresh, Andy Lester, Aristotle
      Pagaltzis, Chad Granum, Chase Whitener, Chris 'BinGOs' Williams, Chris
      Lamb, Christian Hansen, Christian Millour, Colin Newell, Craig A.
      Berry, Dagfinn Ilmari Mannsaaker, Dan Collins, Daniel Dragan, Dave
      Cross, Dave Rolsky, David Golden, David H.  Gutteridge, David Mitchell,
      Dominic Hargreaves, Doug Bell, E. Choroba, Ed Avis, Father
      Chrysostomos, Francois Perrad, Hauke D, H.Merijn Brand, Hugo van der
      Sanden, Ivan Pozdeev, James E Keenan, James Raspass, Jarkko Hietaniemi,
      Jerry D. Hedden, Jim Cromie, J. Nick Koston, John Lightsey, Karen
      Etheridge, Karl Williamson, Leon Timmermans, Lukas Mai, Matthew
      Horsfall, Maxwell Carey, Misty De Meo, Neil Bowers, Nicholas Clark,
      Nicolas R., Niko Tyni, Pali, Paul Marquess, Peter Avalos, Petr PisaX,
      Pino Toscano, Rafael Garcia-Suarez, Reini Urban, Renee Baecker, Ricardo
      Signes, Richard Levitte, Rick Delaney, Salvador Fandin~o, Samuel
      Thibault, Sawyer X, Sebastien Aperghis-Tramoni, Sergey Aleynikov,
      Shlomi Fish, Smylers, Stefan Seifert, Steffen Mueller, Stevan Little,
      Steve Hay, Steven Humphrey, Sullivan Beck, Theo Buehler, Thomas Sibley,
      Todd Rinaldo, Tomasz Konojacki, Tony Cook, Unicode Consortium, Yaroslav
      Kuzmin, Yves Orton, Zefram.

      The list above is almost certainly incomplete as it is automatically
      generated from version control history.  In particular, it does not
      include the names of the (very much appreciated) contributors who
      reported issues to the Perl bug tracker.

      Many of the changes included in this version originated in the CPAN
      modules included in Perl's core.  We're grateful to the entire CPAN
      community for helping Perl to flourish.

      For a more complete list of all of Perl's historical contributors,
      please see the AUTHORS file in the Perl source distribution.

(ryoon)

2017-06-05 13:39:53 UTC MAIN commitmail json YAML

Note update of www/contao43 package to 4.3.11.

(taca)

2017-06-05 13:39:25 UTC MAIN commitmail json YAML

Update contao43 to 4.3.11.

* Switch to standard-editon to managed-edition since official distfile
  is managed-edition for now.
* The bugfix release fixes an incompatibility with Symfony 3.3.

(taca)

2017-06-05 12:30:56 UTC MAIN commitmail json YAML

Extend last year fix to PostgreSQL 9.5 and 9.6.

Override INSTALL script to make sure owner/group args do not leak
into the binary packages. Fixes problems with building software
manually against binary postgresql*-client packages built under an
unprivileged setup. Bump respective PKGREVISION.

(fhajny)

2017-06-05 11:40:41 UTC MAIN commitmail json YAML

2017-06-05 11:38:51 UTC MAIN commitmail json YAML

gdb: update to 8.0

GDB 8.0 brings the following fixes and improvements, including:

  * C++: Support for rvalue references

  * Python scripting enhancements:
    ** New functions to start, stop and access a running btrace recording.
    ** Rvalue reference support in gdb.Type.

  * GDB commands interpreter:
    ** User commands now accept an unlimited number of arguments.
    ** The "eval" command now expands user-defined arguments.

  * DWARF version 5 support
    (note that its .debug_names index is not supported yet).

  * GDB/MI enhancements:
    ** New -file-list-shared-libraries command to list the shared
      libraries in the program.
    ** New -target-flash-erase command, to erase flash memory.

  * Support for native FreeBSD/mips (mips*-*-freebsd)

  * Support for the following targets:
    ** Synopsys ARC (arc*-*-elf32)
    ** FreeBSD/mips (mips*-*-freebsd)

  * Miscellaneous enhancements:
    ** Command-line redirection now supported on MS-Windows hosts.
    ** Support for thread names on MS-Windows.
    ** Support for the PKU register on GNU/Linux.
    ** Support for Target descriptions on sparc32 and sparc64.
    ** New GDB/CLI command to erase flash memory
    ** rdrand and rdseed instructions record/replay support.

Support for the following features have been removed:

  * Support for Java programs compiled with gcj

  * Support for the following configurations:
    ** FreeBSD/alpha (alpha*-*-freebsd*)
    ** GNU/kFreeBSD/alpha (alpha*-*-kfreebsd*-gnu)

For a complete list and more details on each item, please see the
gdb/NEWS file.

(maya)

2017-06-05 10:42:12 UTC MAIN commitmail json YAML

2017-06-05 10:41:28 UTC MAIN commitmail json YAML