New features and improvements
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Only pass clang-specific `-stdlib=` to the preprocessor.
- Improved handling of stale NFS handles.
- Made it harder to misinterpret documentation of boolean environment settings'
semantics.
Bug fixes
~~~~~~~~~
- Include m4 files used by configure.ac in the source dist archives.
- Corrected "Performance" section in the manual regarding `__DATE_`, `__TIME__`
and `__FILE__` macros.
- Fixed build on Solaris 10+ and AIX 7.
- Fixed failure to create directories on QNX.
- Don't (try to) update manifest file in ``read-only'' and ``read-only direct''
modes.
- Fixed a bug in caching of `stat` system calls in ``file_stat_matches
sloppiness mode''.
- Fixed bug in hashing of clang plugins, leading to unnecessary cache misses.
- Fixed --print-config to show ``pch_defines sloppiness''.
- The man page is now built when running ``make install'' from Git repository
sources.
Hypothesis is a library for testing your Python code against a much
larger range of examples than you would ever want to write by hand.
It's based on the Haskell library, Quickcheck, and is designed to
integrate seamlessly into your existing Python unit testing work
flow.
Hypothesis is both extremely practical and also advances the state
of the art of unit testing by some way. It's easy to use, stable,
and extremely powerful. If you're not using Hypothesis to test your
project then you're missing out.
v1.6
====
* `ffi.list_types()`_
* `ffi.unpack()`_
* `extern "Python+C"`_
* in API mode, ``lib.foo.__doc__`` contains the C signature now. On
CPython you can say ``help(lib.foo)``, but for some reason
``help(lib)`` (or ``help(lib.foo)`` on PyPy) is still useless; I
haven't yet figured out the hacks needed to convince ``pydoc`` to
show more. (You can use ``dir(lib)`` but it is not most helpful.)
* Yet another attempt at robustness of ``ffi.def_extern()`` against
CPython's interpreter shutdown logic.
.. _`ffi.list_types()`: ref.html#ffi-list-types
.. _`ffi.unpack()`: ref.html#ffi-unpack
.. _`extern "Python+C"`: using.html#extern-python-c
XXX: standard download URL doesn't work?!
copied to ftp.netbsd.org manually
2016-04-20 meld 3.15.3
======================
Fixes:
* Handle unicode dates in non-UTF8 locales in folder comparison (Kai
Willadsen)
* Try to auto-detect when to use our dark theme highlighting (Kai Willadsen)
* After undo/redo actions in file comparisons, place the cursor at the
natural edit position and scroll it back onscreen (Kai Willadsen)
2.000003 - 2016-04-21
- don't install subs if importing into a package that is already a role. This
can happen if the module previously imported Moo::Role.
Upsream changes:
v1.3.0 2015-09-24 (Evan Giles <egiles@cpan.org>)
- Announce the deprecation the procedural functions (long overdue)
- Remove the 'use 5.6.2' statements, they cause warnings.
- Fix some spelling errors
- Move the xt tests into the main test directory
- Protect against strange PERL5LIB chars (CPAN-RT 104614)
Overview of changes between 1.40.0 and 1.40.1
=============================================
- Use $PKG_CONFIG consistently (#764191)
- Fix a PangoWin32FontMap memory leak (#7640890)
number of fixes and improvements compared to the GSOC version. Most
notable for pkgsrc is that it now works on Darwin.
However it is still version 0, and there is no ChangeLog, so all we
can do is bump the PKGREVISION.
2.000003 - 2016-04-19
- update bundled ExtUtils::HasCompiler to 0.013 to fix potential false
negative (RT#113637)
- list optional XS dependencies as suggests rather than recommends (RT#107393)
2.000002 - 2016-04-19
- restore compatibility with Moo versions pre 1.004_003
- delay loading Class::Method::Modifiers until applying modifiers to a package
- use croak rather than die for reporting errors
- apply method modifiers only once, even if they are applied via multiple
composition paths (RT#106668)
2016-04-19 k <andk@cpan.org>
* release 2.17
* address #113886: unlist Compress::Zlib as a prereq, it was and
still is only used by a test that won't run for normal user
installs (Thanks to Graham Knop for reporting)
- Switch to using XZ-compressed tarballs
- build: Make autoconf-archive support optional
- gnome-autogen: Fix the package name tracing
- macros2: Deprecate GNOME_CXX_WARNINGS
- macros2: Deprecate GNOME_COMPILE_WARNINGS
- gnome-autogen: warn if $srcdir is undefined
- gnome-autogen: warn for deprecated variables
- gnome-autogen: infer PKG_NAME from configure.ac
- README: remove check for automake version in example
- compiler-warnings: Add -Werror=return-type
- macros2: Remove support for gnome-doc-utils
- doc-build: Remove doc-build infrastructure
- macros2: Remove GNOME_COMMON_INIT
XXX is SUBST_CLASSES.Darwin+= libname still needed?
Changes in dconf 0.26.0
========================
None.
Changes in dconf 0.25.1
========================
- the libdbus-1 backend has been removed. dconf now always uses GDBus.
- support has been added for system administration frameworks to set up
the dconf profile via a file placed in the XDG_RUNTIME_DIR or in
/run/dconf/. In the case of the file in /run/dconf/, it is not
possible to modify the proile, even via the DCONF_PROFILE environment
variable, which makes it slightly more difficult to evade lockdown.
- directory resets are now implemented properly in DConfChangeset which
means that reading a subkey through a changeset that contains a reset
for a parent directory of that key will return TRUE with a NULL value
- a new API dconf_client_read_full() has been added which allows
reading the user value, the default value, or querying what the
effective value would be if a changeset were to be applied.
- a new API has been added for listing the locks that are in effect:
dconf_client_list_locks()
- DConfClient has a new "writability-changed" signal
- support for reading default values and listing locks have been added
to the dconf commandline tool
- support for g_autoptr() has been added for DConfClient and
DConfChangeset
- the handling of reading of default values via GSettings has been made
more efficient. More major changes to the GSettings backend are
expected in the near future.
Changes in dconf 0.24.0
========================
The version number was increased and a new entry was added to the NEWS.
Changes in dconf 0.23.2
========================
- remove dconf-editor manpage (accidentally missed during the split)
- fix whitespace issues in 'dconf --help'
Changes in dconf 0.23.1
========================
- dconf-editor is now in a separate package
- portability improvements
-------------------------------------------------
7.12 Tue Apr 19 12:24:41 BST 2016
Enhancements:
- version ranges are now supported for PREREQS, etc.
- Metadata is now represented internally as Meta Spec 2.0
- ExtUtils::Command has been re-incorporated at 1.19 of that module
- Refactored XS handling
- XSMULTI=>1 - put multiple *.xs under lib, it "just works" and XSBUILD
for refined control of XSMULTI
- can do "make test" without first doing "make"
Bug fixes:
- Handle new warnings from File::Path
- Resolve RT#106572 specifying AUTHOR via command-line is broken
- Warning on missing TEST_REQUIRES and CONFIGURE_REQUIRES
- Sanitise make_type on Win32
- Cygwin rebase fixes
- Makefile starting comments reflect decoded @ARGV, not raw
- Add various targets to .PHONY to avoid disk IO with dmake
- Fixed race condition in realclean
- improve static-build lib detection
- Eliminate non-error STDERR
- Make WriteEmptyMakefile Makefile functional when called in subdir
- manifypods fixes
- perllocal.pod generation "Perl in Space" fix
- PASTHRU fixes
- Fix distsignature dependencies for parallel make
- Check exit status for commands in "make ci" target
- Less noisey output during building sub-modules
- Fix dos2unix() on Windows
- stop makeaperl from polluting @ARGV in cases where ARGV contains args with spaces
- Fix regression when both test.pl and t/*.t are present
- Refactored internals to remove DirHandle usage
- MM_Unix::find_perl() dont repeatedly stat the same path in a loop
- No longer repeatedly attempt to load CPAN::Meta if it is now available
VMS fixes:
- Made MM_VMS::oneline build continuation lines properly
- Implemented XSMULTI and XSBUILD
- Resurrect PASTHRU on VMS
- make_macro should handle multiple macros
- Fix regression with File::Spec changes in previous release
Win32 fixes:
- t/echo.t needs SHELL env for Win32 gmake
Dist fixes:
- Made %ExtraPrereqs match bundled prereqs
- Included MANIFEST.SKIP from ExtUtils::Manifest
- The bundled Encode::Locale has been updated to 1.04
Test fixes:
- test PL_FILES of a "module"
- Various tests no longer require a separate .pm file for testing
- Support v5.6.1 in various tests
- test static build if $ENV{AUTHOR_TESTING}
- XS tests now pluggable
- test for "Perl in Space"
Doc fixes:
- better document for PL_FILES, oneliner method
- FAQ updated
7.11_06 Tue Mar 29 19:22:38 BST 2016
Bug fixes:
- Backported change from blead for Win32 miniperl
7.11_05 Sat Mar 19 09:41:02 GMT 2016
Bug fixes:
- Less noisey output during building sub-modules
- Fix dos2unix() on Windows
VMS fixes:
- Implemented XSMULTI and XSBUILD
- Resurrect PASTHRU on VMS
7.11_04 Mon Feb 15 11:20:14 GMT 2016
Enhancements:
- Added find_tests_recursive_in() method
Bug fixes:
- Fix regression when both test.pl and t/*.t are present
- Refactored internals to remove DirHandle usage
VMS fixes:
- make_macro should handle multiple macros
- Fix regression with File::Spec changes in previous release
Win32 fixes:
- t/echo.t needs SHELL env for Win32 gmake
7.11_03 Wed Nov 25 15:23:25 GMT 2015
Bug fixes:
- stop makeaperl from polluting @ARGV in cases where ARGV contains args with spaces
7.11_02 Sat Nov 21 20:05:45 GMT 2015
Bug fixes:
- MM_Unix::find_perl() dont repeatedly stat the same path in a loop
7.11_01 Thu Nov 12 11:58:58 GMT 2015
Includes all the Changes noted for v7.06 and the following:
Bug fixes:
- Use of intermediate files to store META.* and *.ppd files has been reverted
- No longer repeatedly attempt to load CPAN::Meta if it is now available
- Made MM_VMS::oneline build continuation lines properly
-------------------------------------------
version 2.64 at 2016-04-14 09:58:30 +0000
-----------------------------------------
Updated for v5.24.0-RC1
-----------------------------------------------------------------
Overview of changes in Glib::Object::Introspection 0.040
========================================================
* Makefile.PL: Make G:O:I dependent on Glib version >= 1.320
* Makefile.PL: Module marked as "stable"; no code changes since 0.033
Overview of changes in Glib::Object::Introspection 0.033
========================================================
* perli11ndoc: Add a path bar to the results display
Overview of changes in Glib::Object::Introspection 0.032
========================================================
* Makefile.PL/setup.pl: change library extensions on OS X
* Docs: include an edited version of Gtk2::api
* perli11ndoc: hush a warning when run in text mode without args
* perli11ndoc: display a synopsis for callables
* perli11ndoc: make Control+k focus the search entry
* perli11ndoc: make Right/Left expand/collapse rows in the tree view
* perli11ndoc: make the result text view readonly
* Croak in setup for unhandled modificaion combinations
* perli11ndoc: when filtering, show children of matching elements
Overview of changes in Glib::Object::Introspection 0.031
========================================================
* Add perli11ndoc, an interactive documentation viewer
* Added perli11ndoc to MANIFEST
Overview of changes in Glib::Object::Introspection 0.030
========================================================
* Avoid repeating setting up a library as this can lead to issues.
* Add some missing version guards to the tests
* Added t/variants.t to MANIFEST
Overview of changes in Glib::Object::Introspection 0.029
========================================================
* Hook up the generic Glib::Boxed constructor on load, not in setup()
* Add support for marshalling GVariants.
* Add support for marshalling GArrays, GPtrArrays and GByteArrays.
* Support flat arrays when converting from C to Perl.
Overview of changes in Glib::Object::Introspection 0.028
========================================================
* Harmonize the format of type names in error messages
* Add a FIXME about a corner case of GInitiallyUnowned handling
--------------------------------------------
Overview of changes in Glib 1.321 (stable)
============================================
* GBoxed.xs: Check Glib version prior to enabling GBoxed types; RT#111150;
patch submitted by Torsten Lüttgert <st-cpan at luettgert.de>
Overview of changes in Glib 1.320 (stable)
============================================
* Makefile.PL: added IRC link to metadata block
* Glib::GenPod: supply default namespace blurbs (Bugzilla #753468)
* Properly escape dash in generated API index POD (Bugzilla #753467)
Overview of changes in Glib 1.314 (unstable)
============================================
* Fix compilation on libglib < 2.26
* Add Glib::Bytes, a wrapper for GBytes
* Sort the output of podify_signals in a deterministic way (BZ#743863)
Overview of changes in Glib 1.313 (unstable)
============================================
* Fix compilation and test issues with Glib::Variant::ParseError
* Add a short explanatory comment to the lazy-loading system
Overview of changes in Glib 1.312 (unstable)
============================================
* Fix incorrect version in Glib::GenPod
Overview of changes in Glib 1.311 (unstable)
============================================
* Add the GVariant converters to the win32 linker export list
* Fix Glib::Variant::classify on big-endian machines
Overview of changes in Glib 1.310 (unstable)
============================================
* Add Glib::Variant and Glib::VariantType
* Add char_byte and char_byte_ornull typemaps
* Reorder and reformat our header and typemap a little
* Hush a compiler warning in GMainLoop.xs
Overview of changes in Glib 1.308 (stable)
==========================================
* Release to replace stable version of Glib deleted from CPAN by accident;
closes RT#110119. There are no code changes for this release.
+BUILD_DEPENDS+= sqlite3>=3.10.0:../../databases/sqlite3
+BUILD_DEPENDS+= p5-Mouse-[0-9]*:../../devel/p5-Mouse
- Move following lines to be grouped with other lines
BUILD_DEPENDS+= p5-Test-Compile>=0.11:../../devel/p5-Test-Compile
- Add following lines for make test
+BUILD_DEPENDS+= p5-Test-Requires-Git-[0-9]*:../../devel/p5-Test-Requires-Git
+BUILD_DEPENDS+= p5-namespace-clean-[0-9]*:../../devel/p5-namespace-clean
(upstream)
- Updated to devel/p5-Git-Repository-Plugin-AUTOLOAD-1.003
-----------------------------------------------------------
1.003 2016-04-06 BOOK
- Use test_requires_git (Test::Requires::Git) instead of has_git (Test::Git)
- Replaced t/00-load.t by the Test::Compile dzil plugin
- Updated release header lines to follow CPAN::Changes::Spec
Changelog:
The NSS team has released Network Security Services (NSS) 3.23, which is a minor
release.
The following security-relevant bug has been resolved in NSS 3.23.
Users are encouraged to upgrade immediately.
* Bug 1245528 (CVE-2016-1950):
Fixed a heap-based buffer overflow related to the parsing of certain ASN.1
structures. An attacker could create a specially-crafted certificate which,
when parsed by NSS, would cause a crash or execution of arbitrary code with
the permissions of the user.
New functionality:
* ChaCha20/Poly1305 cipher and TLS cipher suites now supported
(bug 917571, bug 1227905)
* Experimental-only support TLS 1.3 1-RTT mode (draft-11).
This code is not ready for production use.
New Functions:
* SSL_SetDowngradeCheckVersion - Set maximum version for new ServerRandom
anti-downgrade mechanism
Notable Changes:
* The copy of SQLite shipped with NSS has been updated to version 3.10.2
(bug 1234698)
* The list of TLS extensions sent in the TLS handshake has been reordered
to improve compatibility of the Extended Master Secret feature
with servers (bug 1243641)
* The build time environment variable NSS_ENABLE_ZLIB has been renamed
to NSS_SSL_ENABLE_ZLIB (Bug 1243872).
* The build time environment variable NSS_DISABLE_CHACHAPOLY was added,
which can be used to prevent compilation of the ChaCha20/Poly1305 code.
* The following CA certificates were Removed
- Staat der Nederlanden Root CA
- NetLock Minositett Kozjegyzoi (Class QA) Tanusitvanykiado
- NetLock Kozjegyzoi (Class A) Tanusitvanykiado
- NetLock Uzleti (Class B) Tanusitvanykiado
- NetLock Expressz (Class C) Tanusitvanykiado
- VeriSign Class 1 Public PCA – G2
- VeriSign Class 3 Public PCA
- VeriSign Class 3 Public PCA – G2
- CA Disig
* The following CA certificates were Added
- SZAFIR ROOT CA2
- Certum Trusted Network CA 2
* The following CA certificate had the Email trust bit turned on
- Actalis Authentication Root CA
The full release notes, including the SHA256 fingerprints of the changed
CA certificates, are available at
https://developer.mozilla.org/en-US/docs/Mozilla/Projects/NSS/NSS_3.23_release_notes
This module is meant as a debugging aid. It can be used to make
a script complain loudly with stack backtraces when warn()ing
or die()ing. Unlike other similar modules (e.g. Carp::Always),
it includes stack traces even when exception objects are thrown.
pkg-config 0.29.1
=================
- Fixed a regression from 0.29 with unquoting values queried with
--variable. In some cases, this would cause shell special characters to
be escaped in ways they weren't before. Instead, the unquoting only
occurs if the value appears to be quoted. (#93284)
- Add support for building pkg-config with Microsoft Visual Studio.
Thanks to Chun-wei Fan for the fix. (#92489)
- Allow overriding pkg-config variables with environment variables. By
setting an environment variable of the form
PKG_CONFIG_$PACKAGE_$VARIABLE, a pkg-config variable can be set
globally without always having to pass --define-variable. Thanks to
Alex Larsson for the fix. (#90917)
- Honor -Wl,-framework in addition to -framework so that multiple
frameworks are handled on OSX. (#1278)
- Fix the OSX build using --with-internal-glib. Thanks to Rudá Moura for
the initial fix and Adam Mercer for testing the final patch. (#92902)
Changes from Ant 1.9.6 TO Ant 1.9.7
===================================
Changes that could break older environments:
-------------------------------------------
* <exec> and <apply> used to ignore the dir attribute if it was the
same as the current working directory. They now no longer do,
which changes the behavior for vmlauncher="false" which would have
used the project's basedir rather than the current working
directory in that case.
Bugzilla Report 58555
Fixed bugs:
-----------
* ZipOutputStream could cause an ArrayIndexOutOfBoundsException when
adding entries with comments. This never happens when using Ant as
a build tool but may affect users using Ant's zip package as a
library.
* <gunzip> and <bunzip2> didn't work for non-filesystem resources.
* <jar> ignored the zip64Mode attribute when creating manifest-only
jars. This resulted in jar files that couldn't be read by Java5.
Bugzilla Report 58428
* <untar> will now detect GNU tar longname/link records even if they
don't use the names used by GNU tar itself. star is known to
create archives of that kind.
https://issues.apache.org/jira/browse/COMPRESS-324
* <script> could fail to find javax.script on JDK9 with Jigsaw.
Bugzilla Report 58271
* <import> sometimes failed to normalize file names which could lead
to files getting imported twice which in turn could lead to targets
getting executed twice.
Bugzilla Report 58886
* TarInputStream now properly extraxt directory entries with a
non-zero size.
* <union> would drop file resources pointing to the same file but
using different names.
Bugzilla Report 57965
* <replace>'s result was undefined when using the replacefilterfile
attribute and some keys inside the properties files have been
substrings of other keys. Keys are now sorted by descending size.
Bugzilla Report 58997
* Ant fails to run when arguments contain double-quote character.
Bugzilla Report 58898
* Ant fails to run if ANT_HOME contains a double-quote character.
Bugzilla Report 58874
* Definer's way to parse URLs from classloader breaks with
recent Java 9 builds (b108).
Bugzilla Report 59130
* <ant> and <antcall> now preserve the ext status set by an <exit>
task.
Bugzilla Report 59228
Other changes:
--------------
* <scp> now supports (filesystem-only) resource collections in
addition to filesets.
Bugzilla Report 50769
* The <http> condition has a new optional attribute followRedirects.
Bugzilla Report 58840
* <java> now supports Java9 modules.
https://github.com/apache/ant/pull/15
* <javac> now supports Java9 modules.
https://github.com/apache/ant/pull/16
* <sshexec> and <scp> have two new attributes serverAliveInterval and
serverAliveCountMax that can be used to keep a intermediaries from
closing idle connections.
Bugzilla Report 59162
v20.9.0
-------
* #548: Update certify version to 2016.2.28
* #545: Safely handle deletion of non-zip eggs in rotate
command.
v20.8.1
-------
* Issue #544: Fix issue with extra environment marker
processing in WorkingSet due to refactor in v20.7.0.
v20.8.0
-------
* Issue #543: Re-release so that latest release doesn't
cause déjà vu with distribute and setuptools 0.7 in
older environments.
v20.7.0
-------
* Refactored extra enviroment marker processing
in WorkingSet.
* Issue #533: Fixed intermittent test failures.
* Issue #536: In msvc9_support, trap additional exceptions
that might occur when importing
``distutils.msvc9compiler`` in mingw environments.
* Issue #537: Provide better context when package
metadata fails to decode in UTF-8.
v20.6.8
-------
* Issue #523: Restored support for environment markers,
now honoring 'extra' environment markers.
1.9.1
-----
* Handle IndexError during version string parsing
1.9.0
-----
* Updated from global requirements
* package: fix wrong catch in email parsing
* Do not convert git tags when searching history
* Include wsgi_scripts in generated wheels
* Correct the indentation in the classifiers example
* Do not silently drop markers that fail to evaluate
* Clarifications around tags and version numbers
* Correct typo - s/enabeld/enabled/
* Use apt-cache generated packages to provide build deps
* fix some variable names
* Don't attempt to test with 0.6c11 with Py3
* Support entry point patching on setuptools < 12
* Updated from global requirements
* Split changelog on nulls instead of (
* Add libjpeg and liberasurecode for tests
* Handle markers to support sdist on pip < 6
* Deprecated tox -downloadcache option removed
* passenv integration environment variables re-enabling integration tests
* Enable pep8 H405 tests
* Add patch to properly get all commands from dist
* doc: Remove 'MANIFEST.in'
* doc: Trivial cleanup of 'index.rst'
* doc: Add deprecation note for 'requirements-pyN'
* doc: Restructure 'Requirements' section
* doc: Restructure 'Usage' section
* doc: Add details of manifest generation
* Support git://, git+ssh://, git+https:// without -e flag
* More support Sphinx >=1.3b1 and <1.3.1
* Fix docs for markers
* Do not error when running pep8 with py3
* Ensure changelog log output is written if it already exists
* Cleanup jeepyb and pypi-mirror special casing
1.20 2015-10-04
- Fix RT #107304: Newer versions of File::Path cause warning "_Inline for _Inline: No such file or directory at Module/ScanDeps.pm line 1339."
- drop the dubious call to rmtree()
- Fix RT106142: Preload dependencies for PDL and PDL::NiceSlice
- adopted from a patch by Shawn Laffan, thanks Shawn!
- Fix RT#106144: Preload dependencies for File::BOM)
- adopted from a patch by Shawn Laffan, thanks Shawn!
- Revise our stance on utf8.pm:
- A line of "use utf8;" just means "this file is encoded in UTF-8"
and should _not_ result in scanning utf8.pm which will pull in
the whole Unicode shebang (propery tables and what not).
Yes, utf8.pm _doesn contain "require utf8_heavy.pl", but only inside
an AUTOLOAD() that is _not_ triggered by calling functions
like utf8::is_utf8().
- OTOH the innocently looking one-liner
perl -ne 'print if /\pN/'
implicitly loads utf8.pm and triggers the AUTOLAD().
- So prevent utf8.pm from being scanned and make utf8_heavy.pl
the indicator for "I need the Unicode stuff" instead.
- Cache the results of _get_preload('utf8_heavy.pl').
- Make %Preload "transitive" so that given
my %Preload = (
'Foo.pm' => [ 'Bar.pm' ],
'Bar.pm' => [ 'Quux.pm' ],
...
);
scan_deps_static() register a dependency on Bar.pm _and_
Quux.pm when it saw "use Foo;"
- Minor changes:
- drop dubious %Preload of utf8.pm for SOAP::Lite and XML::Parser::Expat
- drop code for Perl < 5.008 as we require 5.8.1 already
- rework the implementation of -x/-c
- add add_preload_rule() to dynamically add a %Preload rule
- recognize constructs like "open FH, '<:via(Foo)', ..."
- upgrade to Module::Install 1.16
Upstream changes:
Version 0.8.3
------------------------------------------------------------------------------
* Added an argument `filename` to evaluate() and parse_all() (thanks,
@flying-sheep, #58).
Version 0.8
------------------------------------------------------------------------------
* Changed package license to MIT.
Version 0.7.2
------------------------------------------------------------------------------
* replay() fails to replay certain objects such as NULL (#53).
Version 0.7
------------------------------------------------------------------------------
* R 3.0.2 is the minimal required version for this package now.
Version 0.6
------------------------------------------------------------------------------
* Plots are no longer recorded when the current graphical device has been
changed, which may introduce issues like yihui/knitr#824.
* `parse_all()` can parse R code that contains multibyte characters correctly
now (#49, yihui/knitr#988)
+# for make test
+BUILD_DEPENDS+= p5-Git-Version-Compare-[0-9]*:../../devel/p5-Git-Version-Compare
+BUILD_DEPENDS+= p5-namespace-clean-[0-9]*:../../devel/p5-namespace-clean
Test::Requires::Git checks if the version of Git available for testing
meets the given requirements.
The "current git" is obtained by running git --version (so the first git
binary found in the PATH will be tested).
If the checks fail, then all tests will be skipped.
Cyclic dependency issue is now solved.
0.16 2016-03-21
- The subs installed into the caller are now named with Sub::Name. This makes
these subs appear to be part of the caller, as opposed to an import, which
is what we want, since each installed sub is constructed uniquely for a
given package.
Simply loading this module makes your constructors "strict". If
your constructor is called with an attribute init argument that
your class does not declare, then it dies. This is a great way to
catch small typos.
0.026 2016-01-27 03:13:25Z
- fixed use of modules loaded by t/zzz-check-breaks.t
- fix stderr suppression on perl 5.6
0.025 2016-01-24 02:59:28Z
- prereqs lowered from required to suggested:
- CPAN::Meta::Check
- CPAN::Meta::Requirements
- tests are now quieter to avoid causing confusion, by suppressing
the printing of stderr in tests
0.120 2015-11-27
- no changes since 0.119_01
0.119_01 2015-11-19
- add none() test; it's like any(), but negative
- fix stringification of any() expectations
t/01-simple.t ..... ld: cannot find -lfoo
error building /tmp/DGF3X5mKcu/foo from /tmp/DGF3X5mKcu/test.o
at /usr/pkg/lib/perl5/5.22.0/ExtUtils/CBuilder/Base.pm line
320, <DATA> line 226.
(upstream)
- Updated to devel/p5-ExtUtils-LibBuilder-0.08
--------------------------------------------
0.08 Mon Oct 19 17:46:02 WEST 2015
- Minor spelling fix (thanks to Gregor Herrmann,
from Debian community).
0.07 Mon Sep 28 14:53:40 WEST 2015
- Added minimum required Perl Version (kwalitee++).
0.5.8 -> 0.6.0
=================
Akira TAGOH (20):
Initialize databases just once.
Fix a build fail when --enable-debug
Support BCP47-like locale representation
Add deprecation macros
Use deprecation macros
Add a lookup table to catch up a script/language from language/script
Update to add more files into .gitignore
Fix the behavior on lt_tag_transform() according to tr#35
Fix a segfault if no territory or script is given to lt_tag_convert_from_locale_string()
Add missing header file
Add lt_db_set_val() to change the place of database
Fix memory leaks
missing gobject function for lt_relation_db thing
Support the language alias for the canonicalization
Fix lt_trie_keys() not working
Fix OOM handling
Improve the bootup time
Get rid of deprecated macro
Add docs
Update the library versioning
----------------------------------------
2016-04-11 k <andk@cpan.org>
* release 2.16
* docs only change: create a real link to perlref.pod (Slaven Rezić)
1.06 Wed Jan 27 22:36:55 CST 2016
This release has no changes in functionality.
[FIXES]
* One of the tests used to use CGI.pm, but it was not a declared
dependency, so Perl 5.22+ would fail because CGI.pm is no longer
shipped as core. The test now uses a local module for testing.
Thanks, Graham Ollis.
2.00 2015-10-28
The default client is now '2_0::Direct', for use with Elasticsearch 2.x.
Specify client '1_0::Direct' if using with Elasticsearch 1.x.
Breaking:
* The field parameter to indices.get_field_mapping() has been renamed to fields
New features:
* Added fields param to Bulk helper
* The name parameter to indices.get_template() can accept multiple options
* Added indices.forcemerge() and deprecated indices.optimize()
* The index parameter to indices.open() and indices.close() is required
* Added allow_no_indices, expand_wildcards, and ignore_unavailable params
to indices.flush_synced()
* Added the timeout param to cluster.stats(), nodes.hot_threads(),
nodes.stats(), and nodes.info()
* cluster.health() can accept multiple indices
* Added cat.repositories() and cat.snapshots()
* Added detect_noop param to update()
* search_shards() accepts multi values for index/type
* delete_template() requires an id
* Add fork protection to Scroll and Async::Scroll
Bug fix:
* Added missing debug QS param
0.59 2015-10-05 19:05 UTC
+ Chg : The magic callbacks are now always executed in a new stack.
This should have no negative impact (performance-wise or
behaviour-wise) on your code, but may actually fix some obscure
bugs.
+ Fix : [RT #107294] : Crash during install of Variable-Magic-0.58
running "t\18-opinfo.t".
A recent change in the core implementation of transliterations
caused an undefined behaviour when it interacted with this
module, which resulted in visible stack corruption on Win32
perls compiled with MSVC when the op_info feature was enabled.
This was fixed by using a separate stack for the magic
callbacks.
Thanks Peter Stokes for reporting and providing greatly
helpful feedback.
+ Fix : The object op_info feature will now handle correctly Unicode
transliteration, multideref and custom ops.
0.43 [2015-12-28]
- No Changes from developer build 0.42_1
0.42_1 [2015-12-28]
- Remove Test2/Test-Stream special cases, they are not needed
0.41 [2015-12-21]
- Updated for Test2
-----------------------------------------
0.313 2016-04-11
- release 0.312_001 without further changes
0.312_001 2016-04-07
- fix "check_member" test (RT#113271)
- update Copyright notice
- add license files
- improve compile_if_else/link_if_else by proving whether object
or binary file (compile result) exists
- fix some typos and other POD quirks
- Update README.md from updated POD
---------------------------------------
2016-04-09 k <andk@cpan.org>
* Version 2.11
* address #113615: extend the production rules for compressed file to
.tgz and .tbz. Up to now those did not get a checksum for the uncompress
file (requested by Ralf Neubauer)
* bump the Module::Signature version that decides whether to test or not
to test (Salve J. Nilsen)
* never let t/00signature.t actually fail
so updated ( perl-5.22.0 and perl-5.22.1 does not make difference)
at t/001_load.t line 6.
Tried to use 'CppGuessTest'.
Error: Can't load
for module CppGuessTest:
/WRKOBJDIR/devel/p5-ExtUtils-CppGuess/work/ExtUtils-CppGuess-0.09/
t/makemaker/blib/arch/auto/CppGuessTest/CppGuessTest.so:
Undefined PLT symbol "_ZNSsC1ERKSs" (symnum = 7) at
/usr/pkg/lib/perl5/5.22.0/x86_64-netbsd-thread-multi/DynaLoader.pm
line 193.
(upstream)
- Updated to devel/p5-ExtUtils-CppGuess-0.11
---------------------------------------------
0.11 Thu Sep 10 13:12:17 MDT 2015
- Add leading whitespace to $cflags in _get_cflags (bulk88)
0.10 Tue Sep 08 21:15:30 MDT 2015
- Config's ccflags must always be used. (bulk88)
attrs is an MIT-licensed Python package with class decorators that ease the
chores of implementing the most common attribute-related object protocols.
You just specify the attributes to work with and attrs gives you:
* a nice human-readable __repr__,
* a complete set of comparison methods,
* an initializer,
* and much more
without writing dull boilerplate code again and again.
This gives you the power to use actual classes with actual types in your code
instead of confusing tuples or confusingly behaving namedtuples.
So put down that type-less data structures and welcome some class into your
life!
=============
Version 3.14.1
=============
Release bump to 3.14.1
Fixes
* WnckWindow: Add a new 'type-changed' signal (Sebastian Geiger)
* don't use deprecated style properties (Alberts Muktupāvels)
* pager: use correct state flags (Alberts Muktupāvels)
* selector: avoid warnings from GtkStyleContext (Alberts Muktupāvels)
* tasklist: avoid warnings from GtkStyleContext (Alberts Muktupāvels)
* Back to dependency on gtk+-3.0 >= 3.10
Translators
* Aurimas Černius (lt)
* Cédric Valmary (oc)
* Cheng-Chia Tseng (zh_TW)
* Jordi Mas (ca)
* Maria Mavridou (el)
* Mario Blättermann (de)
* Muhammet Kara (tr)
* Pedro Albuquerque (pt)
* Piotr Drąg (pl)
* Samir Ribic (bs)
2.9.1
=====
**Bug Fixes**
* Improve error message when a plugin fails to load.
Thanks `@nicoddemus`_ for the PR.
* Fix (`#1178 <https://github.com/pytest-dev/pytest/issues/1178>`_):
``pytest.fail`` with non-ascii characters raises an internal pytest error.
Thanks `@nicoddemus`_ for the PR.
* Fix (`#469`_): junit parses report.nodeid incorrectly, when params IDs
contain ``::``. Thanks `@tomviner`_ for the PR (`#1431`_).
* Fix (`#578 <https://github.com/pytest-dev/pytest/issues/578>`_): SyntaxErrors
containing non-ascii lines at the point of failure generated an internal
py.test error.
Thanks `@asottile`_ for the report and `@nicoddemus`_ for the PR.
* Fix (`#1437`_): When passing in a bytestring regex pattern to parameterize
attempt to decode it as utf-8 ignoring errors.
* Fix (`#649`_): parametrized test nodes cannot be specified to run on the command line.
.. _#1437: https://github.com/pytest-dev/pytest/issues/1437
.. _#469: https://github.com/pytest-dev/pytest/issues/469
.. _#1431: https://github.com/pytest-dev/pytest/pull/1431
.. _#649: https://github.com/pytest-dev/pytest/issues/649
.. _@asottile: https://github.com/asottile
2.9.0
=====
**New Features**
* New ``pytest.mark.skip`` mark, which unconditionally skips marked tests.
Thanks `@MichaelAquilina`_ for the complete PR (`#1040`_).
* ``--doctest-glob`` may now be passed multiple times in the command-line.
Thanks `@jab`_ and `@nicoddemus`_ for the PR.
* New ``-rp`` and ``-rP`` reporting options give the summary and full output
of passing tests, respectively. Thanks to `@codewarrior0`_ for the PR.
* ``pytest.mark.xfail`` now has a ``strict`` option, which makes ``XPASS``
tests to fail the test suite (defaulting to ``False``). There's also a
``xfail_strict`` ini option that can be used to configure it project-wise.
Thanks `@rabbbit`_ for the request and `@nicoddemus`_ for the PR (`#1355`_).
* ``Parser.addini`` now supports options of type ``bool``.
Thanks `@nicoddemus`_ for the PR.
* New ``ALLOW_BYTES`` doctest option. This strips ``b`` prefixes from byte strings
in doctest output (similar to ``ALLOW_UNICODE``).
Thanks `@jaraco`_ for the request and `@nicoddemus`_ for the PR (`#1287`_).
* Give a hint on ``KeyboardInterrupt`` to use the ``--fulltrace`` option to show the errors.
Fixes `#1366`_.
Thanks to `@hpk42`_ for the report and `@RonnyPfannschmidt`_ for the PR.
* Catch ``IndexError`` exceptions when getting exception source location.
Fixes a pytest internal error for dynamically generated code (fixtures and tests)
where source lines are fake by intention.
**Changes**
* **Important**: `py.code <http://pylib.readthedocs.org/en/latest/code.html>`_ has been
merged into the ``pytest`` repository as ``pytest._code``. This decision
was made because ``py.code`` had very few uses outside ``pytest`` and the
fact that it was in a different repository made it difficult to fix bugs on
its code in a timely manner. The team hopes with this to be able to better
refactor out and improve that code.
This change shouldn't affect users, but it is useful to let users aware
if they encounter any strange behavior.
Keep in mind that the code for ``pytest._code`` is **private** and
**experimental**, so you definitely should not import it explicitly!
Please note that the original ``py.code`` is still available in
`pylib <http://pylib.readthedocs.org>`_.
* ``pytest_enter_pdb`` now optionally receives the pytest config object.
Thanks `@nicoddemus`_ for the PR.
* Removed code and documentation for Python 2.5 or lower versions,
including removal of the obsolete ``_pytest.assertion.oldinterpret`` module.
Thanks `@nicoddemus`_ for the PR (`#1226`_).
* Comparisons now always show up in full when ``CI`` or ``BUILD_NUMBER`` is
found in the environment, even when ``-vv`` isn't used.
Thanks `@The-Compiler`_ for the PR.
* ``--lf`` and ``--ff`` now support long names: ``--last-failed`` and
``--failed-first`` respectively.
Thanks `@MichaelAquilina`_ for the PR.
* Added expected exceptions to ``pytest.raises`` fail message.
* Collection only displays progress ("collecting X items") when in a terminal.
This avoids cluttering the output when using ``--color=yes`` to obtain
colors in CI integrations systems (`#1397`_).
**Bug Fixes**
* The ``-s`` and ``-c`` options should now work under ``xdist``;
``Config.fromdictargs`` now represents its input much more faithfully.
Thanks to `@bukzor`_ for the complete PR (`#680`_).
* Fix (`#1290`_): support Python 3.5's ``@`` operator in assertion rewriting.
Thanks `@Shinkenjoe`_ for report with test case and `@tomviner`_ for the PR.
* Fix formatting utf-8 explanation messages (`#1379`_).
Thanks `@biern`_ for the PR.
* Fix `traceback style docs`_ to describe all of the available options
(auto/long/short/line/native/no), with `auto` being the default since v2.6.
Thanks `@hackebrot`_ for the PR.
* Fix (`#1422`_): junit record_xml_property doesn't allow multiple records
with same name.
.. _`traceback style docs`: https://pytest.org/latest/usage.html#modifying-python-traceback-printing
.. _#1422: https://github.com/pytest-dev/pytest/issues/1422
.. _#1379: https://github.com/pytest-dev/pytest/issues/1379
.. _#1366: https://github.com/pytest-dev/pytest/issues/1366
.. _#1040: https://github.com/pytest-dev/pytest/pull/1040
.. _#680: https://github.com/pytest-dev/pytest/issues/680
.. _#1287: https://github.com/pytest-dev/pytest/pull/1287
.. _#1226: https://github.com/pytest-dev/pytest/pull/1226
.. _#1290: https://github.com/pytest-dev/pytest/pull/1290
.. _#1355: https://github.com/pytest-dev/pytest/pull/1355
.. _#1397: https://github.com/pytest-dev/pytest/issues/1397
.. _@biern: https://github.com/biern
.. _@MichaelAquilina: https://github.com/MichaelAquilina
.. _@bukzor: https://github.com/bukzor
.. _@hpk42: https://github.com/hpk42
.. _@nicoddemus: https://github.com/nicoddemus
.. _@jab: https://github.com/jab
.. _@codewarrior0: https://github.com/codewarrior0
.. _@jaraco: https://github.com/jaraco
.. _@The-Compiler: https://github.com/The-Compiler
.. _@Shinkenjoe: https://github.com/Shinkenjoe
.. _@tomviner: https://github.com/tomviner
.. _@RonnyPfannschmidt: https://github.com/RonnyPfannschmidt
.. _@rabbbit: https://github.com/rabbbit
.. _@hackebrot: https://github.com/hackebrot
1.040 2016-02-24 17:47:00-05:00 America/New_York
[Fixed]
- Fixed duplicated documentation sections.
1.038 2016-02-10 14:15:31-07:00 America/Mazatlan
- No changes from 1.037
1.037 2016-02-05 20:22:34-05:00 America/New_York (TRIAL RELEASE)
[Fixed]
- Fixed t/filescreen.t Unicode string tests to use a backwards
compatible form. Should fix tests before 5.16.
1.035 2016-02-04 14:47:20-05:00 America/New_York (TRIAL RELEASE)
[Changed]
- The default formatter now replaces a code reference argument with
the results of calling the code reference ONLY when it is the first
argument (in place of a format string). Code references in
subsequent arguments (to sprintf) are not executed, as this would
break backwards compatibility.
[Documented]
- Noted that repeatedly calling 'set' to set an adapter without calling
'remove' or using the 'lexically' feature will leak memory.
1.033 2016-02-03 10:32:57-05:00 America/New_York (TRIAL RELEASE)
[Added]
- The default formatter now expands code references. If the first
argument is a code reference, it is expanded and returned. If an
argument to "sprintf" style formatting is a code reference, it is
expanded.
[Changed]
- The File adapter now opens files with the ":utf8" layer. It also
takes a 'binmode' attribute to change the default.
[Fixed]
- does_not_contain_ok test adapter function now gives proper
diagnostic message
- all diagnostic messages that output the captured log now
correctly dump the log with pretty formatting
[Documented]
- documented the 'proxy_class' argument to `get_logger`
[~Internal~]
- Data::Dumper is loaded lazily, to reduce module load times
for programs that don't need it.
0.6.0 (2016-XX-XX)
------------------
- Fix working with recent versions of setuptools.
- Python versions < 2.7 support was dropped.
Currently Python 2.7 and Python 3.4, 3.5 versions are supported
(should work in other Python 3 releases too, but it is not tested).
- Package "zip-safety" is determined by setuptools.
- New package home page is https://github.com/rutsky/setuptools-trial,
please report issues there!
Internal changes:
- Getting version from Darcs VCS was removed.
- Remove trash files (prebuild eggs, PKG-INFO).
- Source code is not PEP8 and pyflakes conformant.
- Simple integration tests were added.
v20.6.7
-------
* Issue #523: Disabled support for environment markers
introduced in v20.5.
v20.6.6
-------
* Issue #503: Restore support for PEP 345 environment
markers by updating to Packaging 16.6.
v20.6.0
-------
* New release process that relies on
`bumpversion <https://github.com/peritus/bumpversion>`_
and Travis CI for continuous deployment.
* Project versioning semantics now follow
`semver <https://semver.org>`_ precisely.
The 'v' prefix on version numbers now also allows
version numbers to be referenced in the changelog,
e.g. https://pythonhosted.org/setuptools/history.html#v20-6-0.
20.5
----
* BB Pull Request #185: Add support for environment markers
in requirements in install_requires, setup_requires,
tests_require as well as adding a test for the existing
extra_requires machinery.
20.4
----
* Issue #422: Moved hosting to
`Github <https://github.com/pypa/setuptools>`_
from `Bitbucket <https://bitbucket.org/pypa/setuptools>`_.
Issues have been migrated, though all issues and comments
are attributed to bb-migration. So if you have a particular
issue or issues to which you've been subscribed, you will
want to "watch" the equivalent issue in Github.
The Bitbucket project will be retained for the indefinite
future, but Github now hosts the canonical project repository.
20.3.1
------
* Issue #519: Remove import hook when reloading the
``pkg_resources`` module.
* BB Pull Request #184: Update documentation in ``pkg_resources``
around new ``Requirement`` implementation.
20.3
----
* BB Pull Request #179: ``pkg_resources.Requirement`` objects are
now a subclass of ``packaging.requirements.Requirement``,
allowing any environment markers and url (if any) to be
affiliated with the requirement
* BB Pull Request #179: Restore use of RequirementParseError
exception unintentionally dropped in 20.2.
0.24 (2016-03-xx)
=================
Features added
--------------
* PEP 498: Literal String Formatting (f-strings).
Original patch by Jelle Zijlstra.
https://www.python.org/dev/peps/pep-0498/
* PEP 515: Underscores as visual separators in number literals.
https://www.python.org/dev/peps/pep-0515/
* Parser was adapted to some minor syntax changes in Py3.6, e.g.
https://bugs.python.org/issue9232
* The embedded C code comments that show the original source code
can be discarded with the new directive ``emit_code_comments=False``.
* Cpdef enums are now first-class iterable, callable types in Python.
* Ctuples can now be declared in pure Python code.
* Posix declarations for DLL loading and stdio extensions were added.
Patch by Lars Buitinck.
* The Py2-only builtins ``unicode()``, ``xrange()``, ``reduce()`` and
``long`` are now also available in compile time ``DEF`` expressions
when compiling with Py3.
* Exception type tests have slightly lower overhead.
This fixes ticket 868.
* @property syntax fully supported in cdef classes, old syntax deprecated.
* C++ classes can now be declared with default template parameters.
Bugs fixed
----------
* C++ exceptions raised by overloaded C++ operators were not always
handled. Patch by Ian Henriksen.
* C string literals were previously always stored as non-const global
variables in the module. They are now stored as global constants
when possible, and otherwise as non-const C string literals in the
generated code that uses them. This improves compatibility with
strict C compiler options and prevents non-const strings literals
with the same content from being incorrectly merged.
* Compile time evaluated ``str`` expressions (``DEF``) now behave in a
more useful way by turning into Unicode strings when compiling under
Python 3. This allows using them as intermediate values in expressions.
Previously, they always evaluated to bytes objects.
* ``isinf()`` declarations in ``libc/math.pxd`` and ``numpy/math.pxd`` now
reflect the actual tristate ``int`` return value instead of using ``bint``.
* Literal assignments to ctuples avoid Python tuple round-trips in some
more corner cases.
* Iteration over ``dict(...).items()`` failed to get optimised when dict
arguments included keyword arguments.
* cProfile now correctly profiles cpdef functions and methods.
0.23.5 (2016-03-26)
===================
* Compile errors and warnings in integer type conversion code. This fixes
ticket 877. Patches by Christian Neukirchen, Nikolaus Rath, Ian Henriksen.
* Reference leak when "*args" argument was reassigned in closures.
* Truth-testing Unicode strings could waste time and memory in Py3.3+.
* Return values of async functions could be ignored and replaced by ``None``.
* Compiler crash in CPython 3.6.
* Fix prange() to behave identically to range(). The end condition was
miscalculated when the range was not exactly divisible by the step.
* Optimised ``all(genexpr)``/``any(genexpr)`` calls could warn about unused
code. This fixes ticket 876.
2.54 2016-01-18
- Fixed File::Locked even more. While some deadlocks were eliminated, there
was still an issue when a Log::Dispatch object was created in a parent
process and then recreated in the children. If the close_after_write
parameter was false, then there would be a deadlock.
- Made Syslog work on a threaded Perl when the threads module was not
explicitly loaded. Fixed by Konrad Bucheli. GitHub #13.
- The log() method now accepts a level parameter as an integer. Fixed by Steve
Bertrand. GitHub #15.
2.53 2016-01-14
- Actually fix File::Locked, this time with some actual tests.
2.52 2016-01-13
- Make File::Locked output do an explicit unlock if the close_after_write
option is not set. Without this it would open the file once in each process,
attempting to get a lock at open time, pretty much guaranteeing a deadlock
in a multiprocess environment. Now it should work sanely whether
close_after_write is true or not.
2.51 2015-09-19
- Fixed t/08-screen.t on Windows platforms.
1.62 2015-12-13 NEILB
- Fixed Hack Kampbjørn's surname -- I'd previously switched the ø to an
upper-case Ø -- thanks to pink-mist for pointing this out.
- Updated github repo URL after changing my github username
Library
-------
- Issue #26323: Add Mock.assert_called() and Mock.assert_called_once()
methods to unittest.mock. Patch written by Amit Saha.
- Issue #22138: Fix mock.patch behavior when patching descriptors. Restore
original values after patching. Patch contributed by Sean McCully.
- Issue #24857: Comparing call_args to a long sequence now correctly returns a
boolean result instead of raising an exception. Patch by A Kaptur.
- Issue #23004: mock_open() now reads binary data correctly when the type of
read_data is bytes. Initial patch by Aaron Hill.
- Issue #21750: mock_open.read_data can now be read from each instance, as it
could in Python 3.3.
- Issue #18622: unittest.mock.mock_open().reset_mock would recurse infinitely.
Patch from Nicola Palumbo and Laurent De Buyst.
- Issue #23661: unittest.mock side_effects can now be exceptions again. This
was a regression vs Python 3.4. Patch from Ignacio Rossi
- Issue #23310: Fix MagicMock's initializer to work with __methods__, just
like configure_mock(). Patch by Kasia Jachim.
- Issue #23568: Add rdivmod support to MagicMock() objects.
Patch by Håkan Lövdahl.
- Issue #23581: Add matmul support to MagicMock. Patch by Håkan Lövdahl.
- Issue #23326: Removed __ne__ implementations. Since fixing default __ne__
implementation in issue #21408 they are redundant. *** NOT BACKPORTED ***
- Issue #21270: We now override tuple methods in mock.call objects so that
they can be used as normal call attributes.
- Issue #21256: Printout of keyword args should be in deterministic order in
a mock function call. This will help to write better doctests.
- Issue #21262: New method assert_not_called for Mock.
It raises AssertionError if the mock has been called.
- Issue #21238: New keyword argument `unsafe` to Mock. It raises
`AttributeError` incase of an attribute startswith assert or assret.
- Issue #21239: patch.stopall() didn't work deterministically when the same
name was patched more than once.
- Issue #21222: Passing name keyword argument to mock.create_autospec now
works.
- Issue #17826: setting an iterable side_effect on a mock function created by
create_autospec now works. Patch by Kushal Das.
- Issue #17826: setting an iterable side_effect on a mock function created by
create_autospec now works. Patch by Kushal Das.
- Issue #20968: unittest.mock.MagicMock now supports division.
Patch by Johannes Baiter.
- Issue #20189: unittest.mock now no longer assumes that any object for
which it could get an inspect.Signature is a callable written in Python.
Fix courtesy of Michael Foord.
- Issue #17467: add readline and readlines support to mock_open in
unittest.mock.
- Issue #17015: When it has a spec, a Mock object now inspects its signature
when matching calls, so that arguments can be matched positionally or
by name.
- Issue #15323: improve failure message of Mock.assert_called_once_with
- Issue #14857: fix regression in references to PEP 3135 implicit __class__
closure variable (Reopens issue #12370)
- Issue #14295: Add unittest.mock
NEW IN WAF 1.8.20
-----------------
* Detection SUN cc compiler version 5.0 #1697
* Initialize 'VALAFLAGS' with OS flags #1703
* Convert --perlbinary data to list #1705
* Cppcheck fixes#1708
* Boost module enhancements and fixes#1709
* Various cpplint enhancements #1711
* Workaround for unreadable home folders
* Renderer: Added get/set_alpha().
* Attribute: Added create_attr_background_alpha() and
create_attr_foreground_alpha().
* Reduce the cairomm dependency back to 1.2.2.
Build:
* MacOS X: Add alternative #ifdefed code when thread_local
is not supported by the compiler.
* Replace most typedefs with the C++11 using keyword.
* Documentation: Mention use with CMake.
2.7.2 (unstable):
* Deprecate sigc::ref() and sigc::reference_wrapper(),
adding support instead for std::ref(), std::cref(),
and std::reference_wrapper().
* mem_fun(): Deprecate mem_fun(pointer, func).
Leaving just mem_fun(reference, func).
Please let us know if you disagree strongly with this.
* Make all operator bool() explicit. (A C++11 feature.)
* Build: Remove some now-unnecessary configure checks.
But please let us know if this causes problems for you.
* Build: Update MSVC project.
* Documentation: Improve the documentation of mem_fun(), making it clear that
mem_fun() does not return a slot.
2.7.1 (unstable):
* signal: Add a moving connect() method, taking an rvalue reference
to a slot.
* can_deduce_result_type_with_decltype: Rename the check() methods to
checksize(). check() is a preprocessor macro in Mac OS X.
* Temporarily undefine the nil macro, if it's defined.
nil is a preprocessor macro in Objective-C++ and Mac OS X.
* Correct the mem_functor<> declarations.
- Revive make test (remove TEST_TARGET= # empty line)
Found that expect segfaults at make test if hostname lookup fails,
but it is environment issue only.
(upstream)
- Updated to devel/dejagnu-1.5.3
-------------------------------
2015-03-30 Ben Elliston <bje@gnu.org>
* DejaGnu 1.5.3 released.
2015-03-30 Ben Elliston <bje@gnu.org>
* runtest.exp (frame_version): Set to 1.5.3.
* configure.ac (AC_INIT): Set version to 1.5.3.
* configure: Regenerate.
* doc/dejagnu.xml (appversion, version): Set to 1.5.3.
* doc/dejagnu.texi: Regenerate.
2015-03-30 Ben Elliston <bje@gnu.org>
Revert the following change which causes breakage reported here:
https://lists.gnu.org/archive/html/dejagnu/2015-03/msg00003.html
2011-03-14 Jan Kratochvil <jan.kratochvil@redhat.com>
* runtest.exp (runtest): Call reset_vars at initialisation so that
perror does not affect test results in other test scripts.
2015-02-26 H.J. Lu <hjl.tools@gmail.com>
* lib/target.exp (default_target_compile): Prepend multilib_flags,
instead of appending it.
2015-02-18 Ben Elliston <bje@gnu.org>
* lib/framework.exp (log_and_exit): Set tool to "testrun" if
unset.
2015-02-05 Ben Elliston <bje@gnu.org>
* doc/runtest.1: Fix a spelling error.
2015-02-04 Ben Elliston <bje@gnu.org>
* config.guess: Update to version 2015-01-01.
* config.sub: Likewise.
* Fix Python plugin loader module paths
* Do not add pyexecdir to Python's paths at runtime
* Add support for embedded C plugins
* Order the PeasEngine:plugin-list by dependencies
* Keep plugins order in PeasExtensionSet
* Add g_autoptr() autocleanup support
* Fix Python tests on OpenBSD
* Fix libpeas crash on windows
* Simplify __all__ support in the Python plugin loader
* Removed support for Seed plugins
* Added support for Lua 5.1 plugins using the LGI bindings
* PeasEngine is now thread-safe
* Added support for non-global plugins loaders
* Load C plugins with local linkage
* Use Python to implement the plugin loader's logic
* Support Python plugins that define __all__
* Use Lua to implement the plugin loader's logic
* Prevent the accidental escaping of globals with Lua plugins
* Misc bugfixes
* Many translation updates
Before, at-spi2-core installed accessibility.conf in $(sysconfdir)/at-spi2,
so we used the CONF_FILES mechanism to place it in
share/examples/at-spi2-core/at-spi2, so the administrator can make local
adjustments.
Now, at-spi2-core works with sensible defaults without an
accessibility.conf file, and places it in $(datadir)/defaults/at-spi2,
essentially to be used as an example.
Choose to revert to the old behaviour.
* Don't drop the main context when p2p socket configuration fails
(bgo#763273).
* Make sure the runtime directory exists (bgo#763274).
* Avoid lagging with large gtk menus (bgo#754048).
* Always accept connections as root if unable to find an owning user.
* Test fixes for atk-adaptor and cache changes.
* Fixes for Text, Value, and Table Cell adaptors (bgo#755724).
* Many new tests, thanks to Patryk Kaczmarek of Samsung.
* Update for new caching API (bgo#650090).
* Eliminate some console warnings due to NULL gobjects.
* Tests: fix out-of-source and parallel build (bgo#751137).
* Add some missing files to the distribution (bgo#751138).
* Support a stateless configuration by default (bgo#763540).
* Don't display warnings when connecting to an app that no longer exists.
* Don't display warning if unable to connect when logged in via ssh
(bgo#761600).
* at-spi-bus-launcher: register with session manager (bgo#753931).
* Disable xevie by default--it probably doesn't do anything anyhow.
* get_index_in_parent: Don't crash if parent is defunct.
* Don't crash when trying to set an invalid state (bgo#757915).
* Don't crash if we get a children-changed event with a non-existent
child (bgo#755951).
* atspi_hyperlink_get_index_range: don't return random values if the
call fails (bgo#755727).
* Fixed some atspi_text_ functions (bgo#755731).
* Fixed crash during removal of last application in registryd (bgo#756513).
* Modified the cache API to specify an object's index and child count
rather than its children. This eliminates the need for the application
to enumerate its children, improving performance in some places with
large lists (bgo#650090).
* Functions shouldn't try to return values (bgo#749330).
* Fix atspi_table_cell_get_position.
Library to access the DWARF debugging file format which supports source level
debugging of a number of procedural languages, such as C, C++, and Fortran.
This package contains utility dwarfdump.
- Add following line for make test
+BUILD_DEPENDS+= p5-Scalar-List-Utils>=1.45:../../devel/p5-Scalar-List-Utils
(upstream)
- Updated to devel/p5-Dist-Zilla-5.044
------------------------------------
5.044 2016-04-06 20:32:14-04:00 America/New_York
- require a newer List::Util to avoid a dumb bug caused by relying on
side effects of loading Moose (thanks, Karen Etheridge!)
5.043 2016-01-04 22:54:56-05:00 America/New_York
- dzil test now supports --extended to set EXTENDED_TESTING (thanks,
Philippe Bruhat)
5.042 2015-11-26 09:05:37-05:00 America/New_York
- try to avoid testing errors when the local time zone can't be
determined (https://github.com/rjbs/Dist-Zilla/issues/497)
5.041 2015-10-27 22:07:54-04:00 America/New_York
- add 'static_attribution' attribution to MakeMaker plugin
- fix prereqs for App::Cmd and Config::MVP::Reader::INI
5.040 2015-10-13 11:42:25-04:00 America/New_York
- the distribution tarball name no longer includes -TRIAL if the
version contains an underscore
- [PkgVersion] adds "$VERSION = $VERSION_SANS_UNDERSCORES" when
version contains an underscore
- made the PodCoverageTests and PodSyntaxTests plugins generate author
tests, not release tests. These are tests you want passing on every
commit, not just before a release (Dave Rolsky)
---------------------------------------------------
0.10 2015-09-27 04:28:27Z
- fix invalid prereq specification for Heavy implementation (changed
in 0.09)
----------------------------------------
2.08 2015-10-30 NEILB
- Switched to Dist::Zilla
- Updated the github repo URL after changing my github username
- Removed a spurious =back from Graph::ReadWrite.
Thanks to Lucas Kanashiro for RT#106063
--------------------------------
0.9704 2015-10-07 Jarkko Hietaniemi <jhi@iki.fi>
- rt.cpan.org 107567: edges() missing on undirected multiedged graph:
was broken in 0.96, had been fixed somewhere there and here,
added the test case
- rt.cpan.org 107600: no modify Storable $VERSION
-------------------------------------------------------
1.002 Tue Feb 2 2016
- Prevent system/global configs from messing with test expectations
(Fix for RT #111645, reported by DBOOK)
- Load Config when using %Config
(Fix for RT #100806, reported by PPISAR)
----------------------------------------
2.2.1 2016-02-06
[ BUG FIXES ]
- Test was failing if color.ui was set to always. (GH#29, reported by
grinnz)
[ STATISTICS ]
- code churn: 3 files changed, 24 insertions(+), 13 deletions(-)
----------------------------------
0.11 2015-11-26 NEILB
- Document when reference gras the timestamps, and give example.
Thanks to SMYLERS for the PR with this.
0.10 2015-10-24 NEILB
- Added [MetaJSON] to dist.ini, so we get a META.json in the release
- Auto tag and push to github on release
- Updated repo URL after changing my github username
-------------------------------------
0.8 2016-01-26 14:44:57
- rt 88849: allow multiple processes to operate on the same paths (thanks HELENA!)
- Add github to POD
- rt 96842: Fix mk() to work correctly w/ UNC paths on windows (thanks MITHUN!)
- Add following line for make test
BUILD_DEPENDS+= p5-Test-Exception-[0-9]*:.....
(upstream)
- Update to devel/p5-File-BOM-0.15
--------------------------------
0.15 - Mon Dec 14 2015
- Fix spelling errors spotted by debian team. Thanks to Gregor Herrmann.
0.55 January 03, 2016
Prodded by Chris Kirke during the CPAN Pull Request Challenge, standardized
$VERSION numbers and release dates across all files. No changes in
functionality or tests.
------------------------------------
v1.19 released Dec 12th 2015
----------------------------
* Improved gfortran linking on OS X Yosemite+ which avoids PERL_DL_NONLAZY meltdowns
- Add c++ for USE_LANGUAGE for make test
(upstream)
- Update to devel/p5--ExtUtils-CBuilder-0.28.22.24
------------------------------------------------
0.282224 - 2015-10-09
Enhncements:
- Use warnings/strict on all modules.
----------------------------
1.08 Sun Sep 26, 2015, joern
Notes:
- Just a stable release without changes. I thought
CPAN testers would pick up developer releases
too, but was wrong ;)
1.08_01 Sat Sep 26, 2015, joern
Features:
- Support for multiple serialisers: Sereal, CBOR::XS,
JSON::XS and Storable. That's because Storable is
known to be insecure, so this module should not
rely on it.
Great care has been taken to make these changes
up- and downwards compatible, so old clients still
can connect to new servers and vice versa.
Check documentation chapters MESSAGE FORMAT OPTIONS
in Event::RPC, Event::RPC::Server and Event::RPC::Client
manpages for details.
Thanks to mala <mala AT ma.la> for his hint about the
security issues of Storable. You find more details in
this article:
http://www.masteringperl.org/2012/12/the-storable-security-problem/
Bugfixes:
- Proper handling of exceptions which occur at the
server before a remote method is really executed,
e.g. when lookup of the class failed due to
incorrect @INC path.
- Fixed encoding of source files and corrected
the year of all Copyright notices.
--------------------------
4.22 Sun Dec 20 02:34:39 CET 2015
- (libev) when epoll detects unremovable fds in the fd set, rebuild
only the epoll descriptor, not the signal pipe, to avoid
SIGPIPE in ev_async_send. This doesn't solve it on fork,
so document what needs to be done in ev_loop_fork
(analyzed by Benjamin Mahler).
- (libev) remove superfluous sys/timeb.h include on win32
(analyzed by Jason Madden).
- updated libecb.
-----------------------------------
0.096 2016-03-02
* eval_string() now preserves the value of $@ both inside the eval'd
code and outside (if no exception is thrown).
+# for make test
+BUILD_DEPENDS+= p5-Test-Warnings-[0-9]*:../../devel/p5-Test-Warnings
+# following introduce cyclic dependency
+#UILD_DEPENDS+= p5-Package-Stash-[0-9]*:../../devel/p5-Package-Stash
---------------------------------------
2.094 2016-03-20 BOOK
- Disabled extra debugger features (HAARG)
- Added SEE ALSO section with a discussion of the main modules
on CPAN for determining dependencies. (NEILB)
- Tweaked formatting of this file as per CPAN::Changes::Spec, in the
process ensuring all dates are UTC (like CPAN & PAUSE). (NEILB)
- Simplified the data structure and the caller() call
- Switched to Dist::Zilla for packaging
-----------------------------------------
2.01 2016-03-02
- Fixed the frames method when it is called with arguments. Previously this
did not work if it was called before the method was called as a
reader. Fixed by Mark Fowler. PR #8.
--------------------------------------
Changes in Devel::NYTProf 6.03 - 25th March 2016
Fix to account for the deep stack optimizations in perl 5.23.8
by Dave Mitchell, with my great gratitude!
Fix SIGUSR2 on MSWin32 and div by zero #78 PR#79
with thanks to Nicolas Georges.
Fix for perl 5.8.8 but adding definition for tTHX.
Unify 'Profile format error' messages.
Improve the TROUBLESHOOTING docs.
Added some more docs for the start=init option #85
Corrected URL of KCachegrind home page #87
with thanks to Denis Ibaev.
Corrected URL of sample report #80
Removed dead code.nytimes.com URL and redundant history #84
------------------------
New in 1.34 (Released April 2, 2016)
Several bugs fixed in demo programs (which also function as the
documentation) for forms and menus and comments added explaining need to
keep Perl variables around to avoid invalid memory references.
New in 1.33 (Released October 3, 2015)
Package contains META files.
-------------------------
Version 0.78 (2016-02-14)
-------------------------
* added Clang support to ccconfig
* fix cpan #109978: improve wording/grammar in POD; thanks to
gregor herrmann for providing a patch
* fix cpan #95716: Add support for #warn/#warning; thanks to
Jon DuSaint for providing an initial patch
* update module metadata
----------------------------------------
0.004008 Sat Sep 12 13:16:30 2015
- Promoted $VERSION variable to earlier in source to
attempt to placate cpanminus (thanks, Karen!)
- Added prototype to overridden caller()
- Changed way caller() is overridden, hopefully will no longer
clash with Sub::Uplevel
--------------------------------
2.60 - fix rt.cpan.org#107929: added missing test config.
2.59 - fix rt.cpan.org#107108 by adding support for IncludeOptional.
- clarified documentation on StoreDelimiter.
----------------------------
0.27 2016-03-31
- depend on Module::Pluggable::Object directly instead of assuming it is
bundled with Module::Pluggable (RT #113148)
Upstream changes:
Changes in Rcpp version 0.12.4 (2016-03-22)
Changes in Rcpp API:
New accessors as() and clone() were added to the Nullable class (Dan in PR #423 closing #421)
The Nullable<>::operator SEXP() and Nullable<>::get() now also work for const objects (Kirill Mueller in PR #417).
A subsetting error was fixed (Qiang via #432 closing #431).
Changes in Rcpp Sugar:
Added new Sugar function median() (Nathan in PR #425 closing #424)
Added new Sugar function cbind() (Nathan in PR #447 closing #407)
Changes in Rcpp Attributes:
A plugin for C++14 was added (Dan in PR #427)
Changes in Rcpp Documentation:
An entry was added to the Rcpp-FAQ vignette describing the required packages for vignette building (#422).
Use on OS X was further detailed (James Balamuta in #433 with further review by Bob Rudis).
An entry was added concerning the hard-code limit of arguments to some constructor and function (cf #435).
The Rcpp-FAQ vignette now contains a table of content.
Typos and indentation were corrected in the Rcpp Sugar vignette (#445 by Colin Gillespie).
Changes in Rcpp version 0.12.3 (2016-01-10)
Changes in Rcpp API:
Const iterators now CharacterVector now behave like regular iterators (PR #404 by Dan fixing #362).
Math operators between matrix and scalars type have been added (PR #406 by Qiang fixing #365).
A missing std::hash function interface for Rcpp::String has been addded (PR #408 by Qiang fixing #84).
Changes in Rcpp Attributes:
Avoid invalid function names when generating C++ interfaces (PR #403 by JJ fixing #402).
Insert additional space around & in function interface (PR #400 by Kazuki Fukui fixing #278).
Changes in Rcpp Modules:
The copy constructor now initialized the base class (PR #411 by Joshua Pritikin fixing #410)
Changes in Rcpp Repository:
Added a file Contributing.md providing some points to potential contributors (PR #414 closing issue #413)
Changes in Rcpp version 0.12.2 (2015-11-14)
Changes in Rcpp API:
Correct return type in product of matrix dimensions (PR #374 by Florian)
Before creating a single String object from a SEXP, ensure that it is from a vector of length one (PR #376 by Dirk, fixing #375).
No longer use STRING_ELT as a left-hand side, thanks to a heads-up by Luke Tierney (PR #378 by Dirk, fixing #377).
Rcpp Module objects are now checked more carefully (PR #381 by Tianqi, fixing #380)
An overflow in Matrix column indexing was corrected (PR #390 by Qiang, fixing a bug reported by Allessandro on the list)
Nullable types can now be assigned R_NilValue in function signatures. (PR #395 by Dan, fixing issue #394)
operator<<() now always shows decimal points (PR #396 by Dan)
Matrix classes now have a transpose() function (PR #397 by Dirk fixing #383)
operator<<() for complex types was added (PRs #398 by Qiang and #399 by Dirk, fixing #187)
Changes in Rcpp Attributes:
Enable export of C++ interface for functions that return void.
Changes in Rcpp Sugar:
Added new Sugar function cummin(), cummax(), cumprod() (PR #389 by Nathan Russell fixing #388)
Enabled sugar math operations for subsets; e.g. x[y] + x[z]. (PR #393 by Kevin and Qiang, implementing #392)
Changes in Rcpp Documentation:
The NEWS file now links to GitHub issue tickets and pull requests.
The Rcpp.bib file with bibliographic references was updated.
add experimental SunOS backend support adapted from the upstream RTI
submission from Oracle and initial illumos support from OpenIndiana/Hipster
From the Changelog:
For detailed information about the changes below, please see the git log or
visit: http://log.libusb.info
2015-09-13: v1.0.20
* Add Haiku support
* Fix multiple memory and resource leaks (#16, #52, #76, #81)
* Fix possible deadlock when executing transfer callback
* New libusb_free_pollfds() API
* Darwin: Fix devices not being detected on OS X 10.8 (#48)
* Linux: Allow larger isochronous transfer submission (#23)
* Windows: Fix broken builds Cygwin/MinGW builds and compiler warnings
* Windows: Fix broken bus number lookup
* Windows: Improve submission of control requests for composite devices
* Examples: Add two-stage load support to fxload (#12)
* Correctly report cancellations due to timeouts
* Improve efficiency of event handling
* Improve speed of transfer submission in multi-threaded environments
* Various other bug fixes and improvements
The (#xx) numbers are libusb issue numbers, see ie:
https://github.com/libusb/libusb/issues/16
MAKE_JOBS_SAFE=no given build issues when enabled.
---------------------------------------------
2015-11-15 Jarkko Hietaniemi <jhi@iki.fi>
* The added RLIMIT_ values were not added quite right.
Our own tests didn't detect this, but mod_perl2 ones did
[cpan #108955]
2015-10-21 Jarkko Hietaniemi <jhi@iki.fi>
* Add more RLIMIT_ values from various systems,
especially from Linux 2.6:
RLIMIT_MSGQUEUE
RLIMIT_NICE
RLIMIT_NPTS
RLIMIT_PTHREAD
RLIMIT_RTPRIO
RLIMIT_RTTIME
RLIMIT_SBSIZE
RLIMIT_SIGPENDING
RLIMIT_SWAP
Check your system's setrlimit() documentation,
and BSD::Resource::get_rlimits().
* Add PRIO_THREAD, if available.
Check your system's setpriority() documentation.
* Release 1.2908.
--------------
2.03 2015-10-25 NEILB
- Trimmed down the README and moved some of the history into Changes
- Updated github repo URL after changing my github username
RBTools 0.7.5 Release Notes
Release date: October 26, 2015
New Features
Users can now specify the search path for .reviewboardrc files.
The $RBTOOLS_CONFIG_PATH environment variable now controls the search path.
It’s a list of paths separated by the system’s native path separator
(':' on Linux/MacOS X, ';' on Windows).
These paths are searched prior to any default paths.
This is handy for forcing the .reviewboardrc file in your home directory
to be searched first.
Patch by Nils Philippsen.
Bug Fixes
General
Fixed displaying error output containing non-ASCII characters. (Bug #3907)
On Windows, home directory detection now prefers the %HOME% environment
variable to %APPDATA%.
Prior to this release, %HOME% was always ignored in favor of %APPDATA%,
which caused problems in certain setups.
Patch by Vadim Zeitlin.
rbt api-get
The help output no longer mistakenly shows that – is required before
query arguments.
rbt post
Fixed a crash when using --diff-filename. (Bug #3908)
Git
Git version checking is no longer performed unless Git is installed,
preventing a harmless but visible error message for systems without Git.
Mercurial
Fixed path separator differences when posting a change against Mercurial
on Windows. (Bug #3894)
Patch by Ali Kamali.
Fixed a regression with looking for empty files.
When looking for empty files in a change (which must be handled specially),
RBTools would get a null result, which wasn’t being handled correctly.
Perforce
Fixed posting submitted Perforce changes.
Posting submitted changes for review regressed in 0.7.x. They can now be
posted again.
Fixed posting Perforce paths that are not included in the client view.
(Bug #3864)
It’s once again possible to post a committed revision for review when the
commit contains files not in the client view.
Subversion
Fixed Unicode issues when parsing repository info when using
--repository-url. (Bug #3969)
Patch by Griffin Myers.
Fixed issues processing when a changeset has history scheduled with a
commit.
Patch by Griffin Myers.
Team Foundation Server
Fixed a number of issues with diff generation on TFS.
The method previously used to generate diffs didn’t deal with a lot of
cases that could come up when using different editors or platforms, such
as files that don’t end with a newline or files that start with a
byte-order marker.
RBTools now requires GNU diff for TFS diff generation.
Contributors
Ali Kamali
Anthony Hache
Barret Rennie
Christian Hammond
David Trowbridge
Griffin Myers
Nils Philippsen
Vadim Zeitlin
--------------
1.10 2015-10-26 NEILB
- Switched to Dist::Zilla
- Got rid of tab characters
- Updated github repo URL after changing my github username
- Made the SEE ALSO entries linked, and simplified the formatting
-------------------
0.75: # 2016-03-16T23:41:41+0100
- Thanks to our contributors: Dave Cross, Dean Hamstead, Mark J. Reed, Mohammed El-Afifi,
- Add '--no-patchperl' install option for skip running patchperl.
- Add '--destdir' install option.
- "uninstall" command now also delete associated libs.
- "mirror" command is removed.
- "config" command is removed.
0.74: # 2015-11-24T20:55:02+0100
- A big thanks for our contributers: Aristotle Pagaltzis, Dave Jacoby,
Dylan Cali, John Krull, Kevin Johnson, Nathan Day, Paul Cochrane, Tudor
Constantin, Yanick Champoux, tm604
- support PERLBREW_LIB_PREFIX.
c29813cd46
- prefer https and remove insecure download cli flags.
f49a908c17
- perlbrew now dies when switching to/use a lib that does not exist.
https://github.com/gugod/App-perlbrew/issues/432
- RHEL vendorlib fix https://github.com/gugod/App-perlbrew/issues/409
- improvements with fish integration.
- unbreak when running with perl 5.14 or earlier:
"list" command now correctly display lib names.
- retaining flags in upgrade-perl.
b8d466670b
- variour document clean up and improvements.
------------------
0.010005 - 2015-12-14
- fix fatpacked code to work properly on perl 5.6 in all cases
- removed x_static_install metadata (RT#110299)
This is an out of cycle release to address three security issues:
CVE-2016-3630 Mercurial: remote code execution in binary delta decoding
Mercurial prior to 3.7.3 contained two bounds-checking errors in its binary delta decoder that may be exploitable via clone, push, or pull.
CVE-2016-3068 Mercurial: arbitrary code execution with Git subrepos
Mercurial prior to 3.7.3 allowed URLs for Git subrepos that could result in arbitrary code execution on clone. This is a further side-effect of Git CVE-2015-7545. Reported by Blake Burkhart.
CVE-2016-3069 Mercurial: arbitrary code execution when converting Git repos
Mercurial prior to 3.7.3 allowed arbitrary code execution when converting Git repos with hostile names. This could affect automated conversion services. Reported by Blake Burkhart.
bdiff: (pure) support array.array arrays (issue5130)
convert: add new, non-clowny interface for shelling out to git (SEC)
convert: dead code removal - old git calling functions (SEC)
convert: rewrite calls to Git to use the new shelling mechanism (SEC)
convert: rewrite gitpipe to use common.commandline (SEC)
convert: test for shell injection in git calls (SEC)
files: don't recurse into subrepos without a path or -S (issue5127)
hg: perform update after pulling during clone with share (issue5103)
mq: restrict generated patch name to 75 characters (issue5117)
obsolete: fix n^2 marker computation behavior
parsers: detect short records (SEC)
parsers: fix list sizing rounding error (SEC)
streamclone: fix error when store files grow while stream cloning
subrepo: adapt to git's recent renames-by-default
subrepo: set GIT_ALLOW_PROTOCOL to limit git clone protocols (SEC)
Short for "Metamodel On A Runtime", MoarVM is a virtual machine built
especially for Rakudo Perl 6 and the NQP Compiler Toolchain. The primary
backend target for the Christmas 2015 release of Perl 6, MoarVM stands
out amongst the various Rakudo and NQP compilation targets by:
- Running the Perl 6 specification test suite in the least time
- Passing more of the Perl 6 specification tests than Rakudo Perl 6 on
other backends
- Having the lowest memory usage
- Having the best startup time
- Being fastest to build both NQP and Rakudo - and thus your Perl 6 and
NQP programs too!
- Being the only backend to support strings at grapheme level
Version 6.4
-----------
(bugfix release; released on March 24th 2016)
- Fix bug in bash completion where click would discard one or more trailing
arguments. See #471.
0.16 2016-03-21
- The subs installed into the caller are now named with Sub::Name. This makes
these subs appear to be part of the caller, as opposed to an import, which
is what we want, since each installed sub is constructed uniquely for a
given package.
1.45 -- 2016/03/25 16:09:40
[CHANGES]
* Renamed existing uniq() to uniqstr()
* Canonicalise undef to {empty string,zero} in uniq{str,num}()
* Add a new uniq() with more DWIMish semantics around undef
[BUGFIXES]
* Fix uses of GET magic by the uniq*() family of functions. GET magic
is now always invoked exactly once if it exists.
1.44 -- 2016/03/17 23:08:46
[CHANGES]
* Added List::Util::uniq() and uniqnum()
(with thanks to randir and ilmari for assistance writing code and
tests)
1.23 2016-03-26
- Fixed some Perl stack corruption bugs. Based on a proposed PR from Tony Cook
plus some additional changes. GH #8.
- Fixed tests with Carp 1.01 (shipped with Perl 5.8.3). Patch by Andreas
Koenig. RT #113318.
Version 6.5.4 - Mar 27 2016
This is a bug fix release.
[CHANGES]
Changed:
o gtags.conf: The entries which include '.xml' were commented out,
because they require libxml.a. If you use them, please uncomment them.
[FIXED BUGS]
Serious bug:
o universal-ctags.la: Universal ctags plug-in parser did not work
on almost files. Now it works.
$ echo 'void a() {}' >a.inl
$ echo 'void b() {}' >b.cpp
$ gtags --gtagslabel=new-ctags
$ global -x '.*'
a 1 a.inl void a() {}
(b.cpp is ignored.)
Version 6.5.3 - Mar 18 2016
[CHANGES]
New facilities:
o Added new plug-in parser 'universal-ctags.la'.
o gtags.conf:
- Added new entry 'universal-ctags'.
- Added new variable 'ctagscom'. You can overwrite the path of ctags(1)
dynamically for exuberant-ctags.la, universal-ctags.la and pygments-parser.la.
Changed:
o gtags.conf: Added the following patterns to the skip list.
'*.o,*.a,*.so,*.lo,*.zip,*.gz,*.bz2,*.xz,*.lzh,*.Z,*.tgz,*.min.js,*min.css'
o gtags: Added some explanations for the --explain option.
[FIXED BUGS]
o global: Global generated incorrect path with the --color option.
Now, it works well.
[bad case]
+- GTAGS
+- hi/
|- foo.cpp
+- what/
$ cd hi/what
$ global -P hi/foo.cpp --color
foo.cpp # It should be '../foo.cpp'.
o global: Global aborted in obstack_free() in some cases. Now it works well.
o htags: Htags ended abnormally when there is a file which includes single
quote characters in its file name. Now, it works.
0.15 2016-03-15 22:17:38Z
- fix uninitialized warning in test on perls < 5.8.6 (RT#104510)
- repository moved to the github p5sagit organization (the primary is on
shadowcat, mirrored to github)
--------------
Version 2.10b:
--------------
- Fixed a minor core counting glitch, reported by Tyler Nighswander.
--------------
Version 2.09b:
--------------
- Made several documentation updates.
- Added some visual indicators to promote and simplify the use of -Z.
--------------
Version 2.08b:
--------------
- Added explicit support for -m32 and -m64 for llvm_mode. Inspired by
a request from Christian Holler.
- Added a new benchmarking option, as requested by Kostya Serebryany.
--------------
Version 2.07b:
--------------
- Added CPU affinity option (-Z) on Linux. With some caution, this can
offer a significant (10%+) performance bump and reduce jitter.
Proposed by Austin Seipp.
- Updated afl-gotcpu to use CPU affinity where supported.
- Fixed confusing CPU_TARGET error messages with QEMU build. Spotted by
Daniel Komaromy and others.
2011 Sun Mar 13 04:41:48 CET 2016
- the year the terrorists won and the free world sacrificed the free.
- added more explanatory blurb, based on ideas by Kent Fredric.
- colour wasn't reset in time to catch further Makefile.PL
output in some cases.
2016-03-28 meld 3.15.2
======================
Features:
* Port to use GtkSourceView's file loader and saver helpers, bringing
several benefits (Kai Willadsen)
* Loading and saving speed are significantly improved, and are
asychronous
* We try a wider range of locale-specific file encodings by default,
and still support user-specified default encodings (requires
GtkSourceView 3.18 or later)
* We now support saving files with inconsistent newline types
* Binary files will now be loaded on a best-effort basis
* Reinstate support for Darcs; this is currently limited to very new
(>2.10.3) versions (Guillaume Hoffmann)
* Folder comparisons now apply text filters in the same way as file
comparisons (David Rabel)
* Add support for showing new shortcuts help dialog (Kai Willadsen, Piotr
Drąg)
* Preliminary support for building Meld as an xdg-app (Kai Willadsen)
Fixes:
* Many fixes for GTK+ 3.20 theming changes and widget compatibility issues
(Kai Willadsen)
* All accelerators should now consistently use the Primary key, for better
integration on OSX systems (Youssef Adnan)
* Significantly improve text filter performance (Jason Gauci)
* Fix performance regression where scanning for missing files was extremely
slow, especially in large repositories (Kai Willadsen)
* Fix truncated line number display in short file (Kai Willadsen)
* Fix filter dropdown not appearing below button (Kai Willadsen)
* Add more documentation on setup and installation requirements (Jesse
Sheridan, Martin Thoma)
* Update the AppData to spec version 0.7+ (Richard Hughes)
Translations:
* Anders Jonsson (sv)
* Benjamin Steinwender (de)
* Bernd Homuth (de)
* Cédric Valmary (oc)
* Daniel Mustieles (es)
* Gábor Kelemen (hu)
* Jiri Grönroos (fi)
* Luisa Cerón Perea (es)
* Marek Černocký (cs)
* Mario Blättermann (de)
* Muhammet Kara (tr)
* Pedro Albuquerque (pt)
* Piotr Drąg (pl)
* Rafael Fontenelle (pt_BR)
* Мирослав Николић (sr, sr@latin)
Upstream changes:
1.2017 2016-03-02 15:49:43-05:00 America/New_York
[CHANGED]
- Distributions that appear to be in a Perl6/ directory are not
graded and reports are discarded.
1.2016 2016-02-28 22:39:04-05:00 America/New_York
[FIXED]
- Improve prerequisite version detection for dynamic versions
- Add support for static Config like cperl and XSConfig
Upstream changes:
version 2.60 at 2016-03-20 18:29:24 +0000
-----------------------------------------
Change: b898659a1ced0560f65088da253ccd1cf343953c
Author: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Date : 2016-03-20 18:29:24 +0000
Updated for v5.23.9
-----------------------------------------
version 2.58 at 2016-02-20 21:58:41 +0000
-----------------------------------------
Change: 7abfc549182e85b37835ab155202476274b6028d
Author: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Date : 2016-02-20 21:58:41 +0000
Updated for v5.23.8
-----------------------------------------
version 2.56 at 2016-01-20 22:29:45 +0000
-----------------------------------------
Change: e49a8ccdcab90da90f431b5852fa77a762fa2737
Author: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Date : 2016-01-20 22:29:45 +0000
Updated with v5.23.7
Change: eea8677292ef0f82c5375362311e27cbf517995f
Author: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Date : 2016-01-20 22:29:45 +0000
Remove RJBS release candidates and v5.7.1
As per
http://www.nntp.perl.org/group/perl.perl5.porters/2016/01/msg233715.h
tml
v5.7.1 as well as it is AWOL
Upstream changes:
0.36 2016-02-28 21:36:57-05:00 America/New_York
[Docs]
- Fixed typos.
0.34 2016-02-18 23:26:13-05:00 America/New_York
[Fixed]
- Removed spurious JSON::PP dependency added by a broken
Dist::Zilla plugin.
0.32 2016-02-18 10:12:02-05:00 America/New_York
[Docs]
- Changed internal formatting of documentation
[Changes]
- No functional changes from 0.31
0.31 2016-02-14 07:33:50-07:00 America/Mazatlan (TRIAL RELEASE)
[Fixed]
- Application of layers to handles during and after capture now attempts
to more accurately duplicate the original layers, including potential
duplicate layers. Because of the unusual ways that layers are ordered
and applied, exact duplication is not guaranteeed, but this should be
better that what Capture::Tiny did before.
- Avoids a hard crash on Windows with Perl < 5.20 if a fork occurs in a
capture block. Also documented the risks and lack of support for
forks in capture blocks.
Update DEPENDS
Upstream changes:
1.32
Docs
- Corrected spelling mistake (thanks again Gregor Herrmann)
1.31
Docs
- Corrected spelling mistake (thanks Gregor Herrmann)
- Document that strings work as regexen (thanks jarich)
New Features
- Call backtick() qx() instead (but keep backtick around) (thanks jarich)
Incompatible Changes
- s() now returns the modified string or success/fail depending on context
Fix
- m() and nm() were failing to signal failure (thanks grtodd!)
3.20.0 21-Mar-2016
3.19.92 15-Mar-2016
3.19.91 01-Mar-2016
- Fix marshaling of GError stored in GValue
(Simon Feltman) (Thibault Saunier) (#761592)
- Fix marshaling or GError from Python to C from function calls
(Simon Feltman) (#685197)
- Error handling/reporting fixes (Christoph Reiter) (#751956)
- Fix crash due to GVariant implemented as PyGBoxed not PyGIStruct
(Christoph Reiter) (#751956)
- Fix crash with GValueArray stored in GValue
(Mikhail Fludkov) (#754359)
3.19.90 20-Feb-2016
- tests: Set the active style context state before retrieving values
(Simon Feltman)
- tests: Fix crash with empty drag source icon names
(Simon Feltman) (#762392)
- Try to import GdkX11 in Gdk overrides (Christoph Reiter) (#673396)
- Fix import warnings pointing to the wrong code with CPython 3.3/3.5
(Christoph Reiter) (#757184)
3.19.2 31-Oct-2015
- tests: Fix failure due to new GTK+ warning regarding size_allocate()
(Simon Feltman)
- Fix build warnings regarding _POSIX_C_SOURCE redefinition
(Simon Feltman)
- Drop -std=c90 for now (Matthias Clasen)
3.19.1 30-Oct-2015
- Use a named tuple for returning multiple values (Christoph
Reiter) (#727374)
- enum/flags: use gir info for type names and __repr__
instead of the gtype name (Christoph Reiter) (#657915)
- Improve and unify __repr__ format for PyGObject, PyGBoxed
and PyGIStruct (Christoph Reiter) (#657915)
- Don't leak internal RepositoryError on import. (Christoph
Reiter) (#756033)
- Import dependencies when importing typelibs from
gi.repository (Christoph Reiter) (#656314)
- Fix Gdk.rectangle_intersect/rectangle_union missing with
GTK+ 3.18 (Christoph Reiter) (#756364)
- Don't import inspect at module level (Christoph Reiter)
- invoke state: add a free memory cache for PyGIInvokeArgState
(Christoph Reiter) (#750658)
- invoke/closure: reduce g_slice_alloc usage (Christoph
Reiter) (#750658)
- pep8: ignore new errors reported by pep8 1.6 (Christoph Reiter)
- Bump g-i dependency to latest stable (Garrett Regier)
- Avoid calling g_slist_length() during invoke (Garrett Regier)
- Simplify closure_convert_arguments() (Garrett Regier)
- Remove a level of indentation in convert_ffi_arguments()
(Garrett Regier)
- Prevent passing the user data multiple times to callbacks
(Garrett Regier) (#750347)
- Support throwing exceptions in closures (Garrett Regier) (#710671)
- Don't emit require_version warning if namespace was loaded
previously using g_irepository_require (Christoph Reiter)
(#754491)
- configure.ac: post release version bump to 3.19.1 (Garrett Regier)
3.18.0 22-Sep-2015
3.17.90 19-Aug-2015
- Allow passing unicode lists to GStrv properties on Python 2
(Christoph Reiter) (#744745)
- Avoid a silent long to int truncation (Rui Matos) (#749698)
- Handle gtype marshalling (Mathieu Bridon) (#749696)
- pygi-foreign-cairo.c: fix include for py3cairo.h
(Daniel Hahler) (#746742)
- tests: Silence various error messages and warnings
(Christoph Reiter) (#751156)
- Fix test regression when xdg-user-dirs is not installed
(Christoph Reiter) (#751299)
- Explicitly check if an override exists instead of ImportError
(Garrett Regier) (#749532)
3.17.1 15-Jun-2015
- Add gi.PyGIWarning used when import version is not specified
(Christoph Reiter) (#727379)
- Remove Gdk.Rectangle alias with newer gobject-introspection and GTK+
(Christoph Reiter) (#749625)
- overrides: Provide _overrides_module attribute
(Christoph Reiter) (#736678)
- overrides: Conditionalize touch override support in Gdk
(Simon Feltman) (#747717)
- Field setters: Remove unneeded type/range checks and unused code
(Christoph Reiter) (#746985)
- pygi-argument: Remove unused imports/includes
(Christoph Reiter) (#746985)
- Improve test coverage for field setters/getters
(Christoph Reiter) (#746985)
Overview of changes between 1.39.1 and 1.40.0
=============================================
- Fix typos
- Don't treat OGHAM SPACE MARK as whitespace
Overview of changes between 1.38.1 and 1.39.0
=============================================
- Update PangoScript to Unicode 8.0
Overview of changes between 1.38.0 and 1.38.1
=============================================
- Require 0.9.30
- Fix make check with --enable-debug=no
- Win32 build fixes
Overview of changes between 1.37.5 and 1.38.0
=============================================
- No changes
Overview of changes between 1.37.4 and 1.37.5
=============================================
- Small optimizations
Overview of changes between 1.37.3 and 1.37.4
=============================================
- Use grapheme boundaries in justification
- Support coloring marks and bases independently (#541608)
- Fix a crash (#753167)
- Make Thai and Arabic support thread-safe
- Fix justification (#753772)
Overview of changes between 1.37.2 and 1.37.3
=============================================
- Fix pango_glyph_item_iter_prev
- Add alpha attributes
- Add support for alpha in markup
Overview of changes between 1.37.1 and 1.37.2
=============================================
- Don't use hb_glib_get_unicode_funcs
- Warning fixes
- removed patch-gio_gtestdbus.c: applied upstream.
- removed patch-gio_gunixmounts.c: can no longer reproduce problem.
Overview of changes in GLib 2.48.0
==================================
* a minor build fix in the name of determinism
* a few coverity fixes
Bugs fixed:
763617 giotypefuncs.c: Sort _get_type functions in the 'C' locale
Translations updated:
Danish
Italian
Overview of changes in GLib 2.47.92
===================================
* gdbus-codegen now supports g_autoptr()
* g_get_user_runtime_dir() now reliably returns an existing directory
* g_array_remove_range() can now remove 0 items from the end of an array
* Many fixes for Windows
* build fixes
* file monitoring
* gsettings backend
* streams
* random numbers
* wide character support
* documentation improvements
* other small bugfixes
Bugs fixed:
724847 Segmentation fault on "gsettings list-recursively"
743933 gapplication: add --app-id command line option
756706 [PATCH] gio/gtestdbus.c: don't use non-standard %m printf modifier
757506 gsettings: schema_list should use the passed schema's source
760694 W32: Apps linked with -mwindows make cursor busy sometimes
762202 g_win32_error_message improvements
762637 build: Unconditionally dist tapset files
762748 Undefined behavior
762937 Mention that g_clear_error can be used with an "empty" GError
763339 array: Support clearing an empty array with g_array_remove_range()
763344 g_get_user_runtime_dir(): ensure directory exists
763379 codegen: Add support for g_autoptr to gdbus-codegen–generated objects
Translations updated:
Brazilian Portuguese
Czech
Finnish
French
Galician s
German
Greek
Hebrew
Hungarian
Italian
Kazakh
Korean
Latvian
Lithuanian
Occitan
Polish
Russian
Serbian
Slovak
Slovenian
Spanish
Swedish
Overview of changes in GLib 2.47.6
==================================
* Windows usupport:
- Fixes and improvements to the GSettings registry backend
- Handle readability and writability of registry keys
- Use Unicode registry APIs
* Bugs fixed:
760852 744772 761126 747927 761337 744570 761504 761550 761843
744570 GString is missing (transfer none) annotations on many of its methods
744772 systemtap and gdb scripts install in wrong place
747927 Documentation: various small improvements
760852 gdbusobjectmanagerserver: Clarify recommended ObjectManager paths
761126 winiconv: update to upstream version
761337 Fix some annotations
761504 W32 registry GSettings backend does not use Unicode
761550 Cannot build with default flags under Fedora rawhide (-Werror=format-...
761843 gmacros.h is testing attributes with __has_feature (when compiling wi...
* Translation updates:
Brazilian Portuguese
Bulagarian
Chinese (Taiwan)
Hungarian
Polish
Slovak
Slovenian
Spanish
Swedish
Overview of changes in GLib 2.47.5
==================================
* the system copy of PCRE is now used by default to implement GRegex.
Configure with --with-pcre=internal if a system PCRE version
is unavailable or undesired.
* interfaces for DTLS support have been added. A new version of
glib-networking will also be required.
* GDBusMethodInvocation now drops replies if the sender set the
NO_REPLY_EXPECTED flag
* several GApplication fixes, including fixes for commandline arguments
in interpreted languages on Windows
Bugs fixed:
624186 Deprecate glib-gettext macros
734095 gtk-demo.py of PyGObject fails to run on Windows (and likely other binding scripts using g_application_run())
735754 Implement close on TLS GOutputStream
748064 gnulib vfprintf returns desired (not actual) number of bytes, ignores errors
752240 Add DTLS support to GIO
755421 GDBus ignores NO_REPLY_EXPECTED flag in messages, leading to warnings on system bus
756875 Include ntdef.h for NTSTATUS
759554 g_application_run() calls g_main_context_default() repeatidly
760199 gsettings: Install gettext ITS rules
760215 G_LIKELY/_UNLIKELY macros need more parentheses
760683 regex test: Check the expected PCRE exceptions at runtime
Translations updated:
Brazilian Portuguese
Czech
German
Lithuanian
Swedish
Overview of changes in GLib 2.47.4
==================================
* The GApplication documentation has been improved in several areas.
* Bugs fixed:
749092 gdb pretty-printers fail on Python 3 with a TypeError...
757374 macros: clean up "inline" mess
758641 Memory leak in g_dbus_proxy_new_for_bus_sync()
759134 Add missing checks for gnulib vasnprintf()
759408 Do not use uninitialized var
756475 Stop supporting non-POSIX getpwuid_r, getgrgid_r
757372 GApplication: destroy the impl on shutdown
728099 macros: add G_GNUC_CHECK_VERSION() for compiler checks
757299 glib-compile-resources: do not leak c_name
758553 Fix gettext use
758823 file monitors: reorder some code to avoid segfault
756214 gsettings: Don't translate ""
710243 Add GParamSpec object ref management annotations
735696 xdgmime: Finer handling for cases where mmap() is not available
752983 gapplication: Acquire the main context before running
* Translation updates:
Swedish
Overview of changes in GLib 2.47.3
==================================
The inline cleanup in the last release accidentally removed three
symbols from libglib-2.0.so. It is unlikely that this will have caused
any problems because these symbols were only backup symbols for
definitions exported as inlines in the header files, but ABI is ABI.
This release corrects only this problem.
Overview of changes in GLib 2.47.2
==================================
* We have formalised the assumption that all compilers that are
interested in support 'static inline' and simplified the macros around
this considerably. Please watch for and report unintentional fallout.
* New API: hardware-assisted helpers for overflow-checked integer math.
* other fixes
Bugs fixed:
696324 gtester-report doesn't work with Python 3.x
719966 glib: Add missing (nullable) and (optional) annotations
752837 gobject and glib-compile-resources rely on .CRT$XCU section, no longer works with Win 10 UCRT (VS 2015)
755364 make gtkdoc-check happy again
756134 Segmentation fault on calling g_simple_action_group_add_action with bad action constructor call
756179 gwin32.c: Replace VerifyVersionInfoW() with RtlGetVersion() due to API deprecation
756988 GSequence should document each function's complexity
757294 Move G_POLLFD_FORMAT to glibconfig.h
757374 macros: clean up "inline" mess
757451 doc: fix g_task_attach_source() example
757628 gio tests fail to build when cross compiling 2.46.1
757693 Invalid free in g_local_file_trash()
757742 Fix up annotations in ghash.c
758181 GTask: fix wrong example code
Translations updated:
Greek
Hebrew
Hungarian
Norwegian bokmål
Portuguese
Scottish Gaelic
Simplified Chinese
Spanish
Overview of changes in GLib 2.47.1
==================================
* The Unicode support has been updated to version 8.0 of the Unicode standard
* GDesktopAppInfo no longer sets the DISPLAY environment variable when
launching apps. This is now done in the GAppLaunchContext implementations
when appropriate
* Bug fixes:
664740 Key-value file parser, space after integer
687223 cleverer GThreadPool management
692085 stderr and stdout are not always file descriptors 1 and 2
697907 Add interface for socket-like things (GSocket, DTLS, etc)
735754 Implement close on TLS GOutputStream
737116 Add functions to print GSocketConnectables and addresses as strings
743011 Minor additions to GError documentation
749161 undefined reference to `__imp__stat32i64'
749314 Cannot restore a just-trashed file
751924 Add recvmmsg()-like API on GSocket
752240 Add DTLS support to GIO
752837 gobject and glib-compile-resources rely on .CRT$XCU section, no longer...
753310 Remove `#pragma GCC system_header` from gmessages.h
753935 Update example namespace and class names in GObject tutorial
754855 Object instantiation documentation refers to example that no longer ex...
754983 Wayland: g_desktop_app_info_launch_uris_with_spawn() forces DISPLAY va...
754994 g_date_time_get_second () sometimes returns an off-by-one result
755083 Clarify in G_ADD_PRIVATE that it is safe to call _get_instance_private...
755351 Example still contains g_autoptr(gchar)
755355 Move GStrv to glib.h so it can be used with g_auto()
755374 g_variant_get_child(): flatten-first logic on '&'
755496 glib 2.46 fails GStreamer test suite
755609 glib 2.46.0 breaks Sun Java JVM 1.8.0.60
755766 gvalue: The g_auto cleanup function assert if value is G_VALUE_INIT
755795 2.46 considers empty files as octet-stream rather than text (leads to...
755961 Fix up annotations in gbytes.c
756053 MSVC doesn't understand the symbol 'msghdr'
756054 MSVC linker error due to 'g_socket_send_message_with_timeout()'
756077 testutils: remove internal ABI comment
756099 g_main_context_query(): Annotate @n_fds as (in) parameter
756139 musl: ctors called in the wrong order
756179 gwin32.c: Replace VerifyVersionInfoW() with RtlGetVersion() due to AP...
756251 The documentation of G_DBUS_PROXY_FLAGS_DO_NOT_AUTO_START is confusing
756255 GOutputStream swallowing errors in splice with G_OUTPUT_STREAM_SPLICE...
756316 GSequence should provide fast api to check if empty
756382 snprintf used on Windows with VS2015 doesn't support %n
756477 gio/gthreadedresolver.c has outdated copy of bionic headers (for android)
756550 gtypes.h: Make G_MININTn literals negative
756875 Include ntdef.h for NTSTATUS
756952 giomodule: return a copy of module name
* Translation updates:
Basque
Czech
Serbian
Serbian Latin
Vietnamese
Overview of changes in GLib 2.46.0
==================================
* Disable runtime-deprecation warnings
* Fix marshalling of flags on bigendian 64bit architectures
* Translation updates
Brazilian Portuguese
Danish
German
Latvian
Russian
Turkish
Overview of changes in GLib 2.45.8
==================================
* utf8 validation and utf8-to-ucs4 conversion are faster
* Small speedups to property change notification
* Various other small optimizations for GQuark, GData
* Bugs fixed:
696426 GParamSpecTypeInfo do not need to be static
735429 Cleanup MSVC Project Files Generation
738504 Optimize UTF-8 decoding by unrolling branches and expressions
742903 Add missing (transfer) annotation to GString
748633 g_set_object order of operations
754431 Fix build of glib/gstrfuncs.c on Windows
754560 gioerror: Add more mappings for WinSock error codes
754582 Glib cannot compile
754601 Make g_strerror work with non-glibc POSIX systems
754636 tests/unicode-encoding test fails for glib 2.45.7 on x86-64
754788 more g_strerror stuff
754831 autocleanups: Add GString type
754924 Improve test coverage of g_utf8_validate() by added known-...
754986 Avoid unnecessary signal emission during draw
* Translation updates:
Italian
Kazakh
Korean
Lithuanian
Slovenian
Swedish
Overview of changes in GLib 2.45.7
==================================
* Add G_FILE_ATTRIBUTE_STANDARD_IS_VOLATILE for use by non-POSIX-like
backends (e.g. cloud storage).
* GFileMonitor: Make the inotify backend work with atomic renames again
* GSettings: change notification is again working unconditionally
* GListStore has a sort function now
* Test infrastructure:
- Tests are now required to have unique names
- TAP support has been improved
- A macro for asserting that two memory regions have identical content
has been added
* Bugs fixed:
708525 A "g_file_query_info" on the file path "/sys/kernel/debug/hid"...
742849 inotify: send paired events to both sides
744060 Update GObject tutorial documentation to use G_DECLARE_FINAL_T...
747364 Fix GError leak in g_file_query_writable_namespaces()
749492 Support file creation time on FreeBSD and NetBSD
752769 (g_socket_receive_message | g_socket_send_message) performance
753745 glib-genmarshal still needed for cross-compilation
754152 Add g_list_store_sort
754211 Memory leak in g_file_enumerator_iterate ()
754264 GLib 2.44 certificate chain construction fails if the PEM incl...
754283 gtestutils: add g_assert_cmpmem()
754284 gtestutils: print the TAP test plan first, not last
754286 misc gtestutils fixes
754307 size of array '_GStaticAssertCompileTimeAssertion_3387' is neg...
* Translation updates:
Chinese (Taiwan)
French
Galician
Greek
Hebrew
Hungarian
Indonesian
Polish
Portuguese
Overview of changes in GLib 2.45.6
==================================
* Fix a test failure and a build failure
Overview of changes in GLib 2.45.5
==================================
* GNetworkMonitor now provides information about metered networks
* g_mem_set_vtable has been deprecated; it has not been working for
quite a while. The recommendation is to use valgrind, or replace
malloc itself.
* Bugs fixed:
656325 Make GDBusInterfaceVTable binding friendly
741779 Documentation tweaks addressing real-world API misuses
741822 Fails to build with VS 2015
742386 gdbusconnection: Don't g_printerr() when exiting
743018 gobject: Add more cross-links between documentation pages
750282 Add g_network_monitor_get_network_metered() to get if the connection...
751358 GFileMonitor doesn't react to "mv some-file watched-file"
751592 Stop using GMemVtable
751598 Stop 'handle-local-options' propagation when callback reports an err...
751610 g_str_hash produces collisions with strings of length 2
751751 Wrong docs of g_async_queue_remove
752210 gdbus command crashes with SIGSEGV
752656 gdbusconnection: Fix signal subscription documentation
752767 Fix typo in g_hash_table_replace() documentation
753278 gdbus: Don't use g_assert_no_error() GDBusObjectManagerServer
753285 g_menu_item_set_icon fails if called with NULL icon
* Translation updates:
Catalan
Czech
French
Indonesian
Lithuanian
Norwegian bokmål
Slovak
Spanish
Thai
Turkish
Overview of changes in GLib 2.45.4
==================================
* Bugs fixed:
727829 win32: glibconfig.h.win32 updates
741901 Clang cannot know that g_error don't return
746339 GSocket kills process when fd is not a socket
747676 gio/tests/socket fails: test_fd_roundtrip
748610 Some tests fail with non-English locales
749911 g_inet_address_to_string broken on XP/2003
749912 g_inet_address_new_from_string broken on XP/2003
750625 Should dismiss Software Updates Available notification after...
750807 G_BREAKPOINT doesn't work as intended on Darwin
751160 gtask does unnecessary work
751672 -Wduplicate-decl-specifier in glib/tests/keyfile.c
751731 GFile/DirectoryMonitor emit move events with other_file=NULL
751737 gio/tests/appmonitor test fails in 2.45.3
751798 Wrong enum type used in some test-cases
752089 make gsocketservice::active a property
752293 small cleanup: use list_free_full
* Translation updates:
Greek
Hebrew
Portuguese
Overview of changes in GLib 2.45.3
==================================
* Improve performance of g_signal_handler_disconnect for signals
with many handlers
* GDBus has gained a new call flag to allow interactive authorization
* GSettings:
- New API: g_settings_schema_list_keys
- Deprecated: g_settings_list_keys
* OS X:
- Implement GNotification
- Bump the OS X requirement to 10.9
* Windows:
- Add registry reading API
- Reimplement GAppInfo using registry information
* Bugs fixed:
666831 Support URI opening on W32
728489 property action with inverted boolean state
730168 Incorrect annotation on g_action_group_get_action_state_type return...
733325 Several regex tests fail with pcre3 8.35
734888 GLib has no helper functions to work with W32 Registry
737009 signal handler lookup doesn't scale
738185 Misleading language about "file name encoding" in the docs on g_env...
738504 Optimize UTF-8 decoding by unrolling branches and expressions
739122 glib not handling -1 return ("no limit") from sysconf (_SC_THREAD_S...
739424 gnome-shell crashes when files are added, deleted, or modified in $...
739616 DBus; Add new call flag to allow interactive authorization
740308 Add g_settings_schema_list_keys() method
740516 RFE: please provide an introspectible version of g_log_set_handler
741788 Document GSettings build system integration
745013 GBinding not thread safe
747146 Implement GNotification on OSX
747941 try XDG_RUNTIME_DIR/bus before falling back to X11 autolaunch (dbus...
748727 Filechooser dialog shows no icons for directories on W32
749693 GActionGroupExporter: flush queue on requests
750203 GNetworkMonitorNetlink hangs in user namespace
750322 gapplication: Make sure --help output is translated
750344 GTlsInteractionClass is missing from doc
750369 Various GBinding cleanups
750386 Race condition in g_io_condition_get_type
750399 Typo "equilalent" in glib documentation's glib-Error-Reporting.html...
750573 GTlsDatabaseClass is not documented
750918 genmarshal: silence register storage class warnings
751122 gsocket: avoid unnecessary g_socket_cond_wait() in _send_messages()
479730 The "g_key_file_set_comment" interface prepends '#' character to...
* Translation updates:
Hungarian
Spanish
Overview of changes in GLib 2.45.2
==================================
* Improve error reporting in glib-compile-schemas.
* Add introspection annotations to GListStore.
* Bugs fixed:
696749 win32 : failed to compile because of careless mistake in the code
723394 const parameter to GtkPopover gtk_popover_set_pointing_to
724113 gdbus-connection-loss test can fail on slow machines
725981 tap-driver.sh: internal error getting exit status
733325 Several regex tests fail with pcre3 8.35
744895 Unknown or unsupported transport 'this-should-not-be-used-and-will...
747882 gtype: Bump allowed number of children
748534 gtest: if a subprocess assertion fails, print its stdout and stderr
748612 de_DE locale used in option-context test is not supported by FreeBSD
748614 Double unref in g_socket_listener_add_inet_port
748834 glocalfilemonitor: Emit notification on rate limit change
749079 gdbus-peer test: TCP tests can fail with ECONNRESET due to a race...
749080 gdatetime test: fails if close to rollover between seconds
749180 gnetworkaddress: add return type annotation to parse methods
749352 g_binding_unbind() fails when source is also the target
749353 GBinding does not connect to the detailed notify signal
* Translation updates:
Catalan
French
Slovak
Thai
Overview of changes in GLib 2.45.1
==================================
* The GSettings schema compiler, glib-compile-schemas has been changed
to reject schema xml that has duplicate <summary> or <description>
elements. Such elements typically occur when translations are merged
into the schema, with xml:lang attributes. This is not the correct
way to translate schemas. Instead keep the translations in the .mo
file and set the gettext-domain attribute on the <schemalist> element.
* The file monitoring infrastructure has been rewritten, and all backends
have seen major improvements.
The inotify backend is reporting events with less delay (no event will
be delayed more than 10ms) and wakeups due to file monitoring have been
significantly reduced. A CHANGES_DONE event will also be sent when new
files appear.
The poll implementation is now using the thread default main context.
The fam implmentation is now running in the worker thread.
The fen implementation has been removed, since it was unmaintained.
* The GSettings schema compiler, glib-compile-schemas, is more strict
about rejecting schemas with xml:lang style merged translations.
Schema translations should be done by specifying the gettext domain
in the xml, and keeping the translations in gettext. To avoid breaking
already-installed schemas, this change is only taking effect when
you use the --strict option.
* The hardcoded 10-thread limit of GTask's thread pool has been removed,
since it was prone to causing deadlocks. The thread pool is now allowed
to grow dynamically and will shrink back over time.
* GSimpleAsyncResult has been deprecated in favor of GTask.
* The algorithm used by GAppInfo to find default handlers for mime types
has been tweaked to prefer apps that handle the specific subtype over
default handlers for a generic supertype.
* Bug fixes:
627285 inotify file monitor hardwired delay
631597 Segmentation fault in append_escaped_text
661767 merge/improve various bits of run-in-thread functionality
687223 cleverer GThreadPool management
711547 win32: silence some build warnings
719966 glib: Add missing (nullable) and (optional) annotations
726447 Possibly an error in text string
728663 W32: wrong stat struct is used when built with MinGW-w64
728669 W32: GLocalFile can't measure size of files larger than 2^32...
730188 gsocket: Document FD ownership with g_socket_new_from_fd()
733325 Several regex tests fail with pcre3 8.35
738207 Add a way to set SO_SENDBUF and SO_RECVBUF on listener (and...
739850 GClosure: add valgrind hints
741791 gmain: Save errno when handling unix signals
744282 gvfs-open for application/x-virt-viewer changed behaviour bet...
745255 Add support for copying sessions between GTlsClientConnections
745745 gdbus: fix out-of-bound array access
745821 Don't use __alloc_size__ attribute with clang
746749 GLib-GIO:ERROR:inotify-kernel.c:327:ik_source_dispatch: ass...
746753 Glib-compile-resources --generate-header not using ".h" as ...
747209 glib-compile-schemas ought to reject repeated <summary> and...
747349 Conversion of gdbus to use GTask causes deadlocks
747363 gatomic: Add missing new line in API doc comment
747472 Don't ignore already-installed schemas with multiple <summa...
747541 gdbus segfaults with invalid --dest
747772 Having hardcoded utf8 strings in the source code does not p...
748019 gsocketconnection: Fix copy-pasto in documentation
748177 not all test schemas are distributed, "make distcheck" fails
* Translation updates:
Basque
Czech
Danish
Finnish
German
Hebrew
Icelandic
Norwegian bokmål
Russian
Turkish
* "make rpmbuild" target was broken as its input, git.spec.in, was
not updated to match a file it describes that has been renamed
recently. This has been fixed.
build, but not linked at runtime. Fixes PLIST conflict with
devel/binutils and subsequently makes pbulk on (at least) SunOS
happy. This follows what e.g. ports and homebrew already do.
Remove a leftover patch file while at it.
Bump PKGREVISION.
In-freeze, OK by wiz.
This fixes the build where the C extension built successfully. In turn, it
will break where the C extension failed to build. I believe it is better to
fail when actually failing, and this is a leaf package, so I am going ahead
with this - and then the issues when the C extension should be corrected
instead (when observed).
Submitted by Terin Stock terinjokes AT gmail com
Closes PR pkg/50875
Reviewed by wiz@
2013-08-30: StGit-0.17.1 released
Fix dirty index errors when resolving conflicts.
Test suite improvements.
Print tracebacks to stderr.
Run test suite in parallel.
Fix --authdate date parsing.
