----------------------------------------------
2.46 2016-09-21 Increase timeout for running Makefile.PL, to
cope with really slow machines with pitiful
I/O
You can write a setup.py with no version information specified, and vcversioner
will find a recent, properly-formatted VCS tag and extract a version from it.
It's much more convenient to be able to use your version control system's
tagging mechanism to derive a version number than to have to duplicate that
information all over the place. I eventually ended up copy-pasting the same
code into a couple different setup.py files just to avoid duplicating version
information. But, copy-pasting is dumb and unit testing setup.py files is hard.
This code got factored out into vcversioner.
A priority queue which implements a lazy binomial heap. It supports the
change priority operation, being suitable for algorithms like Dijkstra's
shortest path and Prim's minimum spanning tree. It can be instantiated as a
min-priority queue as well as a max-priority queue.
----------------------------------------
1.302059 2016-09-25 12:32:21-07:00 America/Los_Angeles
- No changes from last trial
1.302058 2016-09-21 10:46:13-07:00 America/Los_Angeles (TRIAL RELEASE)
- Mask warning when comparing $@ in Test2::API::Context
1.302057 2016-09-18 12:12:18-07:00 America/Los_Angeles (TRIAL RELEASE)
- Doc fixes
- Win32 color support in Test::Builder::Tester
- Support v-strings in is_deeply
- A streamed subtest run inside a buffered subtest will be automatically
converted to a buffered subtest. Otherwise the output from inside the
subtest is lost entirely. (#721)
-------------------------------------------
1.007 2016-10-02 BOOK
- Add an extra diagnostic reporting the git version used for testing
(with a scheme to limit the number of times it's output, so as not
to pollute test output too much)
---------------------------------------
0.002005 - 2016-09-27
- fix skipping when Test::Tester (Test::More < v2) is loaded
0.002004 - 2016-08-18
- fix test counts when Test2 not available
0.002003 - 2016-08-18
- fix loud warnings when aborting a test under Test2
- some kwalitee improvements
------------------------------------
1.442 2016-09-22T15:35:50Z
* Update to stable release. This fixes an upcoming Perl @INC
issue that removes the current directory from the default list.
This affects only the test files which use a private library.
If you already have this installed you don't need to update.
1.441_01 2016-09-19T21:44:45Z
* Fix for missing . in
@INC (Github #14https://github.com/briandfoy/test-file/issues/14)
This relates to CVE-2016-1238
----------------------------------
0.101 2016-08-29
* A test accidentally added a dependency on Readonly.pm. Fixed!
(GH #30)
* Update README.
0.100 2016-08-27
* Support situations in op-code implementation where the parameters
do not come as a list. (Zefram)
* Fix memory leak in dangling op. (Zefram)
* Support magic (tied variables). (Tony Cook)
* Rework op implementation. (Zefram)
Thanks to Christian Walde for porting and merging Zefram's work
from Hash::SharedMem.
* Speed up by changing the top of the stack instead of POPing and
PUSHing. (Sergey Aleynikov)
* Update ppport.h file from Devel::PPPort and remove the copy of
SVRXOK since it's now available.
(Christian Walde)
* Add license in Pod. (Pete Houston)
* Specify minimum version of perl (5.6.2). (Pete Houston)
-----------------------------
PDCurses 3.4 - 2008/09/08
=========================
Nothing much new this time, but I've been sitting on some bug fixes for
almost a year, so it's overdue. Apart from bugs, the main changes are in
the documentation.
New features:
- setsyx() is now a function rather than a macro.
Bug fixes and such:
- In x11, the xc_atrtab table size was under-calculated by half,
resulting in crashes at (oddly) certain line counts. (It should've
crashed a lot more.) Reported by Mark Hessling.
- Test for moved cursor was omitting the window origin offset. Reported
by Carey Evans.
- Is DOS and OS/2, the value for max items in key_table was still wrong.
Reported by C.E.
- Changed isendwin() so it won't crash after delscreen().
- Ensure zero-termination in PDC_mbstowcs() and PDC_wcstombs().
- Disable QuickEdit Mode when enabling mouse input for the Win32
console; reported by "Zalapkrakna".
- Fix for building under Innotek C (I hope). Report by Elbert Pol, fix
courtesy of Paul Smedley.
- Unified exports list with no duplicates -- pdcurses.def is now built
from components at compile time.
- Don't install curspriv.h, and don't include it with binary
distributions.
- Building DLLs with LCC is no longer supported, due to the primitive
nature of its make.exe.
- Export the terminfo stub functions from the DLLs, too.
- Added support for Apple's ".dylib" in configure. Suggested by Marc
Vaillant (who says it's needed with OS 10.5.)
- In sdl1/Makefile.mng, ensure that CC is set.
- In the gcc makefiles, "$?" didn't really have the desired effect --
_all_ the dependencies showed up on the command line, including
curses.h, and pdcurses.a twice. And apparently, this can mess up some
old version (?) of MinGW. So, revert to spelling out "tuidemo.o
tui.o". Reported by "Howard L."
- Extensive documentation revision and reorganizing. More to do here.
For example, I moved the build instructions from INSTALL (which never
really described installation) to the platform-specific READMEs.
- New indentation standard: four spaces, no tabs.
---------------------------------
0.236 20161009
. Disable benchmark test rt62341.t as it runs out of memory on
many smoker systems; closes RT #118310
. No code changes, no need to upgrade
0.235 20161009
. Add Travis test configuration (contributed by James McCoy)
. Make test for PMC availability more reliable (contibuted by E. Choroba)
. No code changes, no need to upgrade
---------------------------------
0.3.4 (stable)
Some fixes for filterdiff and rediff. The splitdiff utility
now has an option to write output to a subdirectory. Ranges
can now be prefixed by a modifier ('x') to exclude spans
instead of include them. Extended headers produced by git are
now handled by filterdiff and interdiff.
-----------------------------
(From README)
0.9 (February 29, 2016):
* Lots of new features. Please refer to the Git commit log:
https://github.com/NixOS/patchelf/commits/master
This release has contributions from Aaron D. Marasco, Adrien
Devresse, Alexandre Pretyman, Changli Gao, Chingis Dugarzhapov,
darealshinji, David Sveningsson, Eelco Dolstra, Felipe Sateler,
Jeremy Sanders, Jonas Kuemmerlin, Thomas Tuegel, Tuomas Tynkkynen,
Vincent Danjean and Vladimír Čunát.
TortoiseHg 3.9.2 is a regularly scheduled bug-fix release.
There were no changes to TortoiseHg itself in this release, but
the installer packages include Mercurial 3.9.2 and many updates to
the packaged software libraries. The Windows installers now come
with the latest Python release (2.7.10), the latest Qt4 and PyQt4
releases, and updates to several Mercurial extensions including
keyring, evolve, and dulwich (hg-git).
v1.13.0
=======
* fix regression caused by the fix of #101
* assert types for version dumping
* strictly pass all versions trough parsed version metadata
v1.12.0
=======
* fix issue #97 - add support for mercurial plugins
* fix issue #101 - write version cache even for pretend version
(thanks anarcat for reporting and fixing)
v28.3.0
#809: In find_packages(), restore support for excluding a parent
package without excluding a child package.
#805: Disable -nspkg.pth behavior on Python 3.3+ where PEP-420
functionality is adequate. Fixes pip #1924.
New Features
============
Configuration
-------------
- Cmakehdf5: Added Ability to Run Multiple Make Commands
Added option --njobs to specify up to how many jobs to launch during
build (cmake) and testing (ctest).
(AKC - 2015/12/13, HDFFV-9612)
- Cmakehdf5: Added Szip Support and Verbose Option
Added --with-szlib to support the Szip library; and
--enable/disable-verbose to display all CMake process output.
(AKC - 2015/11/16, HDFFV-8932 and DAILYTEST-195)
- CMake minimum is now 3.1.0. (ADB - 2015/11/14)
- Large File System (LFS) Support has Changed in the Autotools
We assume that fseeko and ftello exist.
The *64 I/O functions and types are no longer explicitly used.
We now rely on a mapping provided by _FILE_OFFSET_BITS (or its
equivalent).
_LARGEFILE(64)_SOURCE is no longer exposed via AM_CPPFLAGS.
(DER - 2016/03/29, HDFFV-9626 and HDFFV-9541)
Library
-------
- New API Calls for Searching for External Dataset Storage
API calls that determine the search path for dataset external
storage were added. H5Pset/get_efile_prefix() API calls were added
to the library. These functions give control over the search path
for dataset external storage that has been configured with
H5Pset_external().
Additionally, the HDF5_EXTFILE_PREFIX environment variable can be
used to control the search path.
(DER - 2016/04/20, HDFFV-8740)
High-Level APIs
---------------
C Packet Table API
------------------
- Replacement of a Public Function with H5PTcreate
The existing function H5PTcreate_fl limits applications so they
can use the deflate compression only. The public function
H5PTcreate has been added to replace H5PTcreate_fl. H5PTcreate
takes a property list identifier to provide flexibility on
creation properties. This also removes the following warning:
"deprecated conversion from string constant to "char*"
[-Wwrite-strings]".
(BMR - 2016/04/25, HDFFV-9708, HDFFV-8615)
- New Public Functions: H5PTget_dataset and H5PTget_type
Two accessor functions have been added. H5PTget_dataset returns
the identifier of the dataset associated with the packet table,
and H5PTget_type returns the identifier of the datatype used by
the packet table.
(BMR - 2016/04/25, HDFFV-8623 patch 3)
- Regarding #ifdef VLPT_REMOVED
The #ifdef VLPT_REMOVED blocks have been removed from the packet
table (PT) library source except for the following functions:
+ H5PTis_varlen() has been made available again
+ H5PTfree_vlen_readbuff() is now H5PTfree_vlen_buff()
(BMR - 2016/04/25, HDFFV-442)
C++ Packet Table API
--------------------
- New Constructor in FL_PacketTable
An overloaded constructor has been added to FL_PacketTable and
takes a property list identifier to provide flexibility on
creation properties such as compression.
FL_PacketTable(hid_t fileID, const char* name, hid_t dtypeID,
hsize_t chunkSize = 0, hid_t plistID = H5P_DEFAULT)
(BMR - 2016/04/25, HDFFV-8623 patch 5)
- New Member Functions in PacketTable
Two accessor wrappers were added to class PacketTable.
PacketTable::GetDataset() returns the identifier of the dataset
associated with the packet table, and PacketTable::GetDatatype()
returns the identifier of the datatype that the packet table uses.
(BMR - 2016/04/25, HDFFV-8623 patch 4)
- New Member Functions with "char*" as an Argument
Overloaded functions were added to provide the "const char*"
argument; the existing version will be deprecated in future
releases. This also removes the following warning:
"deprecated conversion from string constant to "char*"
[-Wwrite-strings]".
(BMR - 2016/04/25, HDFFV-8623 patch 1, HDFFV-8615)
- Regarding #ifdef VLPT_REMOVED
The #ifdef VLPT_REMOVED blocks have been removed from the packet
table library source code except for the following functions:
+ VL_PacketTable::IsVariableLength() was moved to PacketTable
+ VL_PacketTable::FreeReadBuff() is now PacketTable::FreeBuff()
(BMR - 2016/04/25, HDFFV-442)
C++ API
-------
- New Member Function in DSetCreatPropList
DSetCreatPropList::setNbit() was added to setup N-bit compression for
a dataset.
(BMR - 2016/04/25, HDFFV-8623 patch 7)
- New Overloaded "const" Member Functions in ArrayType
The two following functions were added:
ArrayType::getArrayNDims() const
ArrayType::getArrayDims() const
to provide const version, and the non-const version was marked
deprecated. In-memory array information, ArrayType::rank and
ArrayType::dimensions, were removed. This is an implementation
detail and should not affect applications.
(BMR, 2016/04/25, HDFFV-9725)
- New member function added
The assignment operator ArrayType::operator= is added because ArrayType
has pointer data members.
(BMR, 2016/03/07, HDFFV-9562)
Support for New Platforms, Languages, and Compilers
===================================================
- Mac OS X El Capitan 10.11.4 with compilers Apple clang/clang++
version 7.3.0 from Xcode 7.3, gfortran GNU Fortran (GCC) 5.2.0
and Intel icc/icpc/ifort version 16.0.2
Bug Fixes since HDF5-1.8.16
===========================
Configuration
-------------
- Updated Linux Language Level Flags to Match the Autotools. Removed
Linux-specific Flags from OS X.
An addition to the flags simply being out of sync with the Autotools,
the Linux flags were used on OS X builds which led to symbols not being
found. Although this was non-fatal and compilation continued (implicit
definitions were used by the compiler and the symbols resolved at link
time), a large number of warnings were raised.
Linux changes:
* CHANGED: _POSIX_C_SOURCE (from 199605 to 200112L)
* ADDED: _GNU_SOURCE
* REMOVED: _BSD_SOURCE
* REMOVED: _DEFAULT_SOURCE
(DER - 2015/12/08, HDFFV-9627)
- The --enable-clear-file-buffers configure Option was Non-functional
so the Feature was Always Enabled (its default value).
Regardless of the configure flag, the setting was always enabled when
the Autotools were used to configure HDF5. This was due to the "no"
option being processed after the "*" option in configure.ac so "*"
matched first. CMake was unaffected.
The option now works correctly.
NOTE that builders are always advised to leave this option enabled.
When disabled, buffers that are written to disk may contain the
memory's previous contents, which may include secure information.
The performance overhead of the feature (a single memset call per
allocation) is minimal.
(DER - 2016/02/03, HDFFV-9676)
- Added a patch to remove '"'s from arguments for MPI compilers that
were causing errors compiling H5lib_settings.c with SGI MPT.
(LRK - 2016/04/20, HDFFV-9439)
Library
-------
- Fixed shared file pointer problem which caused a crash when running a
program provided by a user.
(VC - 2016/04/01, HDFFV-9469)
- Fixed some format string warnings that prevent compiling with
-Werror=format-security on gcc.
These only appeared in error messages and would not cause problems
under normal operation.
(DER - 2016/01/13, HDFFV-9640)
- Fixed a library segmentation fault when accessing a corrupted
file provided by a user.
(MSC - 2016/02/19, HDFFV-9670)
Tools
-----
- h5dump: Sub-setting Fixed for Dimensions Greater than Two
When a dataset has more than two dimensions, sub-setting would
incorrectly calculate the data that needed to be displayed.
Added in block and stride calculations that account for dimensions
greater than two. NOTE: lines that have line breaks inserted
because of display length calculations may have index info that
is incorrect until the next dimension break.
(ADB - 2016/03/07, HDFFV-9698)
- h5dump: Issue with Argument Segmentation Fault
When an argument with an optional value was at the end of the command
line with a value, h5dump would crash. Reworked check for remaining
arguments.
(ADB - 2016/03/07, HDFFV-9570, HDFFV-9684)
- h5dump: Issue with Default Fill Value
Added all default cases of fill value to the display of fill value.
(ADB -, 2016/03/07, HDFFV-9241)
- h5dump: Clarified Help
Clarified usage of -O F option in h5dump utility help.
(ADB - 2016/03/07, HDFFV-9066)
- h5dump: Issue with Double Free Fault
Added a check for filename not null before calling free().
(ADB - 2016/01/27, HDFFV-9639)
- VS2015 Release Changed how Timezone was Handled
Created a function, HDget_timezone, in H5system.c. Replaced
timezone variable usage with function call.
(ADB - 2015/11/02, HDFFV-9550)
C++ API
-------
- Removal of Obsolete Methods
The overloaded methods which had parameters that should be const
but were not have been removed.
(BMR - 2016/01/13, HDFFV-9789)
High-Level APIs:
---------------
- Fixed Memory Leak in Packet Table API
Applied user's patch to fix memory leak in the creation of a
packet table.
(BMR - 2016/04/25, HDFFV-9700)
Known Problems
==============
* On windows platforms in debug configurations, the VFD flush1 tests will fail
with the split and multi VFD drivers. These tests will display a modal debug
dialog which must be answered or wait for the test timeout to expire.
(ADB - 2014/06/23 - HDFFV-8851)
* CLANG compiler with the options -fcatch-undefined-behavior and -ftrapv
catches some undefined behavior in the alignment algorithm of the macro DETECT_I
in H5detect.c (Issue 8147). Since the algorithm is trying to detect the alignment
of integers, ideally the flag -fcatch-undefined-behavior shouldn't to be used for
H5detect.c. In the future, we can separate flags for H5detect.c from the rest of
the library. (SLU - 2013/10/16)
* Make provided by Solaris fails in "make check". Solaris users should use
gmake to build and install the HDF5 software. (AKC - 2013/10/08 - HDFFV-8534)
* The C++ and FORTRAN bindings are not currently working on FreeBSD with the
native release 8.2 compilers (4.2.1), but are working with gcc 4.6 from the
ports (and probably gcc releases after that).
(QAK - 2012/10/19)
* The following h5dump test case fails in BG/P machines (and potentially other
machines that use a command script to launch executables):
h5dump --no-compact-subset -d "AHFINDERDIRECT::ah_centroid_t[0] it=0 tl=0"
tno-subset.h5
This is due to the embedded spaces in the dataset name being interpreted
by the command script launcher as meta-characters, thus passing three
arguments to h5dump's -d flag. The command passes if run by hand, just
not via the test script.
(AKC - 2012/05/03)
* The STDIO VFD does not work on some architectures, possibly due to 32/64
bit or large file issues. The basic STDIO VFD test is known to fail on
64-bit SunOS 5.10 on SPARC when built with -m64 and 32-bit OS X/Darwin
10.7.0. The STDIO VFD test has been disabled while we investigate and
a fix should appear in a future release.
(DER - 2011/10/14 - HDFFV-8235)
* h5diff can report inconsistent results when comparing datasets of enum type
that contain invalid values. This is due to how enum types are handled in
the library and will be addressed in a future release.
(DER - 2011/10/14 - HDFFV-7527)
* The links test can fail under the stdio VFD due to some issues with external
links. This will be investigated and fixed in a future release.
(DER - 2011/10/14 - HDFFV-7768)
* After the shared library support was fixed for some bugs, it was discovered
that "make prefix=XXX install" no longer works for shared libraries. It
still works correctly for static libraries. Therefore, if you want to
install the HDF5 shared libraries in a location such as /usr/local/hdf5,
you need to specify the location via the --prefix option during configure
time. E.g, ./configure --prefix=/usr/local/hdf5 ...
(AKC - 2011/05/07 - HDFFV-7583)
* The parallel test, t_shapesame, in testpar/, may run for a long time and may
be terminated by the alarm signal. If that happens, one can increase the
alarm seconds (default is 1200 seconds = 20 minutes) by setting the
environment variable, $HDF5_ALARM_SECONDS, to a larger value such as 3600
(60 minutes). Note that the t_shapesame test may fail in some systems
(see the "While working on the 1.8.6 release..." problem below). If
it does, it will waste more time if $HDF5_ALARM_SECONDS is set
to a larger value.
(AKC - 2011/05/07)
* Shared Fortran libraries are not quite working on AIX. While they are
generated when --enable-shared is specified, the fortran and hl/fortran
tests fail. the issue. HL and C++ shared libraries should now be
working as intended, however.
(MAM - 2011/04/20)
* While working on the 1.8.6 release of HDF5, a bug was discovered that can
occur when reading from a dataset in parallel shortly after it has been
written to collectively. The issue was exposed by a new test in the parallel
HDF5 test suite, but had existed before that. We believe the problem lies with
certain MPI implementations and/or file systems.
We have provided a pure MPI test program, as well as a standalone HDF5
program, that can be used to determine if this is an issue on your system.
They should be run across multiple nodes with a varying number of processes.
These programs can be found at:
http://www.hdfgroup.org/ftp/HDF5/examples/known_problems/
(NAF - 2011/01/19)
* All the VFL drivers aren't backward compatible. In H5FDpublic.h, the
structure H5FD_class_t changed in 1.8. There is new parameter added to
get_eoa and set_eoa callback functions. A new callback function
get_type_map was added in. The public function H5FDrealloc was taken
out in 1.8. The problem only happens when users define their own driver
for 1.6 and try to plug in 1.8 library. Because there's only one user
complaining about it, we (Elena, Quincey, and I) decided to leave it as
it is (see bug report #1279). Quincey will make a plan for 1.10.
(SLU - 2010/02/02)
* The --enable-static-exec configure flag will only statically link libraries
if the static version of that library is present. If only the shared version
of a library exists (i.e., most system libraries on Solaris, AIX, and Mac,
for example, only have shared versions), the flag should still result in a
successful compilation, but note that the installed executables will not be
fully static. Thus, the only guarantee on these systems is that the
executable is statically linked with just the HDF5 library.
(MAM - 2009/11/04)
* A dataset created or rewritten with a v1.6.3 library or after cannot be read
with the v1.6.2 library or before when the Fletcher32 EDC filter is enabled.
There was a bug in the calculation of the Fletcher32 checksum in the
library before v1.6.3; the checksum value was not consistent between big-
endian and little-endian systems. This bug was fixed in Release 1.6.3.
However, after fixing the bug, the checksum value was no longer the same as
before on little-endian system. Library releases after 1.6.4 can still read
datasets created or rewritten with an HDF5 library of v1.6.2 or before.
(SLU - 2005/06/30)
Upstream changes:
(from https://github.com/ruby-gnome2/ruby-gnome2/blob/3.0.9/NEWS)
== Ruby-GNOME2 3.0.9: 2016-08-12
=== Changes
==== All
* Update Realease date in NEWS.
* Use the latest Rubies.
* Readme: update ruby versions supported.
[Patch by cedlemo]
* travis: install experimental gems dependent packages for GSF and GOffice.
[patch by Hiroshi Hatake]
* add GSF, GOffice and Gnumeric in the main Rakefile.
[patch by cedlemo]
* ignore no-GI version of gdk_pixbuf2 in main run-test.rb and in the main extconf.rb.
* remove conditionnal macro based on HAVE_RUBY_ENCODING_H.
* clean code in main Rakefile.
* define top level windows:version:update in main Rakefile.
==== Ruby/GIO2
* Improvements
* Fixes
* remove needless lazy initialization.
* follow GI API improvement.
==== Ruby/GLIB2
* Improvements
* support for GLib 2.30 on CentOS 6.
[ruby-list:50310][5.5]
[ruby-list:50316][5.5]
* add GLib 2.30 check.
* add GLib 2.34 check.
* support customizing signal callback.
* support (({GObject#bind_property_full})).
[patch by cedlemo]
* clean code and style issue.
* fix GC bug that alive proc may be GC-ed.
* Fixes
* windows: improve rake.
* split g-ir-compiler args.
* add debug option as comment.
* fix wrong define location.
* use predicate style.
* fix wrong function in rbglib-variant.c.
* bump version.
* use SPDX format for license.
* homebrew:
* add workaround for libffi.pc isn't found on OS X.
* enable libffi workaround for gi.
* remove checks for old Ruby.
* Ruby 2.1 or later has (({rb_str_new_cstr})).
* set UTF-8 encoding to (({GLIB.XXX_to_utf8})) return value.
* remove unused variables in rbglib_convert.c.
[patch by cedlemo]
* support Ruby 2.4 Integer unification.
[patch Hiroshi Hatake]
* gnome2 rake add windows:version:update task.
* gnome2 rake reduce too much update.
* gnome2 rake accept the latest version isn't found case.
* gnome2 rake ignore development series.
* windows: update dependencies.
* support :freedesktop download site.
* support :freedesktop_gstreamer download site.
* ignore development version in :freedesktop_gstreamer.
==== Ruby/GStreamer
* Improvements
* Fixes
* windows: improve rake file.
* update patches.
* follow file name change.
* specify x86_64 as arch for 64bit Windows.
* stop splitting g-ir-compiler args by default.
* fix typo.
* windows: update dependencies.
* use :freedesktop_gstreamer download_site.
* update external packages.
==== Ruby/Poppler
* Improvements
* Fixes
* windows: improve rake file.
* use https.
* add missing Makefile existence check.
* segmentation fault in (({Poppler::Page#text_layout})).
[patch by YAMAMOTO Masayuki]
* use (({GdkPixbuf::Pixbuf})) form in sample files.
[patch by cedlemo]
* remove needless gdk_pixbuf2 build dependency.
* add missing make file existence check in run-test.rb.
==== Ruby/GObjectIntrospection
* Improvements
* support dispatching by Flags and Enums.
[GitHub#745][Reported by Mamoru TASAKA]
* cache results of methods of (({GI::CallableInfo})).
[GitHub#749][Reported by rafagf1]
* cache collection reader result.
[GitHub#749][Reported by rafagf1]
* reduce needless function call.
[GitHub#749][Reported by rafagf1]
* cache (({GI::AgInfo#gclosure?})).
* add (({GI::Loader#rubyish_class_name})).
[patch by cedlemo]
* accept on_XXX as predicate method name.
* support gslist of structure values as return value.
[patch by cedlemo]
* make constant name easy to customize.
* Fixes
* fix a typo in loader.rb.
* fix a bug that int32[] returns wrong values.
[GitHub#758][reported by TASAKA]
* don't return array length output argument for return value.
* windows: update dependencies.
==== Ruby/GDK3
* Improvements
* Fixes
* windows:
* support symbolic icons.
[GitHub#750][Abby Archer]
* use (({GdkPixbuf::Pixbuf})) form in gdk3 library and test files.
[patch by cedlemo]
* Rake file update GTK+ version.
==== Ruby/GTK3
* Improvements
* update gtk-demo textscroll.
[patch by cedlemo]
* update gtk-demo data files to Gtk 3.20.
[patch by cedlemo]
* cache (({Gtk::TreeModel#get_column_type})).
[GitHub#749][Reported by rafagf1]
* update gtk-demo main.rb to Gtk 3.20.
[patch by cedlemo]
* update theming_style_classes gtk demo to 3.20.
[patch by cedlemo]
* update assistant gtk demo to 3.20.
[patch by cedlemo]
* update css related gtk demos to 3.20.
[patch by cedlemo]
* update markup gtk demo to 3.20.
[patch by cedlemo]
* update font-features demo to 3.20.
[patch by cedlemo]
* make (({Gtk::TreeView#expand_row})) rubyish.
* support "row-collapsed" and "row-expanded" signals for (({Gtk::TreeView})).
[GitHub#762][Reported Eric Cunningham]
* update misc sample bindings.rb to Gtk 3.20.
[patch by cedlemo]
* update rotated text gtk demo.
[patch by cedlemo]
* update pixbufs gtk demo.
[patch by cedlemo]
* hide "found" return value from (({Gtk::TextView#get_iter_at_XXX})).
[GitHub#794][Reported by Miguel Hernández]
* add (({Gtk::TextView#get_iter_at}))
* Fixes
* guard text marks from GC.
[GitHub#743][Reported by cedlemo]
* fix a bug that GC-ed (({Gtk::TextTag})) may be reused.
* relax condition in test for (({Gtk::ListStore})).
* use GTK+ version check in test for (({Gtk::Image})).
* fix issue in the gtk3 tutorial.
[patch from Renich Bon Ciric]
* follow (({Gtk::TextBuffer#select_bounds})) API change.
[GitHub#794][Reported by Miguel Hernandez]
* (({Gtk::TextView#get_iter_at*})) return nil when not found.
[GitHub#794][Reported by Miguel Hernandez]
* fix typo at gtk-demo.
[patch by tSU_RooT]
* use (({GdkPixbuf::Pixbuf})) form in gtk-demo files.
[patch by cedlemo]
* use (({GdkPixbuf::Pixbuf})) form in sample misc files.
[patch by cedlemo]
* use (({GdkPixbuf::Pixbuf})) form in test files.
[patch by cedlemo]
* gtk3 sample: follow (({Gtk::TextBuffer#select_bounds})) API change.
[GitHub#794][Reported by Miguel Hernández]
* (({Gtk::TextView#get_iter_at*})) return nil when not found.
[GitHub#794][Reported by Miguel Hernández]
* fix wrong method used in order to set window size in tutorial samples.
[patch by Renich Bon Ciric]
* fix label in ((#Gtk::TextView})) tests.
* fix typo at gtk-demo.
[patch by tSU_RooT]
* update all samples, tests, demo, tutorial to use (({GdkPixbuf#Pixbuf})).
[patch by cedlemo]
* add gtk3 demo transparent.rb.
[patch by cedlemo]
* update sample with (({Gtk::SeclectionData})) API changes.
[patch by cedlemo]
* lib and tests follow GI API improvement.
* test omit on old GTK+.
* require more newer GTK+ version for (({Gtk::TextView})) test.
==== Ruby/Pango
* Improvements
* add tests for (({Pango::Context})).
[patch by cedlemo]
* improve tests for (({Pango::Context})).
* implement (({Pango::Context#set_chape_renderer})).
* add (({Pango::AttrShape#data})).
[GitHub#766][Reported by cedlemo]
* Fixes
* fix error with MACRO usage in rbpangoattrlist.c.
[patch by cedlemo]
* fix a bug that (({Pango#GlyphGeometry#geometry=})) does nothing.
* fix GC related problem in rbpangocontext.c.
* windows: update external packages.
==== Ruby/GSF
* Improvements
* start GSF module.
[patch by cedlemo]
* Fixes
* follow GI API improvement.
* windows: fix external package name.
* windows: update external packages.
==== Ruby/GOffice
* Improvements
* start GOffice module.
[patch by cedlemo]
* support version.
* Fixes
* add tests for GOffice.
* use (({GI::Loader#rubyish_class_nme})) in loader.rb.
[patch by cedlemo]
* add rules in order to fix issues when building objects.
[patch by cedlemo]
* add availability check in tests.
* fix condition issue in test.
* make some tests omitted when run by Travis CI.
* windows: update external packages.
==== Ruby/Gnumeric
* Improvements
* start Gnumeric module.
[patch by cedlemo]
* Fixes
* add missing dependency in run-test.rb.
* use (({GI::Loader#rubyish_class_nme})) in loader.rb.
[patch by cedlemo]
* add rules in order to fix issues when building objects.
[patch by cedlemo]
* travis : add gnumeric as package to install instead of relying on dep.
[patch by cedlemo]
* improve renaming rules for C structures to ruby Class in GI loader.
[patch by cedlemo]
* add rules in order to avoid 2 objects information beeing reloaded.
[patch by cedlemo]
* windows: fix external package version.
==== Ruby/GdkPixbuf2
* Improvements
* replace C implementation with one based on GObject-Introspection.
[patch by cedlemo]
* implement backward compatibility for (({GdkPixbuf::Pixbuf.new})).
[patch by cedlemo]
* add rubyish methods.
* add (({GdkPixbuf::Pixbuf#composite!})).
* implement rubyish methods:
* (({GdkPixbuf#dup}))
* (({GdkPixbuf#fill!}))
* (({GdkPixbuf#rotate}))
* (({GdkPixbuf#saturate_and_pixelate}))
* (({GdkPixbuf#save}))
* (({GdkPixbuf#scale}))
* (({GdkPixbuf#scale!}))
* support version information.
* support options in (({GdkPixbuf::Pixbuf#composite})) and (({GdkPixbuf::Pixbuf#composite!})).
* Fixes
* remove needless lazy initialization.
* improve tests.
* add deprecated message for backward compatibility.
* add tests for (({GdkPixbuf::Pixbuf#new})).
[patch by cedlemo]
* add tests for (({GdkPixbuf::Pixbuf#dup})) and (({GdkPixbuf::Pixbuf#fill!})).
[patch by cedlemo]
* ignore no-gi version of gdk_pixbuf2 in run-test.rb and extconf.rb.
* add tests for (({GdkPixbuf::Pixbuf#rotate})) and (({GdkPixbuf::Pixbuf#new})).
[patch by cedlemo]
* update and fix issues in the samples.
[patch by cedlemo]
* improve gdk_pixbuf2 libs and tests style, and add missing copyright header.
* add missing GObjectIntrospection dependency in tests.
* add backward compatibility between the old Gdk::Pixbuf and GdkPixbuf::Pixbuf.
* supports running on different directory for gdk_pixbuf2 samples.
* add (({PixbufLoader#last_write})).
* add tests for (({GdkPixbuf#new})).
[patch by cedlemo]
* add tests for (({GdkPixbuf#new})) with :file and :scale parameters.
[patch by cedlemo]
* add tests for (({GdkPixbuf#new})) from subpixbuf.
[patch by cedlemo]
* add backward compatibily for (({GdkPixbuf::Pixbuf#new})) and (({GdkPixbuf::Pixbuf#new_subpixbuf})).
* simplify and improve tests.
* use warn and raise instead of puts messages.
* add test for (({GdkPixbuf::Pixbuf#dup})).
[patch by cedlemo]
* add test for (({GdkPixbuf::Pixbuf#new})) from data.
[patch by cedlemo]
* add test for (({GdkPixbuf::Pixbuf#fill!})).
[patch by cedlemo]
* clean (({GdkPixbuf::Pixbuf#fill!})) tests.
* add test for (({GdkPixbuf::Pixbuf#new})) from bytes.
[patch by cedlemo]
* clean (({GdkPixbuf::Pixbuf#new})) tests.
* add (({GdkPixbuf::Pixbuf#rotate})) tests.
[patch by cedlemo]
* clean (({GdkPixbuf::Pixbuf#rotate})) tests.
* add (({GdkPixbuf#new})) from resource test.
[patch by cedlemo]
* add (({GdkPixbuf::Pixbuf#new})) from resource at scale tests.
[patch by cedlemo]
* clean (({GdkPixbuf::Pixbuf#new})) tests.
* update sample with the GdkPixbuf::Pixbuf form.
* add deprecated hook for (({GdkPixbuf::PixbufFormat#signature})).
* add tests for (({GdkPixbuf::Pixbuf#saturate_and_pixelate})).
[patch by cedlemo]
* keep backward compatibility for (({Gdk::PixbufError})).
==== Ruby/VTE3
* Improvements
* Fixes
* remove needless lazy load.
* windows: update external package.
==== Ruby/GTK2
* Improvements
* Fixes
* use (({GdkPixbuf::Pixbuf})) form in library, test and sample files.
[patch by cedlemo]
==== Ruby/RSVG2
* Improvements
* Fixes
* use (({GdkPixbuf::Pixbuf})) form in library, test and sample files.
[patch by cedlemo]
* windows: update external packages.
==== Ruby/Clutter
* Improvements
* Fixes
* use (({GdkPixbuf::Pixbuf})) form in library, test and sample files.
[patch by cedlemo]
* windows update version.
==== Ruby/ClutterGtk
* Improvements
* Fixes
* use (({GdkPixbuf::Pixbuf})) form in library, test and sample files.
[patch by cedlemo]
* Fix issue with initialization of (({Gtk::IconView})) without hash argument.
[patch by cedlemo]
==== Ruby/ClutterGstreamer
* Improvements
* Fixes
* add missing gdk_pixbuf2 dependency in tests.
==== Ruby/GtkSourceView3
* Improvements
* Fixes
* windows: update dependencies.
==== Ruby/GSF
* Added.
==== Ruby/GOffice
* Added.
==== Ruby/Gnumeric
* Added.
==== Thanks
* Mamoru TASAKA
* rafagf1
* 5.5
* Abby Archer
* Eric Cunningham
* Hiroshi Hatake
* Renich Bon Ciric
* Miguel Hernández
* tSU_RooT
* YAMAMOTO Masayuki
* cedlemo
2016-08-04 Release 1.12.0
* Add Tins::Unit.parse? method to check for units in strings.
2016-07-12 Release 1.11.0
* Provide Tins::Unit.format and Tins::Unit.parse methods.
2016-05-25 Release 1.10.2
* Avoid some warnings.
2016-04-15 Release 1.10.1
* Fix bug in go if a concatenated argument comes last.
2016-04-12 Release 1.10.0
* Support mixed rest arguments in go argument parsing.
## 3.2.1 - 2016-07-19 {#version-3-2-1}
### Improvements
* Clarified lib/test/unit/diff.rb license. It's a triple license of
the Ruby license, PSF license and LGPLv2.1 or later.
[Reported by Luisa Pace]
* Reported norification when data driven test doesn't have
parameter.
[GitHub#122][Reported by Satoshi "Moris" Tagomori]
### Thanks
* Luisa Pace
* Satoshi "Moris" Tagomori
## 3.2.0 - 2016-06-12 {#version-3-2-0}
### Improvements
* Supported rxvt family terminals as color available terminals.
[GitHub#121][Reported by Ippei Kishida]
### Thanks
* Ippei Kishida
## 3.1.9 - 2016-05-20 {#version-3-1-9}
### Fixes
* Fixed conflict with test-unit-power_assert.
[GitHub#120][Patch by Kazuki Tsujimoto]
* Fixed a bug that path in `$LOAD_PATH` may be removed.
### Thanks
* Kazuki Tsujimoto
## 3.1.8 - 2016-03-19 {#version-3-1-8}
### Improvements
* Added `--stop-on-failure` command line option. With this option,
running test suite is stopped immediately when one test is failed
or an error is raised in one test.
## 1.4.3 20160821
* Quick fix of install failures. Do not try to use install 1.4.2.
## 1.4.2 20160820
* Refine SSL examples.
* Address issue #124 with additional RDOC.
* spec for Stomp::Client - check that headers passed to connection contain
required values as well as given custom and that given hash is not modified.
* Stomp::Client now does not modify given headers hash
* spec description enhancement.
* fix build_subscription_id - symbol and string were mixed up.
* STOMP_TESTSSL flag should enable all SSL tests.
* Add a basic Gemfile.
* Fix a memory leak in receipt implementation.
* Add unit test helper script.
## 1.4.1 20160623
* Add call to #post_connection_check to normal SSL processing. This change
further validates the name of the broker connected to. This change adds to
the current SSL connection processing logic, and is **highly recommended**. In the
case a client cannot tolerate this logic, it can be disabled by adding
:ssl_post_conn_check => false to the connection hash.
* Fix typo in SSL failure recovery processing.
## 1.4.0 20160608
* Connection parameter :parse_timeout now means IO:select wait time for socket
reads. Consumer clients should see a significantly reduced memory
footprint. If the default (5 seconds) is not used by your client,
we suggest you test carefully.
* Add example programs for sending / receiving large messages.
* Changelog format is changed from .rdoc to .md.
* README format is changed from .rdoc to .md.
* README format change of contributor's list.
* Add example utilities for generating the contributor's list.
* Eliminate many warnings when using 'rake test', mostly from the 2.x Ruby series.
* Get rakefile up to date.
* Add the 'adhoc' subdirectory, an area for experiments and issue recreation code.
### 3.5.2
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.1...v3.5.2)
Bug Fixes:
* Stop unnecessarily loading `rspec/core` from `rspec/rails` to avoid
IRB context warning. (Myron Marston, #1678)
* Deserialize arguments within ActiveJob matchers correctly.
(Wojciech Wnętrzak, #1684)
### 3.5.1 / 2016-07-08
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.0...v3.5.1)
Bugfixes:
* Only attempt to load `ActionDispatch::IntegrationTest::Behavior` on Rails 5,
and above; Prevents possible `TypeError` when an existing `Behaviour` class
is defined. (#1660, Betesh).
### 3.5.0 / 2016-07-01
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.0.beta4...v3.5.0)
**No user facing changes since beta4**
### 3.5.0.beta4 / 2016-06-05
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.0.beta3...v3.5.0.beta4)
Enhancements:
* Add support for block when using `with` on `have_enqueued_job`. (John Schroeder, #1578)
* Add support for `file_fixture(...)`. (Wojciech Wnętrzak, #1587)
* Add support for `setup` and `teardown` with blocks (Miklós Fazekas, #1598)
* Add `enqueue_job ` alias for `have_enqueued_job`, support `once`/`twice`/
`thrice`, add `have_been_enqueued` matcher to support use without blocks.
(Sergey Alexandrovich, #1613)
Bug fixes:
* Prevent asset helpers from taking precendence over route helpers. (Prem Sichanugrist, #1496)
* Prevent `NoMethodError` during failed `have_rendered` assertions on weird templates.
(Jon Rowe, #1623).
### 3.5.0.beta3 / 2016-04-02
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.0.beta2...v3.5.0.beta3)
Enhancements:
* Add support for Rails 5 Beta 3 (Sam Phippen, Benjamin Quorning, Koen Punt, #1589, #1573)
Bug fixes:
* Support custom resolvers when preventing views from rendering.
(Jon Rowe, Benjamin Quorning, #1580)
### 3.5.0.beta2 / 2016-03-10
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.5.0.beta1...v3.5.0.beta2)
Enhancements:
* Include `ActionDispatch::IntegrationTest::Behavior` in request spec
example groups when on Rails 5, allowing integration test helpers
to be used in request specs. (Scott Bronson, #1560)
Bug fixes:
* Make it possible to use floats in auto generated (scaffold) tests.
(Alwahsh, #1550)
### 3.5.0.beta1 / 2016-02-06
[Full Changelog](http://github.com/rspec/rspec-rails/compare/v3.4.2...v3.5.0.beta1)
Enhancements:
* Add a `--singularize` option for the feature spec generator (Felicity McCabe,
#1503)
* Prevent leaking TestUnit methods in Rails 4+ (Fernando Seror Garcia, #1512)
* Add support for Rails 5 (Sam Phippen, #1492)
Bug fixes:
* Make it possible to write nested specs within helper specs on classes that are
internal to helper classes. (Sam Phippen, Peter Swan, #1499).
* Warn if a fixture method is called from a `before(:context)` block, instead of
crashing with a `undefined method for nil:NilClass`. (Sam Phippen, #1501)
* Expose path to view specs (Ryan Clark, Sarah Mei, Sam Phippen, #1402)
* Prevent installing Rails 3.2.22.1 on Ruby 1.8.7. (Jon Rowe, #1540)
* Raise a clear error when `have_enqueued_job` is used with non-test
adapter. (Wojciech Wnętrzak, #1489)
### 3.5.0 / 2016-07-01
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.5.0.beta4...v3.5.0)
**No user facing changes since beta4**
### 3.5.0.beta4 / 2016-06-05
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.5.0.beta3...v3.5.0.beta4)
Bug Fixes:
* Fix `include` matcher so that it provides a valid diff for hashes. (Yuji Nakayama, #916)
### 3.5.0.beta3 / 2016-04-02
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.5.0.beta2...v3.5.0.beta3)
Enhancements:
* Make `rspec/expectations/minitest_integration` work on Minitest::Spec
5.6+. (Myron Marston, #904)
* Add an alias `having_attributes` for `have_attributes` matcher.
(Yuji Nakayama, #905)
* Improve `change` matcher error message when block is mis-used.
(Alex Altair, #908)
### 3.5.0.beta2 / 2016-03-10
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.5.0.beta1...v3.5.0.beta2)
Enhancements:
* Add the ability to raise an error on encountering false positives via
`RSpec::Configuration#on_potential_false_positives = :raise`. (Jon Rowe, #900)
* When using the custom matcher DSL, support new
`notify_expectation_failures: true` option for the `match` method to
allow expectation failures to be raised as normal instead of being
converted into a `false` return value for `matches?`. (Jon Rowe, #892)
Bug Fixes:
* Allow `should` deprecation check to work on `BasicObject`s. (James Coleman, #898)
### 3.5.0.beta1 / 2016-02-06
[Full Changelog](http://github.com/rspec/rspec-expectations/compare/v3.4.0...v3.5.0.beta1)
Enhancements:
* Make `match_when_negated` in custom matcher DSL support use of
expectations within the match logic. (Chris Arcand, #789)
Bug Fixes:
* Return `true` as expected from passing negated expectations
(such as `expect("foo").not_to eq "bar"`), so they work
properly when used within a `match` or `match_when_negated`
block. (Chris Arcand, #789)
### 3.5.4 / 2016-09-30
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.3...v3.5.4)
Bug Fixes:
* Remove accumulated `ExampleGroup` constants when reseting RSpec,
preventing a memory leak. (TravisSpangle, #2328)
### 3.5.3 / 2016-09-02
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.2...v3.5.3)
Bug Fixes:
* When applying shared group metadata to a host group, overwrite
conflicting keys if the value in the host group was inherited from
a parent group instead of being specified at that level.
(Myron Marston, #2307)
* Handle errors in `:suite` hooks and provide the same nicely formatted
output as errors that happen in examples. (Myron Marston, #2316)
* Set the exit status to non-zero when an error occurs in an
`after(:context)` hook. (Myron Marston, #2320)
### 3.5.2 / 2016-07-28
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.1...v3.5.2)
Bug Fixes:
* Wait to report `example_finished` until the example's `execution_result`
has been completely filled in. (Myron Marston, #2291)
* Make sure example block is still available when using `duplicate_with`
to clone examples. (bootstraponline, #2298)
* Don't include the default `--pattern` in the Rake task when
`rspec_opts` specifies its own. (Jon Rowe, #2305)
### 3.5.1 / 2016-07-06
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.0...v3.5.1)
Bug Fixes:
* Ensure that config hooks that are added to existing example groups are
added only once. (Eugene Kenny, #2280)
### 3.5.0 / 2016-07-01
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.0.beta4...v3.5.0)
Enhancements:
* Include any `SPEC_OPTS` in reproduction command printed at the end of
a bisect run. (Simon Coffey, #2274)
Bug Fixes:
* Handle `--bisect` in `SPEC_OPTS` environment variable correctly so as
to avoid infinite recursion. (Simon Coffey, #2271)
### 3.5.0.beta4 / 2016-06-05
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.0.beta3...v3.5.0.beta4)
Enhancements:
* Filter out bundler stackframes from backtraces by default, since
Bundler 1.12 now includes its own frames in stack traces produced
by using `bundle exec`. (Myron Marston, #2240)
* HTML Formatter uses exception presenter to get failure message
for consistency with other formatters. (@mrageh, #2222)
* Load spec files in the order of the directories or files passed
at the command line, making it easy to make some specs run before
others in a one-off manner. For example, `rspec spec/unit
spec/acceptance --order defined` will run unit specs before acceptance
specs. (Myron Marston, #2253)
* Add new `config.include_context` API for configuring global or
filtered inclusion of shared contexts in example groups.
(Myron Marston, #2256)
* Add new `config.shared_context_metadata_behavior = :apply_to_host_groups`
option, which causes shared context metadata to be inherited by the
metadata hash of all host groups and examples instead of configuring
implicit auto-inclusion based on the passed metadata. (Myron Marston, #2256)
Bug Fixes:
* Fix `--bisect` so it works on large spec suites that were previously triggering
"Argument list too long errors" due to all the spec locations being passed as
CLI args. (Matt Jones, #2223).
* Fix deprecated `:example_group`-based filtering so that it properly
applies to matching example groups. (Myron Marston, #2234)
* Fix `NoMethodError` caused by Java backtraces on JRuby. (Michele Piccirillo, #2244)
### 3.5.0.beta3 / 2016-04-02
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.0.beta2...v3.5.0.beta3)
Enhancements:
* Add new `config.filter_run_when_matching` API, intended to replace
the combination of `config.filter_run` and
`config.run_all_when_everything_filtered` (Myron Marston, #2206)
Bug Fixes:
* Use the encoded string logic for source extraction. (Jon Rowe, #2183)
* Fix rounding issue in duration formatting helper. (Fabersky, Jon Rowe, #2208)
* Fix failure snippet extraction so that `def-end` snippets
ending with `end`-only line can be extracted properly.
(Yuji Nakayama, #2215)
### 3.5.0.beta2 / 2016-03-10
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.5.0.beta1...v3.5.0.beta2)
Enhancements:
* Remove unneeded `:execution_result` example group metadata, saving a
bit of memory. (Myron Marston, #2172)
* Apply hooks registered with `config` to previously defined groups.
(Myron Marston, #2189)
* `RSpec::Core::Configuration#reporter` is now public API under SemVer.
(Jon Rowe, #2193)
* Add new `config.when_first_matching_example_defined` hook. (Myron
Marston, #2175)
### 3.5.0.beta1 / 2016-02-06
[Full Changelog](http://github.com/rspec/rspec-core/compare/v3.4.4...v3.5.0.beta1)
Enhancements:
* Add `RSpec::Core::ExampleGroup.currently_executing_a_context_hook?`,
primarily for use by rspec-rails. (Sam Phippen, #2131)
Bug Fixes:
* Ensure `MultipleExceptionError` does not contain a recursive reference
to itself. (Sam Phippen, #2133)
2016-05 Release 0.5.2
Horst Duchene <monora@gmail.com>
* Issue #21: Use new method vertex_id instead of object_id to identify vertices in dot export. (fa7592)
* Integrate Code Climate's test coverage reporting (0ab722)
* Clarify traversal order of DFS search (see #20). (afa788)
Chase Gilliam <chase.gilliam@gmail.com>
* drop 1.9.3 add newer jruby and rubinius (fad333)
Mat«¿as Battocchia <matias@riseup.net>
* Switched to a different heap implementation. (bd7c13)
gorn <j@kub.cz>
* Adding failing test for issue #24 (1f6204)
== 1.8.0 - 19-Jun-2016
* Changed license to Apache 2.0.
* Some cleanup and updates to the Rakefile and benchmarks.
* Refactored some realpath tests so they're not touching system files like
/dev/stdin any more. Thanks go to Michael R. Crusoe for pointing out the
potential pitfalls of doing that.
=== 5.9.1 / 2016-09-25
* 2 bug fixes:
* Re-release to refresh gem certificate signing. ugh.
* Fixed hoe/minitest to not augment load path if we're actually testing minitest.
=== 5.9.0 / 2016-05-16
* 8 minor enhancements:
* Added Minitest.info_signal accessors to customize signal for test run info. (nate)
* Added assert_mock to make it more clear that you're testing w/ them.
* Added negative filter by test name. (utilum)
* Added warning to README that 1.8 and 1.9 support will be dropped in minitest 6.
* Automatically activate minitest/hell if $MT_HELL is defined.
* Improved default error messages for assert and refute. (bhenderson)
* minitest/hell now tries to require minitest/proveit
* mu_pp for strings prints out non-standard encodings to improve assert_equal diffs.
* 1 bug fix:
* Removed Interrupt from PASSTHROUGH_EXCEPTIONS (already handled). (waldyr)
* Add hamlit parser support for haml.
* Filter out options unknown to msgcat.
* Fix issue requiring libraries when they were already loaded.
* Fix Rails 5 compatibility.
## 3.2.2: 2016-04-17 {#version-3-2-2}
### Improvements
* Supported non POSIX locale format such as "zh-Hant" for
.mo search path.
[GitHub#45][Patch by Micha«³l Hoste]
2.2.0:
- #new_instances now mocks the #initialize method instead of mocking after the
allocation was done. This allows to do mock methods called by #initialize
itself. Behaviour when the allocator is explicitely provided is left
unchanged, which means that the old behaviour is still available by passing
:new to new_instances.
2.1.0:
- added `#and_iterates` to fix some shortcomings of `#and_yield` without
breaking backward compatibility
- strict partial mocks (and "based mocks" if `FlexMock.partials_are_based` is
set to true) are now based on the object's singleton class, instead of its
class.
* Minor documentation and code style improvements.
* Support Ruby 2.1.8.
* Support Ruby 2.3.0.
* Fix library loading symbols.
For more detail, please refer:
<https://github.com/ffi/ffi/compare/1.9.10...1.9.14>.
2.37 - 2016-10-07
* Make minimum version of ExtUtils::MakeMaker 6.58
* Better support parallel builds by making 'dynamic' a dependency of the ReadKey.pm
* Add a 'provides' META section to help indexing
2.36 - 2016-10-07
* Add a package declaration to the ReadKey_pm.PL to aid indexing
2.35 - 2016-10-06
* Use FULLPERLRUNINST to run the ReadKey_pm.PL to get the 'blib'
2.34 - 2016-10-06
* Fix GetTerminalSize() argument handling
* Fix miniperl build on Win32
* Remove unused Cwd
* Fix TAP output
* Fix warnings in XS file
* Don't attempt some tests if STDIN is not a terminal
* Generate ReadKey.pm from ReadKey_pm.PL to expand blockoptions
* demand-load Carp,
* remove unneeded AutoLoader,
* Improve genchars.pl, support make -s silence.
* Improve Makefile.PL add SIGN, fix pure_site_install, support dual-life, fix realclean
Extends nose.plugins.cover to add Cobertura-style XML reports.
Includes nose-xmlcover, a companion to the built-in nose.plugins.cover,
this plugin will write out an XML coverage report to a file named
coverage.xml.
Changelog:
The NSS team has released Network Security Services (NSS) 3.27.1.
This is a patch release to address a TLS compatibility issue
that some applications experienced with NSS 3.27.
Notable Changes:
Availability of the TLS 1.3 (draft) implementation has been re-disabled
in the default build.
Previous versions of NSS made TLS 1.3 (draft) available only when compiled
with NSS_ENABLE_TLS_1_3. NSS 3.27 set this value on by default, allowing
TLS 1.3 (draft) to be disabled using NSS_DISABLE_TLS_1_3, although the
maximum version used by default remained TLS 1.2.
However, some applications query the list of protocol versions that are
supported by the NSS library, and enable all supported TLS protocol versions.
Because NSS 3.27 enabled compilation of TLS 1.3 (draft) by default, it caused
those applications to enable TLS 1.3 (draft). This resulted in connectivity
failures, as some TLS servers are version 1.3 intolerant, and failed to
negotiate an earlier TLS version with NSS 3.27 clients.
New Libraries
-------------
Fiber:
* Framework for userland-threads/fibers, from Oliver Kowalke.
QVM:
* Boost QVM is a generic library for working with quaternions, vectors and matrices of static size with the emphasis on 2, 3 and 4-dimensional operations needed in graphics, video games and simulation applications, from Emil Dotchevski.
Updated Libraries
-----------------
Atomic:
* Improved support for Oracle Studio and SPARC. The library now provides native atomic operations on SPARCv8+.
Chrono:
* 11330 boost::chrono::duration default constructor doesn't initialize rep_
* 11618 Chrono IO V2 doc ios_state.hpp does not exist
* 11631 boost chrono io v2 does not let you support custom clocks
Circular Buffer:
* Debug implementation is no longer used by default in debug mode. Debug implementation does not have the same thread safety guarantees as the normal (release) implementation or other containers. As a result of this change BOOST_CB_DISABLE_DEBUG macro is no longer used. BOOST_CB_ENABLE_DEBUG=1 should be defined instead to enable debug support.
Container:
* Fixed bugs:
Trac 9481: "Minor comment typo in Boost.Container".
Trac 9689: "Add piecewise_construct to boost::container".
Trac 11170: "Doc slip for index_of".
Trac 11802: "Incorrect ordering after using insert() with ordered_range_t on a flat_multiset with a non-default sort order".
Trac 12117: "flat_set constructor with ordered_unique_range".
Trac 12177: "vector::priv_merge uses unqualified uintptr_t".
Trac 12183: "GCC 6.1 thinks boost::container::string violates strict aliasing".
Trac 12256: "set<std::pair<int,int>>::insert cause compilation error in debug configuration in Visual Studio 2012".
Trac 12273: "static_vector max_size() and capacity() should be constant expressions". Added constant static_vector<>::static_capacity to use the configured capacity in constant expressions.
Trac 12286: "PMR flat_map from Boost Container does not compile".
Trac 12296: "{deque,string} combine for a memory leak".
Trac 12319: "flat_set should be nothrow move constructible".
* Revised noexcept expressions of default and move constructors in all containers.
* Implemented C++17 insert_or_assign/try_emplace for map and flat_map.
* Implemented C++17 extract/insert(node) for map, multimap, set, multiset.
Context:
* 12215 all stack corrupted on Windows using default fixedsize_stack
* 12242 build issue with Intel C++ compiler for MacOS
* fix stack unwinding for execution_context_v1
Coroutine:
* deprecated in favour of Coroutine2
Coroutine2:
* 12221 coroutine<>::push_type function starts without pushing
* do not swallow exceptions thrown at coroutine<>::push_type construction
DLL:
* Mangled symbols and classes loading was implemented by Klemens Morgenstern
* Suppress a 'unused parameter' warning in detail::aggressive_ptr_cast pull-request 9
* Allowed to query non-existant sections pull-request 32
* More tests and docs
Functional/Forward:
* Fix C++11 compile error.
Interprocess:
* Fixed bug GitHub Pull 27 ("Fix undefined behavior").
Intrusive:
* Fixed bugs:
Boost Trac 11476: has_member_function_callable_with.hpp is massively broken with BOOST_NO_CXX11_DECLTYPE
Boost Trac 11994: Support intrusive container key extractors that return the key by value
Boost Trac 12184: clang -Wdocumentation warning
Boost Trac 12190: Intrusive List + Flat Map combination crashes
Boost Trac 12229: intrusive::unordered_set<T>::rehash() broken
Boost Trac 12245: bstree uses a shared static size_traits for constant_time_size<false>
Lexical Cast:
* Fix incorrect static assertion 11759
Log:
* New features:
Added new tools for inter-process logging on a local machine. The implementation includes a resource name wrapper, an inter-process message queue and a sink backend.
Added a new character decorator called max_size_decor. The new decorator allows to limit the output of its adopted formatter up to the specified length.
* Bug fixes:
Fixed that logging streams could retain formatting settings across different log records. (12178)
See changelog for more details.
Math:
* New Features:
Enabled all the special function code to work correctly with types whose precision can change at runtime: for example type mpfr_float from Boost.Multiprecision.
* Patches:
Fix tgamma_delta_ratio for cases where the delta is small compared to the base.
Fix misc GCC-4.4 test failures.
Optional:
* Fixed Trac 12179.
Phoenix:
* Fix compilation for MSVC 10. pull-request 18
* Remove unused preprocessed files. pull-request 28
* Fix erroneous doc for switch_ statement. 8156
* Cleanup config unordered. pull-request 33
* Suppress unused placeholder warnings.
* Fix too few arguments error on binary math functions.
Regex:
* Fix buffer over-run error when parsing certain invalid regexes, see 12222.
* Fix detection of ICU in library build, see 12152.
* Fix bug in case sensitivity change, see 11940.
* Allow types wider than int in \x{} expressions (for char32_t etc), see 11988.
Test:
* Boost.test v3.3 see the Change log section for more details.
* New features
dataset driven test-cases have now a unique name and all sample tests of a dataset live inside the same test suite
Boost.test learned a new command line switch, --logger for setting a test logger as well as its parameter in one command
Boost.test learned to have several loggers at the same time, each of which having their own log level and output stream
Boost.test learned a new logger, JUNIT, that is able to output Junit/xUnit compatible streams
It is now possible to specify several test filters through the environment variable BOOST_TEST_RUN_FILTERS
* Bug fixes
Trac tickets 8707, 8834, 11128, 11845, 11859, 12024,
Trac tickets 12093, 12103, 12224, 12241, 12257, 12378
Thread:
* 12102 condition_variable_fwd.hpp fails to compile when BOOST_THREAD_PROVIDES_INTERRUPTIONS is disabled
* 12120 Performance improvement in thread/barrier.hpp
* 12146 make_exceptional_future is not mentioned in the docs
* 12202 shared_lock should be in shared_mutex header
* 12371 boost thread/future.hpp fails to build
* 88 fix typos in boost::upgrade_lock
* 89 fix a bug in upgrade_to_unique_lock<>::operator=()
* 90 fix a bug in try_lock_wrapper<>::operator=()
* 91 Add shared_lock_guard to the included lock types
* 92 Fixed compilation with MSVC-8.
* 93 Fix variable shadowing warnings (Clang)
* 94 fix bugs in boost::barrier
* 95 fix a mistake in boost::completion_latch
* 96 rename async_func.hpp to invoker.hpp.
* 97 fix a mistake in sync_timed_queue<>::pull_until()
TypeIndex:
* ctti_type_index was made constexpr in C++14. Now it is possible to retrieve actual name of the type as const char* at compile time. So we can do a lot of interesting things at compile time: check namespace of a type, sort types according to their lexical representation, check that type is a template type and so on... A few examples were provided.
* Fixed issue with noexcept on MSVC in type names and issue with space in type name (pulled from klemens-morgenstern/develop)
* Fixed typos pull-request 7
* Improved CI testing
Unordered:
* Remove use of deprecated boost::iterator.
* Fixed an exception safety issue in assignment of unordered_multiset and unordered_multimap.
* See the changelog for more details.
Variant:
* Fixed variant construction from classes derived from variant 7120, 10278, 12155
* Variant constructors and assignment operators now do not participate in overload resolutions if variant can not hold the input type 5871, 11602
* Fixed the issue with implicit conversion operator in C++11 8555
* Fixed comparisons when variant holds non-const reference 11751
* Fixed noexcept for variant's move-assignment 11696
* Fixed double quotes in includes 12057
* Dropped BOOST_VARIANT_NO_REFERENCE_SUPPORT macro and dropped support for compilers without SFINAE 12250
* Suppress implicit instantiation of MPL end iterator
* Multiple minor fixes (typos, missing includes)
What's new since 2.3.0:
* New test suite based on Busted; runs on Linux, OSX and Windows. This
test suite was developed by Robert Karasek as his Google Summer of
Code project for LabLua. You can check out the combined coverage of
all platforms at CodeCov:
https://codecov.io/gh/keplerproject/luarocks/
* git+ssh:// fetch protocol
* Improved behavior preserving permissions
* Improved listing of dependencies on installation
* Improved behavior of argument handling in `pack`
* MSYS and Haiku platform detection
* Feature-based detection of internal bit32 and utf8 modules
* Internal reorganization of luarocks.fs code
* `remove` option --force=fast renamed to --force-fast
* Plus assorted bugfixes and cleanups
What's new since 2.4.0:
* Avoid coroutine use inside luarocks.loader, that apparently affected
OpenResty's require()
* Fix upgrade issues for old versions (e.g. when upgrading from LuaRocks 2.1.0)
-----------------
* Clarify various ways to specify the "revision ranges" in the
documentation.
* "diff-highlight" script (in contrib/) learned to work better with
"git log -p --graph" output.
* The test framework left the number of tests and success/failure
count in the t/test-results directory, keyed by the name of the
test script plus the process ID. The latter however turned out not
to serve any useful purpose. The process ID part of the filename
has been removed.
* Having a submodule whose ".git" repository is somehow corrupt
caused a few commands that recurse into submodules loop forever.
* "git symbolic-ref -d HEAD" happily removes the symbolic ref, but
the resulting repository becomes an invalid one. Teach the command
to forbid removal of HEAD.
* A test spawned a short-lived background process, which sometimes
prevented the test directory from getting removed at the end of the
script on some platforms.
* Update a few tests that used to use GIT_CURL_VERBOSE to use the
newer GIT_TRACE_CURL.
* Update Japanese translation for "git-gui".
* "git fetch http::/site/path" did not die correctly and segfaulted
instead.
* "git commit-tree" stopped reading commit.gpgsign configuration
variable that was meant for Porcelain "git commit" in Git 2.9; we
forgot to update "git gui" to look at the configuration to match
this change.
* "git log --cherry-pick" used to include merge commits as candidates
to be matched up with other commits, resulting a lot of wasted time.
The patch-id generation logic has been updated to ignore merges to
avoid the wastage.
* The http transport (with curl-multi option, which is the default
these days) failed to remove curl-easy handle from a curlm session,
which led to unnecessary API failures.
* "git diff -W" output needs to extend the context backward to
include the header line of the current function and also forward to
include the body of the entire current function up to the header
line of the next one. This process may have to merge to adjacent
hunks, but the code forgot to do so in some cases.
* Performance tests done via "t/perf" did not use the same set of
build configuration if the user relied on autoconf generated
configuration.
* "git format-patch --base=..." feature that was recently added
showed the base commit information after "-- " e-mail signature
line, which turned out to be inconvenient. The base information
has been moved above the signature line.
* Even when "git pull --rebase=preserve" (and the underlying "git
rebase --preserve") can complete without creating any new commit
(i.e. fast-forwards), it still insisted on having a usable ident
information (read: user.email is set correctly), which was less
than nice. As the underlying commands used inside "git rebase"
would fail with a more meaningful error message and advice text
when the bogus ident matters, this extra check was removed.
* "git gc --aggressive" used to limit the delta-chain length to 250,
which is way too deep for gaining additional space savings and is
detrimental for runtime performance. The limit has been reduced to
50.
* Documentation for individual configuration variables to control use
of color (like `color.grep`) said that their default value is
'false', instead of saying their default is taken from `color.ui`.
When we updated the default value for color.ui from 'false' to
'auto' quite a while ago, all of them broke. This has been
corrected.
* A shell script example in check-ref-format documentation has been
fixed.
* "git checkout <word>" does not follow the usual disambiguation
rules when the <word> can be both a rev and a path, to allow
checking out a branch 'foo' in a project that happens to have a
file 'foo' in the working tree without having to disambiguate.
This was poorly documented and the check was incorrect when the
command was run from a subdirectory.
* Some codepaths in "git diff" used regexec(3) on a buffer that was
mmap(2)ed, which may not have a terminating NUL, leading to a read
beyond the end of the mapped region. This was fixed by introducing
a regexec_buf() helper that takes a <ptr,len> pair with REG_STARTEND
extension.
* The procedure to build Git on Mac OS X for Travis CI hardcoded the
internal directory structure we assumed HomeBrew uses, which was a
no-no. The procedure has been updated to ask HomeBrew things we
need to know to fix this.
* When "git rebase -i" is given a broken instruction, it told the
user to fix it with "--edit-todo", but didn't say what the step
after that was (i.e. "--continue").
* "git add --chmod=+x" added recently lacked documentation, which has
been corrected.
* "git add --chmod=+x <pathspec>" added recently only toggled the
executable bit for paths that are either new or modified. This has
been corrected to flip the executable bit for all paths that match
the given pathspec.
* "git pack-objects --include-tag" was taught that when we know that
we are sending an object C, we want a tag B that directly points at
C but also a tag A that points at the tag B. We used to miss the
intermediate tag B in some cases.
* Documentation around tools to import from CVS was fairly outdated.
* In the codepath that comes up with the hostname to be used in an
e-mail when the user didn't tell us, we looked at ai_canonname
field in struct addrinfo without making sure it is not NULL first.
Also contains minor documentation updates and code clean-ups.
0.21 2016-10-02 05:00:12Z
- no change since 0.20
0.20 2016-09-11 00:33:10Z (TRIAL RELEASE)
- fix occasional segmentation fault on OpenBSD when malloc randomization
causes nameptr to be at the beginning of the page (RT#117072, Alexander
Bluhm)
0.28 2016-10-02
- Added a Test::Specio module to provide helpers for testing Specio libraries.
- Fixed another bug with a subtype of special types and inlining.
0.27 2016-10-01
- Cloning a type with coercions defined on it would cause an exception.
- Creating a subtype of a special type created by *_isa_type, *_can_type, or
*_does_type, or enum would die when trying to inline the type constraint.
- Removed the never-documented Any type.
- Added documentation for each type in Specio::Library::Builtins.
1.46 -- 2016/09/28 23:17:07
[CHANGES]
* Remember to add unpairs and uniq to SYNOPSIS
* Document the behaviour of uniqnum() on NaN
[BUGFIXES]
* C89 fixes
* Various internal bugfixes backported from perl core
2016-09-26 meld 3.16.3
======================
Fixes:
* Fix bad scroll syncing and inability to move between comparison chunks in
file comparisons (Vasily Galkin)
* Improve version control view behaviour when opening missing paths (Vasily
Galkin)
* Fix saving of text created in blank comparison (Vasily Galkin)
* GTK+ 3.22 compatibility fixes for saving window size (Kai Willadsen)
* Fix formatted patches missing newlines in certain cases (Kai Willadsen)
* Silence GTK+ warning logging if not running in uninstalled (i.e.,
development) mode (Kai Willadsen)
* Update AppData (Kai Willadsen)
Translations:
* Anders Jonsson (sv)
* Balázs Meskó (hu)
* Daniel Mustieles (es)
* Marek Černocký (cs)
* Mario Blättermann (de)
* Piotr Drąg (pl)
* Rafael Fontenelle (pt_BR)
* Tiago Santos (pt)
* Мирослав Николић (sr, sr@latin)
v28.1.0
-------
* #803: Bump certifi to 2016.9.26.
v28.0.0
-------
* #733: Do not search excluded directories for packages.
This introduced a backwards incompatible change in ``find_packages()``
so that ``find_packages(exclude=['foo']) == []``, excluding subpackages of ``foo``.
Previously, ``find_packages(exclude=['foo']) == ['foo.bar']``,
even though the parent ``foo`` package was excluded.
* #795: Bump certifi.
* #719: Suppress decoding errors and instead log a warning
when metadata cannot be decoded.
v27.3.1
-------
* #790: In MSVC monkeypatching, explicitly patch each
function by name in the target module instead of inferring
the module from the function's ``__module__``. Improves
compatibility with other packages that might have previously
patched distutils functions (i.e. NumPy).
v27.3.0
-------
* #794: In test command, add installed eggs to PYTHONPATH
when invoking tests so that subprocesses will also have the
dependencies available. Fixes `tox 330
<https://github.com/tox-dev/tox/issues/330>`_.
* #795: Update vendored pyparsing 2.1.9.
fixes build for 32bit when passing USE_64 (which is questionable)...
in pkgsrc we declare all mips64* platforms as 64bit, and use USE_64.
However, netbsd/mips64 is using a 32bit ABI, so it is akin to passing
USE_64=1 for 32bit.
perhaps not declaring it a 64bit platform is correct, but this package
is one of the only few using this logic, and it's unfeasible to have
correct logic for 32bit/64bit.
this package has considerably more logic for USE_64 than for USE_NSS_64,
so to avoid inadvertent damage to other platforms, retain the USE_64=1
logic.
feel free to object to this option in the discussion on tech-pkg.
This is a regularly-scheduled bugfix release.
annotate: correct output in some merge cases (issue5360)
crecord: properly handle files with No newline at eof (issue5268)
grep: rewrite help to better document current (confusing) behavior
Asio 1.10.8
* Added compatibility with OpenSSL 1.1.0.
* Fixed out-of-bounds iterator use in `asio::connect()` when the
`connect_condition` returns an end iterator.
* Added a workaround for a move detection problem on MSVC 2015 Update 2.
* Changed a workaround that was previously added for broken Windows firewalls
to only bind to 127.0.0.1 if `getsockname` reports 0.0.0.0.
* Added call to `SSL_COMP_free_compression_methods` to fix two memory leaks
reported at shutdown, for OpenSSL versions >= 1.0.2 and < 1.1.0.
* Fixed `use_future` compile error encountered on some standard library
implementations, by changing `std::allocator<void>` use to a non-void
template parameter.
* Enabled use of native `getaddrinfo` by default on Apple OSes, rather than
emulation in terms of `getipnodebyname`.
Asio 1.10.7
* Added support for Windows 8.1 Store apps.
* Fixed macro multiple definition error on Microsoft Visual Studio 2015.
* Changed Asio's SSL wrapper to respect OpenSSL's `OPENSSL_NO_SSL3` feature
test `#define`.
* Changed Asio's SSL wrapper to use OpenSSL's new `SSL_CTX_clear_chain_certs`
function, if available.
* Suppressed a clang 3.6+ warning about unused typedefs.
* Regenerated certificates used by SSL examples.
* Fixed buffer sizes passed to `strncat` in the `getaddrinfo` emulation and in
the SSL wrapper's password handling.
* Changed Windows backend to use non-macro `CreateEventW` rather than
`CreateEvent`.
libbson 1.4.2
- No change since 1.4.1; released to keep pace with libmongoc's
version.
libbson 1.4.1
- This release improves the HTML documentation's Makefile.
Upstream changes:
2.1806 2016-09-25
[BUG FIXES}
- A role generated on the fly (as opposed to one in a file on disk) could
not be used as a trait. Fixed by Aaron Cohen. (RT#92089, PR#47).
[ENHANCEMENTS]
- When a required attribute's init_arg differs from the attribute's name, we
now include both the attribute name and init_arg in the exception thrown
when that attribute is not provided. (RT#116814, Gordon Fisher)
- The unimport sub generated by Moose::Exporter can now be called with
parameters like { into => 'Package' } or { into_level => 1 }, just like
the import sub it generates. Requested by Dmytro Zagashev. (RT #93666).
- Moose::Meta::Attribute objects created from role attributes keep a
reference to the Moose::Meta::Role::Attribute which created
them. Moose::Meta::Attribute now has ->role_attribute and
->has_role_attribute methods. RT #84563 and #117586. (Dave Rolsky)
[DOCUMENTATION]
- Updated Moose::Manual::Roles with more detail on role consumption and how
to deal with various corner cases involving required methods, multiple
roles, etc.
Jenkins 2 is a drop-in replacement of the Jenkins 1.x series of
releases and fully backward compatible.
2.7.4 (2016-09-08)
- Prevent File descriptor leaks when reading plugin manifests. It
causes failures during the upgrade of detached plugins on
Windows.
2.7.3 (2016/08/31)
- Stop A/B testing of the remoting JNLP3 protocol due to the known
issues. The protocol can be enabled manually via the
jenkins.slaves.JnlpSlaveAgentProtocol3.enabled system property.
- When checking Update Center, append ?uctest parameter to HTTP
and HTTPS URLs only.
- Ensure that detached plugins are always at least their minimum
version.
- Remove trailing space from Hudson.DisplayName in Spanish, which
resulted in problems with Blue Ocean.
- Make sure that the All view is created.
- Incorrect formatting of messages in the Update Center and Setup
Wizard.
- Underprivileged users were unable to use the default value of a
password parameter.
- Properly handle exceptions during global configuration form
submissions when SCM Retry Count field is empty.
- Do not allow disabled project to be triggered remotely.
- Ensure that SCMDescriptor.newInstance overrides are honored when
creating new SCM entries.
- Add a cache for user information to fix performance regression
due to SECURITY-243.
- Performance: Disable AutoBrowserHolder by default to improve the
changelog rendering performance.
- Honor non-default update sites in setup wizard.
2.7.2 (2016/08/03)
- Always send usage statistics over HTTPs to the new
usage.jenkins.io hostname.
- Fix issues in file management in hudson.remoting.Launcher (main
executable class).
- Remoting 2.60: Fix potential file handle leaks during the build
agent (FKA slave) startup. issue 35190)
- Remoting 2.60: Proper handling of the no_proxy environment
variable.
- Performance: Improve configuration page load times by removing
the CodeMirror reloading cycle.
- Remoting 2.60: hudson.Remoting.Engine#waitForServerToBack now
uses credentials for connection.
- IllegalStateException under certain conditions when reloading
configuration from disk while jobs are in the queue.
- Allow keeping builds forever with custom build retention
strategies.
- Remoting 2.60: Make the channel reader tolerant against Socket
timeouts.
2.7.1 (2016/07/06)
Changes from 2.7:
- Installation Wizard: Do not offer creating new admin user if the
security is preconfigured.
- API: Make it easier for UpdateSites to tweak the
InstallationJob.
- Fix the repeatable item delete button layout in Safari.
Addresses Build Steps and other such configuration items.
- Prevent NullPointerException on startup after update from
Jenkins 2.5.
- Explicitly declare compatibility of Windows build agent service
with .NET Framework 4.
- Honor noProxy settings from "Manage Jenkins > Manage Plugins >
Advanced".
- API: Restrict external usages of
jenkins.util.ResourceBundleUtil.
- Internal: Upgrade Groovy to 2.4.7 to finalize the fix in Jenkins
2.7.
Notable changes since 1.651.3:
- More detailed information about the new features in Jenkins 2 on
the overview page. Note that AJP support has been removed, if
your service script enables it, Jenkins will fail to start.
- New password-protected setup wizard shown on first run to guide
users through installation of popular plugins and setting up an
admin user.
- Plugin bundling overhaul: Bundled plugins are only installed if
necessary when upgrading, all plugins can be uninstalled.
- Redesigned job configuration form makes it easier to understand
the option hierarchy, and to navigate the form.
- Richer 'Create Item' form with job icons and job categories
(once a threshold of three categories has been reached).
- Support encrypted communication between master and JNLP slaves.
- Enable disabled dependencies during plugin installations.
- Force ordering between GPG and jarsigner to ensure correct GPG
signature.
- Secured Jenkins installations didn't properly save the queue on
shutdown.
- Upgrade wizard encourages installation of Pipeline related
plugins when upgrading from 1.x.
- Jenkins now requires Servlet 3.1. Upgraded embedded
Winstone-Jetty to Jetty 9 accordingly. This removes AJP support
when using the embedded Winstone-Jetty container.
- Bundled Groovy updated from 1.8.9 to 2.4.7.
- Moved tools configuration from Configure Jenkins to separate
dialog.
- Added option to prohibit anonymous access to security realm
"Logged in users can do anything", enable by default.
- Renamed 'slave' to 'agent' on the UI.
- Improvements to inline documentation of numerous form fields in
Jenkins global and job configuration.
- Change default CSRF protection crumb name to Jenkins-Crumb for
nginx compatibility.
- Add symbol annotations on core.
- Workaround for unpredictable Windows file locking.
- Remove the historical initialization of CVS changelog parser for
jobs without explicit SCM definition. Warning! This change may
potentially cause a regression if a Jenkins plugin depends on
this default behavior and injects changelogs without SCM.
- Add the JOB_BASE_NAME environment variable to builds (job name
without path).
- Allow overriding Jenkins UpdateCenter by a custom
implementation.
- Allow overriding Jenkins PluginManager by a custom
implementation.
- Allow setting of properties from context.xml and web.xml in
addition to setting system properties from the command line.
- Remoting: Allow Jenkins admins to adjust the socket timeout.
(Controlled by hudson.remoting.Engine.socketTimeout)
- Remoting: Allow disabling the remoting protocols individually.
Allows working around compatibility issues like JENKINS-34121.
(Controlled by PROTOCOL_CLASS_NAME.disabled)
- Remoting, scalability: Ensure that the unexporter cleans up
whatever it can each GC sweep.
- Remoting: Force class load on UserRequest to prevent deadlocks
on Windows nodes agents in the case of multiple classloaders.
(Controlled by hudson.remoting.RemoteClassLoader.force)
- Make ToolInstallers to follow HTTP 30x redirects.
- Disable JSESSIONID in URLs when running in the JBoss web
container. It prevents Error 404 due to invalid links starting
from Jenkins 1.556. More info: WFLY-4782
- Allow starting non-AbstractProject (e.g. Pipeline) jobs from
CLI.
- Plugin Manager was building incorrect list of bundled plugins
for nested dependencies.
- Developer API: Add WorkspaceList.tempDir(...).
- Developer API: Allow putting @Initializer annotations on
instance methods.
- Developer API: Allow specifying custom AbortExceptions.
1.651.3 (2016/06/08)
- Cannot enable disabled dependencies.
- Listed Parameters should reflect what was used when the build
ran.
- Installation Wizard: SEVERE errors in logs, enabling of the
enabled plugin.
- Check Updates PeriodicWork dies horribly in the case of invalid
signature.
- RSS ID duplication for items with same name in different
folders.
- Remoting, scalability: Ensure that the unexporter cleans up
whatever it can each GC sweep.
- Remoting: Force class load on UserRequest to prevent deadlocks
on Windows nodes agents in the case of multiple classloaders.
(Controlled by hudson.remoting.RemoteClassLoader.force)
- Remoting: Allow Jenkins admins to adjust the socket timeout.
(Controlled by hudson.remoting.Engine.socketTimeout)
- Remoting: Allow disabling the remoting protocols individually.
Allows working around compatibility issues like JENKINS-34121.
(Controlled by PROTOCOL_CLASS_NAME.disabled)
1.651.2 (2016/05/11)
- Important security fixes
- Update remoting to 2.57.
- Pipeline runs not reliably started after restart when using
Build after other projects are built.
- Prevent badges in build history sidepanel widget from
overlapping page contents.
- Do not hardcode .bat extension for Maven on Windows.
- Don't store redundant build causes, make list of build causes
immutable.
- Make context meny link Delete Project work with CSRF protection
enabled.
1.651.1 (2016/04/14)
Changes from 1.651:
- Honor the option to opt out of usage statistics submission.
- Plugin filters were failing to be removed and blocking restart.
- Do not fail update center check if there are no tool installers
defined.
- Fix argument masking for sensitive build variables on Windows.
- Under some conditions Jenkins startup could fail because of
incorrectly linked extensions; now recovering more gracefully.
- Multiple bug fixes related to shutdown sequence.
Notable changes since 1.642.3:
- Move periodic task log files from JENKINS_HOME/*.log to
JENKINS_HOME/logs/tasks/*.log and rotate them periodically
rather than overwrite every execution.
- Allow changing the directory used for the extraction of plugin
archives via the --pluginroot CLI option (also controllable via
the hudson.PluginManager.workDir system property / context
parameter. Also document the --webroot CLI parameter in java
-jar jenkins.war --help
- Unify CLI exit code semantics.
- Add time zone to generation date in footer in most locales.
- The Windows service wrapper now specifies the --webroot argument
to extract the war file into %BASE%.
- Allow retrying core update when the first attempt failed.
- Allow specifying the default TCP slave agent listener port via
system property.
- Fix documentation of proxy configuration.
- Retrieve tool installer metadata from all update sites.
- Fields on the parameters page are no longer aligned at the
bottom.
- Cleanup of CLI error handling and return codes.
- Boot failure hook script did not work,
WebAppMain.contextDestroyed produces weird errors.
- ArrayIndexOutOfBoundsException when parsing range set.
- Generate new instance identity file when the existing one is
found to be corrupt.
- Developer: The official parent POM for plugins is now hosted in
the plugin-pom repository, starting with version 2.0.
- API changes: Add a reusable implementation of IdleOfflineCause
class.
- Developer: Split test harness into separate artifact.
- Developer: Pass $it to contents of dropdownDescriptorSelector.
1.642.4 (2016/03/31)
- Honor the option to opt out of usage statistics submission.
1.642.3 (2016/03/16)
- Fields on the parameters page are no longer aligned at the
bottom.
- Under some conditions a build record could be loaded twice,
leading to erratic behavior.
1.642.2 (2016/02/24)
- Important security fixes
- Don't submit usage statistics while Jenkins hasn't finished
loading.
- Performance regression when setting JDK installations.
- Renaming a node over another was possible and destroys both
configurations.
- A CloudProvisioningListener can prevent provisioning of all
clouds instead of just the targeted cloud.
- GroovyHookScript needs Jenkins to be initialized but should not
(for e.g. boot failure script).
- Don't show "termination trace" as warning in the log as
it's not necessarily an error condition.
1.642.1 (2016/01/20)
- No changes compared to 1.642
--------------
Version 2.35b:
--------------
- Fixed a minor cmdline reporting glitch, spotted by Leo Barnes.
- Fixed a silly bug in libdislocator. Spotted by Johannes Schultz.
Changelog:
The NSS team has released Network Security Services (NSS) 3.27,
which is a minor release.
Below is a summary of the changes.
Please refer to the full release notes for additional details,
including the SHA256 fingerprints of the changed CA certificates.
New functionality:
* Allow custom named group priorities for TLS key exchange handshake
(SSL_NamedGroupConfig).
* Added support for RSA-PSS signatures in TLS 1.2 and TLS 1.3
New Functions:
* SSL_NamedGroupConfig
Notable Changes:
* NPN can not be enabled anymore.
* Hard limits on the maximum number of TLS records encrypted with the same
key are enforced.
* Disabled renegotiation in DTLS.
* The following CA certificates were Removed
- CN = IGC/A, O = PM/SGDN, OU = DCSSI
- CN = Juur-SK, O = AS Sertifitseerimiskeskus
- CN = EBG Elektronik Sertifika Hizmet Sağlayıcısı
- CN = S-TRUST Authentication and Encryption Root CA 2005:PN
- O = VeriSign, Inc., OU = Class 1 Public Primary Certification Authority
- O = VeriSign, Inc., OU = Class 2 Public Primary Certification Authority - G2
- O = VeriSign, Inc., OU = Class 3 Public Primary Certification Authority
- O = Equifax, OU = Equifax Secure Certificate Authority
- CN = Equifax Secure eBusiness CA-1
- CN = Equifax Secure Global eBusiness CA-1
The full release notes are available at
https://developer.mozilla.org/en-US/docs/Mozilla/Projects/NSS/NSS_3.27_release_notes
Changelog:
NSPR 4.13 has the following bug fixes:
* PL_strcmp (and others) were fixed to return consistent results
when one of the arguments is NULL.
* PollEvent was fixed to not block on clear.
* Pipes are always nonblocking.
* PR_GetNameForIdentity: added thread safety lock and bound checks.
* Removed the PLArena freelist.
* Avoid some integer overflows.
* fixed several comments.
For the full list see:
https://bugzilla.mozilla.org/buglist.cgi?resolution=FIXED&query_format=advanced&product=NSPR&target_milestone=4.13
0.6.0 -> 0.6.1
=================
Akira TAGOH (2):
Fix the build issue regarding to 'index' (fixes: issue#15)
Get rid of the arch-dependent code from lt-config.h
David Tardon (1):
enum must be declared before it's used
Version 6.5.5 - Sep 21 2016
This is a bug fix release.
[FIXED BUGS]
o exuberant-ctags.la: The path name of ctags command couldn't changed.
Now, it can be changed by configuration variable 'ctagscom'.
o Tag files got corrupted when the built-in C++ parser gave up parsing and aborted.
Now, it is avoided.
o Some inappropriate error messages was changed.
Error1: NULL file (file size == 0)
$ cp /dev/null GTAGS
$ global -x main
(GLOBAL-6.5.4) global: /tmp/test/GTAGS seems older format. Please remake tag files.
(GLOBAL-6.5.5) global: /tmp/test/GTAGS seems to be corrupted.
Error2: Corrupted file
$ echo aaa >GTAGS
$ global -x main
(GLOBAL-6.5.4) global: GTAGS not found.
(GLOBAL-6.5.5) global: /tmp/test/GTAGS seems to be corrupted.
0.26 2016-09-24
- Require Role::Tiny 1.003003. This should fix the test failures some
CPANTesters reported with this error:
Can't resolve method "???" overloading "&{}" in package
"Specio::Constraint::Simple" at Specio::Constraint::Simple->new line 35.
2012 Sun Sep 25 01:45:56 CEST 2016
- another year where the UNFCCC had to adjust their predictions
in the worse direction.
- do not print the big intro blob in yellow.
2.3.0 - 2015-01-04
------------------
- **Feature**: Add ``--output-file`` option to specify a file to write to
instead of ``stdout``.
- **Bug** Fix interleaving of output while using multiprocessing
(`GitLab#17`_)
.. _GitLab#17: https://gitlab.com/pycqa/flake8/issues/17
2.4.0 - 2015-03-07
------------------
- **Bug** Print filenames when using multiprocessing and ``-q`` option.
(`GitLab#31`_)
- **Bug** Put upper cap on dependencies. The caps for 2.4.0 are:
- ``pep8 < 1.6`` (Related to `GitLab#35`_)
- ``mccabe < 0.4``
- ``pyflakes < 0.9``
See also `GitLab#32`_
- **Bug** Files excluded in a config file were not being excluded when flake8
was run from a git hook. (`GitHub#2`_)
- **Improvement** Print warnings for users who are providing mutually
exclusive options to flake8. (`GitLab#8`_, `GitLab!18`_)
- **Feature** Allow git hook configuration to live in ``.git/config``.
See the updated `VCS hooks docs`_ for more details. (`GitLab!20`_)
2.4.1 - 2015-05-18
------------------
- **Bug** Do not raise a ``SystemError`` unless there were errors in the
setuptools command. (`GitLab#39`_, `GitLab!23`_)
- **Bug** Do not verify dependencies of extensions loaded via entry-points.
- **Improvement** Blacklist versions of pep8 we know are broken
2.5.0 - 2015-10-26
------------------
- **Improvement** Raise cap on PyFlakes for Python 3.5 support
- **Improvement** Avoid deprecation warnings when loading extensions
(`GitLab#59`_, `GitLab#90`_)
- **Improvement** Separate logic to enable "off-by-default" extensions
(`GitLab#67`_)
- **Bug** Properly parse options to setuptools Flake8 command (`GitLab!41`_)
- **Bug** Fix exceptions when output on stdout is truncated before Flake8
finishes writing the output (`GitLab#69`_)
- **Bug** Fix error on OS X where Flake8 can no longer acquire or create new
semaphores (`GitLab#74`_)
2.5.1 - 2015-12-08
------------------
- **Bug** Properly look for ``.flake8`` in current working directory
(`GitLab#103`_)
- **Bug** Monkey-patch ``pep8.stdin_get_value`` to cache the actual value in
stdin. This helps plugins relying on the function when run with
multiprocessing. (`GitLab#105`_, `GitLab#107`_)
2.5.2 - 2016-01-30
------------------
- **Bug** Parse ``output_file`` and ``enable_extensions`` from config files
- **Improvement** Raise upper bound on mccabe plugin to allow for version
0.4.0
2.5.3 - 2016-02-11
------------------
- **Bug** Actually parse ``output_file`` and ``enable_extensions`` from config
files
2.5.4 - 2016-02-11
------------------
- **Bug** Missed an attribute rename during the v2.5.3 release.
2.5.5 - 2016-06-14
------------------
- **Bug** Fix setuptools integration when parsing config files
- **Bug** Don't pass the user's config path as the config_file when creating a
StyleGuide
2.6.0 - 2016-06-15
------------------
- **Requirements Change** Switch to pycodestyle as all future pep8 releases
will use that package name
- **Improvement** Allow for Windows users on *select* versions of Python to
use ``--jobs`` and multiprocessing
- **Improvement** Update bounds on McCabe
- **Improvement** Update bounds on PyFlakes and blacklist known broken
versions
- **Improvement** Handle new PyFlakes warning with a new error code: F405
2.6.1 - 2016-06-25
------------------
- **Bug** Update the config files to search for to include ``setup.cfg`` and
``tox.ini``. This was broken in 2.5.5 when we stopped passing
``config_file`` to our Style Guide
2.6.2 - 2016-06-25
------------------
- **Bug** Fix packaging error during release process.
3.0.0 -- 2016-07-25
-------------------
- Rewrite our documentation from scratch! (http://flake8.pycqa.org)
- Drop explicit support for Pythons 2.6, 3.2, and 3.3.
- Remove dependence on pep8/pycodestyle for file processing, plugin
dispatching, and more. We now control all of this while keeping backwards
compatibility.
- ``--select`` and ``--ignore`` can now both be specified and try to find the
most specific rule from each. For example, if you do ``--select E --ignore
E123`` then we will report everything that starts with ``E`` except for
``E123``. Previously, you would have had to do ``--ignore E123,F,W`` which
will also still work, but the former should be far more intuitive.
- Add support for in-line ``# noqa`` comments to specify **only** the error
codes to be ignored, e.g., ``# noqa: E123,W503``
- Add entry-point for formatters as well as a base class that new formatters
can inherit from. See the documentation for more details.
- Add detailed verbose output using the standard library logging module.
- Enhance our usage of optparse for plugin developers by adding new parameters
to the ``add_option`` that plugins use to register new options.
- Update ``--install-hook`` to require the name of version control system hook
you wish to install a Flake8.
- Stop checking sub-directories more than once via the setuptools command
- When passing a file on standard-in, allow the caller to specify
``--stdin-display-name`` so the output is properly formatted
- The Git hook now uses ``sys.executable`` to format the shebang line.
This allows Flake8 to install a hook script from a virtualenv that points to
that virtualenv's Flake8 as opposed to a global one (without the virtualenv
being sourced).
- Print results in a deterministic and consistent ordering when used with
multiprocessing
- When using ``--count``, the output is no longer written to stderr.
- AST plugins can either be functions or classes and all plugins can now
register options so long as there are callable attributes named as we
expect.
3.0.1 -- 2016-07-25
-------------------
- Fix regression in handling of ``# noqa`` for multiline strings.
(See also `GitLab#177`_)
- Fix regression in handling of ``--output-file`` when not also using
``--verbose``. (See also `GitLab#180`_)
- Fix regression in handling of ``--quiet``. (See also `GitLab#180`_)
- Fix regression in handling of ``--statistics``. (See also `GitLab#180`_)
3.0.2 -- 2016-07-26
-------------------
- Fix local config file discovery. (See also `GitLab#181`_)
- Fix indexing of column numbers. We accidentally were starting column indices
at 0 instead of 1.
- Fix regression in handling of errors like E402 that rely on a combination of
attributes. (See also `GitLab#186`_)
3.0.3 -- 2016-07-30
-------------------
- Disable ``--jobs`` for any version of Python on Windows.
(See also `this Python bug report`_)
- Raise exception when entry_point in plugin not callable.
This raises an informative error when a plugin fails to load because its
entry_point is not callable, which can happen with a plugin which is buggy or
not updated for the current version of flake8. This is nicer than raising a
`PicklingError` about failing to pickle a module (See also `GitLab#164`_)
- Fix ``# noqa`` comments followed by a ``:`` and explanation broken by
3.0.0 (See also `GitLab#178`_)
- Always open our output file in append mode so we do not overwrite log
messages. (See also `GitLab#193`_)
- When normalizing path values read from configuration, keep in context the
directory where the configuration was found so that relative paths work.
(See also `GitLab#194`_)
- Fix issue where users were unable to ignore plugin errors that were on
by default. (See also `GitLab#195`_)
- Fix our legacy API StyleGuide's ``init_report`` method to actually override
the previous formatter. (See also `GitLab#200`_)
3.0.4 -- 2016-08-08
-------------------
- Side-step a Pickling Error when using Flake8 with multiprocessing on Unix
systems. (See also `GitLab#164`_)
- Fix an Attribute Error raised when dealing with Invalid Syntax. (See also
`GitLab#203`_)
- Fix an unhandled Syntax Error when tokenizing files. (See also
`GitLab#205`_)
* Fixed a problem in the “multiple -arch options” support introduced in 3.3. When using the direct mode (the default), different combinations of -arch options were not detected properly.
* Fixed an issue when compiler option -Wp,-MT,path is used instead of -MT path (and similar for -MF, -MP and -MQ) and run_second_cpp (CCACHE_CPP2) is enabled.
-------------------------------
Overview of changes in Pango 1.227
==================================
* Ensure word separation in linker arguments with pkg-config >= 0.29
------------------------------------
Overview of changes between 1.40.2 and 1.40.3
=============================================
- Fix export of pango_cairo_win32_font_map_get_type (#771004)
- Fix build on OS X (#770729)
Overview of changes between 1.40.1 and 1.40.2
=============================================
- Improve introspection builds in msvc (#764984)
- Properly handle UTF32 chars in coretext
- Add rebeccapurple to list of colors
- Correct font weight mapping in coretext (#766148)
- Use version macros for exports (#767587)
Upstream changes:
This is a maintenance release, source code is unchanged compare to the previous
release. This release introduces a guile-lib-1.0.pc, so libs and apps that needs
guile-lib may now use pkgconfig to check if it's available, what version.
Note that starting with this release, the installation location for modules and
compiled files has changed: make sure you uninstall any previous version.
- Modules installation location changed:
from $(prefix)/share/guile/site
to $(GUILE_GLOBAL_SITE)
You will need write privileges to this directory to install guile-lib. You may
check its location using the folowing command:
guile -c "(display (%global-site-dir)) (newline)"
- Compiled files installation location changed:
from $(prefix)/share/guile/site
to $(GUILE_SITE_CCACHE)
You will need write privileges to this directory to install guile-lib. You may
check its location using the folowing command:
guile -c "(display (%site-ccache-dir)) (newline)"
-------------------------------
Overview of changes in Glib 1.322 (stable)
============================================
* Win32: only export the GVariant conversion symbols when available
∗ Glib.exports: updated copyright date
-----------------------------
1.011 2016-09-18
- Move to GitHub. Thanks, OpenFoundry, for years of service.
- Suppress warning "Use of uninitialized value in do "file" ..."
- Put a description of its purpose into the canary file
- Remove all traces of Internals::PAR::CLEARSTACK
- Remove all references to http://par.perl.org/, doesn't exist anymore
pkgsrc change: switch from qt4 to qt5 for GUI option.
Cppcheck-1.75 has been released
General changes:
- Replaced internal preprocessor by the brand-new preprocessor 'simplecpp'
- Improved Windows installer: Install a copy of the license instead of asking to accept it
- The Windows x64 binaries are now compiled with profile guided optimization, resulting in a speedup of 11%
- Improved manual, especially the chapter about Libraries
- Improved CWE mapping
- --append is deprecated and will be removed in 1.80
New checks:
- Detect passed by value for non-const variables and print message only if type size justifies optimization
Checking improvements:
- Implemented support for trailing return types (C++11)
- Improved support for digit separators (C++14)
- Improved support for enum types in buffer overflow checking
- Better handling of volatile variables when checking for redundant assignments
- Properly support integer suffixes i64 and ui64
- Support function arguments with default value in Libraries
- Always set file0 attribute of error messages to identify the source file cppcheck was checking
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
Cppcheck-1.74 has been released
General changes:
- Improved CWE mapping of messages
- Git pre-commit hook checks only added or modified files
Checking improvements:
- Replaced simplification of enums by keeping and parsing them in the SymbolDatabase
- Added support to Library for specifying the parameter used by allocating/deallocating functions
- Improved support for integers defined in Libraries
- Improved accuracy of ValueType analysis
- Improved accuracy of VarID assignment, especially when dealing with structs and unions
- Improved performance of VarID assignment, checking for struct member usage, buffer overrun checking and several simplifications
- Added support for lots functions to windows.cfg and posix.cfg
- Better support for operator overloads
- Detect buffer overflows when %c is used with a width
- Improved checking for sizeof() taken of wrong type
- Support char literals when checking for conditions being always true or false
- Reimplemented check for usage of boolean results used in bitwise operations based on ValueType
- Improved checking for c_str() usage
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
Cppcheck-1.73 has been released
General changes:
- CWE mapping of messages
- Translated manual to Japanese language
Removed checks:
- Checks for variables hiding enums or typedefs have been removed
New checks:
Checking improvements:
- Improved ValueType a lot, use it in more checks
- Improved VarId support for template constructors, namespaces and references as class members
- Improved libraries, especially gnu.cfg, posix.cfg and windows.cfg
- Improved simplification of enums and templates
- Better distinguishing between possible and known null pointer dereferenciations
- Assume integers to be signed by default
- better support for cplusplus macro in preprocessor
- Preprocessor directives for addons
- New tools: times-vs.py, reduce.py
GUI:
- Detect Geany and QtCreator
- Make statistics dialog shown when checking is finished optional
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
Cppcheck-1.72 has been released.
General changes:
- Added platform files to customize characteristics of different target platforms
- Added setting to configure the default sign of integral variables and bit size of type 'char'
- Added option -E (only preprocess the code)
- Option --dump now outputs data for each configurations, not only the last one
- Several qmake project files have been removed
Removed checks:
- Check for unnecessary forward declaration has been removed
New checks:
- Warn about shifting negative values
- Detect statements with undefined execution order (pre-C++11 code)
- Added a generic check to discourage usage of specific functions, replacing CheckNonReentrantFunctions and CheckObsoleteFunctions
Checking improvements:
- Several improvmenets to ValueFlow analysis
- Improved ValueType handling, refactored several checks to make use of it
- Improved memory leak checking
- Improved Container specification in Library files, rewrote some checks based on it
- Improved handling of character literals
- Improved checking for non-interlocked accesses
- Properly support <cstdint> types in namespace std
- Added some validation mechamisms to avoid crashs
GUI:
- Improved Library editor
- Support "Enter" key in results tree
- Properly detect editors on 64-Bit Windows
- Added Japanese translation
- Allow opening project file without extra parameter from command line
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
Cppcheck-1.71 has been released
General changes:
- The deprecated options --suppressions and --exitcode-suppressions have been removed
- .hh and .hxx files are now treated as C++ files
- Improved documentation (readmes and manuals)
- Internal errors (for example syntax error) are now suppressable
Removed checks:
- Check for unnecessary qualification has been removed
Checking improvements:
- Support range-based for-loops (C++11)
- Better support for C++11 style initialization
- Better support for lambdas (C++11)
- Support digit separators (C++14)
- Added determination of the type of an expression (ValueType)
- Support case ranges (GCC extension)
- Improved AST generation and added validation
- Improved value flow analysis
- Improved checking for unitialized arrays
- Improved VarId assignment for member variables
- Rewrote checking of char variables
- Known variable simplification has been removed partially in favour of value flow analysis
GUI:
- Added library editor
- Improved display of inconclusive messages
- Added option to enforce parsing as C or C++ code
- Show file that included a header in details view
- Added "Recheck all files" button, "Recheck files " renamed to "Rechecked modified files"
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
Cppcheck-1.70 has been released
General changes:
- New version of .cfg file format, adding support for namespaces and declaring several functions at once
- Support building x64 installer for Windows; Both x64 and x86 builds are now provided
- Warn about deprecated options --suppressions and --exitcode-suppressions. They will be removed in future
- Added debugging option --debug-normal, which prints out debug output before second stage simplification
New checks:
- Warn about modifying string literals
- Detect declaration of variable length arrays with negative size
- Warn about implicit type conversion from int to long
- Warn about redundant code in conditions like (y || (!y && z))
- Detect conditions with known result
- Race condition: non-interlocked access after InterlockedDecrement()
- Detect unused 'goto' labels
Removed checks:
- Do no longer warn about a bug in scanf implementation of ancient glibc versions
- Multifile analysis (added in 1.69) has been removed because it did not work
- Removed ExecutionPath checker, which has been superseeded by ValueFlow analysis
Improvements:
- Support hexadecimal floating point numbers (C99)
- Support [[deprecated]] (C++14)
- Improved handling of sizeof()
- Improved handling of reserved keywords
- Better handling of C declaration specifiers; support complex/_Complex (C99)
- Better handling of ternary operator in ValueFlow analysis
- Lots of improvements to .cfg files, especially std.cfg, which now supports namespace std
- Improved performance and memory usage of Preprocessor
- Improved performance of matchcompiler
- Reduced Disk IO when ignoring paths
- Removed several obsolete simplifications
- Added python addons: naming.py, threadsafety.py and cert.py
GUI:
- Support printing
- Added item "Open containing folder" to context menu
Additionally, lots of false positives and bugs have been fixed and several existing checks have been improved.
---------------------------------
0.24 - 2016-09-01 - SAPER
[Distribution]
- fixed the way the module's version is detected in order to correctly
generate META.* files
0.23 - 2016-08-31 - SAPER
[Makefile.PL updated from perl blead 9e9fbcaf153d3c9eb897b5cab54a102632df4c78]
- 2016-08-05 note a possible pitfall
- 2016-08-05 add "provides" metadata for XSLoader (RT#138836)
0.22 - 2016-07-05 - SAPER
[Code updated from perl blead a651dcdf6a9151150dcf0fb6b18849d3e39b0811]
- 2016-07-04 Fix XSLoader to recognize drive letters
- 2016-07-03 Don’t let XSLoader load relative paths
[Code updated from perl 5.24.0]
- 2015-11-10 use dl_find_symbol 3rd optional argument
- 2015-11-02 remove XSLoader and DynaLoader OS specific code on NA OSes
[Distribution]
- Converted the Changes file to CPAN::Changes::Spec format.
0.20 - 2016-05-05 - SAPER
[Code updated from perl 5.22.1]
- 2014-12-06 0.20 Don't test for bootstrap file twice
- 2014-10-31 0.19 Amend the mod2fname generation to detect more cases
- 2014-10-03 0.18 Allow XSLoader to load modules from a different namespace
- 2014-01-03 0.17 DynaLoader: Introduce d_libname_unique for Android
[Distribution]
- Fixed a stupid mistake in Makefile.PL
---------------------------------------------------
0.15 2016-05-04T08;15:00
- Correct physical address of Free Software Foundation in LICENSE file. See RT#114149.
Thanx to Petr Pisar.
----------------------------------------
1.302056 2016-09-12 09:03:49-07:00 America/Los_Angeles
- Minor typo fix
- No logic chnges since last trial
1.302055 2016-08-30 12:13:32-07:00 America/Los_Angeles (TRIAL RELEASE)
- Fix special case of ok line ending in \
- Improve a test that captures STDERR/STDOUT (Thanks HAARG)
1.302054 2016-08-20 16:21:44-07:00 America/Los_Angeles (TRIAL RELEASE)
- Allow '#' and '\n' in ok names
1.302053 2016-08-17 21:22:55-07:00 America/Los_Angeles (TRIAL RELEASE)
- Fix skip_all in require in intercept (#696)
- Documentation of what is better in Test2 (#663)
- Document Test::Builder::Tester plan limitations
- Document limitations in is_deeply (#595)
- Better documentation of done_testing purpose (#151)
- Make ctx->send_event detect termination events (#707)
----------------------------------------
1.22 2016-09-17
- Fix RT#117887: Not parsing new release of Net::DNS::Resolver
add %Preload rule for Net/DNS/Resolver.pm
Snappy v1.1.3, July 6th 2015:
This is the first release to be done from GitHub, which means that
some minor things like the ChangeLog format has changed (git log
format instead of svn log).
* Add support for Uncompress() from a Source to a Sink.
* Various minor changes to improve MSVC support; in particular,
the unit tests now compile and run under MSVC.
Snappy v1.1.2, February 28th 2014:
This is a maintenance release with no changes to the actual
library
source code.
* Stop distributing benchmark data files that have unclear
or unsuitable licensing.
* Add support for padding chunks in the framing format.
------------------------------------
- ChangeLog unknown
(pkgsrc changes)
- Add following line for make test
BUILD_DEPENDS+= p5-Test-Toolbox-[0-9]*:../../devel/p5-Test-Toolbox
This is a terse description of the new features added to readline-7.0 since
the release of readline-6.3.
New Features in Readline
a. The history truncation code now uses the same error recovery mechansim as
the history writing code, and restores the old version of the history file
on error. The error recovery mechanism handles symlinked history files.
b. There is a new bindable variable, `enable-bracketed-paste', which enables
support for a terminal's bracketed paste mode.
c. The editing mode indicators can now be strings and are user-settable
(new `emacs-mode-string', `vi-cmd-mode-string' and `vi-ins-mode-string'
variables). Mode strings can contain invisible character sequences.
Setting mode strings to null strings restores the defaults.
d. Prompt expansion adds the mode string to the last line of a multi-line
prompt (one with embedded newlines).
e. There is a new bindable variable, `colored-completion-prefix', which, if
set, causes the common prefix of a set of possible completions to be
displayed in color.
f. There is a new bindable command `vi-yank-pop', a vi-mode version of emacs-
mode yank-pop.
g. The redisplay code underwent several efficiency improvements for multibyte
locales.
h. The insert-char function attempts to batch-insert all pending typeahead
that maps to self-insert, as long as it is coming from the terminal.
i. rl_callback_sigcleanup: a new application function that can clean up and
unset any state set by readline's callback mode. Intended to be used
after a signal.
j. If an incremental search string has its last character removed with DEL, the
resulting empty search string no longer matches the previous line.
k. If readline reads a history file that begins with `#' (or the value of
the history comment character) and has enabled history timestamps, the history
entries are assumed to be delimited by timestamps. This allows multi-line
history entries.
l. Readline now throws an error if it parses a key binding without a terminating
`:' or whitespace.
Test::Toolbox provides (as you might guess) tools for automated
testing. Test::Toolbox is much like some other testing modules, such as
Test::More and Test::Simple. Test::Toolbox provides a different flavor of
tests which may or may not actually be to your preference.
The tools in Test::Toolbox have a standard format. Commands start with (the
command (of course), followed by the test name. Then there is usually the
value being tested, or values being compared, then other options. So, for
example, this command checks compares two values:
rtcomp 'test name', $val, $other_val;
In some cases it's preferable to flip the logic of the test, so that, for
example, two values should not be the same. In that case, you can add the
should option:
rtcomp 'test name', $val, $other_val, should=>0;
All test commands require a test name as the first param.
NEW IN WAF 1.9.4
----------------
* Enable 'waf dist' in arbitrary paths #1806
* Handle subprocess timeouts in Python 3.3 #1807
* Set the configuration test build class on conf.run_build_cls
* Provide execution order constraints in parallel configuration tests
* Accept Task.cwd as str type
TortoiseHg 3.9.1
TortoiseHg 3.9.1 is a regularly scheduled bug-fix release.
Bug Fixes
graphopt: show working copy if any of its parents are in a branch filtering
sync: reorder SecureDialog.init to avoid too early _updateUi() (fixes#4559)
toposort: avoid crash when tip revision is hidden
v27.2.0
-------
* #520 and #513: Suppress ValueErrors in fixup_namespace_packages
when lookup fails.
* Nicer, more consistent interfaces for msvc monkeypatching.
v1.8.3
======
* When passing a ``void *`` argument to a function with a different
pointer type, or vice-versa, the cast occurs automatically, like in C.
The same occurs for initialization with ``ffi.new()`` and a few other
places. However, I thought that ``char *`` had the same
property---but I was mistaken. In C you get the usual warning if you
try to give a ``char *`` to a ``char **`` argument, for example.
Sorry about the confusion. This has been fixed in CFFI by giving for
now a warning, too. It will turn into an error in a future version.
* Noteworthy changes in release 2016.09.16 (2016-09-16) [stable]
The following macros were added: AX_BUILD_DATE_EPOCH, AX_CHECK_X86_FEATURES,
AX_GCC_X86_CPU_SUPPORTS, AX_OPEN62541_CHECK_H, AX_OPEN62541_CHECK_LIB, and
AX_LUAROCKS_ROCK.
While updating AX_CODE_COVERAGE, the variable CODE_COVERAGE_LDFLAGS was
renamed to CODE_COVERAGE_LIBS.
AX_VALGRIND_CHECK has been extended considerably. In this process. the
undocumented internal rule check-valgrind-tool was removed.
It checks if the host cpu supports various instruction set: mmx, sse, popcnt,
avx, avx2, etc. If the instruction were supported, it would export
HAVE_XXX_INSTRUCTIONS C preprocessor macros and add the instruction set flag to
X86_FEATURES_CFLAGS variable. By default it would also set CFLAGS, but this can
be disabled by the ACTION-IF-FOUND hook.
AX_WITH_CURSES and AX_WITH_CURSES_* macros have been modernized, in line
with PKG_CHECK_MODULES's style. This change breaks the interface, but
makes AX_WITH_CURSES compatible and even completely replaceable with the
interface of PKG_CHECK_MODULES. Ultimately, these macros should be
replaced in the future by calls to PKG_CHECK_MODULES. Variables in
AX_WITH_CURSES have been renamed
CURSES_LIB -> CURSES_LIBS
CURSES_CPPFLAGS -> CURSES_CFLAGS
and for the AX_WITH_CURSES_* macros
PANEL_LIB -> PANEL_LIBS
MENU_LIB -> MENU_LIBS
FORM_LIB -> FORM_LIBS
such that once the variables have been renamed in configure.ac and Makefile.am,
the calls AX_WITH_CURSES and PKG_CHECK_MODULES([CURSES], [ncurses]) are
completely interchangeable.
A minimal, zero-config, BSD licensed, readline replacement.
Noteworthy features:
- Single and multi line editing mode with the usual key bindings implemented.
- History handling.
- Completion.
- Hints (suggestions at the right of the prompt as you type).
- About 1,100 lines of BSD license source code.
- Only uses a subset of VT100 escapes (ANSI.SYS compatible).
TinyCThread is a minimalist, portable, threading library for C, intended to
make it easy to create multi threaded C applications.
The library is closesly modeled after the C11 standard, but only a subset is
implemented at the moment.
The Specio distribution provides classes for representing type constraints
and coercion, along with syntax sugar for declaring them.
Note that this is not a proper type system for Perl. Nothing in this
distribution will magically make the Perl interpreter start checking a
value's type on assignment to a variable. In fact, there's no built-in way
to apply a type to a variable at all.
Instead, you can explicitly check a value against a type, and optionally
coerce values to that type.
My long-term goal is to replace Moose's built-in types and MooseX::Types
with this module.
Create a customized, optimized, non-lobotomized, uncompromised, and
thoroughly specialized parameter checking subroutine.
(make test is NOT confirmed yet)
--------------------------------------------
1.025000 30.8.16
* Added some version requirements to improve SSL over
HTTP::Tiny (Mickey)
* Added default values for distribution keys with no content
(Mickey, per Tux request)
1.024000 28.08.16
* Try to fetch clientinfo from https://clientinfo.metacpan.org
to get default production version (Mickey)
1.023000 27.08.16
* Added support for version by env METACPAN_VERSION (Mickey)
* Added tests for version argument (Mickey)
(pkgsrc-changes)
- Add following line for make test
BUILD_DEPENDS+= p5-Ref-Util-[0-9]*:../../devel/p5-Ref-Util
-------------------------------------
0.046 2016-09-12 20:13:04-07:00 America/Los_Angeles
No changes since last development release.
0.045_091 2016-09-10 14:26:13-07:00 America/Los_Angeles (TRIAL RELEASE)
Properly handle scores in raw log files (Atoomic)
Update error message for inability to run 'git config --list'
0.045_090 2016-09-08 21:59:58-07:00 America/Los_Angeles (TRIAL RELEASE)
Fix tabs being expanded in log output in git 2.9+ (Graham Knop)
Misc POD fixes (gregor herrmann/Debian, Philippe Bruhat)
----------------------------
CHANGES since 3.4 reaches 280 lines, 11.2 kbytes, so only recent
Announce is listed here, thank you.
----------------------------------------------------------------
August 31, 2016
Announcing : PLY-3.9 (Python Lex-Yacc)
http://www.dabeaz.com/ply
I'm pleased to announce PLY-3.9--a pure Python implementation of the
common parsing tools lex and yacc. PLY-3.9 is a minor bug fix
release. It supports both Python 2 and Python 3.
If you are new to PLY, here are a few highlights:
- PLY is closely modeled after traditional lex/yacc. If you know how
to use these or similar tools in other languages, you will find
PLY to be comparable.
- PLY provides very extensive error reporting and diagnostic
information to assist in parser construction. The original
implementation was developed for instructional purposes. As
a result, the system tries to identify the most common types
of errors made by novice users.
- PLY provides full support for empty productions, error recovery,
precedence rules, and ambiguous grammars.
- Parsing is based on LR-parsing which is fast, memory efficient,
better suited to large grammars, and which has a number of nice
properties when dealing with syntax errors and other parsing
problems. Currently, PLY can build its parsing tables using
either SLR or LALR(1) algorithms.
More information about PLY can be obtained on the PLY webpage at:
http://www.dabeaz.com/ply
PLY is freely available.
The program contains a collection of tools for analyzing Perl
code. Analyzing a static program runs headlong into the halting problem --
no automatic process can find all infinite loops, or fully model dynamic
data generated within an arbitrary program. If you need to solve that kind
of problem, a runtime debugger, human insight, and good programming
practices go a long way. On the other hand, if you're reading a new Perl
code for the first time, you've come to the right place.
------------------------------------------
0.3.0 2016-09-11
- Converted the build system to Dist-Zilla.
- Part of the impetus was removing the warning about
Test::TrailingSpace needed for the trailing whitespace tests
-------------------------------------------------
Version 2.34b:
--------------
- Added a note about afl-tmin to technical_details.txt.
- Added support for AFL_NO_UI, as suggested by Leo Barnes.
-----------------------------------------
1.042 2016-08-26 23:37:33-05:00 America/Chicago
[Added]
- Default adapters can now be configured with arguments (thanks
@bjakubski!)
1.041 2016-08-18 00:00:10-05:00 America/Chicago (TRIAL RELEASE)
[Added]
- Logging methods now return the formatted log string so that it can
be used in a `die` or `warn` call.
[Changed]
- A new default log proxy (Log::Any::Proxy::Null) is used when there
are no adapters configured (and so no place for logs to go). This
proxy does no processing and is about 1000% percent faster on my
laptop.
[Fixed]
- Suppress 'redundant argument' warnings if too many arguments are
given to a log formatting string.
Add TEST_TARGET. Some tests fail, but the calendar example runs.
Upstream changes:
Guile-Gnome-Platform 2.16.4 - David Pirotte, 2015-12-08
=======================================================
This release fixes a serious bug, leading to a runtime error for
Guile-Gnome built against Gcc >= 5.0.
Glib
----
* glib/gnome/gobject/gclosure.c: gcc_please_look_the_other_way must be
declared volatile.
guile-gnome-platform 2.16.3 - David Pirotte, 2015-07-30
=======================================================
The primary objectives, working towards this release, were (1) to
remove all code specific to the support of guile versions prior to
guile 2, (2) getting rid of all guile related deprecated messages
[that second part of the mission almost completed, there is still one
corba related deprecated warning message, help to resolve that last
one is welcome!], and (3) wrap gdk_pixbuf_get_pixels required by
guile-clutter 1.12.
All modules
-----------
* Guile-Gnome now requires Guile 2.0, allows Guile 2.2
* Requires G-Wrap >= 1.9.15
* Requires Guile-Cairo >= 1.10
Note that at the time we write, Guile-Cairo 1.10 has not been
released yey, it will hopefully soon be the case. In the mean
time, the required bleeding edge Guile-Cairo stable version must
be obtained from the source, see:
http://www.nongnu.org/guile-cairo/dev/
* toplevel calls moved into an eval-when conditions exp...
* scm-error incorrect arg list, various calls, all fixed.
* all [but 1 Corba related] Guile deprecated calls removed
* remove use of some undocumented GOOPS internals
* adapt to GOOPS changes in Guile 2.2
* make check and make distcheck both have been fixed to work with
recent automake versions [thanks to Ludovic Courtés for his help wrt
to make check]
Corba
-----
* fixing allocate-instance for <CORBA:Object>
* gnome-corba-error, gruntime-error, don't call save-stack
[deprecated]
Glib
----
* g_get_current_time now [temporarily] ignored, it was not properly
wrapped anyway.
* <glib-wrapset> missing some basic pointer types: guint8*, guchar*,
gunichar*, gunichar2* added and wrapped as opaque pointers.
* scm_c_gparam_construct 4th argument type definition fixed
* use &gw-bad-typespec, not &gw:bad-typespec, fixed.
Gdk-Pixbuf
----------
* gdk_pixbuf_get_pixels removed from the ignore list, it returns a
guchar*, just added, no specific wrapper needed.
Gtk
---
* gtk_recent_info_get_application_info should [temporarily] be
ignored, it needs a special wrapper, to be done still.
------------------------------------
1.006 2016-09-10 14:47:49-04:00 America/New_York
- No changes since 1.005
1.005 2016-09-06 12:12:06-04:00 America/New_York (TRIAL RELEASE)
[INTERNAL]
- Optimize generated accessors (5% to 25% faster depending on
combination of read vs write and default vs no-default).
----------------------------------------------
2.45 2016-09-07 No functional changes, bump the required version
of YAML::Tiny to cope with some modern META.yml
files
-------------------------------------------
2.40
Fixup issues with JSON and with non-Padwalker builds (thanks to FC)
various other tweaks and cleanups to build on newer Perls. Updated
meta files, etc.
--------------------------------------
* Fixes 'Name "Config::Grammar::File" used only once' (patch by Salvatore Bonaccorso)
* Fix spelling error in manpage (patch by Salvatore Bonaccorso)
* Use Dist::Zilla to prepare the Perl distribution for CPAN
CVS: ----------------------------------------------------------------------
This package had been held at 1.9.14 because 1.9.15 requires guile
2.0.
Changes from 1.9.14:
G-Wrap now requires Guile 2.0, allows Guile 2.2;
Requires autoconf >= 2.61, automake >= 1.12, adding /build-aux
We no longer update Changelog, see G-Wrap's git logs instead
No free/mark on WCT objects, and remove use of scm_gc_free
Use scm_module_variable instead of the deprecated scm_sym2var
The documentation has been restructured, updated and fixed to work with texinfo >= 5.2;
Removing doc/version.texi - it is generated/updated
Bug Fixes
* Require an updated version of Net::DNS to avoid
"resolver->dnssec(1) without Net::DNS::SEC installed"
(Schwern) [github #4]
Test Fixes
* Fix our fake random number generator to avoid cross platform issues.
[rt.cpan.org 71064] [github #7] (Lisa Hare)
Distribution
* Kwalitee fixups (BooK) (Lisa Hare)
* Added a LICENSE file (Lisa Hare)
* Now using Travis CI for Unix testing (Schwern)
* Now using AppVeyor for Windows testing (Schwern)
