Release 2.2.1
* Fix executable resolution on Windows when a directory exists with the same
name as the command to run
Release 2.2.0
* Remove windows-pr dependency
=== 5.8.0 / 2015-08-06
* 2 minor enhancements:
* Add optional delegation mechanism to extend object with a mock. (zamith)
* Return early if there are no filtered methods. (jeremyevans)
* 1 bug fix:
* Don't extend io with pride if io is not a tty. (toy)
2.2.2 (07/15/2015)
Bugs fixed
* Fix uninitialized constant StringIO error in C extension
2.2.1 (07/13/2015)
Bugs fixed
* Aaron Lasseigne: Thanks for fixing my bad .travis.yml
* Kouhei Sutou: Numerous circular argument reference fixes, method redefined,
assigned but unused warnings, superclass mismatch warnings, etc.
* Fixed uninitialized variable when using C-ext.
* Fixed README require
* Fixed superclass mismatch in rbx when using C-ext.
* Lots of rubocop fixes, and added rubocop to travis
Version 0.1.8
Move flush into own method
Version 0.1.7
Don't use closed fobj
Add makefile
Version 0.1.6
Actually call fsync
Remove useless file
Simplify CI setup
Adapt to internal tox changes
Fix docs
Fix link
Split alternatives and credit
Update alternatives
Fix comment
TortoiseHg 3.5.1
TortoiseHg 3.5.1 is a regularly scheduled bugfix release.
Bugs Fixes
visdiff: preserve the executable bit after editing a file
Installer
setup: exclude setup.py and distutils from py2exe build
wix: add mercurial/help/scripting.txt (fixes#4264)
------
18.3.1
------
* Issue #425: Correct regression in setuptools.findall.
----
18.3
----
* Setuptools now allows disabling of the manipulation of the sys.path
during the processing of the easy-install.pth file. To do so, set
the environment variable ``SETUPTOOLS_SYS_PATH_TECHNIQUE`` to
anything but "rewrite" (consider "raw"). During any install operation
with manipulation disabled, setuptools packages will be appended to
sys.path naturally.
Future versions may change the default behavior to disable
manipulation. If so, the default behavior can be retained by setting
the variable to "rewrite".
* Issue #257: ``easy_install --version`` now shows more detail
about the installation location and Python version.
MAJOR CHANGES:
Now builds with CMake 2.8.9 and newer (previously required
3.0.0)
CHANGES IN THE SOFTWARE CONFIGURATION:
CMakeLists.txt / CMake
Supports CMake 2.8.9 and later.
Add missing file which wasn't being distributed, causing
unit tests to fail.
Make shared/static library building configurable.
CMake reads all version information directly from configure.ac
to avoid duplication of values.
CMake builds are now included in 'distcheck' target.
Makefile.am
Autotools 'make distcheck' now tests the CMake-based build
if CMake is available.
CHANGES IN LIBTIFF:
Fixes to avoid undefined behaviour of signed types (C standard
compliance).
Fixes to avoid possible isses when casting to unsigned char.
Fixes to avoid undefined behaviour with shifts.
Fix generation of output with 16 bit or 32 bit integer, when
byte swapping is needed, in horizontal predictor (#2521).
Fix decoding when there is a single pixel to decode (unlikely
case...) and byte swapping is involved.
Add add explicit masking with 0xff before casting to uchar in
floating-point horizontal differencing and accumulation routines.
Eliminate requirement for and use of 64-bit constant values.
CHANGES IN THE TOOLS:
tiffgt
Silence glut API deprecation warnings on MacOS X.
fax2ps
Detect failure to write to temporary file.
Fixes since v2.5.1
------------------
* "git init empty && git -C empty log" said "bad default revision 'HEAD'",
which was found to be a bit confusing to new users.
* The "interpret-trailers" helper mistook a multi-paragraph title of
a commit log message with a colon in it as the end of the trailer
block.
* When re-priming the cache-tree opportunistically while committing
the in-core index as-is, we mistakenly invalidated the in-core
index too aggressively, causing the experimental split-index code
to unnecessarily rewrite the on-disk index file(s).
* "git archive" did not use zip64 extension when creating an archive
with more than 64k entries, which nobody should need, right ;-)?
* The code in "multiple-worktree" support that attempted to recover
from an inconsistent state updated an incorrect file.
* "git rev-list" does not take "--notes" option, but did not complain
when one is given.
* Because the configuration system does not allow "alias.0foo" and
"pager.0foo" as the configuration key, the user cannot use '0foo'
as a custom command name anyway, but "git 0foo" tried to look these
keys up and emitted useless warnings before saying '0foo is not a
git command'. These warning messages have been squelched.
* We recently rewrote one of the build scripts in Perl, which made it
necessary to have Perl to build Git. Reduced Perl dependency by
rewriting it again using sed.
* t1509 test that requires a dedicated VM environment had some
bitrot, which has been corrected.
* strbuf_read() used to have one extra iteration (and an unnecessary
strbuf_grow() of 8kB), which was eliminated.
* The codepath to produce error messages had a hard-coded limit to
the size of the message, primarily to avoid memory allocation while
calling die().
* When trying to see that an object does not exist, a state errno
leaked from our "first try to open a packfile with O_NOATIME and
then if it fails retry without it" logic on a system that refuses
O_NOATIME. This confused us and caused us to die, saying that the
packfile is unreadable, when we should have just reported that the
object does not exist in that packfile to the caller.
* An off-by-one error made "git remote" to mishandle a remote with a
single letter nickname.
* A handful of codepaths that used to use fixed-sized arrays to hold
pathnames have been corrected to use strbuf and other mechanisms to
allow longer pathnames without fearing overflows.
Also contains typofixes, documentation updates and trivial code
clean-ups.
--------------
Version 1.94b:
--------------
- Changed allocator alignment to improve support for non-x86 systems (now
that llvm_mode makes this more feasible).
- Fixed a minor typo in afl-cmin. Spotted by Jonathan Neuschafer.
- Fixed an obscure bug that would affect people trying to use afl-gcc
with $TMP set but $TMPDIR absent. Spotted by Jeremy Barnes.
--------------
Version 1.93b:
--------------
- Hopefully fixed a problem with MacOS X and persistent mode, spotted by
Leo Barnes.
1.9.10
* Simplify clean list and add libtest build files.
* methodHandleParamTypes needs to refer to valid memory after rbffi_MethodHandle_Init
* tweak travis to allow failure in 2.2 on osx
* allow failure on osx 2.2 and head builds
* CLEAN tweak [ci skip]
1.9.9
* workaround bundler problem
* fix struct specs on 1.8 sets them to pending rather than checking for a raise but...
* better clean/clobber config
* Simplify cross build tasks and add rake-compiler-dock for building Windows binary gems.
* Added Platform.solaris? singleton method.
* fix library load... forever
* better native_type
* enable and fix warnings in specs.
* add compiler and platform based tests
* Fix compatibility with newer libffi on Windows/MSVC
* Handle better RUBY_VERSION.
New features and improvements
* Added support for compiler option -gsplit-dwarf.
Bug fixes
* Support external zlib in nonstandard directory.
* Avoid calling exit() inside an exit handler.
* Let exit handler terminate properly.
* Bail out on compiler option --save-temps in addition to -save-temps.
* Only log "Disabling direct mode" once when failing to read potential include files.
1.967_010 Sun Jul 7 11:23:53 2013
- Base the standalone precompiled parser's runtime name on the
parser's class, rather than use the fixed
"Parse::RecDescent::_Runtime". This prevents "already defined"
warnings when two standalone precompiled parsers are used.
- Add support for allowing precompiled parsers to share a common
runtime via the Precompile({-runtime_class}) option and the
PrecompiledRuntime() function.
- Warn on creation of Precompiled parsers that depend on
Parse::RecDescent.
- *** NON-BACKWARDS COMPATIBLE CHANGE! *** Change global the
<skip:> directive to eval similar to other <skip:> directives,
rather than being single-quoted in the resulting parser.
1.967011 Sat Sep 12 16:42:01 2015
- Correct some typos in the documentation. (RT.cpan.org #87185,
thanks dsteinbrunner!)
- Sort hash keys and rulenames when generating code. This keeps
the output text for a given input text the same, reducing
differences in automated builds. (RT.cpan.org #102160, thanks
Reiner!)
- Precompiled parsers now document which
$Parse::RecDescent::VERSION was used to generate
them. (RT.cpan.org #77001)
LGOGDownloader 2.25
- Added options to login separately to website and API
* --login-api
* --login-website
* Makes it possible to use cookies exported from browser (helps if
user gets reCAPTCHA on login form)
- Added detection of reCAPTCHA during login
- Changed path for game specific config files
* Game specific config files are now located in
$XDG_CONFIG_HOME/lgogdownloader/gamespecific/gamename.conf
- Added support for ignoring dlc count information using game
specific config file with --list option
- Removed --no-deb and --no-targz options
* GOG no longer provides deb and tarball archives
- Implemented gogdownloader:// protocol download (patch by: Sandy
Carter)
* Implements the following commands:
* lgogdownloader gogdownloader://GAME_NAME/FILE_ID
* lgogdownloader --download-file gogdownloader://GAME_NAME/FILE_ID
* lgogdownloader
gogdownloader://GAME_NAME_1/FILE_ID_1,GAME_NAME_2/FILE_ID_2,...
* lgogdownloader --download-file
gogdownloader://GAME_NAME_1/FILE_ID_1,GAME_NAME_2/FILE_ID_2,..
- Added option to specify output filename (patch by: Sandy Carter)
* Adds -o and --output-file option to specify path and downloaded
file name when using --download-file
- Removed tarball specific code from API implementation
- Fixed priority handling with duplicate values
- Added support for using strings to set language, language-priority,
platform and platform-priority instead of using numbers
- Fixed retry on operation timeout
# Changelog
## 1.0.8 (August 6, 2015)
* fix kqueue assertion failed, postpone ArmKqueueWriter until all events are processed [#51, #176, #372, #401, #619]
* fix Rubinius GC, crank the machine from Ruby space when running Rubinius [#201, #202, #617]
* test to show that LineText2 preserves whitespace and newlines [#32, #622]
* bump up compiler warnings and resolve them [#616]
* fix Windows x64 use uintptr_t instead of unsigned long for binding pointers [#612, #615]
* fix linetext2 unroll tail recursion to avoid stack level too deep [#609]
* fix for compilation with SSL on windows [#601]
* open file descriptors and sockets with O_CLOEXEC where possible [#298, #488, #591]
* fix SmtpClient: send second EHLO after STARTTLS. [#589]
* fix nul-terminated strings in C, use StringValueCStr instead of StringValuePtr
* 0.7.0 *
* Refactor build of FreeTDS & Iconv recipes. Add OpenSSL. Merged #207.
* Ensure zero terminated strings, where C-str pointers are expected. Use StringValueCStr() Fixes#208.
* Revert 999fa571 so timeouts do not kill the client. Fixes#179.
* Remove `sspi_w_kerberos.diff` patch. Not needed anymore.
* Tested again on Azure. Added notes to README on recommended settings.
* Replace `rb_thread_blocking_region` (removed in Ruby 2.2.0) w/`rb_thread_call_without_gvl`. Fixes#182.
* Remove 30 char password warning. Fixes#172.
* Remove Ruby 1.8.6 support. We always use Time vs edge case DateTime.
=== 4.26.0 (2015-09-01)
* Make Dataset#== not consider frozen status in determining equality (jeremyevans)
* Support :if_exists option to drop_column on PostgreSQL (jeremyevans)
* Add Dataset#grouping_sets to support GROUP BY GROUPING SETS on PostgreSQL 9.5+, MSSQL 2008+, Oracle, DB2, and SQLAnywhere (jeremyevans)
* Fix handling of Class.new(ModelClass){set_dataset :table} on ruby 1.8 (jeremyevans)
* Use range function constructors instead of casts for known range types in pg_range (jeremyevans) (#1066)
* Make class_table_inheritance plugin work without sti_key (jeremyevans)
* Detect additional disconnect errors when using the tinytds adapter (jeremyevans)
* Make offset emulation without order but with explicit selection handle ambiguous column names (jeremyevans)
* Allow preparing already prepared statements when emulating limits and/or offsets (jeremyevans)
* Have Sequel::NoMatchingRow exceptions record the dataset related to the exception (pedro, jeremyevans) (#1060)
== v0.18.3 [2015-09-03] Michael Granger <ged@FaerieMUD.org>
Enhancements:
- Use rake-compiler-dock to build windows gems easily.
- Add CI-tests on appveyor and fix test cases accordingly.
Bugfixes:
- Fix data type resulting in wrong base64 encoding.
- Change instance_of checks to kind_of for subclassing. #220
- TextDecoder::Date returns an actual Ruby Date instead of a Time
(thanks to Thomas Ramfjord)