Age | Commit message (Collapse) | Author | Files | Lines |
|
libtool-base.
|
|
|
|
== [7.4.2] 2014-06-03 ==
* Add config option to use STGRTMIN-based signals for thread suspend/resume.
* Allow parallel mark to be enabled on powerpc-linux systems.
* Check for Fujitsu compiler in builtin_unwind logic (enable FX10/K-Computer).
* Fix 'Array subscript is above array bounds' GCC warning in GC_new_kind/proc.
* Fix 'attribute declaration must precede definition' warning (clang-3.1).
* Fix (enable) Cygwin-64 build.
* Fix GC_finalized_malloc failure on disclaim_test.
* Fix GC_sig_suspend initialization when non-constant SIGRTMIN used.
* Fix MS VC redefinition warning for functions declared with GC_ATTR_MALLOC.
* Fix TEXT() usage for concatenated strings in GC_CreateLogFile (Win32).
* Fix data roots registration for Android/x86 and NDK ARM 'gold' linker.
* Fix find stackbottom on BlueGene P/Q systems.
* Fix machdep .lo files path in configure (SPARC, IA-64).
* Fix ok_init assignment (missing cast) in GC_new_kind_inner.
* Fix typos in names in AUTHORS and ChangeLog files.
* Remove barrett_diagram file duplicated by tree.html.
* Remove non-existing DISCARD_WORDS from GC data structure ASCII diagram.
* Restore contribution information for ancient releases in ChangeLog.
Also, includes 7.2f changes.
|
|
-----
5.0.1
-----
* Renamed script templates to end with .tmpl now that they no longer need
to be processed by 2to3. Fixes spurious syntax errors during build/install.
---
5.0
---
* Issue #218: Re-release of 3.8.1 to signal that it supersedes 4.x.
|
|
|
|
PBR is a library that injects some useful and sensible default behaviors into
your setuptools run. It started off life as the chunks of code that were copied
between all of the OpenStack projects. Around the time that OpenStack hit 18
different projects each with at least 3 active branches, it seemed like a good
time to make that code into a proper reusable library.
PBR is only mildly configurable. The basic idea is that there's a decent way to
run things and if you do, you should reap the rewards, because then it's simple
and repeatable. If you want to do things differently, cool! But you've already
got the power of Python at your fingertips, so you don't really need PBR.
PBR builds on top of the work that d2to1 started to provide for declarative
configuration. d2to1 is itself an implementation of the ideas behind distutils2.
Although distutils2 is now abandoned in favor of work towards PEP 426 and
Metadata 2.0, declarative config is still a great idea and specifically
important in trying to distribute setup code as a library when that library
itself will alter how the setup is processed. As Metadata 2.0 and other modern
Python packaging PEPs come out, PBR aims to support them as quickly as possible.
|
|
|
|
RELEASENOTES:
== 2.1.11 ==
(2014-01-31)
* hostmask parsing bug fixed by pmarks (a nearly complete rewrite of the mask parsing code)
* i97, incorrectly parses some v6 addresses.
* docstring typos.
* i95, refer to the nets in the exception raised by collapse_address_list
* add license to boilerplate to test-2to3.sh
|
|
implemented a signature-preserving contexmanager decorator. Fixed a bug
with the signature f(**kw) in Python 3 and fixed a couple of doctests
broken by Python 3.3.
3.3.3 Fixed a bug with kwonlyargs for Python 3
|
|
|
|
flexmock is a testing library for Python that makes it easy to create mocks,
stubs and fakes.
The API is inspired by a Ruby library of the same name, but Python flexmock is
not a clone of the Ruby version. It omits a number of redundancies in the Ruby
flexmock API, alters some defaults, and introduces a number of Python-only
features.
flexmock's design focuses on simplicity and intuitiveness. This means that the
API is as lean as possible, though a few convenient short-hand methods are
provided to aid brevity and readability.
flexmock declarations are structured to read more like English sentences than
API calls, and it is possible to chain them together in any order to achieve
high degree of expressiveness in a single line of code.
In addition, flexmock integrates seamlessly with all major test runners to
reduce even more mock-related boilerplate code.
|
|
Changelog:
Fixed in Firefox ESR 24.6
MFSA 2014-52 Use-after-free with SMIL Animation Controller
MFSA 2014-49 Use-after-free and out of bounds issues found using Address Sanitizer
MFSA 2014-48 Miscellaneous memory safety hazards (rv:30.0 / rv:24.6)
|
|
- Fix mimeinfo not adhering to the priorization of globs -
kudos Stefan Seifert.
- Fixed URL in POD. Only run POD 404 tests with extended testing.
|
|
Version 6.3 - June 9 2014
[CHANGES]
New facilities:
o htags-server: A private HTTP/CGI web server for a hyper-text
generated by htags(1).
o gtags.conf: New project based configuration mechanism.
You can make a configuration file for each project.
This is Leo Liu's idea.
o gtags, htags: New environment variables which have default
options for each command: GTAGS_OPTIONS, HTAGS_OPTIONS
o global: Added support of GREP_COLORS environment variable.
o global: Added new options:
-F (--first-match), -M (--match-case), -E (--extended-regexp)
[DEPRECATED FEATURES]
The following features of htags(1) are now deprecated.
They will be removed in the future.
o The -c (--compact) option.
o The --system-cgi option.
o The --overwrite-key option.
o The -x (--xhtml[=version]) option.
All files will be 1.0.
o The following configuration variables:
colorize-warned-line (substitute: --colorize-warned-line)
gzipped_suffix
htags_options (substitute: HTAGS_OPTIONS)
ncol (substitute: -n, --line-number [n])
normal_suffix (will be always '.html')
no_order_list (substitute: --no-order-list)
script_alias
tabs (substitute: --tabs n)
xhtml_version (will be always 1.0)
[FIXED BUGS]
o global: The highlight of symbols in library paths does not
work. Now it works.
o htags: Htags with the --suggest option didn't find GTAGS
in the obj directories. Now it works.
|
|
|
|
- Highlight parameters properly in TermColor output using @+ and @-
https://github.com/sheriff/test-bdd-cucumber-perl/issues/24
|
|
Changes in 0.8.2
----------------
* A fix for BCON when used from C++.
* Change bson_next_power_of_two() to accept size_t. This should not be
an ABI break since it is static inline.
Changes in 0.8.0
----------------
This cycle includes much, much more documentation for your perusing. There is
much more cross-referencing and structure for your navigation pleasure.
We've improved support for Libbson on a few more exotic platforms. SPARC
support is looking pretty good these days.
You'll also find some new examples in this release to help you get started a
bit faster. If there is something you'd like to see, just ask!
There are a few ABI breaks this cycle, as we are well on the road to a 1.0 and
would like things as clean as possible. I anticipate a few more during the next
couple of cycles, but we will try to keep them to a minimum. With that said,
you *WILL* need to recompile your application against 0.8.0.
|
|
|
|
* Update HOMEPAGE
Changelog:
Not provided.
https://www.mozilla.org/security/announce/2014/mfsa2014-55.html
Mozilla Foundation Security Advisory 2014-55 has been fixed
in this release.
|
|
Bump PKGREVISION.
|
|
|
|
|
|
- Check for valid module names. RT#94866 from TOBYINK.
- Changed used of die() to croak() RT#23113
- Fixed typo in pod (fschlich++)
- Added README
- Specified min version of perl 5.6.0
- Now "use warnings"
|
|
- eliminated a UNIVERSAL::isa warning
- added =encoding to POD (thanks to Debian Perl Group)
|
|
- add optional test deps as recommended prereqs (Karen Etheridge, #18)
- also skip the test if Capture::Tiny is too old (Martin Popel, #17)
- documentation updates (Flimm, #15)
|
|
This release fixes a memory leak with the default tempfile backend
layer implementation for trapping output. The systemsafe alternative
backend had the same kind of leak, and has also been fixed. (The
perlio alternative backend was not affected.) Thanks go again to
Felipe Gasper for discovering this!
Test::Trap::Builder:
- Add a method ExceptionFunction to make it easier to avoid
circular references arising from registered closures.
Document and export this method.
Test::Trap::Builder::TempFile:
- Break a circular reference chain.
Test::Trap::Builder::SystemSafe:
- Break a similar but slightly more complicated circular
reference chain.
Tests:
- Added t/14-leaks.t to check that the reference count is as
expected, at least with trivial code and standard setup for
all three output layer backends.
- Regression test for the stale references.
Library files:
- RT #87263: typo fixes, courtesy dsteinbrunner@pobox.com.
- Localize the trap coderef, so that it does not hold on to
what it closes over, creating stale references, but lets
them go out of scope; thanks go to Felipe Gasper.
Documentation:
- Some clarification for RT #91687: Wrong exit code after die.
Also:
- Use ebhanssen@cpan.org for my email address.
- Update copyright years.
- Update MANIFEST.SKIP &c for use of git (goodbye subversion).
|
|
- added repository metadata for CPAN
Contributed by David Steinbrunner (issue #13)
|
|
- Fixed "(in cleanup) Can't use an undefined value as a subroutine
reference" error.
- Remove Build.PL.
|
|
- rebuild MANIFEST, adding two forgotten files
- When printing diagnostics, differentiate the type of a blessed object
from the name of the class itself (RT#78288, caused by changes to how
blessed objects are treated in 0.109) (thanks, Karen Etheridge)
- Typo fixes (thanks, David Steinbrunner)
- Fixes to clarity and accuracy of documentation (thanks, Michael Hamlin)
- Add metadata links to repo and issue tracker
- Added obj_isa for testing ->isa without falling back to ref($x)
- Added the *experimental* ":v1" export group to skip importing
Isa, isa, and blessed
|
|
- properly handle thrown exceptions that stringify to the empty string
before or after chomping (Karen Etheridge, PR#11)
|
|
- Module now maintained by Sanko Robinson. Please see TODO for a possible
set of changes to this module that may effect code written for old, pre-
perl 5.14.0 platforms!!!
|
|
- Require Readonly 1.03+ and Scalar::Util 1.20 for testing with
Readonly. AFAICT, earlier versions of Readonly worked very differently,
and/or this is related to Scalar::Util. Either way, this test isn't that
important anyway. RT #95402.
- A call to validate() where the spec was a reference to a Readonly hash would
segfault. Reported by Salvatore Bonaccorso. RT #80572.
- When using the XS implementation, attempting to validate parameters with
tainted values would cause an "Insecure dependency in eval_sv() while
running with -T switch" exception from Perl *if the validation failed*. This
is fixed, but only for Perl 5.14+. Older Perls do not include the necessary
XS API. RT #70174.
- Fixed some potential segfaults from the XS implementation including the one
reported by Анатолий Гришаев in RT #50412. However, the XS code in general
is pretty crufty and tends to assume that values which could be NULL aren't.
- Fixed bad example of state() in POD. Reported by Salvatore Bonaccorso. RT
#80250.
|
|
- updated for 2014 release
- added support for 5.18
|
|
- Add =encoding directive to POD (RT#85625)
|
|
- switched packaging to just EUMM, as MBT wants the .xs file in a
different place
- switched packaging from MB to EUMM+MBT
- repository moved to the github moose organization
|
|
- repository moved to the github moose organization
|
|
- Replacing string `eval` with block `eval` for requiring test harnesses -
thanks Paul Cochrane
- Module metadata now points to repository and bug tracker - thanks Peter
Rabbitson
- Removed Ouch
- Another attempt to fix up the weird regex test bug
- Remove our experiment with ShareDir
- Some PerlCritic-ish fixes from Paul Cochrane
- Updated copyrights, again from Paul Cochrane
- There's some weird-ass bug with $1 not getting set properly, sometimes,
on perl's older than 5.14. I can't reproduce, and virtually identical
examples act differently. Also I can't reproduce it. Rewritten the test
itself to go via ->matches
- Now works with 5.10.0 again
- Adding missed dependency from Paul Cochrane
- Removed Method::Signatures dependency
- Added C and S step file shortcut subrefs
- Added Internationalization support, thanks to some sterling work
by Gregor Goldbach and Pablo Duboue
|
|
- Assign local abs_path rather than shuffle the symbol table after
using Cwd. Avoids issues with munging symbol table in Cwd itself.
|
|
|
|
Bump PKGREVISION.
|
|
---------------
3.7.1 and 4.0.1
---------------
* Issue #213: Use legacy StringIO behavior for compatibility under pbr.
---
4.0
---
* Issue #210: ``setup.py develop`` now copies scripts in binary mode rather
than text mode, matching the behavior of the ``install`` command.
---
3.8
---
* Extend Issue #197 workaround to include all Python 3 versions prior to
3.2.2.
---
3.7
---
* Issue #193: Improved handling of Unicode filenames when building manifests.
|
|
will now only install those when ncurses is not used. Fixes various
issues with TERM=screen on systems where misc/screen wasn't installed.
Bump PKGREVISION.
|
|
Those are required for the case in readline compatible mode, not for using
builtin editline.
|
|
|
|
Bugs fixed
* Unpacker doesn't increment refcount of ExtType hook.
* Packer raises no exception for inputs doesn't fit to msgpack format.
0.4.1
Changes
* fallback.Unpacker.feed() supports bytearray.
Bugs fixed
* Unpacker doesn't increment refcount of hooks. Hooks may be GCed while
unpacking.
* Unpacker may read unfilled internal buffer.
|
|
Changes are too many to write here.
|
|
Changelog:
3.10: 2014-04-19
Finer control over filtering with caC flags.
New setfield command for tweaking object attributes from lift scripts.
3.9: 2014-04-12
The attribution-parsing code handles odd characters in names better now.
The filter command can operate on email addresses as well as names.
New 'stamp' command to report action stamps of commits.
New 'count' comment reports selection-set counts.
New branchify_mapping option for renaming Subversion branches on analysis.
|
|
|
|
PCRE_INFO_MAXLOOKBEHIND which is introduced in pcre 8.31.
|
|
This should work around PR pkg/48732 until we have a beter solution
e.g. building this package with "libtool".
|