pkgsrc change:
* Add ${GEM_EXTSDIR}/gem.build_complete for new rubygems and updated ruby.
=== unicorn 4.8.3 - the end of an era / 2014-05-07 07:50 UTC
This release updates documentation to reflect the migration of the
mailing list to a new public-inbox[1] instance. This is necessary
due to the impending RubyForge shutdown on May 15, 2014.
The public-inbox address is: unicorn-public@bogomips.org
(no subscription required, plain text only)
ssoma[2] git archives: git://bogomips.org/unicorn-public
browser-friendly archives: http://bogomips.org/unicorn-public/
Using, getting help for, and contributing to unicorn will never
require any of the following:
1) non-Free software (including SaaS)
2) registration or sign-in of any kind
3) a real identity (we accept mail from Mixmaster)
4) a graphical user interface
Nowadays, plain-text email is the only ubiquitous platform which
meets all our requirements for communication.
There is also one small bugfix to handle premature grandparent death
upon initial startup. Most users are unaffected.
[1] policy: http://public-inbox.org/ - git://80x24.org/public-inbox
an "archives first" approach to mailing lists
[2] mechanism: http://ssoma.public-inbox.org/ - git://80x24.org/ssoma
some sort of mail archiver (using git)
=== unicorn 4.8.2 - avoid race condition during worker startup / 2014-02-05 18:24 UTC
We close SELF_PIPE in the worker immediately, but signal handlers
do not get setup immediately. So prevent workers from erroring out
due to invalid SELF_PIPE.
=== unicorn 4.8.1 / 2014-01-29 08:48 UTC
fix races/error handling in worker SIGQUIT handler
This protects us from two problems:
1) we (or our app) somehow called IO#close on one of the sockets
we listen on without removing it from the readers array.
We'll ignore IOErrors from IO#close and assume we wanted to
close it.
2) our SIGQUIT handler is interrupted by itself. This can happen as
a fake signal from the master could be handled and a real signal
from an outside user is sent to us (e.g. from unicorn-worker-killer)
or if a user uses the killall(1) command.
=== unicorn 4.8.0 - big internal changes, but compatible / 2014-01-11 07:34 UTC
This release contains fairly major internal workings of master-to-worker
notifications. The master process no longer sends signals to workers
for most tasks. This works around some compatibility issues with some
versions of the "pg" gem (and potentially any other code which may not
handle EINTR properly). One extra benefit is it also helps stray
workers notice a rare, unexpected master death more easily. Workers
continue to (and will always) accept existing signals for compatibility
with tools/scripts which may signal workers.
PID file are always written early (even on upgrade) again to avoid
breaking strange monitoring setups which use PID files. Keep in mind we
have always discouraged monitoring based on PID files as they are
fragile.
We now avoid bubbling IOError to the Rack app on premature client
disconnects when streaming the input body. This is usually not a
problem with nginx, but may be on some LAN setups without nginx).
Thanks to Sam Saffron, Jimmy Soho, Rodrigo Rosenfeld Rosas,
Michael Fischer, and Andrew Hobson for their help with this release.
Note: the unicorn mailing list will be moved/changed soon due to the
RubyForge shutdown. unicorn will always rely only on Free Software.
There will never be any sign-up requirements nor terms-of-service to
agree to when communicating with us.
=== unicorn 4.8.0pre1 / 2013-12-09 09:51 UTC
Eric Wong (6):
tests: fix SO_REUSEPORT tests for old Linux and non-Linux
stream_input: avoid IO#close on client disconnect
t0300: kill off stray processes in test
always write PID file early for compatibility
doc: clarify SIGNALS and reference init example
rework master-to-worker signaling to use a pipe
=== unicorn 4.7.0 - minor updates, license tweak / 2013-11-04 06:59 UTC
* support SO_REUSEPORT on new listeners (:reuseport)
This allows users to start an independent instance of unicorn on
a the same port as a running unicorn (as long as both instances
use :reuseport).
ref: https://lwn.net/Articles/542629/
* unicorn is now GPLv2-or-later and Ruby 1.8-licensed
(instead of GPLv2-only, GPLv3-only, and Ruby 1.8-licensed)
This changes nothing at the moment. Once the FSF publishes the next
version of the GPL, users may choose the newer GPL version without the
unicorn BDFL approving it. Two years ago when I got permission to add
GPLv3 to the license options, I also got permission from all past
contributors to approve future versions of the GPL. So now I'm
approving all future versions of the GPL for use with unicorn.
Reasoning below:
In case the GPLv4 arrives and I am not alive to approve/review it,
the lesser of evils is have give blanket approval of all future GPL
versions (as published by the FSF). The worse evil is to be stuck
with a license which cannot guarantee the Free-ness of this project
in the future.
This unfortunately means the FSF can theoretically come out with
license terms I do not agree with, but the GPLv2 and GPLv3 will
always be an option to all users.
Note: we currently prefer GPLv3
Two improvements thanks to Ernest W. Durbin III:
* USR2 redirects fixed for Ruby 1.8.6 (broken since 4.1.0)
* unicorn(1) and unicorn_rails(1) enforces valid integer for -p/--port
A few more odd, minor tweaks and fixes:
* attempt to rename PID file when possible (on USR2)
* workaround reopen atomicity issues for stdio vs non-stdio
* improve handling of client-triggerable socket errors
pkgsrc change:
* Add ${GEM_EXTSDIR}/gem.build_complete for new rubygems and updated ruby.
Version 3.2.2
* Consider script as a block-level element so it doesn't get included inside a
paragraph.
Version 3.2.1
* Load RedcarpetCompat when requiring Redcarpet for the sake of backward
compatibility.
Loren Segal
Version 3.2.0
* Add a Safe renderer to deal with users' input. The escape_html and
safe_links_only options are turned on by default.
Moreover, the block_code callback removes the tag's class since the user can
basically set anything with the vanilla one.
Robin Dupret
* HTML5 block-level tags are now recognized
silverhammermba
* The StripDown render object now displays the URL of links along with the
text.
Robin Dupret
* The RedCloth API compatibility layer is now deprecated.
Robin Dupret
* A hyphen and an equal should not be converted to heading.
namusyaka
* Fix emphasis character escape sequence detection while mid-emphasis.
jcheatham
* Add = to the whitelist of escaped chars so it can be used inside highlighted
snippets.
jcheatham
* Convert trailing single quotes to curly quotes. For example, Road Trippin'
now converts to Road Trippin¡Ç.
Kevin Chen
* Allow in-page links (e.g. [headline](#headline)) when :safe_links_only is
set.
jomo
* Enable emphasis inside of sentences in multi-byte languages when
:no_intra_emphasis is set.
Chun-wei Kuo
* Avoid making :no_intra_emphasis only match spaces. This allows using
emphasizes inside quotes when the option is enabled for instance.
Jason Webb and BJ Homer
* The StripDown renderer handles image tags now.
Version 3.1.2
* Remove the yielding of anchors in the header callback. This was a breaking
change between 3.0 and 3.1 as the method's arity changed.
* Compile fix for Ruby 2.2.
* And a few bug fixes.
pkgsrc change
* support pkg_alternatives.
* Add ${GEM_EXTSDIR}/gem.build_complete for new rubygems and updated ruby.
This is the last ordinal release of Ruby 2.0.0. Ruby 2.0.0 goes into the
state of the security maintenance phase, and will never be released unless
any critical regressions or security issues are found. This phase is planned
to be maintained for 1 year. Then, maintenance of Ruby 2.0.0 will be ended
at Feb. 24th, 2016. We recommend to start planning to migrate to newer
versions of Ruby, such as 2.1 or 2.2.
pkgsrc change:
* Change "os" of Gem::Platform, now it change the case of "netbsd" only.
* Reduce patches to builtin rubygems.
Upstream changes:
2.025001 2015-02-28 09:53:36-06:00 America/Chicago
- fix ::RS::Errors test on Windows
2.025000 2015-02-20 22:50:47-06:00 America/Chicago
- Add ::Helper::ResultSet::Errors to help when calling Result methods on an RS
* Simplify Gem::Specification definition.
* Remove some pkgsrc specific modifications.
=== 2.4.6 / 2014-02-05
Bug fixes:
* Fixed resolving gems with both upper and lower requirement boundaries.
Issue #1141 by Jakub Jirutka.
* Moved extension directory after require_paths to fix missing constant bugs
in some gems with C extensions. Issue #784 by Andr«± Arko, pull request
#1137 by Barry Allard.
* Use Gem::Dependency#requirement when adding a dependency to an existing
dependency instance. Pull request #1101 by Josh Cheek.
* Fixed warning of shadowed local variable in Gem::Specification. Pull request
#1109 by Rohit Arondekar
* Gem::Requirement should always sort requirements before coercion to Hash.
Pull request #1139 by Eito Katagiri.
* The `gem open` command should change the current working directory before
opening the editor. Pull request #1142 by Alex Wood.
* Ensure quotes are stripped from the Windows launcher script used to install
gems. Pull request #1115 by Youngjun Song.
* Fixed errors when writing to NFS to to 0444 files. Issue #1161 by Emmanuel
Hadoux.
* Removed dead code in Gem::StreamUI. Pull request #1117 by mediaslave24.
* Fixed typos. Pull request #1096 by hakeda.
* Relaxed CMake dependency for RHEL 6 and CentOS 6. Pull request #1124 by V«¿t
Ondruch.
* Relaxed Psych dependency. Pull request #1128 by V«¿t Ondruch.
2.2. It affects ruby 2.1 only.
* Binary packages' file name start with "ruby21-" instead of "ruby215-".
* ${PREFIX}/include/ruby-2.1.5 are changed from ${PREFIX}/include/ruby-2.1.5
to ${PREFIX}/include/ruby-2.1.0.
* ${PREFIX}/lib/ruby/2.1.5 and ${PREFIX}/lib/ruby/gem/2.1.5 are changed
to ${PREFIX}/lib/ruby/2.1.0 and ${PREFIX}/lib/ruby/gem/2.1.0.
- convert just compile (by cc) into (almost) regular make packaging,
with minor trick, say, renaming WRKSRC etc.
(upstream)
- Update 1.96 to 1.104
(from) https://bitbucket.org/gotoh/connect/commits/all
------------------------------------------------------
Added tag 1.104 for changeset fd6f352325d4 2014-09-24
new version number to 1.104 2014-09-24
allow to use '@' in user name part of host spec string (close#6) 2014-09-24
Clang support on Win32. 2012-09-19
Win32 support for GNU-like toolchains in Makefile 2012-09-19
Fixed typos 2012-09-19
Added return type to domain_match 2012-09-11
Add Interix support 2012-08-25
Fix to support some old Solaris environment. (close#4) 2012-07-26
make Makefile as closs-platform (close#3) 2012-07-26
need ifdef for using make_localnet_as_direct(). 2012-06-20
Added tag 1.103 for changeset c7321796112d 2012-05-14
Use closesocket() on HTTP proxy. 2012-05-14
redefine ECONNRESET only when it is lacked. 2012-04-30
resolve hostname localy before determine direct. 2012-04-21
add feature to make direct access for hosts on the local network. 2012-04-21
doc: fix notes format a little 2012-04-21
Add document and support elisp code. 2009-12-17
Added tag 1.101 for changeset cf9579815b7e 2009-12-17
bump-up to version 1.101. 2009-12-17
fix-up path separator of ssh askpass executable on win32 except cygwin.
2009-12-17
Define version number string explicitly to remove SCM dependency. 2009-12-17
Added tag 1.100 for changeset 7c036cbffb61 2009-12-17
Import connect.c version 1.100. 2009-12-17
- Add LICENSE= gnu-gpl-v2
(upstream)
- Update 3.05 to 3.3
------------------
Following lines are from ChangeLog, but not really clear,
the date is overwrapped from last (pkgsrc) update.
------------------
2001-02-25 lars brinkhoff <lars@nocrew.org>
From Sampo Niskanen <sampo.niskanen@iki.fi>:
* common.c (handle_tunnel_input): write to stdout if fd = 0.
* htc.c (struct Arguments): add use_std and use_daemon.
(usage): add short option -z for --proxy-authorization-file.
add -s, --stdin-stdout option. add -w, --no-daemon option.
(parse_arguments): recognize new options.
(parse_arguments): write diagnostics to stderr instead of stdout.
(main): likewise.
* hts.c (struct Arguments): add use_std and use_daemon.
(usage): add -s, --stdin-stdout option. add -w, --no-daemon option.
(parse_arguments): recognize new options.
(main): write diagnostics to stderr instead of stdout.
2000-09-01 lars brinkhoff <lars@nocrew.org>
From Brian Somers <brian@Awfulhak.org>:
* htc.c (main): correct typo in log message.
* tunnel.c (tunnel_out_connect): use ntohl() to convert IP address.
(tunnel_accept): likewise.
* common.c (set_address): likewise.
2000-08-31 lars brinkhoff <lars@nocrew.org>
* configure.in: version 3.2
* debian/changelog: sync with new version.
2000-07-25 lars brinkhoff <lars@nocrew.org>
From Brian Somers <brian@Awfulhak.org>:
* tunnel.h: update tunnel_new_server prototype.
* tunnel.c (tunnel_is_server): tunnel is server when
server_socket != -1.
(tunnel_out_connect): log the port number.
(tunnel_in_connect): don't call http_destroy_response with
NULL argument.
(tunnel_accept): log client IP number and port.
(tunnel_new_server): accept char *host argument, which is used
to bind the server to a specific network interface. initialize
tunnel->bytes to 0. pass struct in_addr to server_socket.
(tunnel_new_client): initialize tunnel->bytes to 0.
* hts.c (parse_arguments): host defaults to NULL. recognize
[HOST:]PORT syntax.
(main): log host:port, if host was specified. pass host to
tunnel_new_server.
* common.c (server_socket): take a sockaddr_in * argument; improve
sockaddr_in initialization.
(set_address): improve sockaddr_in initialization.
* common.h: update server_socks prototype.
* htc.c (main): pass a struct in_addr to server_socket.
Lua 5.2.4 fixes all bugs listed in
http://www.lua.org/bugs.html#5.2.3 [*]
and is now the current release of Lua 5.2.
Like all minor releases, this is strictly a bug-fix release;
no new features or improvements have been added.
[*] http://www.lua.org/bugs.html#5.2.3
1. Compiler can optimize away overflow check in table.unpack.
2. Ephemeron table can wrongly collect entry with strong key.
3. Chunk with too many lines may crash Lua.
Remove unneeded DEPENDS
Upstream changes:
2015-03-05 Patrick Galbraith, Michiel Beijen, DBI/DBD community (4.031)
* Added LICENSE
* Reworked installation documentation in POD.
* Allow use of bracketed IPv6 addresses in connection string (RT70640),
fix by Tim Mullin @ cPanel).
* Locate mysql_config in path on MSWin32 (RT94838, reported by CHORNY).
* Environment variable DBD_MYSQL_CONFIG actually works now.
* Makefile.PL now correcly handles MariaDB's mysql_config.
* Correct attribution to David Dick #86030: PATCH: adding statistics_info support
* Test suite can now be run in parallel. Fixes RT98994 reported by DOHERTY.
* Fixes for MySQL 5.6+ and for prepared statement handling (stevenh on github).
* Fix for RT100792, New test t/05dbcreate.t fails if user has no permissions.
* Fix for bit test (RT68374, BJDEAN@cpan.org) CaptTofu added 40bit.t to test.
2015-03-02 Patrick Galbraith, Michiel Beijen, DBI/DBD community (4.030_02)
* Reworked installation documentation in POD.
* Allow use of bracketed IPv6 addresses in connection string (RT70640),
fix by Tim Mullin @ cPanel).
* Locate mysql_config in path on MSWin32 (RT94838, reported by CHORNY).
* Environment variable DBD_MYSQL_CONFIG actually works now.
* Makefile.PL now correcly handles MariaDB's mysql_config.
2015-01-27 Patrick Galbraith, Michiel Beijen, DBI/DBD community (4.030_01)
* Correct attribution to David Dick #86030: PATCH: adding statistics_info support
* Test suite can now be run in parallel. Fixes RT98994 reported by DOHERTY.
* Fixes for MySQL 5.6+ and for prepared statement handling (stevenh on github).
* Fix for RT100792, New test t/05dbcreate.t fails if user has no permissions.
* Fix for bit test (RT68374, BJDEAN@cpan.org) CaptTofu added 40bit.t to test.
2014-12-08 Patrick Galbraith (aka CaptTofu) and community (4.029)
* Added fix to tests to create test database if not exists (contstant failure on Travis) (CaptTofu)
* Support the fraction of (Oracle) MySQL Fabric that is supported by the most recent Connector/C (Steffen Mueller smueller<at>cpan<dot>org
* Statistics Info Milanrm <sorm<at>is4u<dot>cz> for work on statistics_info
* Fix for RT 97625, use after free(), Reini Urban rurban<at>cpan<dot>org
2014-08-01 Patrick Galbraith, Michiel Beijen, DBI/DBD community (4.028)
* Fixed bug in mysql.xs where dbh was being used as error code
* RT #97570: fix wrong salloc free in mysql_st_internal_execute - (Reini Urban, cPanel)
* Fix RT #97625 use-after-free in mysql_dr_error, and #86153 - (Reini Urban, cPanel)
* find mysql.h for MariaDB on Win32 (Graham Ollis)
* Update mysql.pm to work with ipv6 and ipv4 addresses (katyavoid)
Core support for multiple instances of the same apphook'ed application
The template tag `render_model_add` can now accept a model class as well as a model instance
Fixes an issue with reverting to Live mode when moving plugins
Fixes a missing migration issue
Fixes an issue when using the PageField widget
Fixes an issue where duplicate page slugs is not prevented in some cases
Fixes an issue where copying a page didn't copy its extensions
Fixes an issue where translations where broken when operating on a page
Fixes an edge-case SQLite issue under Django 1.7
Fixes an issue where a confirmation dialog shows only some of the plugins to be deleted when usingthe "Empty All" context-menu item
Fixes an issue where deprecated 'mimetype' was used instead of 'contenttype'
Fixes an issue where `cms check` erroneous displays warnings when a plugin uses class inheritance
Documentation updates
* "update-index --refresh" used to leak when an entry cannot be
refreshed for whatever reason.
* "git fast-import" used to crash when it could not close and
conclude the resulting packfile cleanly.
* "git blame" died, trying to free an uninitialized piece of memory.
* "git merge-file" did not work correctly in a subdirectory.
* "git submodule add" failed to squash "path/to/././submodule" to
"path/to/submodule".
* In v2.2.0, we broke "git prune" that runs in a repository that
borrows from an alternate object store.
* Certain older vintages of cURL give irregular output from
"curl-config --vernum", which confused our build system.
* An earlier workaround to squelch unhelpful deprecation warnings
from the complier on Mac OSX unnecessarily set minimum required
version of the OS, which the user might want to raise (or lower)
for other reasons.
* Longstanding configuration variable naming rules has been added to
the documentation.
* The credential helper for Windows (in contrib/) used to mishandle
a user name with an at-sign in it.
* Older GnuPG implementations may not correctly import the keyring
material we prepare for the tests to use.
* Clarify in the documentation that "remote.<nick>.pushURL" and
"remote.<nick>.URL" are there to name the same repository accessed
via different transports, not two separate repositories.
* The pack bitmap support did not build with older versions of GCC.
* Reading configuration from a blob object, when it ends with a lone
CR, use to confuse the configuration parser.
* We didn't format an integer that wouldn't fit in "int" but in
"uintmax_t" correctly.
* "git push --signed" gave an incorrectly worded error message when
the other side did not support the capability.
* "git fetch" over a remote-helper that cannot respond to "list"
command could not fetch from a symbolic reference e.g. HEAD.
* The insn sheet "git rebase -i" creates did not fully honor
core.abbrev settings.
* The tests that wanted to see that file becomes unreadable after
running "chmod a-r file", and the tests that wanted to make sure it
is not run as root, we used "can we write into the / directory?" as
a cheap substitute, but on some platforms that is not a good
heuristics. The tests and their prerequisites have been updated to
check what they really require.
* The configuration variable 'mailinfo.scissors' was hard to
discover in the documentation.
* Correct a breakage to git-svn around v2.2 era that triggers
premature closing of FileHandle.
* Even though we officially haven't dropped Perl 5.8 support, the
Getopt::Long package that came with it does not support "--no-"
prefix to negate a boolean option; manually add support to help
people with older Getopt::Long package.
