Age | Commit message (Collapse) | Author | Files | Lines |
|
This release contains a number of fixes for the integration with launchd
on MacOS/darwin, some general code cleanups, and build configuration
improvements.
Packagers should note that this release expands the use of the openssl
program for xauth cookie generation to any platform which has openssl
but not mcookie discoverable by configure. It assumes the presence of
the rand -hex flag originally added for OpenBSD, but now carried in the
upstream OpenSSL. The configure script will output a notice such as:
configure: startx will depend on "/usr/bin/openssl rand -hex 16" to
generate xauth cookies
in order to inform you which was chosen so you may set your package
dependencies appropriately.
Alan Coopersmith (7):
Add printf attributes suggested by gcc
If we have to forcibly kill the Xserver on Solaris, reset kbd_mode
Declare processTimeout() as taking a const string arg
Check for openssl as mkcookie fallback on all platforms, not just OpenBSD
Use pid_t instead of int for type of pids
Add Apple copyright/license notice from launchd files to COPYING
xinit 1.3.3
Egbert Eich (2):
man: Set correct config file location in man page
install: Use sysconfdir instead of libdir for config files.
Jeremy Huddleston (9):
launchd/privileged_startx: Improved error resolution
launchd: Don't provide the LaunchAgent on Tiger
launchd: Provide more verbose error reporting for launchd checkin failures
launchd: Load privileged_startx properly on Tiger
launchd: Fix the destination of moved-aside directories in privileged_startx
darwin: Silence a syslog message from defaults
laucnhd: Silence some syslog spam
Annotate Fatal and Fatalx as _X_NORETURN
XQuartz: Fix an issue where -dpi was not being passed to the server despite being set
Jeremy Huddleston Sequoia (1):
privileged_startx: use root:wheel instead of 0:0
Thomas Klausner (1):
Use xinitrc.d/?*.sh as the matching pattern instead of needing
a fragile sed substitution XSLASHGLOB to produce xinitrc.d/*.sh.
(If that were used literally the slash-star would confuse
cpp.)
|
|
|
|
|
|
|
|
Changes since previous version:
SI6 Networks' IPv6 Toolkit v1.4.1
* frag6: Fixed bug that prevented Ethernet header from being filled
A bug in the code caused Ethernet frames to go on te wire without any of
their header fields completed.
* All: Use of library to avoid code replication
An "libipv6" library was created, such that common functions do not need
to be replicated for each tool. ni6, ns6, rs6, and tcp6 now employ such
library.
pkgsrc changes:
* address6 and its man page are no longer installed
* extend the Makefile changes to include the correct linkage for rs6 and tcp6
i.e. include the libipv6 object mentioned above
|
|
flickcurl-1.25, fontconfig-2.10.95, global-6.2.9, gnumeric112-1.12.7,
graphviz-2.34.0, horde-5.1.4, kde-4.11.1, libkface-3.4.0,
libkgeomap-3.4.0, libkvkontakte-3.4.0, libmediawiki-3.4.0,
libusbx-1.0.17, libvpx-1.2.0, liferea-1.10.2, phpmyadmin-4.0.6,
py-pytz-2013d, py-qt4-4.10.3 [needs py-sip-4.15, see above],
py-setuptools-1.1.4, py-sip-4.15.1 [don't update until pyqt3 and
KDE problems are fixed -- still there in 4.15.1], rhythmbox-3.0,
typo3-6.1.4, x264-devel-20130907.
|
|
|
|
|
|
|
|
Changelog:
unknown.
|
|
|
|
|
|
|
|
|
|
|
|
as necessary. Bump revision.
|
|
* Add FreeMiNT support. This file is needed for bootstrapping pkgsrc.
|
|
|
|
* Fix build on FreeMiNT.
* Tested on some platforms, See files/README.
|
|
|
|
* Fix build with libnbcompat-20120702, PR pkg/42142's side effect is fixed.
Tested on Solaris 10, OpenBSD, FreeBSD etc.
Detect system's sha2.h and SHA2 functions, and include nbcompat/sha2.h
conditionally.
|
|
|
|
Upstream changes:
1.192 2013-09-07 21:00:55 America/New_York
if Net::DNS is not available, do not test null mx validity
1.191 2013-09-04 21:52:50 America/New_York
update bugtracker, repo, etc.
mxcheck fails when remote MX record is "."
drop use of bytes.pm; any bug fixed by that should be fixed again
properly
|
|
|
|
Upstream changes:
2.104 2013-08-16
Documentation improvements (kd).
2.103 2013-05-30
note deprecation in favor of Email::Stuffer
|
|
|
|
Upstream changes:
2.202 2013-09-04 09:48:33 America/New_York
repackage with Dist::Zilla; update repo and bug tracker
2.201 2013-04-16
no changes
2.200_01 2013-04-09
preserve the original header layout unless changed
2.103 2013-04-09
fix a bug in header unfolding where "0\n 1" could be unfolded to "01"
|
|
|
|
Update DEPENDS
UPstream changes:
1.300006 2013-03-17 17:23:21 America/New_York
re-fix the problem with attribute accessors and role methods
clashing, previously fixed in 0.102360
1.300005 2013-02-14 11:37:09 America/New_York
add missing prereq, Sub::Exporter; thanks for report, Esteban
Manchado Velè°©zquez
1.300004 2013-02-09 10:09:02 America/New_York
bump required version of MooX::Types::MooseLike
1.300003 2013-02-06 15:56:53 America/New_York
make Wrapper transports proxy is_simple and allow_partial_success to
the wrapped transport
allow Wrapper transports to construct their own targets; useful for
configuring entire wrapper chain from the environment
1.300002 2013-02-06 14:06:24 America/New_York
THIS IS A BIG DEAL: see v1.300000
this is the first non-trial release using Moo instead of Moose
increase version of Moo to interoperate with ancient Mouse.pm
1.300001 2013-01-07 21:38:31 America/New_York
increase required versions of Throwable and Moo
1.300000 2013-01-02 22:42:58 America/New_York
THIS IS A BIG DEAL:
use Moo instead of Moose (thanks to Justin Hunter and Christian
Walde!)
0.120002 2012-09-11 14:43:07 America/New_York
added shift_deliveries to the Test transport, for treating the
sent messages like a queue of things to inspect
0.120001 2012-07-19 16:28:01 America/Los_Angeles
no changes from 0.120000, just marked as stable release
0.120000 2012-05-08 20:53:43 America/New_York
THIS IS A BIG DEAL:
This release deprecates the scalar context behavior of:
* Failure->recipients
* Failure::Multi->failures
* Failure::Multi->recipients
* Transport::Test->deliveries
* Transport::Failable->failure_conditions
Until now, these methods returned a list in list context and
an array reference in scalar context. This still works, but
scalar context calls will now generate a warning. In one
year (2013-05), they will be become fatal.
0.110005 2012-03-05 21:37:35 America/New_York
binmode output handle to avoid line ending munging in Mbox and
Maildir; thanks, Christian Walde!
0.110004 2012-02-21 17:47:35 America/New_York
import the "Bcc" page from the (otherwise empty) github wiki; this
adds some information to the QuickStart page about why Bcc support
isn't there and why this isn't a bug
0.110003 2012-02-01 15:09:52 America/New_York
add the add_lines_header and add_envelope_headers attributes to
Maildir transports
0.110002 2012-01-31 17:02:35 America/New_York
Maildir deliveries now have a ->filename method to tell you just
where the mail was delivered
|
|
|
|
Update DEPENDS
Upstream changes:
0.200008 2013-04-30 12:50:19 America/New_York
avoid a Moose-using test in the suite of tests run by users (thanks,
Dave Rolsky)
0.200007 2013-04-09 19:20:02 America/New_York
allow previous_exception to be passed to the constructor
0.200006 2013-04-09 14:58:07 America/New_York
lose old $@ in previous_exception less often under 5.17.10 (thanks,
David Wheeler)
0.200005 2013-04-05 15:04:52 America/New_York
Plugged memory leak (thaljef)
0.200004 2013-02-21 16:50:05 America/New_York
update Throwable::Error to use around modifier instead of calling
SUPER:: (thanks, Justin Hunter)
0.200003 2012-10-18 08:52:20 America/New_York
no code changes; just no longer trial
0.200002 2012-10-09 08:56:21 America/New_York
require Moo 1.000001
0.200001 2012-09-22 18:10:29 America/New_York
update a prerequisite (thanks, FREW)
0.200000 2012-09-20 14:12:34 America/New_York
port to Moo (FREW, MSTROUT, ILMARI)
|
|
|
|
Upstream changes:
2.199 2013-08-18
** PLEASE CONSIDER USING EMAIL-SENDER INSTEAD **
stop warning from Return::Value; if it didn't switch you to
Email::Sender by now, it isn't gonna
cope with irritating differences between Net::SMTP and Net::SMTP::TLS
(thanks, Yanick Champoux!)
|
|
|
|
Version 0.1.3 (2012-06-09)
=============
- Migrate away deprecated GTK+ APIs, fixing FTBFS.
|
|
|
|
Add LICENSE
Upstream changes:
1.203 2013-08-16 22:23:32 America/New_York
update repo and bugtracker
reformat docs a bit
|
|
|
|
|
|
Update DEPENDS
Add LICENSE
Upstream changes:
1.317 2013-08-31 23:44:32 America/New_York
update repo and bugtracker information
no significant changes, but please see a note on the future:
https://github.com/rjbs/Email-MIME-Attachment-Stripper/issues/2
|
|
|
|
|
|
Contao 3.0 is obsoleted by Contao 3.1 and it is end of migration period.
|
|
|
|
|
|
|
|
Update DEPENDS
Upstream changes:
1.924 2013-08-10 23:23:55 America/New_York
update use of Email::MIME::ContentType to match new, fixed hash keys:
type/subtype
1.923 2013-08-08 21:59:02 America/New_York
do not consider the part-ending CRLF part of the body
avoid undefined warnings in debug_structure [rt.cpan.org #82388]
(Thanks, Kurt Anderson)
better error message when the given body is a ref but not a scalar
ref [rt.cpan.org #59205]
1.922 2013-07-10 08:45:02 America/New_York
repackage, fixing version number
1.921 2013-07-01 22:51:01 America/New_York
repackage, remove PEP links, update bugtracker
1.920 2013-06-17
do not call parts_set during walk_parts unless the parts have
actually changed
When trying to decode a body, fall back to 7bit if the encoding is
unknown. Trying to create a new body in an unknown encoding is still
forbidden. This should make it easier to handle broken messages
from "the internet." ("Content-Transfer-Encoding: n"!?)
1.912_01 2013-04-08
try to encode headers based on the header structure, if it has one,
rather than treating the header as a big string in all cases; thanks
for this work go to Jesse Luehrs
|
|
|
|
Update LICENSE
Upstream changes:
1.315 2013-08-12 21:11:46 America/New_York
repackage to update repository and bug tracker
1.314 2013-06-17
add a third argument to encode/decode/codec to allow a fallback
codec
|
|
|