Update DEPENDS
Upstream changes:
1.78 2020-12-15
* Fix tests for new failure caused by locale data changes in DateTime::Locale
1.29. Reported by cthulhu8zombie. GH #28.
* Added a warning about using locale-specific patterns. Some of these patterns
can change quite a bit as the locale data is updated, so using them for
parsing does not produce stable results across time. This is what caused the
test failures that this release fixes.
Upstream changes:
1.31 2020-12-22
- The docs for each locale now show examples of its strftime patterns as well.
1.30 2020-12-15
- The previous release included some locale data for locales that had been
deleted from CLDR 38.0.0.
1.29 2020-12-14
- Rebuilt all locale data with the data from CLDR 38.0.0 and glibc 2.32.
Upstream changes:
0.104 2021-03-15 T. R. Wyant
Add rt.cpan.org back to bug reporting methods. Long live RT!
Correct GitHub ticket link.
Get prerequisites up to snuff and add xt/author/prereq.t to ensure
they stay that way.
Upstream changes:
=== Version 1.002 (2020-12-27)
README.md, lib/PDF/Table.pm, Makefile.PL
Miscellaneous updates to clean up support URLs, emails, etc.
util/pod2cpanhtml.pl
Remove use of search.cpan.org CSS reference, as that site no longer exists.
Looking for a good replacement, possibly Bootstrap, that doesn't require
that a phone be turned to landscape orientation in order to read text at
a decent size.
lib/PDF/Table.pm
Cells spanning columns only had the first part (cell) of the bottom
horizontal rule drawn. Reported by "stefanalt". Fixed to draw full spanned
width.
t/PDF-Table.t
One calculated value was considerably different between standard Perl (uses
doubles for floats), long-double Perl, and quad-math Perl. Round the
calculated value to 6 decimal digits so that it works on any system.
Upstream changes:
2.039 2021-03-04
- PDF::API2->open($filename) no longer reads the entire file into memory
before working on it, instead reading from the file as needed. This
results in a substantial reduction in memory usage and start-up time for
large PDF files, particularly in cases where the calling script doesn't
need to access all of the contents of the PDF.
- PDF files containing cross-reference streams can now be modified and saved
normally. Previously, they were read-only.
- $text->paragraph(...) and $text->paragraphs(...) (formerly "section") are
no longer undocumented and experimental. If you were previously using
them at your own risk, note that "-spillover" is no longer an option (text
will always wrap inside the specified width if possible). In addition,
paragraphs(...) no longer collapses consecutive newlines.
- [RT #98546] There is now an -align-last option for justified text to
specify how the final line should be justified. Left is still the
default.
- Code in the PDF::API2::Basic::PDF namespace was originally licensed under
the Perl Artistic License, which has been determined to not be
(L)GPL-compatible. Martin Hosken has graciously granted an MIT license
for use of his code contained in PDF::API2, so the distribution may now be
used solely under the terms of the LGPL rather than requiring both
licenses simultaneously. Many thanks to Petr Pisar for his help in
working through the licensing issues.
- Renamed $text->lead(...) to $text->leading(...). The old name is
deprecated but will continue to work. If you're modifying PDF::API2
objects by editing their hash values instead of using these accessors,
you'll need to update your code.
- Added a -compress option (on by default) to PDF::API2->new(), open(), and
open_scalar(). If disabled, most streams won't be compressed, which can
simplify debugging.
The downstream configure code freaked out when info['define_macros'] was not set.
It is now defined and empty.
I hope it's fine to just push this without further notice.
Build was confirmed by oster on NetBSD-9.1/amd64 and tested by me
on Linux/amd64.
It seemed obvious to me that regen would create the files, so they
could be reviewed, and that one would then commit them manually. Warn
the next person...
Based on tech-pkg traffic over the last few years that appears to have
rough consensus, adjust MESSAGE section to explain that it should only
be used for exceptional circumstances, and specifically not to
substitute for documentation that belongs in the package's docdir.
Changes since 0.31:
Controller Improvements & Rumble Support
This release focuses on Controller Support & Improvements.
Please replace your old config.xml file when upgrading to this release.
* [Controllers] Basic rumble support can now be enabled from the CONTROLS
menu. The strength can be adjusted. The code that controls the rumble is
identical to that of the upright arcade machine motor shaker. The option
will only show in the menu if your controller is supported.
* [Controllers] Auto-detection and assignment of controller buttons and axis
should be much improved. I will update the manual to call out the default
controls in due course. (Right trigger = accelerate, Left trigger = brake,
Start = Start, Back = Menu, A = Gear Change, X = Change View, Y = Coin).
These can be redefined as desired.
* [Controllers] D-Pad can be used in menus.
* [Menus] Gamepad/Joystick settings has its own menu.
* [Menus] Ensure that 'A' button on controllers can be used to select items.
Analog accelerate is now only used when run via original arcade cabs as
it's not optimal.
* [Rom Loader] Sped up CRC32 based loading by using a Hashmap. It was
reportedly slow when used over a network.
* [Rom Loader] It is possible to switch back to the old filename loading
system via a setting in config.xml
* [Config] It is now possible to change and configure the location of the
'res' directory.
Addresses PR pkg/55972, report and patch provided by Greg A. Woods,
additional adjustments by me. (Several people have noted that there
really should be a Makefile.common used here, but I'm not going to
touch more than this leaf package right now, since we're in a freeze.)
This is a micro update that is mostly security fixes and bug fixes
with very small improvements. In addition to this being a security
fix, asterisk16 is a leaf package.
Upstream changes:
Security bugs fixed in this release:
-----------------------------------
* ASTERISK-29305 - ASTERISK-29203 / AST-2021-002 -- Another
scenario is causing a crash
(Reported by Gregory Massel)
* ASTERISK-29260 - sRTP Replay Protection ignored; even tears
down long calls
(Reported by Alexander Traud)
* ASTERISK-29227 - res_pjsip_diversion: sending multiple 181
responses causes memory corruption and crash
(Reported by
Ivan Poddubny)
Bugs fixed in this release:
-----------------------------------
* ASTERISK-29215 - res_pjsip_session: NULL active_media_state
topology caused asterisk crash
(Reported by sungtae kim)
* ASTERISK-29035 - chan_local: Multistream support breaks T.38
faxing
(Reported by Matthias Hensler)
* ASTERISK-29071 - app_confbridge: Memory rises when
jitterbuffer enabled and muting over AMI occurs
(Reported
by Stefan Ruf)
* ASTERISK-29329 - app_dial: DTMF to 'D' option gets duplicated
if there are multiple progress events
(Reported by N A)
* ASTERISK-24434 - Fix differing usage of assignment operators
in modules.conf
(Reported by Rusty Newton)
* ASTERISK-29306 - strings: Incorrect use of
__attribute__((pure)) in ast_str_to_lower definition
(Reported by Vitezslav Novy)
* ASTERISK-29300 - res_rtp_asterisk: When native local bridging
the remote SSRC becomes permanent
(Reported by Sebastian
Damm)
* ASTERISK-29235 - res_pjsip_nat: Contact is rewritten on
REGISTER responses with external_signaling_address
(Reported by Brian Paboojian)
* ASTERISK-29266 - ICE Role conflict with an unauthorized
session
(Reported by Salah Ahmed)
* ASTERISK-29105 - chan_pjsip: 180 Ringing with SDP not changed
into progress
(Reported by Sebastian Damm)
* ASTERISK-29297 - say: Y2021 problem – Asterisk cannot say
year 2021 in Dutch
(Reported by Jacek Konieczny)
* ASTERISK-29315 - res_pjsip: re-registration gets stuck if
setting initial auth credentials fails
(Reported by Nick
French)
* ASTERISK-29312 - res_fax: asterisk fails to publish the
Stasis and ReceiveFax status messages if the remote Station ID
contains invalid UTF-8 characters
(Reported by Alexei
Gradinari)
* ASTERISK-16799 - Callee declined when 'beep' audio file does
not exist
(Reported by IAMJames_)
* ASTERISK-29313 - res_pjsip_refer: Segfault in progress
notify
(Reported by George Joseph)
* ASTERISK-29293 - res_config_pgsql: Limit realtime_pgsql() to
return one (no more) record
(Reported by Boris P. Korzun)
* ASTERISK-29303 - pjsip: Re-invite occurs when it shouldn't
(Reported by Benjamin Keith Ford)
* ASTERISK-29311 - res_odbc_transaction sets forcecommit
default value based on isolation level instead of forcecommit
(Reported by Jaco Kroon)
* ASTERISK-28452 - pjsip: <sess-version> of SDP is not
incremented though SDP may be changed on reinvite without SDP
offer
(Reported by Michael Maier)
* ASTERISK-29287 - app.h: C++ compatibility broken
(Reported by Jean Aunis - Prescom)
* ASTERISK-28369 - app_queue: Member device state "invalid"
when second call is ringing and hint is used
(Reported by
Boolah )
* ASTERISK-29203 - res_pjsip_t38: Crash when changing state
(Reported by Gregory Massel)
* ASTERISK-29205 - res_rtp_asterisk: Asterisk crashes when
making hold/unhold from webrtc client
(Reported by Edvin
Vidmar)
* ASTERISK-29196 - res_pjsip: Segmentation fault
(Reported by Mauri de Souza Meneguzzo (3CPlus))
* ASTERISK-29280 - chan_sip: Allow peers without audio
(text+video).
(Reported by Alexander Traud)
* ASTERISK-29265 - chan_sip: Allow text+video media streams,
again.
(Reported by Alexander Traud)
* ASTERISK-29261 - res_pjsip: user=phone validation fail for
isup numbers containing *#
(Reported by Mark Petersen)
* ASTERISK-29259 - channel: Allow text+video media streams,
again.
(Reported by Alexander Traud)
* ASTERISK-29258 - chan_sip: Audio stream rejected, Other
stream present: Invalid SDP.
(Reported by Alexander Traud)
* ASTERISK-29220 - After T38 reinvite response of 488 a
subsequent G711 reinvite is not processed correctly. Instead the
previous T38 session media is used
(Reported by Robert
Cripps)
* ASTERISK-29248 - res_pjsip_session: res sometimes
uninitialized reported by compiler Clang.
(Reported by
Alexander Traud)
Improvements made in this release:
-----------------------------------
* ASTERISK-29321 - sorcery: Add support for more intelligent
reloading.
(Reported by Joshua C. Colp)
* ASTERISK-29325 - res_pjsip_registrar: Include source IP
address and port in log messages
(Reported by Joshua C.
Colp)
* ASTERISK-29326 - asterisk: Update copyright/company
(Reported by Joshua C. Colp)
* ASTERISK-29244 - Add MixMonitorStart / Stop / Mute AMI
events
(Reported by Sébastien Duthil)
* ASTERISK-29275 - Support of MIME-type for wav16
(Reported by Boris P. Korzun)
* ASTERISK-29252 - TRANSFERSTATUSPROTOCOL variable to report
Transfer (REFER) failure SIP code
(Reported by Dan Cropp)
* ASTERISK-29262 - Support of various URL-schemes by MoH
(Reported by Boris P. Korzun)
This approaches a serious performance issue with 1.26.x when
resampling is used on certain platforms (observed on SPARC).
Floating point denormals crop up. This change approaches that
two-fold
1. Do not disable default optimization flags for GCC.
2. Update to 1.26.5 that contains code to avoid denormals.
The default flags are about code optimization (like -ffast-math),
not the target architecture, so should not affect portability.
They are also rather safe and tested as development of mpg123
always has them in place. This is how it is meant to be played;-)
Approved during freeze by wiz.
Fix openblas build issues, mainly by updating to 0.3.10.
This pulls in these commits from WIP (newest first):
commit 3c6284cba90280bc367cf4d1d8252ae4d6e92e76
Author: Jason Bacon <bacon@NetBSD.org>
Date: Thu Feb 25 11:56:13 2021 -0600
openblas: Update ONLY_FOR_PLATFORMS documentation
commit 8071bf28f3ffc95af046ff3eaaac6983f4f70035
Author: Jason Bacon <bacon@NetBSD.org>
Date: Thu Feb 25 11:51:32 2021 -0600
openblas*: Successful build on NetBSD
commit 056e3d5c972a4b286e8755dbee323a9951855165
Author: Dr. Thomas Orgis <thomas.orgis@uni-hamburg.de>
Date: Wed Feb 24 18:40:17 2021 +0100
openblas: flags from environment again to un-break PICy build
The bug that prompted us to force the compiler flags in the
make arguments is fixed. Now we got a different one: The logic
that decides to add -fPIC where needed is overridden when doing
this, resulting in relocation errors (strangely, not with every
toolchain).
So let's remove that again and take FFLAGS and friends from the
environment again.
commit 86af17db8526e629c2c02c6af1f1ce7db6f6ba6d
Author: Dr. Thomas Orgis <thomas.orgis@uni-hamburg.de>
Date: Thu Nov 12 12:44:39 2020 +0100
openblas: version 0.3.10
This updated fixes the build with gcc 10 (segfault in cblat1 test).
I did not go all the way to the current 0.3.12, as that would
need some hacking of chosen compiler flags. 0.3.13 should be
the next one.
This commit also fixes the ARCH → ARCH_ sed to change all
occurences on a line.
This fulfills pkg/55999 and was approved by wiz during freeze.
The previous state included our BLAS choice but the build
did not honour the selected libs. This patches the config
script to not search for LAPACK and BLAS, using our settings
instead. Hopefull we can get that option upstreamed.
Approved during freeze by wiz.
The previous state included our BLAS choice but the build
did not honour the selected libs. This patches the config
script to not search for LAPACK and BLAS, using our settings
instead. Hopefull we can get that option upstreamed.
Approved during freeze by wiz.
