There are still some packages that fail the strict SUBST check. These
packages should nevertheless be built using the default pkgsrc
configuration, at least in the stable 2020Q2 branch. After 2020Q2 has
been switched, the strict SUBST checks will be activated again in the
default configuration.
Features
Add L3 switching features for USW-Pro series.
Improvements
Add support for MongoDB 3.6.
Auto-optimize enables Wi-Fi AI.
Prioritize busy times in Wi-Fi AI logic.
Update Hotspot 2.0 for WFA requirements.
Add VHT160 option for Japan.
Show client icons on DPI and statistics pages.
Improve SSID validation.
Improve support for USP Plug.
Allow adopting UBB only if both devices are connected.
Allow changing VHT mode for 5GHz radio on UBB.
Adjust speed test statistics to longer intervals.
Add LTS notifications.
Add Analytics & Improvements opt-in modal which is displayed post upgrade.
Add description of SNMPv3 security options.
Allow sending test emails via Cloud Email service.
Make timezones dropdown searchable.
Update IANA timezone database.
Update inter-VLAN routing Network form.
Remove OP mode from Port Profiles.
Add MAC column to Insights/Known Clients.
Update available IDS/IPS categories to lessen false positives and improve resource use.
Rename UDM-Beacon to UAP-BeaconHD.
Improve Property Panel for USW Leaf.
Handle LCM sync event notifications on UDM.
Support U-LTE on UDM devices.
Support WAN failover for UDM devices.
Add UDM boot sound checkbox.
Various fixes and improvements in Beta Settings.
Various performance improvements.
Update translations and add missing events.
Known issues
Don't remind me button for LTS notifications doesn't function properly.
Blocking countries in map is not working, workaround is using the classic settings.
Incorrect DPI Statistics.
Version 0.50.0:
This is a more usual release in comparison to the others that have been made in the last six months. It comprises the result of a number of maintenance tasks along with some new features and a lot of bug fixes.
Highlights of core feature changes include:
The compilation chain is now based on LLVM 9.
The error handling and reporting system has been improved to reduce the size of error messages, and also improve quality and specificity.
The CUDA target has more stream constructors available and a new function for compiling to PTX without linking and loading the code to a device. Further, the macro-based system for describing CUDA threads and blocks has been replaced with standard typing and lowering implementations, for improved debugging and extensibility.
v0.33.0:
This release upgrades to LLVM 9 and drops support for older LLVM versions.
Pull requests:
* Fix CUDA with LLVM9
* Fix meta.yaml
* buildscripts: Unpin wheel
* add python_requires to setup.py
* Adds override for LLVM version check, re-formats docs.
* Add FAQ entry on LLVM version support.
* Trove classifiers may be out of date.
* llvmlite wheel building fixes
* Update the release date
* Upgrade to LLVM9
* Allow instructions to be removed from blocks
4.12.1:
- [_n_a_m_e] Fixed error in ``addMultilingualName`` with one-character names.
Only attempt to recovered malformed UTF-16 data from a ``bytes`` string,
not from unicode ``str``.
4.12.0:
- [otlLib/varLib] Ensure that the ``AxisNameID`` in the ``STAT`` and ``fvar``
tables is grater than 255 as per OpenType spec.
- [docs] Document more modules in ``fontTools.misc`` package: ``filenames``,
``fixedTools``, ``intTools``, ``loggingTools``, ``macCreatorType``, ``macRes``,
``plistlib``.
- [OS/2] Don't calculate whole sets of unicode codepoints, use faster and more memory
efficient ranges and bisect lookups.
- [voltLib] Support writing back abstract syntax tree as VOLT data.
- [voltLib] Accept DO_NOT_TOUCH_CMAP keyword.
- [subset/merge] Fixed a namespace clash involving a private helper class.
4.11.0:
- [feaLib] Introduced ``includeDir`` parameter on Parser and IncludingLexer to
explicitly specify the directory to search when ``include()`` statements are
encountered.
- [ufoLib] Silently delete duplicate glyphs within the same kerning group when reading
groups.
- [ttLib] Set version of COLR table when decompiling COLRv1 (commit 9d8a7e2).
4.10.2:
- [sfnt] Fixed ``NameError: SimpleNamespace`` while reading TTC header. The regression
was introduced with 4.10.1 after removing ``py23`` star import.
4.10.1:
- [sfnt] Make ``SFNTReader`` pickleable even when TTFont is loaded with lazy=True
option and thus keeps a reference to an external file.
- [feaLib.ast] Restore backward compatibility (broken in 4.10 for
``ChainContextPosStatement`` and ``ChainContextSubstStatement`` classes.
Make them accept either list of lookups or list of lists of lookups.
- [docs] Document some modules in ``fontTools.misc`` package: ``arrayTools``,
``bezierTools`` ``cliTools`` and ``eexec``.
- [ttLib._n_a_m_e] Fixed ``findMultilingualName()`` when name record's ``string`` is
encoded as bytes sequence.
4.10.0:
- [varLib] Allow feature variations to be active across the entire space.
- [ufoLib] Added support for ``formatVersionMinor`` in UFO's ``fontinfo.plist`` and for
``formatMinor`` attribute in GLIF file as discussed in unified-font-object/ufo-spec#78.
No changes in reading or writing UFOs until an upcoming (non-0) minor update of the
UFO specification is published.
- [merge] Fixed merging fonts with different versions of ``OS/2`` table.
- [subset] Fixed ``AttributeError`` while subsetting ``ContextSubst`` and ``ContextPos``
Format 3 subtable.
- [ttLib.table._m_e_t_a] if data happens to be ascii, emit comment in TTX.
- [feaLib] Support multiple lookups per glyph position.
- [psCharStrings] Use inheritance to avoid repeated code in initializer.
- [Doc] Improved documentation for the following modules: ``afmLib``, ``agl``
, ``cffLib``, ``cu2qu``, ``encodings``, ``feaLib``, ``merge``.
- [Doc] Split off developer-centric info to new page, making front page of docs more
user-focused. List all utilities and sub-modules with brief descriptions.
Make README more concise and focused.
- [otlLib] Add function to build STAT table from high-level description.
- [ttLib._n_a_m_e] Add ``findMultilingualName()`` method.
- [unicodedata] Update ``RTL_SCRIPTS`` for Unicode 13.0.
- [gvar] Sort ``gvar`` XML output by glyph name, not glyph order.
- [Doc] Added help options to ``fonttools`` command line tool.
Ensure all fonttools CLI tools have help documentation.
- [ufoLib] Only write fontinfo.plist when there actually is content.
4.9.0:
- [subset] Fixed subsetting of FeatureVariations table. The subsetter no longer drops
FeatureVariationRecords that have empty substitutions as that will keep the search
going and thus change the logic. It will only drop empty records that occur at the
end of the FeatureVariationRecords array.
- [subset] Remove FeatureVariations table and downgrade GSUB/GPOS to version 0x10000
when FeatureVariations contain no FeatureVariationRecords after subsetting.
- [agl] Add support for legacy Adobe Glyph List of glyph names in ``fontTools.agl``
.
- [feaLib] Ignore superfluous script statements.
- [feaLib] Hide traceback by default on ``fonttools feaLib`` command line.
Use ``--traceback`` option to show.
- [feaLib] Check lookup index in chaining sub/pos lookups and print better error
message.
- [feaLib] Fix building chained alt substitutions.
- [Doc] Included all fontTools modules in the sphinx-generated documentation, and
published it to ReadTheDocs for continuous documentation of the fontTools project
. Check it out at https://fonttools.readthedocs.io/. Thanks to Chris Simpkins!
- [transform] The ``Transform`` class is now subclass of ``typing.NamedTuple``. No
change in functionality.
Release 2.2.6:
By John A. Magliacane <kd2bd@amsat.org> (20-May-2020):
* The installer program has been replaced by a bash script that
configures, compiles, and installs PREDICT and several companion
client utility applications.
* The SingleTrack and MultiTrack displays have been modified for
more consistent performance across a wider range of terminals.
* PREDICT's "vocalizer" now employs the Advanced Linux Sound
Architecture (ALSA) environment instead of OSS in response
to recent changes in v5.6 and later Linux kernels.
* The handling of geosynchronous and geostationary satellites
has been improved, but still needs to be used with caution.
* 100 MHz normalized downlink Doppler shift information has been
added to all -p and -f command-line outputs.
* The documentation and the transponder database file have been
updated, and the format of the man page has been corrected.
MMDAgent voice data contains three voice data, mei, slt and takumi.
Voice data of mei and slt are female and takumi is male.
The work is released as a part of MMDAgent (http://www.mmdagent.jp/).
HTS voice trained by using the Nitech Japanese Speech Database "NIT
ATR503 M001" is released as a part of Open JTalk
(http://open-jtalk.sourceforge.net/). This voice consists of HMMs
trained by using HMM-based Speech Synthesis System (HTS) version
2.3 alpha (http://hts.sp.nitech.ac.jp/).
The Open JTalk is a Japanese TTS System. It has been being developed
by HTS working group and some graduate students in Nagoya Institute
of Technology.
Release 2.2.5:
By John A. Magliacane <kd2bd@amsat.org> (05-May-2018):
* Fixed coding bugs identified by GCC v8.1.0.
* Re-arranged GCC arguments in installer.c for better
compatibility with more recent versions of GCC.
* Added a -dp command-line option to PREDICT to
provide quick Doppler shift estimates in CSV format.
(Thanks Joeal Subash)
* Updated the documentation to reflect the changes.
----------------------------------------------------------------------------
Release 2.2.4:
By John A. Magliacane <kd2bd@amsat.org> (12-Mar-2018):
* Fixed a bug affecting the -f switch that sometimes caused
a failure to return proper output. (Tnx: Andrej Lajovic)
* Fixed a bug in installer.c that sometimes caused a failure
to properly install PREDICT's executable into /usr/local/bin
by default.
* Added <errno.h> header to gsat-1.1.0/src/callbacks.c
----------------------------------------------------------------------------
XCPC is a portable Amstrad CPC 464/664/6128 emulator written in C.
XCPC is designed to run on any POSIX compliant system, including
Unix, Linux, BSD and having an X11 server.
Passphrase is a zero-dependency passphrase generator.
It is pretty fast (over 6000 times faster than
bitwarden's "bw generate -p",) and you can use any wordlist.
An English wordlist is included, and is the default.
AnsiWeather 1.17.0 (2020-06-16)
- Use cut instead of awk for parsing the config file, allowing to use the
colon character as part of the config value (Thanks Niklas Lindgren)
qutebrowser is a keyboard-focused browser with a minimal GUI. Its based
on Python and PyQt5 and free software, licensed under the GPL.
It was inspired by other browsers/addons like dwb and Vimperator/Pentadactyl.
