py-ipywidgets: updated to 7.6.3

7.6
---

To see the full list of pull requests and issues, see the [7.6.0 milestone](https://github.com/jupyter-widgets/ipywidgets/milestone/31?closed=1) on GitHub.

The main change in this release is that installing `ipywidgets` 7.6.0 will now automatically enable ipywidgets support in JupyterLab 3.0窶蚤 user has no extra JupyterLab installation step and no rebuild of JupyterLab, nor do they need Node.js installed. Simply install the python ipywidgets package with pip (`pip install ipywidgets==7.6.0`) or conda/mamba (`conda install -c conda-forge ipywidgets=7.6.0`) and ipywidgets will automatically work in classic Jupyter Notebook and in JupyterLab 3.0.

This is accomplished with the new python package `jupyterlab_widgets` version 1.0, on which `ipywidgets` 7.6.0 now depends (similar to how `ipywidgets` already depends on the `widgetsnbextension` package to configure ipywidgets for the classic Jupyter Notebook). The `jupyterlab_widgets` Python package is a JupyterLab 3.0 prebuilt extension, meaning that it can be installed into JupyterLab 3.0 without rebuilding JupyterLab and without needing Node.js installed.

Updates for Widget Maintainers

Custom widget maintainers will need to make two changes to update for JupyterLab 3:

1. Update the `@jupyter-widgets/base` dependency version to include `^4` to work in JupyterLab 3.0. For example, if you had a dependency on `@jupyter-widgets/base` version `^2 || ^3`, update to `^2 || ^3 || ^4` for your widget to work in classic Jupyter Notebook, JupyterLab 1, JupyterLab 2, and JupyterLab 3.
2. In the `package.json`, add the following `sharedPackages` configuration inside the `jupyterlab` key. See the [JupyterLab extension documentation](https://jupyterlab.readthedocs.io/en/stable/extension/extension_dev.html#requiring-a-service) for more information.

  ```json
    "jupyterlab": {
      "sharedPackages": {
        "@jupyter-widgets/base": {
          "bundled": false,
          "singleton": true
        }
      }
    }
  ```

Separate from these two steps to update for JupyterLab 3, we also recommend that you make your widget's JupyterLab extension a prebuilt extension for JupyterLab 3.0. Users will be able to install your JupyterLab 3.0 prebuilt extension without rebuilding JupyterLab or needing Node.js. See the [JupyterLab 3 extension developer documentation](https://jupyterlab.readthedocs.io/en/stable/extension/extension_dev.html) or the new [widget extension cookiecutter](https://github.com/jupyter-widgets/widget-ts-cookiecutter/tree/jlab3) for more details.

(adam)

py-nbformat: updated to 5.1.3

5.1.3
=====
- Change id generation to be hash based to avoid problematic word combinations
- Added tests for python 3.9
- Fixed setup.py build operations to include package data

5.1.2
=====
- Fixed missing file in manifest

5.1.1
=====
- Changes convert.upgrade to upgrade minor 4.x versions to 4.5

5.1.0
=====
- Implemented CellIds from JEP-62
- Fixed a regression introduced when using fastjsonschema,
  which does not directly support to validate a "reference"/"subschema"
- Removed unreachable/unneeded code
- Added CI workflow for package release on tag push

(adam)

Updated devel/py-ipython, net/py-softlayer, devel/py-jupyter_core, devel/py-jupyter_client, devel/py-ipykernel, devel/py-jupyter-console

(adam)

py-jupyter-console: updated to 6.4.0

6.4.0:
Unknown changes

(adam)

py-ipykernel: updated to 5.5.5

5.5.5
-----
- Keep preferring SelectorEventLoop on Windows. (:ghpull:`669`)

5.5.4
-----
- Import ``configure_inline_support`` from ``matplotlib_inline`` if available (:ghpull:`654`)

5.5.3
-----
- Revert Backport of 605: Fix Handling of ``shell.should_run_async`` (:ghpull:`622`)

5.5.2
-----
**Note:** This release was deleted from PyPI since it had breaking changes.

- Changed default timeout to 0.0 seconds for stop_on_error_timeout. (:ghpull:`618`)

5.5.1
-----
**Note:** This release was deleted from PyPI since it had breaking changes.

- Fix Handling of ``shell.should_run_async``. (:ghpull:`605`)

5.5.0
-----
- Kernelspec: ensure path is writable before writing ``kernel.json``. (:ghpull:`593`)
- Add ``configure_inline_support`` and call it in the shell. (:ghpull:`590`)

(adam)

py-jupyter_client: updated to 6.1.12

6.1.11
======
- Move jedi pinning to test requirements (:ghpull:`599`)

6.1.10
======
- Add change parameter needed for observer method of kernel_spec_manager trait (:ghpull:`598`)

6.1.9
=====
- Pin jedi<=0.17.2 (:ghpull:`596`)

6.1.8
=====
- Doc updates (:ghpull:`563`, :ghpull:`564`, :ghpull:`587`)
- Fix path to the connection file (:ghpull:`568`)
- Code cleanup (:ghpull:`574`, :ghpull:`579`)
- Silence kill_kernel when no process is present (:ghpull:`576`)
- Remove extra_env and corresponding test (:ghpull:`581`)
- Add documentation dependencies to setup.py (:ghpull:`582`)
- Fix for Windows localhost IP addresses (:ghpull:`584`)
- Drop Travis CI, add GitHub Actions (:ghpull:`586`)
- Adapt KernelManager._kernel_spec_manager_changed to observe (:ghpull:`588`)
- Allow use ~/ in the kernel's command or its arguments (:ghpull:`589`)
- Change wait_for_ready logic (:ghpull:`592`)
- Fix test_session with msgpack v1  (:ghpull:`594`)

(adam)

py-jupyter_core: updated to 4.7.1

4.7.1

- Allow creating user to delete secure file (:ghpull:`213`)

4.7.0

- Add a new ``JUPYTER_PREFER_ENV_PATH`` variable, which can be set to switch the order of the environment-level path and the user-level path in the Jupyter path hierarchy (e.g., ``jupyter --paths``). It is considered set if it is a value that is not one of 'no', 'n', 'off', 'false', '0', or '0.0' (case insensitive). If you are running Jupyter in multiple virtual environments as the same user, you will likely want to set this environment variable.
- Drop Python 2.x and 3.5 support, as they have reached end of life.
- Add Python 3.9 builds to testing, and expand testing to cover Windows, macOS, and Linux platforms.
- ``jupyter --paths --debug`` now explains the environment variables that affect the current path list.
- Update the file hidden check on Windows to use new Python features rather than ctypes directly.
- Add conda environment information in ``jupyter troubleshoot``.
- Update ``_version.version_info`` and ``_version.__version__`` to follow Python conventions.

(adam)

xonsh: py-prompt_toolkit has moved

(adam)

py-softlayer: updated to 5.9.5

[5.9.5] - 2021-05-25

Improvements

Changed a testing domain to one that really doesnt exist
Fix Incomplete notes field for file and block
Show component versions on hw detail
Add the firewall information on slcli firewall detail
Add an --orderBy parameters to call-api
Add image detail transaction data

[5.9.4] - 2021-04-27

New Commands

slcli hw authorize-storage
slcli order quote-save
Improvements

Refactored managers.ordering_manager.verify_quote() to work better with the REST endpoing
Add routers for each DC in slcli hw create-options
Add preset datatype in slcli virtual detail
Add upgrade option to slcli hw.
Ibmcloud authentication support
slcli config setup --ibmid
slcli config setup --sso
slcli config setup --cloud_key
slcli config setup --classic_key
Refactor slcli hw detail prices.
Updated contributing guide
Add the Hardware components on "slcli hardware detail"
Add billing and lastTransaction on hardware detail
Forced reserved capacity guests to be monthly
Removing the rwhois commands
Added automation to publish to test-pypi
Updating author_email to SLDN distro list
Add the option to add and upgrade the hw disk.
Added a utility to merge objectFilters,
Fixes shift+ins when pasteing into a password field for windows users.
Add Billing and lastTransaction on slcli virtual detail
Fixing 'import mock' pylint issues

(adam)

py-click-repl: py-prompt_toolkit has moved

(adam)

py-ipython: updated to 7.24.1

IPython 7.24
============

Third release of IPython for 2021, mostly containing bug fixes. A couple of not
typical updates:

Misc
----

- Fix an issue where ``%recall`` would both succeeded and print an error message
  it failed. :ghpull:`12952`
- Drop support for NumPy 1.16 – practically has no effect beyond indicating in
  package metadata that we do not support it. :ghpull:`12937`

Debugger improvements
---------------------

The debugger (and ``%debug`` magic) have been improved and can skip or hide frames
originating from files that are not writable to the user, as these are less
likely to be the source of errors, or be part of system files this can be a useful
addition when debugging long errors.

In addition to the global ``skip_hidden True|False`` command, the debugger has
gained finer grained control of predicates as to whether to a frame should be
considered hidden. So far 3 predicates are available :

  - ``tbhide``: frames containing the local variable ``__tracebackhide__`` set to
    True.
  - ``readonly``: frames originating from readonly files, set to False.
  - ``ipython_internal``: frames that are likely to be from IPython internal
    code, set to True.

You can toggle individual predicates during a session with

.. code-block::

  ipdb> skip_predicates readonly True

Read-only files will now be considered hidden frames.

You can call ``skip_predicates`` without arguments to see the states of current
predicates:

.. code-block::

    ipdb> skip_predicates
    current predicates:
        tbhide : True
        readonly : False
        ipython_internal : True

If all predicates are set to ``False``,  ``skip_hidden`` will practically have
no effect. We attempt to warn you when all predicates are False.

Note that the ``readonly`` predicate may increase disk access as we check for
file access permission for all frames on many command invocation, but is usually
cached by operating systems. Let us know if you encounter any issues.

As the IPython debugger does not use the traitlets infrastructure for
configuration, by editing your ``.pdbrc`` files and appending commands you would
like to be executed just before entering the interactive prompt. For example:

.. code::

    # file : ~/.pdbrc
    skip_predicates readonly True
    skip_predicates tbhide False

Will hide read only frames by default and show frames marked with
``__tracebackhide__``.

(adam)

py-prompt_toolkit: moved version 3.0.18 from py-prompt_toolkit2

(adam)

py-ipython5: depend on appnope on Darwin, fix ALTERNATIVES

(adam)

Moved devel/py-prompt_toolkit, devel/py-traitlets; Updated devel/py-prompt_toolkit1, devel/py-traitlets, devel/py-ipython5

(adam)

py-ipython5: updated to 5.10.0

IPython 5.10.0
==============
IPython 5.10 pins Pygments dependencies to version lower than ``2.6``, as it is
not compatible Python 2. See :ghpull:`12174`

IPython 5.9.0
=============
IPython 5.9 is a release that accumulated fixes for Python 2 over the past 18
month. Please see source and version control history for more informations.

(adam)

py-traitlets: updated to 5.0.5

5.0.5
*****
- Support deprecated literals for sets, tuples on the command-line: ``nbconvert --TagRemovePreprocessor.remove_cell_tags='{"tag"}'``
- Fix ``from_string_list`` for Tuples in general
- Fix support for ``List(default_value=None, allow_none=True)`` and other Container traits
- Fix help output for nested aliases and tuple traits

5.0.4
*****
- Support deprecated use of byte-literals for bytes on the command-line: ``ipython kernel --Session.key="b'abc'"``. The `b` prefix is no longer needed in traitlets 5.0, but is supported for backward-compatibility
- Improve output of configuration errors, especially when help output would make it hard to find the helpful error message

5.0.3
*****
- Fix regression in handling `--opt=None` on the CLI for configurable traits
  with `allow_none=True`

5.0.2
*****
- Fix casting bytes to unicode

5.0.0
*****
Traitlets 5.0 is a new version of traitlets that accumulate changes over a period of more close to four years; A number of
internal refactoring made the internal code structure cleaner and simpler, and greatly improved the diagnostic error
messages as well has help and documentation generation.

(adam)

moved py-traitlets to py-traitlets4

(adam)

py-prompt_toolkit1: added version 1.0.18

1.0.18: 2019-10-03
Fixes:
- The 1.0.17 build contained files from the 2.0 branch and was broken because
  of this.

1.0.17: 2019-10-03
Bug fixes:
- Handle HANDLE sizes correctly on windows. This made things break randomly on
  64 bit systems.

1.0.16: 2019-04-12
Fixes:
- Fixes DeprecationWarning when trying to import collections.Mapping.

(adam)

Updated print/ghostscript-agpl, devel/py-cachelib

(adam)

py-cachelib: updated to 0.1.1

Version 0.1.1

Fix FileSystemCache on Windows

(adam)

ghostscript-agpl: updated to 9.54.0

The 9.54.0 release is a maintenance release, and also adds new functionality.

Highlights in this release include:

Overprint simulation is now available to all output devices, allowing quality previewing/proofing of PostScript and PDF jobs that rely on overprint. See the -dOverprint option documentation in: Overprint

The "docxwrite" device adds the ability to output to Microsoft Word "docx" format. See: docxwrite

The pdfwrite device is now capable of using the Tesseract OCR engine when it is built into Ghostscript to improve searchability and copy and paste functionality when the input lacks the metadata for that purpose. See: UseOCR

Ghostscript/GhostPDL now includes a "map text to black" function, where text drawn by an input job (except when drawn using a Type 3 font) can be forced to draw in solid black. See: BlackText

Ghostscript/GhostPDL now supports simple N-up imposition "internally". See: NupControl

Our efforts in code hygiene and maintainability continue.

The usual round of bug fixes, compatibility changes, and incremental improvements.

(9.53.0) We have added the capability to build with the Tesseract OCR engine. In such a build, new devices are available (pdfocr8/pdfocr24/pdfocr32) which render the output file to an image, OCR that image, and output the image "wrapped" up as a PDF file, with the OCR generated text information included as "invisible" text (in PDF terms, text rendering mode 3).

Mainly due to time constraints, we only support including Tesseract from source included in our release packages, and not linking to Tesseract/Leptonica shared libraries. Whether we add this capability will be largely dependent on community demand for the feature.

(adam)

Updated net/rabbitmq, devel/py-fastimport

(adam)

py-fastimport: updated to 0.9.13

0.9.13:
Unknown changes

(adam)

rabbitmq: updated to 3.8.17

RabbitMQ 3.8.17

RabbitMQ 3.8.17 is a maintenance release that includes a security patch.

Security Patches

This release addresses an undisclosed vulnerability with CVSS 3.1 score of 2.4 (low).
This section will be updated when the details are made public.

(adam)

rust: fix patch for src/tools/rust-installer/install-template.sh

Better patch for src/tools/rust-installer/install-template.sh
- remove one path component after share/doc/ as the original code does
- do not change spacing for HACK: section

(adam)

Updated devel/protobuf, devel/py-protobuf

(adam)

protobuf py-protobuf: updated to 3.17.3

Protocol Buffers v3.17.3

C++

Introduce FieldAccessListener.
Stop emitting boilerplate {Copy/Merge}From in each ProtoBuf class
Fixed some uninitialized variable warnings in generated_message_reflection.cc.

(adam)

Updated security/nettle, devel/meson

(adam)

meson: updated to 0.58.1

0.58.1:
Bug fixes

(adam)

nettle: updated to 3.7.3

NEWS for the Nettle 3.7.3 release

This is bugfix release, fixing bugs that could make the RSA
decryption functions crash on invalid inputs.

Upgrading to the new version is strongly recommended. For
applications that want to support older versions of Nettle,
the bug can be worked around by adding a check that the RSA
ciphertext is in the range 0 < ciphertext < n, before
attempting to decrypt it.

Thanks to Paul Schaub and Justus Winter for reporting these
problems.

The new version is intended to be fully source and binary
compatible with Nettle-3.6. The shared library names are
libnettle.so.8.4 and libhogweed.so.6.4, with sonames
libnettle.so.8 and libhogweed.so.6.

Bug fixes:

* Fix crash for zero input to rsa_sec_decrypt and
  rsa_decrypt_tr. Potential denial of service vector.

* Ensure that all of rsa_decrypt_tr and rsa_sec_decrypt return
  failure for out of range inputs, instead of either crashing,
  or silently reducing input modulo n. Potential denial of
  service vector.

* Ensure that rsa_decrypt returns failure for out of range
  inputs, instead of silently reducing input modulo n.

* Ensure that rsa_sec_decrypt returns failure if the message
  size is too large for the given key. Unlike the other bugs,
  this would typically be triggered by invalid local
  configuration, rather than by processing untrusted remote
  data.

(adam)

openblas: fix building on Darwin

(adam)

openblas: fix building on Darwin

(adam)

Updated devel/hdf5, devel/hdf5-c++, devel/py-pkgconfig, devel/py-h5py

(adam)

py-Keras-Applications: mark as incompatible with Python < 3.7

(adam)

py-h5py: updated to 3.2.1

What's new in h5py 3.2
======================

New features
------------

* Added support to use the HDF5 ROS3 driver to access HDF5 files on S3
  (:pr:`1755`). This is not enabled in the pre-built packages on PyPI. To use
  it, ensure HDF5 is built with read-only S3 support enabled, and then
  :ref:`build h5py from source <source_install>` using that HDF5 library.

Breaking changes & deprecations
-------------------------------

* Python 3.7 is now the minimum supported version. It may still be possible to
  use this release with Python 3.6, but it isn't tested and wheels are not
  provided for Python 3.6.
* Setting the config option ``default_file_mode`` to values other than ``'r'``
  is deprecated. Pass the desired mode when opening a :class:`~.File` instead.

Exposing HDF5 functions
-----------------------

* ``H5Pset_fapl_ros3`` & ``H5Pget_fapl_ros3`` (where HDF5 is built with
  read-only S3 support).

Bug fixes
---------

* :exc:`OSError` exceptions raised by h5py should now have a useful ``.errno``
  attribute, where HDF5 provides this information. Subclasses such as
  :exc:`FileNotFoundError` should also be raised where appropriate (:pr:`1815`).
* Fix reading data with a datatype of variable-length arrays of fixed length
  strings (:issue:`1817`).
* Fix :meth:`.Dataset.read_direct` and :meth:`.Dataset.write_direct` when the
  source and destination have different shapes (:pr:`1796`).
* Fix selecting data using integer indices in :meth:`.Dataset.read_direct`
  and :meth:`.Dataset.write_direct` (:pr:`1818`).
* Fix exception handling in :meth:`.Group.visititems` (:issue:`1740`).
* Issue a warning when ``File(..., swmr=True)`` is specified with any mode other
  than ``'r'``, as the SWMR option is ignored in these cases (:pr:`1812`).
* Fix NumPy 1.20 deprecation warnings concerning the use of None as shape, and
  the deprecated aliases np.float, np.int and np.bool (:pr:`1780`).

3.2.1 bug fix release
---------------------

* Fix :attr:`.File.driver` when the read-only S3 driver is available (:pr:`1844`).

What's new in h5py 3.1
======================

Bug fixes
---------

* Fix reading numeric data which is not in the native endianness,
  e.g. big-endian data on a little-endian system (:issue:`1729`).
* Fix using bytes as names for :meth:`.Group.create_dataset` and
  :meth:`.Group.create_virtual_dataset` (:issue:`1732`).
* Fix writing data as a list to a dataset with a sub-array data type
  (:issue:`1735`).

Building h5py
-------------

* Allow building against system lzf library by setting ``H5PY_SYSTEM_LZF=1``.
  See :ref:`custom_install`.

Development
-----------

* If pytest is missing `pytest-mpi <https://pytest-mpi.readthedocs.io/en/latest/>`_
  it will now fail with a clear error.
* Fix a test which was failing on big-endian systems.

What's new in h5py 3.0
======================

New features
------------

* The interface for storing & reading strings has changed - see :doc:`/strings`.
  The new rules are hopefully more consistent, but may well require some changes
  in coding using h5py.
* Reading & writing data now releases the GIL, so another Python thread can
  continue while HDF5 accesses data. Where HDF5 can call back into Python, such
  as for data conversion, h5py re-acquires the GIL. However, HDF5 has its own
  global lock, so this won't speed up parallel data access using multithreading.
* Numpy datetime and timedelta arrays can now be stored and read as HDF5
  opaque data (:issue:`1339`), though other tools will not understand them.
  See :ref:`opaque_dtypes` for more information.
* New :meth:`.Dataset.iter_chunks` method, to iterate over chunks within the
  given selection.
* Compatibility with HDF5 1.12.
* Methods which accept a shape tuple, e.g. to create a dataset, now also allow
  an integer for a 1D shape (:pr:`1340`).
* Casting data to a specified type on reading (:meth:`.Dataset.astype`) can now
  be done without a with statement, like this::

      data = dset.astype(np.int32)[:]

* A new :meth:`.Dataset.fields` method lets you read only selected fields from
  a dataset with a compound datatype.
* Reading data has less overhead, as selection has been implemented in Cython.
  Making many small reads from the same dataset can be as much as 10 times
  faster, but there are many factors that can affect performance.
* A new NumPy-style :attr:`.Dataset.nbytes` attribute to get the size of the
  dataset's data in bytes.  This differs from the :attr:`~.Dataset.size`
  attribute, which gives the number of elements.
* The ``external`` argument of :meth:`.Group.create_dataset`, which
  specifies any external storage for the dataset, accepts more types
  (:issue:`1260`), as follows:

  * The top-level container may be any iterable, not only a list.
  * The names of external files may be not only ``str`` but also ``bytes`` or
    ``os.PathLike`` objects.
  * The offsets and sizes may be NumPy integers as well as Python integers.

  See also the deprecation related to the ``external`` argument.
* Support for setting file space strategy at file creation. Includes option to
  persist empty space tracking between sessions. See :class:`.File` for details.
* More efficient writing when assiging a scalar to a chunked dataset, when the
  number of elements to write is no more than the size of one chunk.
* Introduced support for the split :ref:`file driver <file_driver>`
  (:pr:`1468`).
* Allow making virtual datasets which can grow as the source data is resized
  - see :doc:`/vds`.
* New `allow_unknown_filter` option to :meth:`.Group.create_dataset`. This should
  only be used if you will compress the data before writing it with the
  low-level :meth:`~h5py.h5d.DatasetID.write_direct_chunk` method.
* The low-level chunk query API provides information about dataset chunks in an
  HDF5 file: :meth:`~h5py.h5d.DatasetID.get_num_chunks`,
  :meth:`~h5py.h5d.DatasetID.get_chunk_info` and
  :meth:`~h5py.h5d.DatasetID.get_chunk_info_by_coord`.
* The low-level :meth:`h5py.h5f.FileID.get_vfd_handle` method now works for any
  file driver that supports it, not only the sec2 driver.

Breaking changes & deprecations
-------------------------------

* h5py now requires Python 3.6 or above; it is no longer compatible with Python
  2.7.
* The default mode for opening files is now 'r' (read-only).
  See :ref:`file_open` for other possible modes if you need to write to a file.
* In previous versions, creating a dataset from a list of bytes objects would
  choose a fixed length string datatype to fit the biggest item. It will now
  use a variable length string datatype. To store fixed length strings, use a
  suitable dtype from :func:`h5py.string_dtype`.
* Variable-length UTF-8 strings in datasets are now read as ``bytes`` objects
  instead of ``str`` by default, for consistency with other kinds of strings.
  See :doc:`/strings` for more details.
* When making a virtual dataset, a dtype must be specified in
  :class:`.VirtualLayout`. There is no longer a default dtype, as this was
  surprising in some cases.
* The ``external`` argument of :meth:`Group.create_dataset` no longer accepts
  the following forms (:issue:`1260`):

  * a list containing *name*, [*offset*, [*size*]];
  * a list containing *name1*, *name2*, …; and
  * a list containing tuples such as ``(name,)`` and ``(name, offset)`` that
    lack the offset or size.

  Furthermore, each *name*–*offset*–*size* triplet now must be a tuple rather
  than an arbitrary iterable.  See also the new feature related to the
  ``external`` argument.
* The MPI mode no longer supports mpi4py 1.x.
* The deprecated ``h5py.h5t.available_ftypes`` dictionary was removed.
* The deprecated ``Dataset.value`` property was removed.
  Use ``ds[()]`` to read all data from any dataset.
* The deprecated functions ``new_vlen``, ``new_enum``, ``get_vlen`` and
  ``get_enum`` have been removed. See :doc:`/special` for the newer APIs.
* Removed deprecated File.fid attribute. Use :attr:`.File.id` instead.
* Remove the deprecated ``h5py.highlevel`` module.
  The high-level API is available directly in the ``h5py`` module.
* The third argument of ``h5py._hl.selections.select()`` is now an optional
  high-level :class:`.Dataset` object, rather than a ``DatasetID``.
  This is not really a public API - it has to be imported through the private
  ``_hl`` module - but probably some people are using it anyway.

Exposing HDF5 functions
-----------------------

* H5Dget_num_chunks
* H5Dget_chunk_info
* H5Dget_chunk_info_by_coord
* H5Oget_info1
* H5Oget_info_by_name1
* H5Oget_info_by_idx1
* H5Ovisit1
* H5Ovisit_by_name1
* H5Pset_attr_phase_change
* H5Pset_fapl_split
* H5Pget_file_space_strategy
* H5Pset_file_space_strategy
* H5Sencode1
* H5Tget_create_plist

Bug fixes
---------

* Fix segmentation fault when accessing vlen of strings (:issue:`1336`).
* Fix the storage of non-contiguous arrays, such as numpy slices, as HDF5 vlen
  data (:issue:`1649`).
* Fix pathologically slow reading/writing in certain conditions with integer
  indexing (:issue:`492`).
* Fix bug when :meth:`.Group.copy` source is a high-level object and destination
  is a Group (:issue:`1005`).
* Fix reading data for region references pointing to an empty selection.
* Unregister converter functions at exit, preventing segfaults on exit in some
  situations with threads (:pr:`1440`).
* As HDF5 1.10.6 and later support UTF-8 paths on Windows, h5py built against
  HDF5 1.10.6 will use UTF-8 for file names, allowing all filenames.
* Fixed :meth:`h5py.h5d.DatasetID.get_storage_size` to report storage size of
  zero bytes without raising an exception (:issue:`1475`).
* Attribute Managers (``obj.attrs``) can now work on HDF5 stored
  datatypes (:issue:`1476`).
* Remove broken inherited ``ds.dims.values()`` and ``ds.dims.items()`` methods.
  The dimensions interface behaves as a sequence, not a mapping (:issue:`744`).
* Fix creating attribute with :class:`.Empty` by converting its dtype to a numpy
  dtype object.
* Fix getting :attr:`~.Dataset.maxshape` on empty/null datasets.
* The :attr:`.File.swmr_mode` property is always available (:issue:`1580`).
* The :attr:`.File.mode` property handles SWMR access modes in addition to plain
  RDONLY/RDWR modes
* Importing an MPI build of h5py no longer initialises MPI immediately,
  which will hopefully avoid various strange behaviours.
* Avoid launching a subprocess by using ``platform.machine()`` at import time.
  This could trigger a warning in MPI.
* Removed an equality comparison with an empty array, which will cause problems
  with future versions of numpy.
* Better error message if you try to use the mpio driver and h5py was not built
  with MPI support.
* Improved error messages when requesting chunked storage for an empty dataset.
* Data conversion functions should fail more gracefully if no memory is
  available.
* Fix some errors for internal functions that were raising "TypeError:
  expected bytes, str found" instead of the correct error.
* Use relative path for virtual data sources if the source dataset is in the
  same file as the virtual dataset.
* Generic exception types used in tests' assertRaise (exception types changed in
  new HDF5 version)
* Use ``dtype=object`` in tests with ragged arrays

Building h5py
-------------

* The ``setup.py configure`` command was removed. Configuration for the build
  can be specified with environment variables instead. See :ref:`custom_install`
  for details.
* It is now possible to specify separate include and library directories for
  HDF5 via environment variables. See :ref:`custom_install` for more details.
* The pkg-config name to use when looking up the HDF5 library can now be
  configured, this can assist with selecting the correct HDF5 library when using
  MPI. See :ref:`custom_install` for more details.
* Using bare ``char*`` instead of ``array.array`` in h5d.read_direct_chunk since
  ``array.array`` is a private CPython C-API interface
* Define ``NPY_NO_DEPRECATED_API`` to silence a warning.
* Make the lzf filter build with HDF5 1.10 (:issue:`1219`).
* If HDF5 is not loaded, an additional message is displayed to check HDF5
  installation
* Rely much more on the C-interface provided by Cython to call Python and NumPy.
* Removed an old workaround which tried to run Cython in a subprocess if
  cythonize() didn't work. This shouldn't be necessary for any recent version
  of setuptools.
* Migrate all Cython code base to Cython3 syntax

    * The only noticeable change is in exception raising from cython which use bytes
    * Massively use local imports everywhere as expected from Python3
    * Explicitly mark several Cython functions as non-binding

Development
-----------

* Unregistering converter functions on exit (:pr:`1440`) should allow profiling
  and code coverage tools to work on Cython code.

(adam)

py-pkgconfig: updated to 1.5.4

Version 1.5.4

Adjust pyproject.toml and drop Python 2 support

Version 1.5.3

Add configure_extension API

(adam)

hdf5: updated to 1.10.7

HDF5 version 1.10.7 released on 2020-09-11
================================================================================

INTRODUCTION

This document describes the differences between this release and the previous
HDF5 release. It contains information on the platforms tested and known
problems in this release. For more details check the HISTORY*.txt files in the
HDF5 source.

Note that documentation in the links below will be updated at the time of each
final release.

Links to HDF5 documentation can be found on The HDF5 web page:

    https://portal.hdfgroup.org/display/HDF5/HDF5

The official HDF5 releases can be obtained from:

    https://www.hdfgroup.org/downloads/hdf5/

    HDF5 binaries provided are fully tested with ZLIB and the free
    Open Source SZIP successor Libaec (with BSD license).
    The official ZLIB and SZIP/Libaec pages are at:

        ZLIB: http://www.zlib.net/
            http://www.zlib.net/zlib_license.html
        SZIP/Libaec: https://gitlab.dkrz.de/k202009/libaec
            https://gitlab.dkrz.de/k202009/libaec/-/blob/master/Copyright.txt

Changes from Release to Release and New Features in the HDF5-1.10.x release series
can be found at:

    https://portal.hdfgroup.org/display/HDF5/HDF5+Application+Developer%27s+Guide

If you have any questions or comments, please send them to the HDF Help Desk:

    help@hdfgroup.org

CONTENTS

- New Features
- Support for new platforms and languages
- Bug Fixes since HDF5-1.10.6
- Supported Platforms
- Tested Configuration Features Summary
- More Tested Platforms
- Known Problems
- CMake vs. Autotools installations

New Features
============

    Configuration:
    -------------
    - Disable memory sanity checks in the Autotools in release branches

      The library can be configured to use internal memory sanity checking,
      which replaces C API calls like malloc(3) and free(3) with our own calls
      which add things like heap canaries. These canaries can cause problems
      when external filter plugins reallocate canary-marked buffers.

      For this reason, the default will be to not use the memory allocation
      sanity check feature in release branches (e.g., hdf5_1_10_7).
      Debug builds in development branches (e.g., develop, hdf5_1_10) will
      still use them by default.

      This change only affects Autotools debug builds. Non-debug autotools
      builds and all CMake builds do not enable this feature by default.

      (DER - 2020/08/19)

    - Add file locking configure and CMake options

      HDF5 1.10.0 introduced a file locking scheme, primarily to help
      enforce SWMR setup. Formerly, the only user-level control of the scheme
      was via the HDF5_USE_FILE_LOCKING environment variable.

      This change introduces configure-time options that control whether
      or not file locking will be used and whether or not the library
      ignores errors when locking has been disabled on the file system
      (useful on some HPC Lustre installations).

      In both the Autotools and CMake, the settings have the effect of changing
      the default property list settings (see the H5Pset/get_file_locking()
      entry, below).

      The yes/no/best-effort file locking configure setting has also been
      added to the libhdf5.settings file.

      Autotools:

        An --enable-file-locking=(yes|no|best-effort) option has been added.

        yes:          Use file locking.
        no:          Do not use file locking.
        best-effort:  Use file locking and ignore "disabled" errors.

      CMake:

        Two self-explanatory options have been added:

        HDF5_USE_FILE_LOCKING
        HDF5_IGNORE_DISABLED_FILE_LOCKS

        Setting both of these to ON is the equivalent to the Autotools'
        best-effort setting.

      NOTE:
      The precedence order of the various file locking control mechanisms is:

        1) HDF5_USE_FILE_LOCKING environment variable (highest)

        2) H5Pset_file_locking()

        3) configure/CMake options (which set the property list defaults)

        4) library defaults (currently best-effort)

      (DER - 2020/07/30, HDFFV-11092)

    - CMake option to link the generated Fortran MOD files into the include
      directory.

      The Fortran generation of MOD files by a Fortran compile can produce
      different binary files between SHARED and STATIC compiles with different
      compilers and/or different platforms. Note that it has been found that
      different versions of Fortran compilers will produce incompatible MOD
      files. Currently, CMake will locate these MOD files in subfolders of
      the include directory and add that path to the Fortran library target
      in the CMake config file, which can be used by the CMake find library
      process. For other build systems using the binary from a CMake install,
      a new CMake configuration can be used to copy the pre-chosen version
      of the Fortran MOD files into the install include directory.

      The default will depend on the configuration of
      BUILD_STATIC_LIBS and BUILD_SHARED_LIBS:
            YES                  YES        Default to SHARED
            YES                  NO          Default to STATIC
            NO                    YES        Default to SHARED
            NO                    NO          Default to SHARED
      The defaults can be overriden by setting the config option
        HDF5_INSTALL_MOD_FORTRAN to one of NO, SHARED, or STATIC

      (ADB - 2020/07/09, HDFFV-11116)

    - CMake option to use AEC (open source SZip) library instead of SZip

      The open source AEC library is a replacement library for SZip. In
      order to use it for hdf5, the libaec CMake source was changed to add
      "-fPIC" and exclude test files. A new option USE_LIBAEC is required
      to compensate for the different files produced by AEC build.

      Autotools does not build the compression libraries within hdf5 builds,
      but will use an installed libaec when configured as before with the
      option --with-libsz=<path to libaec directory>.

      (ADB - 2020/04/22, OESS-65)

    - CMake ConfigureChecks.cmake file now uses CHECK_STRUCT_HAS_MEMBER

      Some handcrafted tests in HDFTests.c have been removed and the CMake
      CHECK_STRUCT_HAS_MEMBER module has been used.

      (ADB - 2020/03/24, TRILAB-24)

    - Both build systems use same set of warnings flags

      GNU C, C++ and gfortran warnings flags were moved to files in a config
      sub-folder named gnu-warnings. Flags that only are available for a specific
      version of the compiler are in files named with that version.
      Clang C warnings flags were moved to files in a config sub-folder
      named clang-warnings.
      Intel C, Fortran warnings flags were moved to files in a config sub-folder
      named intel-warnings.

      There are flags in named "error-xxx" files with warnings that may
      be promoted to errors. Some source files may still need fixes.

      There are also pairs of files named "developer-xxx" and "no-developer-xxx"
      that are chosen by the CMake option:HDF5_ENABLE_DEV_WARNINGS or the
      configure option:--enable-developer-warnings.

      In addition, CMake no longer applies these warnings for examples.

      (ADB - 2020/03/24, TRILAB-192)

    - Update CMake minimum version to 3.12

      Updated CMake minimum version to 3.12 and added version checks
      for Windows features.

      (ADB - 2020/02/05, TRILABS-142)

    - Fixed CMake include properties for Fortran libraries

      Corrected the library properties for Fortran to use the
      correct path for the Fortran module files.

      (ADB - 2020/02/04, HDFFV-11012)

    - Added common warnings files for gnu and intel

      Added warnings files to use one common set of flags
      during configure for both autotools and CMake build
      systems. The initial implementation only affects a
      general set of flags for gnu and intel compilers.

      (ADB - 2020/01/17)

    - Added new options to CMake for control of testing

      Added CMake options (default ON);
          HDF5_TEST_SERIAL AND/OR HDF5_TEST_PARALLEL
          combined with:
            HDF5_TEST_TOOLS
            HDF5_TEST_EXAMPLES
            HDF5_TEST_SWMR
            HDF5_TEST_FORTRAN
            HDF5_TEST_CPP
            HDF5_TEST_JAVA

      (ADB - 2020/01/15, HDFFV-11001)

    - Added Clang sanitizers to CMake for analyzer support if compiler is clang.

      Added CMake code and files to execute the Clang sanitizers if
      HDF5_ENABLE_SANITIZERS is enabled and the USE_SANITIZER option
      is set to one of the following:
          Address
          Memory
          MemoryWithOrigins
          Undefined
          Thread
          Leak
          'Address;Undefined'

      (ADB - 2019/12/12, TRILAB-135)

    Library:
    --------
    - Add metadata cache optimization to reduce skip list usage

      On file flush or close, the metadata cache attempts to write out
      all dirty entries in increasing address order.  To do this, it needs
      an address sorted list of metadata entries.  Further, since flushing
      one metadata cache entry can dirty another, this list must support
      efficient insertion and deletion.

      The metadata cache uses a skip list of all dirty entries for this
      purpose.  Before this release, this skip list was maintained at all
      times.  However, since profiling indicates that this imposes a
      significant cost, we now construct and maintain the skip list only
      when needed.  Specifically, we enable the skip list and load it with
      a list of all dirty entries in the metadata cache just before a flush,
      and disable it after the flush.

      (JRM - 2020/08/17, HDFFV-11034)

    - Add BEST_EFFORT value to HDF5_USE_FILE_LOCKING environment variable

      This change adds a BEST_EFFORT to the TRUE/FALSE, 1/0 settings that
      were previously accepted. This option turns on file locking but
      ignores locking errors when the library detects that file locking
      has been disabled on a file system (useful on some HPC Lustre
      installations).

      The capitalization of BEST_EFFORT is mandatory.

      See the configure option discussion for HDFFV-11092 (above) for more
      information on the file locking feature and how it's controlled.

      (DER - 2020/07/30, HDFFV-11092)

    - Add H5Pset/get_file_locking() API calls

      This change adds new API calls which can be used to set or get the
      file locking parameters. The single API call sets both the "use file
      locking" flag and the "ignore disabled file locking" flag.

      When opening a file multiple times without closing, the file MUST be
      opened with the same file locking settings. Opening a file with different
      file locking settings will fail (this is similar to the behavior of
      H5Pset_fclose_degree()).

      See the configure option discussion for HDFFV-11092 (above) for more
      information on the file locking feature and how it's controlled.

      (DER - 2020/07/30, HDFFV-11092)

    - Add Mirror VFD

      Use TCP/IP sockets to perform write-only (W/O) file I/O on a remote
      machine. Must be used in conjunction with the Splitter VFD.

      (JOS - 2020/03/13, TBD)

    - Add Splitter VFD

      Maintain separate R/W and W/O channels for "concurrent" file writes
      to two files using a single HDF5 file handle.

      (JOS - 2020/03/13, TBD)

    - Fixed an assertion failure in the parallel library when collectively
      filling chunks. As it is required that chunks be written in
      monotonically non-decreasing order of offset in the file, this assertion
      was being triggered when the list of chunk file space allocations being
      passed to the collective chunk filling routine was not sorted according
      to this particular requirement.

      The addition of a sort of the out of order chunks trades a bit of
      performance for the elimination of this assertion and of any complaints
      from MPI implementations about the file offsets used being out of order.

      (JTH - 2019/10/07)

    Fortran Library:
    ----------------
    - Add wrappers for H5Pset/get_file_locking() API calls

      h5pget_file_locking_f()
      h5pset_file_locking_f()

      See the configure option discussion for HDFFV-11092 (above) for more
      information on the file locking feature and how it's controlled.

      (DER - 2020/07/30, HDFFV-11092)

    - Added new Fortran parameters:

        H5F_LIBVER_ERROR_F
        H5F_LIBVER_NBOUNDS_F
        H5F_LIBVER_V18_F
        H5F_LIBVER_V110_F

    - Added new Fortran API: h5pget_libver_bounds_f

      (MSB - 2020/02/11, HDFFV-11018)

    C++ Library:
    ------------
    - Add wrappers for H5Pset/get_file_locking() API calls

      FileAccPropList::setFileLocking()
      FileAccPropList::getFileLocking()

      See the configure option discussion for HDFFV-11092 (above) for more
      information on the file locking feature and how it's controlled.

      (DER - 2020/07/30, HDFFV-11092)

    Java Library:
    ----------------
    - Add wrappers for H5Pset/get_file_locking() API calls

      H5Pset_file_locking()
      H5Pget_use_file_locking()
      H5Pget_ignore_disabled_file_locking()

      Unlike the C++ and Fortran wrappers, there are separate getters for the
      two file locking settings, each of which returns a boolean value.

      See the configure option discussion for HDFFV-11092 (above) for more
      information on the file locking feature and how it's controlled.

      (DER - 2020/07/30, HDFFV-11092)

    Tools:
    ------
    - h5repack added options to control how external links are handled.

      Currently h5repack preserves external links and cannot copy and merge
      data from the external files. Two options, merge and prune, were added to
      control how to merge data from an external link into the resulting file.
      --merge            Follow external soft link recursively and merge data.
      --prune            Do not follow external soft links and remove link.
      --merge --prune    Follow external link, merge data and remove dangling link.

      (ADB - 2020/08/05, HDFFV-9984)

    High-Level APIs:
    ---------------
    - None

    C Packet Table API
    ------------------
    - None

    Internal header file
    --------------------
    - None

    Documentation
    -------------
    - None

Support for new platforms, languages and compilers.
=======================================
    - None

Bug Fixes since HDF5-1.10.6 release
==================================

    Library
    -------
    - Fix bug and simplify collective metadata write operation when some ranks
        have no entries to contribute.  This fixes parallel regression test
        failures with IBM SpectrumScale MPI on the Summit system at ORNL.

      (QAK - 2020/09/02)

    - Avoid setting up complex MPI types with 0-length vectors, which some
        MPI implementations don't handle well.  (In particular, IBM
        SpectrumScale MPI on the Summit system at ORNL)

      (QAK - 2020/08/21)

    - Fixed use-of-uninitialized-value error

      Appropriate initialization of local structs was added to remove the
      use-of-uninitialized-value errors reported by MemorySanitizer.

      (BMR - 2020/8/13, HDFFV-11101)

    - Creation of dataset with optional filter

      When the combination of type, space, etc doesn't work for filter
      and the filter is optional, it was supposed to be skipped but it was
      not skipped and the creation failed.

      A fix is applied to allow the creation of a dataset in such
      situation, as specified in the user documentation.

      (BMR - 2020/8/13, HDFFV-10933)

    - Explicitly declared dlopen to use RTLD_LOCAL

      dlopen documentation states that if neither RTLD_GLOBAL nor
      RTLD_LOCAL are specified, then the default behavior is unspecified.
      The default on linux is usually RTLD_LOCAL while macos will default
      to RTLD_GLOBAL.

      (ADB - 2020/08/12, HDFFV-11127)

    - Fixed issues CVE-2018-13870 and CVE-2018-13869

      When a buffer overflow occurred because a name length was corrupted
      and became very large, h5dump crashed on memory access violation.

      A check for reading past the end of the buffer was added to multiple
      locations to prevent the crashes and h5dump now simply fails with an
      error message when this error condition occurs.

      (BMR - 2020/7/31, HDFFV-11120 and HDFFV-11121)

    - H5Sset_extent_none() sets the dataspace class to H5S_NO_CLASS which
      causes asserts/errors when passed to other dataspace API calls.

      H5S_NO_CLASS is an internal class value that should not have been
      exposed via a public API call.

      In debug builds of the library, this can cause asserts to trip. In
      non-debug builds, it will produce normal library errors.

      The new library behavior is for H5Sset_extent_none() to convert
      the dataspace into one of type H5S_NULL, which is better handled
      by the library and easier for developers to reason about.

      (DER - 2020/07/27, HDFFV-11027)

    - Fixed the segmentation fault when reading attributes with multiple threads

      It was reported that the reading of attributes with variable length string
      datatype will crash with segmentation fault particularly when the number
      of threads is high (>16 threads).  The problem was due to the file pointer
      that was set in the variable length string datatype for the attribute.
      That file pointer was already closed when the attribute was accessed.

      The problem was fixed by setting the file pointer to the current opened
      file pointer when the attribute was accessed.  Similar patch up was done
      before when reading dataset with variable length string datatype.

      (VC - 2020/07/13, HDFFV-11080)

    -  Fixed issue CVE-2018-17438

      A division by zero was discovered in H5D__select_io() of H5Dselect.c.
        https://security-tracker.debian.org/tracker/CVE-2018-17438

      A check was added to protect against division by zero.  When such
      situation occurs again, the normal HDF5 error handling will be invoked,
      instead of segmentation fault.

      (BMR, DER - 2020/07/09, HDFFV-10587)

    - Fixed CVE-2018-17435

      The tool h52gif produced a segfault when the size of an attribute message
      was corrupted and caused a buffer overflow.

      The problem was fixed by verifying the attribute message's size against the
      buffer size before accessing the buffer.  h52gif was also fixed to display
      the failure instead of silently exiting after the segfault was eliminated.

      (BMR - 2020/6/19, HDFFV-10591)

    - Don't allocate an empty (0-dimensioned) chunked dataset's chunk
      index, until the dataset's dimensions are increased.

      (QAK - 2020/05/07)

    Configuration
    -------------
    - Stopped addition of szip header and include directory path for
      incompatible libsz

      szlib.h is the same for both 32-bit and 64-bit szip, and the header file
      and its path were added to the HDF5 binary even though the configure
      check of a function in libsz later failed and szip compression was not
      enabled.  The header file and include path are now added only when the
      libsz function passes the configure check.

      (LRK - 2020/08/17, HDFFV-10830)

    - Added -fsanitize=address autotools configure option for Clang compiler

      Clang sanitizer options were also added for Clang compilers with CMake.

      (LRK, 2020/08/05, HDFFV-10836)

    - Updated testh5cc.sh.in for functions versioned in HDF5 1.10.

      testh5cc.sh previously tested that the correct version of a function
      versioned in HDF5 1.6 or 1.8 was compiled when one of
      H5_NO_DEPRECATED_SYMBOLS or H5_USE_16_API_DEFAULT were defined.  This
      test was extended for additional testing with H5_USE_18_API_DEFAULT.

      (LRK, 2020/06/22, HDFFV-11000)

    - Fixed CMake include properties for Fortran libraries

      Corrected the library properties for Fortran to use the
      correct path for the Fortran module files.

      (ADB - 2020/02/04, HDFFV-11012)

    Performance
    -------------
    - None

    Java Library:
    ----------------
    - None

    Fortran
    --------
    - Corrected INTERFACE INTENT(IN) to INTENT(OUT) for buf_size in h5fget_file_image_f.

      (MSB - 2020/2/18, HDFFV-11029)

    - Fixed configure issue when building HDF5 with NAG Fortran 7.0.

      HDF5 now accounts for the addition of half-precision floating-point
      in NAG 7.0 with a KIND=16.

      (MSB - 2020/02/28, HDFFV-11033)

    Tools
    -----
    - The tools library was updated by standardizing the error stack process.

      General sequence is:
          h5tools_setprogname(PROGRAMNAME);
          h5tools_setstatus(EXIT_SUCCESS);
          h5tools_init();
          ... process the command-line (check for error-stack enable) ...
          h5tools_error_report();
          ... (do work) ...
          h5diff_exit(ret);

      (ADB - 2020/07/20, HDFFV-11066)

    - h5diff fixed a command line parsing error.

      h5diff would ignore the argument to -d (delta) if it is smaller than DBL_EPSILON.
      The macro H5_DBL_ABS_EQUAL was removed and a direct value comparision was used.

      (ADB - 2020/07/20, HDFFV-10897)

    - h5diff added a command line option to ignore attributes.

      h5diff would ignore all objects with a supplied path if the exclude-path argument is used.
      Adding the exclude-attribute argument will only eclude attributes, with the supplied path,
      from comparision.

      (ADB - 2020/07/20, HDFFV-5935)

    - h5diff added another level to the verbose argument to print filenames.

      Added verbose level 3 that is level 2 plus the filenames. The levels are:
          0 : Identical to '-v' or '--verbose'
          1 : All level 0 information plus one-line attribute status summary
          2 : All level 1 information plus extended attribute status report
          3 : All level 2 information plus file names

      (ADB - 2020/07/20, HDFFV-10005)

    - h5repack was fixed to repack the reference attributes properly.
      The code line that checks if the update of reference inside a compound
      datatype is misplaced outside the code block loop that carries out the
      check. In consequence, the next attribute that is not the reference
      type was repacked again as the reference type and caused the failure of
      repacking. The fix is to move the corresponding code line to the correct
      code block.

      (KY -2020/02/10, HDFFV-11014)

    High-Level APIs:
    ------
    - The H5DSis_scale function was updated to return "not a dimension scale" (0)
      instead of failing (-1), when CLASS or DIMENSION_SCALE attributes are
      not written according to Dimension Scales Specification.

    (EIP - 2020/08/12, HDFFV-10436)

    Fortran High-Level APIs:
    ------
    - None

    Documentation
    -------------
    - None

    F90 APIs
    --------
    - None

    C++ APIs
    --------
    - None

    Testing
    -------
    - Stopped java/test/junit.sh.in installing libs for testing under ${prefix}

      Lib files needed are now copied to a subdirectory in the java/test
      directory, and on Macs the loader path for libhdf5.xxxs.so is changed
      in the temporary copy of libhdf5_java.dylib.

      (LRK, 2020/7/2, HDFFV-11063)

(adam)

Updated multimedia/libogg, net/libtorrent-rasterbar

(adam)

libtorrent-rasterbar: updated to 1.2.14

libtorrent-1.2.14

improve handling of seed flag in PEX messages
fix issue of accruing unlimited DHT node candidates when DHT is disabled
fix bug in parsing chunked encoding
fix incorrect reporting of active_duration when entering graceful-pause
fix python binding for functions taking string_view
fix python binding for torrent_info constructor overloads
issue python deprecation warnings for some deprecated functions in the python bindings
fix python binding for torrent_info::add_url_seed, add_tracker and add_http_seed

(adam)

libogg: updated to 1.3.5

Version 1.3.5 (2020 June 3)
* Fix unsigned typedef problem on macOS.
* Fix overflow check in ogg_sync_buffer.
* Clean up cmake and autotools build files.
* Remove Symbian and Apple XCode build files.
* Fix documentation cross-reference links.

(adam)

Updated security/py-authlib, multimedia/py-m3u8, sysutils/py-magic, textproc/py-docx

(adam)

py-docx: updated to 0.8.11

0.8.11:
- Small build changes and Python 3.8 version changes like collections.abc location.

(adam)

py-magic: updated to 0.4.24

Changes to 0.4.24:
- Fix regression in library loading on some Alpine docker images.

(adam)

py-m3u8: updated to 0.9.0

0.9.0:
Add Github actions to test the code
Treat S3 as a URL instead of a local file

(adam)

py-authlib: updated to 0.15.4

0.15.4:
Unknown changes

(adam)

Updated net/amule, net/wireshark

(adam)

wireshark: updated to 3.4.6

Wireshark 3.4.6 Release Notes

What’s New

  The Windows installers now ship with Npcap 1.31. They previously
  shipped with Npcap 1.10.

  The Windows installers now ship with Qt 5.15.2. They previously
  shipped with Qt 5.12.1.

  Bug Fixes

    • wnpa-sec-2021-04[1] DVB-S2-BB dissector infinite loop

  The following bugs have been fixed:

    • Macro filters can’t handle escaped characters Issue 17160[2].

    • Display filter crashes Wireshark Issue 17316[3].

    • IEEE-1588 Signalling Unicast TLV incorrectly reported as being
      malformed Issue 17355[4].

    • IETF QUIC TLS decryption error with extraneous packets during the
      handshake Issue 17383[5].

    • Statistics → Resolved Addresses: multi-protocol (TCP/UDP/…<U+200B>)
      ports not displayed Issue 17395[6].

  New and Updated Features

  New Protocol Support

  There are no new protocols in this release.

  Updated Protocol Support

  DNP, DVB-S2-BB, ProtoBuf, PTP, QUIC, RANAP, and TACACS

  New and Updated Capture File Support

  Ascend, ERF, K12, NetScaler, and pcapng

(adam)

amule: updated to 2.3.3

2.3.3 - The "delayed too many times, but still not too late" version

Yes, it's been delayed too many times, but finally it's out! With all the new features and bug fixes from the past four years! Let me highlight some for you:

CMake support. CMake provides us platform-independent configuration files and generates native makefiles and workspaces for the compiler of your choice. It's highly experimental in its current state, but it aims to take the place of good old autoconf/automake one day.
Command completion with the TAB key in amulecmd.
Supports compilation with the latest C++ standard (C++17).
Scope graphs now (should) work on Mac.
Notification when a download completes.

(adam)

Updated net/py-smb, misc/py-tqdm, www/py-django3, www/py-django2

(adam)

py-django2: updated to 2.2.24

Django 2.2.24 fixes two security issues in 2.2.23.

CVE-2021-33203: Potential directory traversal via admindocs

Staff members could use the admindocs TemplateDetailView view to check the existence of arbitrary files. Additionally, if (and only if) the default admindocs templates have been customized by the developers to also expose the file contents, then not only the existence but also the file contents would have been exposed.

As a mitigation, path sanitation is now applied and only files within the template root directories can be loaded.

CVE-2021-33571: Possible indeterminate SSRF, RFI, and LFI attacks since validators accepted leading zeros in IPv4 addresses¶

URLValidator, validate_ipv4_address(), and validate_ipv46_address() didn’t prohibit leading zeros in octal literals. If you used such values you could suffer from indeterminate SSRF, RFI, and LFI attacks.

validate_ipv4_address() and validate_ipv46_address() validators were not affected on Python 3.9.5+.

(adam)

py-django3: updated to 3.2.4

Django 3.2.4 fixes two security issues and several bugs in 3.2.3.

CVE-2021-33203: Potential directory traversal via admindocs

Staff members could use the admindocs TemplateDetailView view to check the existence of arbitrary files. Additionally, if (and only if) the default admindocs templates have been customized by the developers to also expose the file contents, then not only the existence but also the file contents would have been exposed.

As a mitigation, path sanitation is now applied and only files within the template root directories can be loaded.

CVE-2021-33571: Possible indeterminate SSRF, RFI, and LFI attacks since validators accepted leading zeros in IPv4 addresses其

URLValidator, validate_ipv4_address(), and validate_ipv46_address() didn���t prohibit leading zeros in octal literals. If you used such values you could suffer from indeterminate SSRF, RFI, and LFI attacks.

validate_ipv4_address() and validate_ipv46_address() validators were not affected on Python 3.9.5+.

Bugfixes

Fixed a bug in Django 3.2 where a final catch-all view in the admin didn���t respect the server-provided value of SCRIPT_NAME when redirecting unauthenticated users to the login page
Fixed a bug in Django 3.2 where a system check would crash on an abstract model
Prevented unnecessary initialization of unused caches following a regression in Django 3.2
Fixed a crash in Django 3.2 that could occur when running mod_wsgi with the recommended settings while the Windows colorama library was installed
Fixed a bug in Django 3.2 that would trigger the auto-reloader for template changes when directory paths were specified with strings
Fixed a regression in Django 3.2 that caused a crash of auto-reloader with AttributeError, e.g. inside a Conda environment
Fixed a regression in Django 3.2 that caused a loss of precision for operations with DecimalField on MySQL

(adam)

py-tqdm: updated to 4.61.0

tqdm v4.61.0 stable

keras support for initial_epoch
misc documentation updates
update & shorten URLs
fix typos
fix & update tests
minor framework updates

(adam)

py-smb: updated to 1.2.7

pysmb-1.2.7, 30 May 2021
========================
- Fix compatibility issues on file retrievals with Likewise servers
- Improve SMBConnection's connect() method to remove the need to provide sock_family
  parameter for IPv6 addresses in Python 3.x

(adam)

Updated finance/py-stripe, sysutils/ansible-lint

(adam)

ansible-lint: updated to 5.0.12

v5.0.12:
Bugfixes

Document syntax-check as unskippable
Fail NoLogPassword only when loops are used
Allow use of ansible debug mode
Fixed typo in .ansible-lint
Remove include missing file rule

(adam)

py-stripe: updated to 2.58.0

2.58.0:
* Add support for new `TaxCode` API.

(adam)

fuse-djmount: mark as broken

(adam)

fuse.buildlink3.mk: on macOS, allow to use macFUSE (formerly OSXFUSE)

(adam)

Updated net/libupnp, net/miniupnpc

(adam)

miniupnpc: updated to 2.2.2

2.2.2:
miniupnpcmodule.c: throw an exception in UPnP_discover()
Fix usage of IP_MULTICAST_IF with struct ip_mreqn

(adam)

libupnp: updated to 1.14.7

Version 1.14.7
Fixes broken windows build.

Version 1.14.6
Fix for a DNS Rebind exploit. A special thanks for the collaboration
of the following people:
- Alaric Senat
- Fabrice Fontaine
- Gabriel Corona
- Ian Whyman
- Jean-Francois Dockes
- Marvin Scholz
- Werner Mahr

When upnp uses ixml to parse SOAP messages which contains too many node,
services are unavailable.

Version 1.14.5
Avoid potential memory leak in http_SendMessage()

Get rid of alloca()'s.

Non-recursive version of ixmlNode_free() avoids stack overflow attack.
Fixes CVE-2021-28302.

Version 1.14.4
Add more missing CMake infrastructure to the tarball.

Version 1.14.3
Add CMake infrastructure to the tarball.

Fix for gena leak.

Version 1.14.2
upnpapi.c: Fix crash in UpnpGetIfInfo

Per getifaddrs documentation, the ifa_addr field of an ifaddrs structure
can be null. In a real world example, an entry may be provided for the
non-existent hardware address of a tunnel device. This behavior was
observed with the netlink based getifaddrs implementation in glibc.

Handle interfaces without address

I got a crash on gerbera startup because at least one interface did not have `ifa_addr` set.

The reason is quite simple: they are the physical interfaces which are part of my channel
```
2: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
    link/ether xx:xx:xx:xx:xx:xx brd ff:ff:ff:ff:ff:ff
3: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
    link/ether xx:xx:xx:xx:xx:xx brd ff:ff:ff:ff:ff:ff
4: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether xx:xx:xx:xx:xx:xx brd ff:ff:ff:ff:ff:ff
```

CMake:
- Fix tests for all occasions. (Hopefully)
- Set right so-version and don't yell commands.
- Reverted libname-change till decided

Fix IPv6 GENA subscription

Use LOCAL_PORT_V6_ULA_GUA or LOCAL_PORT_V6 depending on the IPv6 address

samples: allow the user to specify the control point interface

This is useful to test UPnP on specific interface (e.g. IPv6-only).
This was already possible on the device sample.

upnp/src/api/upnpapi.c: don't set gIF_IPV4 if there is no IPv4

Don't set gIF_IPV4 if no IPv4 is found to keep the default value of '\0'
otherwise SSDP will try to register IPv4 multicast with address 0.0.0.0
which will result in a runtime failure

Version 1.14.1
CMake Github Actions, including windows build.

upnpapi.c: assume that getifaddrs is available

Assume that getifaddrs is available even if it is not POSIX-compliant,
this will simplify the code and, as a side effect, this allow pupnp to
work on an interface that does not have an IPv4 address.

CMake support.

Fixes the inclusion of alloca.h in WIN32.

Remove the now unused file ClientSubscription.c.

Remove port >= 49152 restriction.

Version 1.14.0
Reworked the miniserver code to deal with SO_REUSEADDR

Factored the common socket code and test for EADDRINUSE returning from
listen() when SO_REUSEADDR is turned on.

Removed template classes.

The files generated through template classes had some drawbacks:

1 - You could not read the code.
2 - You could not step through the code with a debugger.
3 - Doxygen was unable to document it.

This patch removes the templates and creates an auto generator for these
boilerplate generated code in the same spirit of the templates, but
solves the above problems.

Still, dealing with documentation still needs some work, but should be
much easier now.

upnpapi.c: retrieve gIF_IPV4_NETMASK with BSD

BSD is using getifaddrs, update the code to retrieve the IPv4 netmask
(used in the CallStranger fix)

Signed-off-by: Fabrice Fontaine fontaine.fabrice@gmail.com

Drop UpnpInit

This function is deprecated since a long time moreover it is vulnerable
to CallStranger a.k.a. CVE-2020-12695 and can't be fixed without
breaking the API as HostIP only allows the user to pass an IP address
and not a netmask.

If we want to discover the netmask from the HostIP provided in UPnPInit,
we'll have to loop through all the available interfaces to find the
interface with the given IP address to finally retrieve the netmask.
This is a lot of work/modification for a deprecated function. Moreover,
in the end UPnPInit will be like a "deprecated" (i.e. IPv4-only) version
of UPnPInit2.

So it is time to remove this deprecated function.

As a result, also remove getlocalhostname and DEFAULT_INTERFACE which
are not needed anymore and replace UpnpInit by UpnpInit2 in comments.

Signed-off-by: Fabrice Fontaine fontaine.fabrice@gmail.com

Partial fix for CallStranger on IPv4

This is a partial fix for CallStranger a.k.a. CVE-2020-12695

Check that DeliveryURLs are in the expected network segment as requested
by the new UPnP UDA:
https://openconnectivity.org/upnp-specs/UPnP-arch-DeviceArchitecture-v2.0-20200417.pdf.

Here is an extract of the new requirement:

The subscription request containing a delivery URL not on the same
network segment as the fully qualified event subscription URL shall not
be accepted. For private networks this means that the delivery URL
provided will adhere to the following IP ranges:

. 10.0.0.0 - 10.255.255.255 (10/8 prefix)
. 172.16.0.0 - 172.31.255.255 (172.16/12 prefix)
. 192.168.0.0 - 192.168.255.255 (192.168/16 prefix)

In the context of pupnp, this means that the IPv4 netmask is now
retrieved when using UPnPInit2. Then, each DeliveryURL is checked
against the device's IPv4 address and netmask. If one of them are not
compliant, the whole subscription is rejected.

This first commit should be enhanced / updated to:

. remove UPnPInit (it is deprecated for a long time) or update it so
. the user can also pass the netmask
. fix IPv6
. fix Windows code
. retrieve the netmask in the BSD code of UPnPInit2

Signed-off-by: Fabrice Fontaine fontaine.fabrice@gmail.com

NULL pointer dereference in FindServiceControlURLPath

Version 1.12.1
setsockopt(IP_MULTICAST_IF) fails on 64-bit CPUs

Do not use usleep when using newer POSIX C source.

usleep is deprecated and is optionally unavailable with uClibc-ng.

Add Os/UserAgent information to UpnpActionRequest and UpnpFileInfo

Add CtrlPtIPAddr to UpnpFileInfo class

List: Add extern C for C++ users

Version 1.12.0
    Reduce spurious HTTP 416 errors due to ill-defined bytes header

    I (re)discovered this behavior trying to use the Gerbera media server
    with Chromecast (built in to my Vizio P55-F1 TV).  Chromecast specifies
    "bytes:0-" with no end range, which caused pupnp to return
    RANGE_NOT_SATISFIABLE.  Jin, the author of MediaTomb, of which Gerbera
    is a continuation, fixed this in 2007 in MediaTomb's fork of pupnp, see
    gerbera@ccd7994d45 "made sure that range requests specified as
    "bytes=0-" do not trigger...", but never passed the fix upstream.  When
    restarted as Gerbera, pupnp was removed from the local tree in favor of
    using the upstream version, and the patch was lost which lead to me
    rediscovering it recently.

    This is Jin's original patch applied to pupnp 1.8.  This teaches pupnp to
    ignore the bytes header if it matches "0-" and return HTTP_OK.

When building v1.8.6 (on GNU/Linux) with --disable-device, pupnp doesn't build:
/bin/bash ../libtool  --tag=CC  --mode=compile gcc -DHAVE_CONFIG_H -I. -I.. -I../upnp/inc  -I./inc -I../ixml/inc  -DNDEBUG -I./src/inc -I./src/threadutil -I/foo/contrib/x86_64-linux-gnu/include -fPIC  -pthread  -I/foo/contrib/x86_64-linux-gnu/include -DNDEBUG=1 -O3  -DUPNP_STATIC_LIB -Os -Wall -c -o src/genlib/util/libupnp_la-util.lo `test -f 'src/genlib/util/util.c' || echo './'`src/genlib/util/util.c
In file included from src/genlib/miniserver/miniserver.c:58:0:
./src/inc/upnpapi.h:208:2: error: unknown type name 'service_info'
service_info **serv_info
^~~~~~~~~~~~

This patch fixes compilation for the --disable-device case.

GitHub PRs:
* Remove unused files
* UpnpEnableWebserver: Error correctly when web server compiled-out
* CI: Remove .travis.yml
* Remove TRUE/FALSE defines and BOOL typedef
* CI: Add address sanitizer and leak sanitizer enabled test run
* Add list test
* Add -Wextra and -Wpedantic to get more warnings
* More warning fixes
* Doxygen changes
* Remove unused headers
* Remove IN, OUT and INOUT defines
* Remove unused defines in uri.h
* Use stdbool.h instead of BOOL typedef and defines
* Various compiler warning fixes
* configure.ac: Fix Windows detection
* configure.ac: Enable silent rules
* LinkedList: Simplify _WIN32 check
* Remove checks of __OSX__ define
* Add Github Actions based CI

Github PR * Add forgotten Windows dll export qualifiers

Github PR * Remove and replace the list.h file

Github PRs:
* Overhaul list.h to fix various issues
* Use rand_s in get_random_info on Windows
* Do not redeclare timezone struct if already defined in Mingw
* Fix interface filtering for Windows
- The mentioned fix is not only needed for MSVC, but when
cross-compiling for Windows with mingw-w64 too.
* config.h: Remove DEBUG_TARGET
- The DEBUG_TARGET is no longer used, it was previously used in
upnpdebug.c to disable logging to a file, which now is done
by just not calling UpnpSetLogFileNames.
* Check for -lpthread too
- Updates pthread m4 check from upstream

Version 1.10.1
Fix format string for ExtraHeaders

Version 1.10.0
Fix sed command for upnp patch in configure.ac

Proper allocation for an array of structs and a check for an unlikely
overflow when calling the SIOCGIFCONF ioctl().

It still bugs me the stack requirements of getlocalhostname() and
UpnpGetIfInfo().

Version 1.8.6
Fix format string for ExtraHeaders

Version 1.8.5
Fixed a wrong ifdef in ssdp_device.c that was causing problems with
ipv6.

(adam)

Updated www/py-websockets, www/py-uvicorn

(adam)

py-uvicorn: updated to 0.14.0

0.14.0

Added

Defaults ws max_size on server to 16MB
Improve user feedback if no ws library installed
Support 'reason' field in 'websocket.close' messages
Implemented lifespan.shutdown.failed

Changed

Upgraded websockets requirements
Switch to asyncio streams API
Update httptools from 0.1.* to 0.2.*
Allow Click 8.0
Add search for a trusted host in ProxyHeadersMiddleware
Up wsproto to 1.0.0

Fixed

Force reload_dirs to be a list
Fix gunicorn worker not running if extras not installed
Fix socket port 0
Prevent garbage collection of main lifespan task

(adam)

py-websockets: updated to 9.1

9.1

Version 9.1 fixes a security issue introduced in version 8.0.
Version 8.0 was vulnerable to timing attacks on HTTP Basic Auth passwords.

(adam)

py-flask-admin: fix building for Python < 3.7

(adam)

py-prettytable: PYTHON_VERSIONS_INCOMPATIBLE should be before pyversion.mk

(adam)

py-immutabledict: not for Python 2.7

(adam)

Updated net/iperf3, graphics/pngquant

(adam)

pngquant: updated to 2.15.1

2.15.1:
Bug fixes

(adam)

iperf3: updated to 3.10.1

iperf-3.10.1 2021-06-03
-----------------------

* Notable user-visible changes

  * Fixed a problem with autoconf scripts that made builds fail in
    some environments.

* Developer-visible changes

  * GNU autoconf 2.71 or newer is now required to regenerate iperf3's
    configure scripts.

(adam)

py-daemon: fix BUILD_DEPENDS

(adam)

Updated databases/redis, audio/faudio

(adam)

faudio: updated to 21.06

21.06

New Features:
Update to FNA3D 21.06
Update to FAudio 21.06

Fixes:
Removed the VS2019 app.config workaround
Fix a possible crash when opening the default Microphone

(adam)

redis: updated to 6.2.4

Redis 6.2.4
===========

Upgrade urgency: SECURITY, Contains fixes to security issues that affect
authenticated client connections. MODERATE otherwise.

Fix integer overflow in STRALGO LCS (CVE-2021-32625)
An integer overflow bug in Redis version 6.0 or newer can be exploited using the
STRALGO LCS command to corrupt the heap and potentially result with remote code
execution. This is a result of an incomplete fix by CVE-2021-29477.

Bug fixes that are only applicable to previous releases of Redis 6.2:
* Fix crash after a diskless replication fork child is terminated
* Fix redis-benchmark crash on unsupported configs

Other bug fixes:
* Fix crash in UNLINK on a stream key with deleted consumer groups
* SINTERSTORE: Add missing keyspace del event when none of the sources exist
* Sentinel: Fix CONFIG SET of empty string sentinel-user/sentinel-pass configs
* Enforce client output buffer soft limit when no traffic

Improvements:
* Hide AUTH passwords in MIGRATE command from slowlog

(adam)

rust: fix-darwin-install-name is no longer needed

(adam)

Updated www/nginx, www/nginx-devel

(adam)

nginx-devel: updated to 1.21.0

Changes with nginx 1.21.0

*) Security: 1-byte memory overwrite might occur during DNS server
  response processing if the "resolver" directive was used, allowing an
  attacker who is able to forge UDP packets from the DNS server to
  cause worker process crash or, potentially, arbitrary code execution
  (CVE-2021-23017).

*) Feature: variables support in the "proxy_ssl_certificate",
  "proxy_ssl_certificate_key" "grpc_ssl_certificate",
  "grpc_ssl_certificate_key", "uwsgi_ssl_certificate", and
  "uwsgi_ssl_certificate_key" directives.

*) Feature: the "max_errors" directive in the mail proxy module.

*) Feature: the mail proxy module supports POP3 and IMAP pipelining.

*) Feature: the "fastopen" parameter of the "listen" directive in the
  stream module.
  Thanks to Anbang Wen.

*) Bugfix: special characters were not escaped during automatic redirect
  with appended trailing slash.

*) Bugfix: connections with clients in the mail proxy module might be
  closed unexpectedly when using SMTP pipelining.

(adam)

nginx: updated to 1.20.1

Changes with nginx 1.20.1

*) Security: 1-byte memory overwrite might occur during DNS server
  response processing if the "resolver" directive was used, allowing an
  attacker who is able to forge UDP packets from the DNS server to
  cause worker process crash or, potentially, arbitrary code execution
  (CVE-2021-23017).

nginx-rtmp-module v1.2.2:
Fixed segfaults.

(adam)

py-rich: depend on typing-extensions only for Python < 3.8

(adam)

Updated net/dante, net/samba4

(adam)

samba4: updated to 4.14.5

Changes since 4.14.4
--------------------
* BUG 14696: s3: smbd: SMB1 SMBsplwr doesn't send a reply packet on success.
* BUG 14708: s3: smbd: Ensure POSIX default ACL is mapped into returned
  Windows ACL for directory handles.
* BUG 14721: s3: smbd: Fix uninitialized memory read in
  process_symlink_open() when used with vfs_shadow_copy2().
* BUG 14689: docs: Expand the "log level" docs on audit logging.
* BUG 14714: smbd: Correctly initialize close timestamp fields.
* BUG 14699: Fix gcc11 compiler issues.
* BUG 14718: docs-xml: Update smbcacls manpage.
* BUG 14719: docs: Update list of available commands in rpcclient.
* BUG 14475: ctdb: Fix a crash in run_proc_signal_handler().
* BUG 14695: s3:winbind: For 'security = ADS' require realm/workgroup to be
  set.
* BUG 14699: lib:replace: Do not build strndup test with gcc 11 or newer.

(adam)

dante: updated to 1.4.3

1.4.3:
Generic changes:
o RHEL8 build fix.

Server changes:
o LDAP module functionality expanded with support for PAC and LDAP based
  authentication.

o Fix an error case where the Dante mother process would never
  terminate a request process that failed handling a client due to
  resource shortage, leading the request process to hang around idle
  forever.

  Will hopefully fix problem reported by ANON.

o Fix for BSD authentication ECHILD error on OpenBSD.

Client changes:
o Work around an issue where "logoutput: syslog" in the client
  could create issues for the application being socksified.

o Added missing NULL argument check in Raccept().

(adam)

Updated fonts/py-fonttools, security/py-ecdsa

(adam)

py-ecdsa: updated to 0.17.0

Relase 0.17.0 (27 May 2021)

New API:
* Keys that use explicit curve parameters can now be read and written.
  Reading of explicit curves can be disabled by using the
  `valid_curve_encodings` keyword argument in `VerifyingKey.from_pem()`,
  `VerifyingKey.from_der()`, `SigningKey.from_pem()`, and
  `SigningKey.from_der()`.
* Keys can now be written with use of explicit curve parameters,
  use `curve_parameters_encoding` keyword argument of `VerifyingKey.to_pem()`,
  `VerifyingKey.to_der()`, `SigningKey.to_pem(), or `SigningKey.to_der()` to
  specify the format. By default `named_curve` will be used, unless the
  curve doesn't have an associated OID (as will be the case for an unsupported
  curve), then `explicit` encoding will be used.
* Allow specifying acceptable point formats when loading public keys
  (this also fixes a minor bug where python-ecdsa would accept raw
  encoding for points in PKCS#8 files). Set of accepted encodings is controlled
  by `valid_encodings` keyword argument in
  `ECDH.load_received_public_key_bytes()`, `VerifyingKey.from_string()`,
  `VerifyingKey.from_pem()`, VerifyingKey.from_der()`.
* `PointJacobi` and `Point` now inherit from `AbstractPoint` that implements
  the methods for parsing points. That added `from_bytes()` and
  `to_bytes()` methods to both of them.
* Curve parameters can now be read and written to PEM and DER files. The
  `Curve` class supports new `to_der()`, `from_der()`, `to_pem()`, and
  `from_pem()` methods.

Doc fix:
* Describe in detail which methods can raise `RSZeroError`, and that
  `SigningKey.sign_deterministic()` won't raise it.

Bug fix:
* Correctly truncate hash values larger than the curve order (only impacted
  custom curves and the curves added in this release).
* Correctly handle curves for which the order is larger than the prime
  (only impacted custom curves and the secp160r1 curve added in this release).
* Fix the handling of `==` and `!=` for `Public_key`, `Private_key`, `Point`,
  `PointJacobi`, `VerifyingKey`, and `SigningKey` so that it behaves
  consistently and in the expected way both in Python 2 and Python 3.
* Implement lock-less algorithm inside PointJacobi for keeping shared state
  so that when calculation is aborted with KeyboardInterrupt, the state doesn't
  become corrupted (this fixes the occasional breakage of ecdsa in interactive
  shells).

New features:
* The `speed.py` script now provides performance for signature verification
  without use of precomputation.
* New curves supported: secp112r1, secp112r2, secp128r1, secp160r1.

Performance:
* Use 2-ary Non-Adjacent Form for the combined multiply-add. This speeds up
  single-shot verify (i.e. without precomputation) by about 4 to 5%.
* Use native Python 3.8 support for calculating multiplicative inverses.

Maintenace:
* Include Python 3.9 in PyPI keywords.
* More realistic branch coverage counting (ignore Python version-specific
  branches).
* Additional test coverage to many parts of the library.
* Migrate to Github Actions for Continuous Testing.

(adam)

py-fonttools: updated to 4.24.4

4.24.4 (released 2021-05-25)
----------------------------

- [subset/instancer] Fixed ``AttributeError`` when instantiating a VF that
  contains GPOS ValueRecords with ``Device`` tables but without the respective
  non-Device values (e.g. ``XAdvDevice`` without ``XAdvance``). When not
  explicitly set, the latter are assumed to be 0.

4.24.3 (released 2021-05-20)
----------------------------

- [otTables] Fixed ``AttributeError`` in methods that split LigatureSubst,
  MultipleSubst and AlternateSubst subtables when an offset overflow occurs.
  The ``Format`` attribute was removed in v4.22.0.

4.24.2 (released 2021-05-20)
----------------------------

- [ttGlyphPen] Fixed typing annotation of TTGlyphPen glyphSet parameter.
- Fixed two instances of DeprecationWarning: invalid escape sequence.

4.24.1 (released 2021-05-20)
----------------------------

- [subset] Fixed AttributeError when SinglePos subtable has None Value (ValueFormat 0)

4.24.0 (released 2021-05-17)
----------------------------

- [pens] Add ``ttGlyphPen.TTGlyphPointPen`` similar to ``TTGlyphPen``.

4.23.1 (released 2021-05-14)
----------------------------

- [subset] Fix ``KeyError`` after subsetting ``COLR`` table that initially contains
  both v0 and v1 color glyphs when the subset only requested v1 glyphs; we were
  not pruning the v0 portion of the table.
- [colorLib] Set ``LayerV1List`` attribute to ``None`` when empty, it's optional
  in CORLv1.

4.23.0 (released 2021-05-13)
----------------------------

- [designspaceLib] Allow to use ``\\UNC`` absolute paths on Windows.
- [varLib.merger] Fixed bug where ``VarLibMergeError`` was raised with incorrect
  parameters.
- [feaLib] Allow substituting a glyph class with ``NULL`` to delete multiple glyphs
.
- [glyf] Fixed ``NameError`` exception in ``getPhantomPoints``.
- [removeOverlaps] Retry pathops.simplify after rounding path coordinates to integers
  if it fails the first time using floats, to work around a rare and hard to debug
  Skia bug.
- [varLib] Added support for building, reading, writing and optimizing 32-bit
  ``ItemVariationStore`` as used in COLRv1 table.
- [otBase/otConverters] Add array readers/writers for int types.
- [feaLib] Allow more than one lookahead glyph/class in contextual positioning with
  "value at end".
- [COLRv1] Default varIdx should be 0xFFFFFFFF.
- [pens] Make RecordingPointPen actually pass on identifiers; replace asserts with
  explicit ``PenError`` exception.
- [mutator] Round lsb for CF2 fonts as well.

4.22.1 (released 2021-04-26)
----------------------------

- [feaLib] Skip references to named lookups if the lookup block definition
  is empty, similarly to makeotf. This also fixes an ``AttributeError`` while
  generating ``aalt`` feature.
- [subset] Fixed bug with ``--no-hinting`` implementation for Device tables.
  The previous code was alwyas dropping Device tables if no-hinting was
  requested, but some Device tables (DeltaFormat=0x8000) are also used to encode
  variation indices and need to be retained.
- [otBase] Fixed bug in getting the ValueRecordSize when decompiling ``MVAR``
  table with ``lazy=True``.
- [varLib/glyf/gvar] Optimized and simplified ``GlyphCoordinates`` and
  ``TupleVariation`` classes, use ``bytearray`` where possible, refactored
  phantom-points calculations. We measured about 30% speedup in total time
  of loading master ttfs, building gvar, and saving.
- [subset] Fixed ``AssertionError`` while pruning unused CPAL palettes when
  ``0xFFFF`` is present.

4.22.0 (released 2021-04-01)
----------------------------

- [ttLib] Remove .Format from Coverage, ClassDef, SingleSubst, LigatureSubst,
  AlternateSubst, MultipleSubst.
  ATTENTION: This will change your TTX dumps!
- [misc.arrayTools] move Vector to its own submodule, and rewrite as a tuple
  subclass.
- [docs] Added a terminology section for varLib.
- [varLib] Move rounding to VariationModel, to avoid error accumulation from
  multiple deltas
- [varLib] Explain merge errors in more human-friendly terms
- [otlLib] Correct some documentation
- [varLib/otlLib] Allow merging into VariationFont without first saving GPOS
  PairPos2
- [subset] Improve PairPosFormat2 subsetting
- [ttLib] TTFont.save: create file on disk as late as possible
- [cffLib] Add missing CFF2 dict operators LanguageGroup and ExpansionFactor
  ATTENTION: This will change your TTX dumps!

(adam)

Updated emulators/wine, net/py-netifaces, devel/py-astroid, devel/py-requests-mock

(adam)

py-requests-mock: updated to 1.9.3

1.9.3
* Release note for 1.9.3 type annotations
* Additional improvements to response matching type stubs
* Improve MockerCore type stubs

(adam)

py-astroid: updated to 2.5.7

What's New in astroid 2.5.7?

* Fix six.with_metaclass transformation so it doesn't break user defined transformations.
* Fix detection of relative imports.
* Fix inference of instance attributes defined in base classes
* Update `infer_named_tuple` brain to reject namedtuple definitions
  that would raise ValueError
* Do not set instance attributes on builtin object()
* Fix some spurious cycles detected in ``context.path`` leading to more cases
  that can now be inferred
* Add ``kind`` field to ``Const`` nodes, matching the structure of the built-in ast Const.
  The kind field is "u" if the literal is a u-prefixed string, and ``None`` otherwise.
* Fix property inference in class contexts for properties defined on the metaclass
* Update enum brain to fix definition of __members__ for subclass-defined Enums
* Update random brain to fix a crash with inference of some sequence elements
* Fix inference of attributes defined in a base class that is an inner class
* Allow inferring a return value of None for non-abstract empty functions and
  functions with no return statements (implicitly returning None)
* scm_setuptools has been added to the packaging.
* Astroid's tags are now the standard form ``vX.Y.Z`` and not ``astroid-X.Y.Z`` anymore.
* Add initial support for Pattern Matching in Python 3.10

(adam)

py-netifaces: updated to 0.11.0

Version 0.11.0

* Added notice about needing a new maintainer.
* Fixed link to Travis CI.
* Added test.py to the source distribution.
* Fixed netmask calculation for IPv6.
* Fixes to gateway detection in some edge cases.
* Build CPython 2.7 wheels for 64-bit Windows (yes, you should be
  using Python 3 now, but still).

(adam)

wine: updated to 5.0.5

5.0.5:
Fixes build failure when v4l is missing

5.0.4:
Various bug fixes

5.0.3:
Various bug fixes

5.0.2:
Various bug fixes

5.0.1:
Fix compilation with gcc 10
Add some timezones
Various bug fixes

5.0:
Builtin modules in PE format.
Multi-monitor support.
XAudio2 reimplementation.
Vulkan 1.1 support.

(adam)

Updated devel/py-boltons, sysutils/py-eliot

(adam)

py-eliot: updated to 1.13.0

1.13.0

Features:

@capture_logging and MemoryLogger now support specifying a custom JSON encoder. By default they now use Eliot窶冱 encoder. This means tests can now match the encoding used by a FileDestination.
Added support for Python 3.9.

Deprecation:

Python 3.5 is no longer supported.

(adam)

py-boltons: updated to 21.0.0

21.0.0

Fix OMD.addlist when the added list is empty
Add funcutils.noop, satisfying PEP 559
Support lists for iterutils.bucketize
Python 3.9 test fixes for OMD (PEP 584)
Make typeutils.make_sentinel more pickleable
jsonutils.reverse_iter_lines now works on Py3 and Windows

(adam)

Updated www/py-urllib3, devel/py-dulwich

(adam)

py-dulwich: updated to 0.20.23

0.20.23
* Fix installation of GPG during package publishing.

(adam)

py-urllib3: updated to 1.26.5

1.26.5 (2021-05-26)
-------------------
* Fixed deprecation warnings emitted in Python 3.10.
* Updated vendored ``six`` library to 1.16.0.
* Improved performance of URL parser when splitting
  the authority component.

(adam)

Updated devel/py-click-threading, devel/py-click-repl

(adam)

py-click-repl: updated to 0.2.0

0.2.0:
Unknown changes

(adam)

py-click-threading: updated to 0.5.0

0.5.0:
Unbreak master

(adam)

py-django-formtools: fix a typo

(adam)

Updated devel/cmake, net/iperf3

(adam)

iperf3: updated to 3.10

iperf 3.10 2021-05-26
---------------------

* Notable user-visible changes

  * Fix a bug where some --reverse tests didn't terminate.
  * Responsiveness of control connections is slightly improved.
  * The allowable clock skew when doing authentication between client
    and server is now configurable with the new --time-skew-threshold.
  * Bitrate throttling using the -b option now works when a burst size
    is specified.
  * A bug with calculating CPU utilization has been fixed.
  * A --bind-dev option to support binding sockets to a given network
    interface has been added to make iperf3 work better with
    multi-homed machines and/or VRFs.
  * --pidfile now works with --client mode.
  * The server is now less likely to get stuck due to network errors
    controlled by the new --rcv-timeout option.
  * Fixed a few bugs in termination conditions for byte or
    block-limited tests.
  * Added tcp_info.snd_wnd to JSON output.
  * Some bugs with garbled JSON output have been fixed.
  * Support for setting the IPv4 don't-fragment (DF) bit has been
    added with the new --dont-fragment option.
  * A failure with not being able to read the congestion control
    algorithm under WSL1 has been fixed.
  * Error handling and error messages now make more sense in cases
    where sockets were not successfully opened.
  * Some buffer overflow hazards were fixed.

* Notable developer-visible changes
  * It is now possible to use the API to set/get the congestion
    control algorithm.

(adam)

cmake: updated to 3.20.3

CMake 3.20.3
* Help: Use relative path for IDE Integration guide link to preset schema
* BinUtils: Use more-private temporary variable names
* ObjectiveC: Respect OSX_ARCHITECTURES for OBJC
* FindBoost: Add support for Boost 1.76
* Ninja: Restore support for Fortran in a symlinked build tree
* Utilities/Sphinx: Update man page config for Sphinx 4
* ExternalProject: Ensure git fetch if updating to hash we don't have yet
* ExternalProject: Only add git config setting with git 1.7.7 or later
* Ninja Multi-Config: Split long command lines by config
* CMP0082: Check EXCLUDE_FROM_ALL property at generate time
* GNU: C++17 default version
* GNU: Final C++20 flags
* GNU: C++23 support
* CUDA: improve regex for CUDA Toolkit root from nvcc verbose output
* cmCommandLineArgument: Correctly record parsing failures
* cmake: `--build` and `--install` error out when encountering bad flags
* cmCommandLineArgument: Provide more information syntax error messages
* NVHPC: Support explicit language flags
* NVHPC: Support Ninja dependency scanning

(adam)

Added textproc/py-bracex, textproc/py-wcmatch

(adam)

py-wcmatch: added version 8.2

Wildcard Match provides an enhanced fnmatch, glob, and pathlib library in order
to provide file matching and globbing that more closely follows the features
found in Bash. In some ways these libraries are similar to Python's builtin
libraries as they provide a similar interface to match, filter, and glob the
file system. But they also include a number of features found in Bash's
globbing such as backslash escaping, brace expansion, extended glob pattern
groups, etc. They also add a number of new useful functions as well, such as
globmatch which functions like fnmatch, but for paths.

(adam)

py-bracex: added version 2.1.1

Bracex is a brace expanding library (a la Bash) for Python. Brace expanding is
used to generate arbitrary strings.

(adam)

Updated devel/py-construct, devel/py-cookiecutter

(adam)

py-cookiecutter: updated to 1.7.3

1.7.3:
Fixed jinja2 and markupsafe dependencies

(adam)

py-construct: updated to 2.10.67

2.10.67:
Unknown changes

(adam)

py-django-taggit: mark as not compatible with Python 2.7

(adam)

py-rich: fix typo

(adam)

Updated audio/faudio, security/sudo

(adam)

sudo: updated to 1.9.7

What's new in Sudo 1.9.7

* The "fuzz" Makefile target now runs all the fuzzers for 8192
  passes (can be overridden via the FUZZ_RUNS variable).  This makes
  it easier to run the fuzzers in-tree.  To run a fuzzer indefinitely,
  set FUZZ_RUNS=-1, e.g. "make FUZZ_RUNS=-1 fuzz".

* Fixed fuzzing on FreeBSD where the ld.lld linker returns an
  error by default when a symbol is multiply-defined.

* Added support for determining local IPv6 addresses on systems
  that lack the getifaddrs() function.  This now works on AIX,
  HP-UX and Solaris (at least).

* Fixed a bug introduced in sudo 1.9.6 that caused "sudo -V" to
  report a usage error.  Also, when invoked as sudoedit, sudo now
  allows a more restricted set of options that matches the usage
  statement and documentation.

* Fixed a crash in sudo_sendlog when the specified certificate
  or key does not exist or is invalid.

* Fixed a compilation error when sudo is configured with the
  --disable-log-client option.

* Sudo's limited support for SUCCESS=return entries in nsswitch.conf
  is now documented.

* Sudo now requires autoconf 2.70 or higher to regenerate the
  configure script.

* sudo_logsrvd now has a relay mode which can be used to create
  a hierarchy of log servers.  By default, when a relay server is
  defined, messages from the client are forwarded immediately to
  the relay.  However, if the "store_first" setting is enabled,
  the log will be stored locally until the command completes and
  then relayed.

* Sudo now links with OpenSSL by default if it is available unless
  the --disable-openssl configure option is used or both the
  --disable-log-client and --disable-log-server configure options
  are specified.

* Fixed configure's Python version detection when the version minor
  number is more than a single digit, for example Python 3.10.

* The sudo Python module tests now pass for Python 3.10.

* Sudo will now avoid changing the datasize resource limit
  as long as the existing value is at least 1GB.  This works around
  a problem on 64-bit HP-UX where it is not possible to exactly
  restore the original datasize limit.

* Fixed a race condition that could result in a hang when sudo is
  executed by a process where the SIGCHLD handler is set to SIG_IGN.

* Fixed an out-of-bounds read in sudoedit and visudo when the
  EDITOR, VISUAL or SUDO_EDITOR environment variables end in an
  unescaped backslash.  Also fixed the handling of quote characters
  that are escaped by a backslash.

* Fixed a bug that prevented the "log_server_verify" sudoers option
  from taking effect.

* The sudo_sendlog utility has a new -s option to cause it to stop
  sending I/O records after a user-specified elapsed time.  This
  can be used to test the I/O log restart functionality of sudo_logsrvd.

* Fixed a crash introduced in sudo 1.9.4 in sudo_logsrvd when
  attempting to restart an interrupted I/O log transfer.

* The TLS connection timeout in the sudoers log client was previously
  hard-coded to 10 seconds.  It now uses the value of log_server_timeout.

* The configure script now outputs a summary of the user-configurable
  options at the end, separate from output of configure script tests.

* Corrected the description of which groups may be specified via the
  -g option in the Runas_Spec section.

(adam)

faudio: updated to 21.05

21.05

New Features:

Added SetAudioTrackEXT extension
Update to FNA3D 21.05
Update to FAudio 21.05
Fixes:

SDL2_FNAPlatform:
Changing window focus no longer unsets fullscreen, except on X11
Using "alt-tab" to switch focus out while fullscreen on Windows no longer causes error beeps when attempting to switch focus back in
Fix mouse support on Wayland

(adam)

Added devel/py-dataclasses, comms/py-rich, devel/py-tenacity, comms/py-enrich, sysutils/ansible-lint; Updated devel/py-parameterized

(adam)

py-parameterized: updated to 0.8.1

0.8.1 (2021-01-09)
* Add README and LICENSE to pypi sdist package

0.8.0 (2021-01-04)
* Handle camelCase test names. This is an ever-so-slightly backwards
  incompatible change, as the method for determining which methods are
  test methods has changed from ``method_name.startswith("test_")`` to
  ``method_name.startswith("test")``. The latter is the behavior of
  the ``unittest`` module.
* Fix arguments to skip_on_empty helper

(adam)

ansible-lint: added version 5.0.11

ansible-lint checks playbooks for practices and behaviour that could
potentially be improved. As a community backed project ansible-lint supports
only the last two major versions of Ansible.

(adam)

py-enrich: added version 1.2.6

Enriched extends rich library functionality with a set of changes that were not
accepted to rich itself.

(adam)

py-tenacity: added version 7.0.0

Tenacity is a retrying library to simplify the task of adding retry behavior to
just about anything.

(adam)

py-rich: fix a typo

(adam)

py-rich: added version 10.2.2

Rich is a Python library for rich text and beautiful formatting in the
terminal.

The Rich API makes it easy to add color and style to terminal output. Rich can
also render pretty tables, progress bars, markdown, syntax highlighted source
code, tracebacks, and more - out of the box.

(adam)

py-dataclasses: added version 0.8

This is an implementation of PEP 557, Data Classes. It is a backport for Python
3.6.

(adam)

Updated net/grpc, net/py-grpcio, net/py-grpcio-testing, net/py-grpcio-tools

(adam)

grpc py-grpcio py-grpcio-testing py-grpcio-tools: updated to 1.38.0

Release v1.38.0

Core

Invalidate ExecCtx now before computing timeouts in all repeating timer events using a WorkSerializer or combiner.
Fix use-after-unref bug in fault_injection_filter.
Fix 25897 to avoid crashes when certificates are not yet updated.
gRPC EventEngine Interface.
Allow the AWS_DEFAULT_REGION environment variable.

C++

s/OnServingStatusChange/OnServingStatusUpdate.
cmake: only install channelz if gRPC_USE_PROTO_LITE is off.

C#

Grpc.Tools: Use x64 protoc on macOS arm64.
Provide runtime/framework info in gRPC C# user agent string.

Python

Add grpcio-admin Python package.
Add CSDS API to Python.
Expose code and details from context on the server side.
Explicitly import importlib.abc; required on Python 3.10.
Fix potential deadlock on the GIL in AuthMetdataPlugin.
Introduce new Python package "xds_protos".
also build python3.6 aarch64 manylinux2014 wheel.
Remove async mark for set_trailing_metadata interface.

Ruby

ruby: use x86_64 binary on M1 macs.
Add ruby 3.0 support for mac binary packages.

(adam)

Updated devel/py-zconfig, databases/sqlitebrowser

(adam)

sqlitebrowser: updated to 3.12.2

DB Browser for SQLite 3.12.2

Fix saving the list of extensions in the Preferences dialog
Corrected a typo in the French translation
Updated the included SQLite and SQLCipher libraries to their latest release (SQLite 3.35.5, SQLCipher 4.4.3)
Updated the "public" certificate, used for communicating with DBHub.io anonymously

(adam)

py-zconfig: updated to 3.6.0

3.6.0 (2021-05-19)
==================

- Added support for Python 3.8, 3.9 and 3.10.  This primarily involves
  avoiding the new-in-3.8 validation of the format string when using the
  'safe-template' format style, since that's not supported in the Python
  standard library.

- Added ``ZConfig.pygments`` module containing a lexer compatible with
  the ``pygments`` library.  Made discoverable via an entry point; use
  **zconfig** as the highlight language for ``code-block`` directives in
  Sphinx documents.

(adam)

Updated finance/bitcoin, security/py-google-auth

(adam)

py-google-auth: updated to 1.30.1

1.30.1

Bug Fixes

allow user to customize context aware metadata path in _mtls_helper
fix function name in signing error message

(adam)

bitcoin: updated to 0.21.1

0.21.1 change log

Consensus

* Speedy trial support for versionbits (ajtowns)
* Speedy trial activation parameters for Taproot (achow101)

P2P protocol and network code

* allow CSubNet of non-IP networks (vasild)
* Avoid UBSan warning in ProcessMessage(…) (practicalswift)

Wallet

* Introduce DeferredSignatureChecker and have SignatureExtractorClass subclass it (achow101)
* Avoid requesting fee rates multiple times during coin selection (achow101)

RPC and other APIs

* Disallow sendtoaddress and sendmany when private keys disabled (achow101)

Build system

* link against -lsocket if required for *ifaddrs (fanquake)
* Fix MSVC build after gui-176 (hebasto)

Tests and QA

* Add fuzzing harness for versionbits (ajtowns)
* fuzz: Bump FuzzedDataProvider.h (MarcoFalke)
* fuzz: Update FuzzedDataProvider.h from upstream (LLVM) (practicalswift)
* Update vcpkg checkout commit (sipsorcery)
* fuzz: Bump FuzzedDataProvider.h (MarcoFalke)
* Fix the unreachable code at feature_taproot (brunoerg)
* Test that a fully signed tx given to signrawtx is unchanged (achow101)
* Make sure non-IP peers get discouraged and disconnected (vasild, MarcoFalke)
* fuzz: cleanups for versionbits fuzzer (ajtowns)

(adam)

Updated graphics/libimagequant, graphics/pngquant, sysutils/ansible-base, devel/py-more-itertools

(adam)

py-more-itertools: updated to 8.8.0

8.8.0
-----

* New functions
    * :func:`countable`

* Changes to existing functions
    * :func:`split_before` was updated to handle empy collections
    * :func:`unique_everseen` got a performance boost
    * The type hint for :func:`value_chain` was corrected

(adam)

ansible-base: updated to 2.10.10

v2.10.10

Bugfixes

- Correctly set template_path and template_fullpath for usage in template lookup and action plugins.
- Fix fileglob bug where it could return different results for different order of parameters (https://github.com/ansible/ansible/issues/72873).
- Improve resilience of ``ansible-galaxy collection`` by increasing the page size to make fewer requests overall and retrying queries with a jittered exponential backoff when rate limiting HTTP codes (520 and 429) occur. (https://github.com/ansible/ansible/issues/74191)
- ansible-test - Use documented API to retrieve build information from Azure Pipelines.
- ansible.builtin.cron - Keep non-empty crontabs, when removing cron jobs (https://github.com/ansible/ansible/pull/74497).
- ansible_test - add constraint for ``MarkupSafe`` (https://github.com/ansible/ansible/pull/74666)
- filter plugins - patch new versions of Jinja2 to prevent warnings/errors on renamed filter decorators (https://github.com/ansible/ansible/issues/74667)
- service - compare version without LooseVersion API (https://github.com/ansible/ansible/issues/74488).

(adam)

pngquant: updated to 2.15.0

version 2.15
- speed and quality improvements

(adam)

libimagequant: updated to 2.15.1

version 2.15
- speed and quality improvements

(adam)

Updated www/py-django-binary-database-files, www/py-django-admin-rangefilter, www/py-django-contrib-comments, www/py-django-taggit, www/py-django-formtools, www/py-django-admin-sortable2

(adam)

py-django-formtools: point to py-django2 in DEPENDS

(adam)

py-django-taggit: point to py-django2 in DEPENDS

(adam)

py-django-contrib-comments: point to py-django2 in DEPENDS

(adam)

py-django-admin-sortable2: updated to 1.0

1.0
Added support for Django 3.2
Drop support for Python-2.7, 3.4 and 3.5.
Drop support for Django-1.10, 1.11, 2.0 and 2.1.
Add Python-3.9 to the testing matrix.
Refactor code base to clean Python-3 syntax.

(adam)

py-django-formtools: updated to 2.3

2.3
- Dropped testing for Django 1.11, 2.0 and 2.1.
- Added support for Django 3.1 and Python 3.9.
- Added support for Django 3.2.
- Dropped support for Django 3.0.
- Dropped support for Python 3.5.

(adam)

py-django-taggit: updated to 1.4.0

1.4.0

* Add Python 3.9 support.
* Remove Python 3.5 support.
* Add Django 3.2 support.
* Remove Django 1.11 and 3.0 support.
* Add Danish translation.
* Fix crashing that could occur with ``similar_objects`` in multi-inheritance contexts.
* Add support for custom fields on through table models with `through_defaults` for ``TaggedManager.add`` and ``TaggedManager.set``.

(adam)

py-django-contrib-comments: updated to 2.1.0

2.1.0

Dropped support for Django 1.11, 2.0, and 2.1.
Added the delete_stale_comments management command.
Added db_index to object_pk and is_removed fields.
Altered object_pk from TextField to CharField(max_length=64) so that the field can be indexed on MySQL, too. Warning: if you attach comments to objects whose primary key is serialized to more than 64 characters, you should provide a custom Comment model (more about that in the documentation) with an appropriate object_pk field.
Confirmed support for Python 3.9.
Added support for Django 3.2.

(adam)

py-django-admin-rangefilter: updated to 0.8.1

0.8.1:
- Fixed problem with `ManifestStaticFilesStorage`

0.8.0:

Added
- Compatibility Django 3.2

Changed
- Module `filter` rename to `filters`

Fix
- Fixed static loads

(adam)

py-django-binary-database-files: updated to 1.0.15

1.0.15:
Unknown changes

(adam)

Updated devel/py-radon, devel/py-dulwich, textproc/py-JWT, devel/py-cpplint

(adam)

py-cpplint: updated to 1.5.5

1.5.5:
Fix 172: Added 'size_t' to typecasts detected by CheckCStyleCast
Fixed wrong CLI help text: Each filter needs + or -
Fix 164: add elif as an exception for CheckSpacingForFunctionCall()
Fix google346: --root option not working on windows due to slashes in path

(adam)

py-JWT: updated to 2.1.0

v2.1.0

Changed
- Allow claims validation without making JWT signature validation mandatory.

Fixed
- Remove padding from JWK test data.
- Make `kty` mandatory in JWK to be compliant with RFC7517.
- Allow JWK without `alg` to be compliant with RFC7517.
- Allow to verify with private key on ECAlgorithm, as well as on Ed25519Algorithm.

Added
- Add caching by default to PyJWKClient
- Add missing exceptions.InvalidKeyError to jwt module __init__ imports
- Add support for ES256K algorithm
- Add `from_jwk()` to Ed25519Algorithm
- Add `to_jwk()` to Ed25519Algorithm
- Export `PyJWK` and `PyJWKSet`

(adam)

py-dulwich: updated to 0.20.22

0.20.22
* Prevent removal of refs directory when the last ref is
  deleted.
* Fix filename: MERGE_HEADS => MERGE_HEAD.
* For ignored directories, porcelain.add and porcelain.status now only return
  the path to directory itself in the list of ignored paths. Previously, paths
  for all files within the directory would also be included in the list.
* Provide depth argument to ``determine_wants``.
* Various tag signature handling improvements.
* Add separate Tag.verify().
* Add support for version 3 index files.
* Fix autocrlf=input handling.
* Attempt to find C Git global config on Windows.

API CHANGES
* The APIs for writing and reading individual index entries have changed
  to handle lists of (name, entry) tuples rather than tuples.

(adam)

py-radon: updated to 4.5.2

4.5.2:
- Add back `flake8-polyfill` dependency because conditional extra does not work
  as intended

(adam)

postgresql95: removed; EOL

(adam)

postgresql95: removed; EOL

(adam)

Removed databases/postgresql95; Updated multimedia/mkvtoolnix

(adam)

mkvtoolnix: updated to 57.0.0

Version 57.0.0 "Till The End" 2021-05-22

New features and enhancements

* mkvmerge: MP4 reader: added support for reading Dolby Vision from MP4 files
  (FourCCs `dvh1` and `dvhe`; configuration records `dvcC`, `dvvC` and `hvcE`
  will be converted into block addition mappings).
* mkvmerge: SRT subtitles: mkvmerge now accepts empty text files with the
  extension `.srt` as SRT subtitle files, enabling the creation of empty SRT
  tracks.
* mkvmerge, mkvextract: HEVC/H.265: access unit delimiter NALUs will no longer
  be discarded, neither during muxing nor during extraction.
* MKVToolNix GUI: preferences: switched the order & wording of controls in the
  "enabling items" panel to make it clearer that certain controls define
  exceptions.

Bug fixes

* mkvmerge: HEVC/H.265 parser: several NALU types, notably the Dolby
  Vision-specific NALUs ("unspecified 62" and "unspecified 63") and suffix SEI
  NALUs, are now stored with the frame they belong to instead of with the next
  frame.
* mkvmerge: HEVC/H.265 packetizer: fixed setting the track's default duration
  when reading HEVC/H.265 from Matroska files that don't have a default
  duration set.
* mkvmerge: HEVC/H.265 packetizer: fixed the calculation of the duration of
  frames so that `SimpleBlock` elements can be used again instead of
  `BlockGroups` with `BlockDuration` elements.
* mkvmerge, mkvextract: HEVC/H.265 parser: fixed issues with ordering &
  duplication of certain NALUs (parameter set & prefix SEI NALUs).
* MKVToolNix GUI: multiplexer: when dragging & dropping files to the
  multiplexer, the source directory will be remembered as the "last open
  directory" again, causing subsequent uses of the "open file" dialog to start
  in the same directory.
* mkvinfo: fixed the position of frames in block groups containing "block
  duration" elements in summary mode (e.g. for subtitle tracks).

Build system changes

* `configure` will now try to detect `libmagic` via `pkg-config` and fall back
  to including & linking directly if it cannot be found via `pkg-config`.

(adam)

Adjust PGSQL_VERSIONS_ACCEPTED

(adam)

postgresql95: removed; EOL

(adam)

Updated www/nginx, biology/py-dnaio

(adam)

py-dnaio: updated to 0.5.1

v0.5.1
Add py.typed and distribute .pyi files

(adam)

nginx: updated to 1.20.0

Changes with nginx 1.20.0
*) 1.20.x stable branch.

(adam)

Updated net/haproxy, time/py-cftime

(adam)

py-cftime: updated to 1.5.0

version 1.5.0 (release tag v1.5.0.rel)
======================================
* clean-up deprecated calendar specific subclasses.
* added string formatting support to `cftime.datetime` objects
  (via `cftime.datetime.__format__`).
* add support for astronomical year numbering (including year zero) for
  real-world calendars using 'has_year_zero' cftime.datetime kwarg.
  Default is False for 'real-world' calendars ('julian', 'gregorian'/'standard',
  'proleptic_gregorian'). Ignored for idealized calendars like '360_day
  (they always have year zero).
* add "change_calendar" cftime.datetime method to switch to another
  'real-world' calendar. Enable comparison of cftime.datetime instances
  with different 'real-world' calendars (using the new change_calendar method)
* remove legacy `utime` class, and legacy `JulianDayFromDate` and
  `DateFromJulianDay` functions (replaced by `cftime.datetime.toordinal`
  and `cftime.datetime.fromordinal`).
* Change ValueError to TypeError in __sub__

(adam)

haproxy: updated to 2.4.0

What's new in 2.4 compared to 2.3 ? Lots of cool stuff! The work was spread
on several fronts:
  - performance and latency: a lot of work was done on locking reduction,
    bringing 2.4 a much lower latency than 2.3 had, and a higher scalability.
    2.4 is NUMA-aware on Linux and will avoid binding to cross-socket cores
    by default. Idle server-side connections are now reusable even when SNI,
    proxy protocol or transparent proxy are used, saving costly round trips.
    Lua now supports lock-free multi-threading. Idle frontend connections
    will now be closed as soon as possible during reloads.

  - reliability: "strict-limits" is on by default, experimental features
    are protected against accidental use from a copy-paste, a diagnostic
    mode was added to report suspicious or uncommon constructs that could
    be mistakes, new live debugging tools were added for better in-field
    debugging. Backend idle connections will be closed before quitting
    so that they don't consume source ports in TIME_WAIT.

  - management and integration: maps and ACLs now support atomic updates,
    server-side SSL certificates can also be updated without reloading,
    servers can be added/removed without reloading (still experimental),
    more server settings can be edited at run time, global variables can
    be listed/edited at run time, "defaults" section can now be named,
    reused an inherited, line numbers and file names can be included in
    any command, file paths can now be relative to the config file, some
    new ".if/.elif/.else/.endif" config directives permit a smoother
    transition between versions using a same config file (particularly
    useful for automated rollbacks),

  - interoperability / protocol support: WebSocket over HTTP/2 (RFC8441)
    is now supported on both sides, regardless of the version on the other
    side. The cache now supports the "Vary" header with a few commonly
    used headers, including "Accept-encoding" which gets normalized for
    optimal cache hit ratio. The Prometheus exporter got a significant
    liftup, requires less tricks on the Prometheus side, and supports
    listing only certain metrics for faster retrieval. Optional native
    support for Opentracing was also integrated (via USE_OT=1). The DNS
    resolvers now support talking to servers over TCP. Basic support for
    extracting information from MQTT and FIX protocol was added. Timeouts
    can now be adjusted on the fly and per-request in order to adapt to
    particuarly slow servers or special protocols.

  - user-friendliness: TCP loggers now do not require a ring section
    declaration anymore, it's transparently done now. Header deletion
    using a pattern matching on the name is now supported instead of
    having to list many names. Checking for real server-side errors got
    easier than before with http_fail_cnt/http_fail_rate. The config
    parser and CLI will now provide suggestions when some keywords are
    misspelled. The CLI's help output can now be filtered to subsets of
    commands and will be sorted for easier finding.

  - processing: the new "wait-for-body" HTTP action allows to wait for
    a request or response body up to a certain size or delay. This is
    convenient to detect POST contents or to detect error patterns or
    information leaks in responses. TCP frontends can now be manually
    upgraded to HTTP, allowing HTTP rules to be used in a frontend in
    a tcp->http scenario. JSON parameters can be decoded to extract some
    fields values. A URL encoder is now available to pass some fields in
    a way that is suitable for query string parameters. Base64 variants
    are now supported (such as the URL-compatible one used by JWT).

  - for distros: building optional addons will be easier thanks to the
    removal of the contrib/ directory and the cleaner integration with
    the regular build process and include files. Libslz was included by
    default, simplifying the packaging work. A few example configs were
    added into examples/ to serve as starters for users. A CPU entry was
    added to produce executables that works well both on old and new
    ARMv8 CPUs.

  - observability: more stats were added (SSL, H2). "show info" can now
    emit floating point values for rates and uptimes if requested.

(adam)

Updated www/libsass, www/sassc

(adam)

sassc: updated to 3.6.2

This is the SassC for LibSass 3.6.5

(adam)

libsass: updated to 3.6.5

3.6.5:
Fix extend edge case going endlessly
Fix source-maps and how we count unicode characters
Fix seed generator if std::random_device fails
Fix url() containing exclamation mark causing an error
Fix Offset initialization when end was not given
Fix obvious backporting error in pseudo extend
Fix obvious identical subexpressions in op_color_number
Fix edge case regarding unit-less number equality as object keys
Revert compound re-ordering for non extended selectors
Prevent compiler warning about unnecessary copy
Replace Travis CI with GitHub Actions

(adam)

Updated lang/nodejs, archivers/unrar

(adam)

unrar: updated to 6.0.6

6.0.6:
Unknown changes

(adam)