doc: Updated lang/python39 to 3.9.9nb2

(wiz)

python38: prepare for upcoming setuptools change

Bump PKGREVISION.

(wiz)

python39: prepare for upcoming setuptools change

Bump PKGREVISION.

(wiz)

doc: Updated emulators/mame to 0.239

(wiz)

mame: update to 0.239.

Did you think we窶囘 let 2021 finish without a parting MAME release?
MAME 0.239 is here, just in time for the new year. This release
includes a fix for many subtle and not-so-subtle sound and music
timing issues in games using Yamaha FM synthesis chips. The frame
rate for Gaelco games has been adjusted to satisfy some wily
protection checks, fixing crashes when continuing in Thunder Hoop
and graphical issues in Squash. A big update for Philips CD-i
emulation just made it in for this release, greatly improving the
experience in a lot of games. Nintendo Famicom Disk System emulation
has also seen some improvements this month.

This release is packed with even more Soviet re-skins of the Game
& Watch Egg program, the latest Apple II dumps and cracks, another
batch of Commodore 64 cassettes, and more exotic NES and Famicom
cartridges. Milan Galcik, who窶冱 been busy with the Elektronika
hand-held games, has also completed a Slovak UI translation and
updated the neglected Czech translation. Both genuine and cloned
Apple II systems have had emulation updates this month, with a
number of unique VTech Laser and Franklin ACE features now supported,
and performance improvements for the Apple IIgs.

(wiz)

bulk-medium: add php80, ride bump

This previously built two versions of php, so let's continue to do that.

(wiz)

doc/TODO: add some

+ avidemux-2.8, calibre-5.34.0, enlightenment-0.25, mame-0.239.

(wiz)

doc/TODO: + lumina-1.6.2, phoronix-test-suite-10.8.

(wiz)

wasi-libcxx: add missing include for wasi-libc

Fixes build.

(wiz)

doc/TODO: add some

+ ImageMagick-7.1.0.19, asciidoc-10.1.1, chromaprint-1.5.1,
  double-conversion-3.1.7, libmicrohttpd-0.9.74, py-async-timeout-4.0.2,
  rust-analyzer-0.0.2021.12.20, vala-0.54.5.

(wiz)

doc: Updated devel/py-boost to 1.77.0nb2

(wiz)

py-boost: revert previous

Re-enable py-numpy dependency, since it's used when it's installed.

However py-numpy for python 2.x seems broken and the numpy extension
is not built for that version, so mark the extension as "for python 3.x"
in the PLIST.

Tested with python27 and python39.

Bump PKGREVISION.

(wiz)

py-pywebdav: remove

Only supports python 2.7, and was only used by trytond packages.
Last release from 2012, dead upstream.

(wiz)

py-trytond-webdav: remove

This module does not exist for the latest trytond release.

(wiz)

py-trytond-calendar: remove

This module does not exist for the latest trytond release.

(wiz)

py-trytond-calendar-classification: remove

This module does not exist for the latest trytond release.

(wiz)

py-trytond-calendar-scheduling: remove

This module does not exist for the latest trytond release.

(wiz)

py-trytond-calendar-todo: remove

This module does not exist for the latest trytond.

(wiz)

py-trytond-party-vcarddav: remove

This module does not exist for the latest trytond release.

(wiz)

py-tryton-platform: mark as not for python 2.x

(wiz)

doc/TODO: + krita-5.0, libarchive-3.5.2, ogre-2.3, supertux-0.6.3.

(wiz)

py-pywebdav: revert incorrect change

(wiz)

py-trytond-webdav: disable py-webdav dependency

The dependency only exists for python 2.7.

Hopefully unbreaks pbulk

(wiz)

fail2ban: fix build with latest setuptools.

Fixes PR 56572 by nia@

Fix pkglint while here.

(wiz)

apache-solr: use BLAKE2s

(wiz)

py-trytond: update options.mk and actually include it

(wiz)

py-trytond: make parallel-installable again, update ALTERNATIVES

ride update

(wiz)

py-trytond: fix dependencies

Fix distfile checksum. Finish update.

(wiz)

py-tryton: fix dependencies so that 'make test' passes

Ride update

(wiz)

py-tryton: remove some options

These three modules are not referenced from the 6.2 sources.

(wiz)

doc/TODO: + gimp-2.10.30, zstd-1.5.1.

(wiz)

doc: Added devel/taplo-lsp version 0.2.6

(wiz)

devel/Makefile: + taplo-lsp

(wiz)

devel/taplo-lsp: import taplo-lsp-0.2.6

Taplo LSP is a language server protocol implementation for the TOML
format.

Packaged by pin and myself for wip.

(wiz)

doc: Added www/py-webdav version 1.1.7

(wiz)

www/Makefile: + py-webdav

(wiz)

www/py-webdav: import py-webdav-1.1.7

WebDAV client library for Python 3

Imported during freeze as possible helper for py-trytond update.

(wiz)

py-proteus: fix build with latest setuptools

(wiz)

py-pip-api: does not support python 2.x

(wiz)

py-flask-flatpages: fix suspected typo

kygments -> pygments

(wiz)

doc: Updated finance/gnucash-docs to 4.9

(wiz)

gnucash-docs: update to 4.9.

4.9  - 19 December 2021
        o Bug 797950 - Reconcilation docs don't mention automatic
      credit card payment feature.
        o Guide:C: Add a directory with datafiles for faster regeneration of
          images
        o Chapter "Online-Quotes" created.
          Description of the installation and configuration of F::Q (moved
          from Help_ch_Account-Actions-xml).
        o Remove several Autotools remains
        o Drop TravisCI as we are using Github Workflows now.
        o Removed the information of HACKING file from README.
        o Add ghelp to the default target
          At some point in the past ghelp didn't have to be built, as one could
          develop and test simply from the source directory. That is no longer
          the case so perform a build by default if ghelp is enabled.

(wiz)

doc: Updated finance/gnucash to 4.9

(wiz)

gnucash: update to 4.9.

4.9 - 19 December 2021
The tenth release of the 4.x Stable Series
Between 4.8 and 4.9, the following bugfixes were accomplished:

    Bug 797502 - (RTL) - Right edge of the reports are locked while scrolling down

    Running gnucash in Hebrew, the right edge of the reports are locked while scrolling down and do not scroll with the rest of the report.
    Bug 797618 - Spanish translation of Shares and Stock
    Bug 797678 - OFX importer should supplement, not replace, existing Notes (and Description and Memo) fields

    Add an "Append" checkbox to the bottom of the "Generic import transaction matcher" window to the left of the "Reconcile after match" checkbox. When ticked, this causes the imported Description/Notes to be appended to the matched transaction Description/Notes respectively. The selected ticked/unticked state of the "Append" checkbox is saved in a key value pair for the import account, so the next import for that account will automatically default it to the saved state. As these mods are limited to the code for the matcher window, this should work for all the imports that use it - ie ofx & csv file imports (both tested) & aqbanking (cannot test).
    Bug 797772 - Allow reconciliation report to double filter by both posting and reconciliation date
    Bug 798222 - Account search dialog: Pushing enter does nothing
    Bug 798266 - Dropdown picker menu for category does not appear
    Bug 798313 - "File/Import/Import Transactions From CSV" crashes Gnucash.
    Bug 798320 - Error message indicating a crash of GNUcash when closing application - MacOS 10.15.7 (19H1419)
    Bug 798325 - New Currency for Venezuela (VED)
    Bug 798327 - Ofx import stops too early for multi-account OFX and one account has no new transactions
    Bug 798335 - No longer possible to change trading account

    Strip trading splits only if "Use Trading Accounts" is enabled for the book so that users who want to manage trading accounts by hand may do so.
    Bug 798341 - Right align in Hebrew - Account screen
    Bug 798346 - crashes when running check & repair
    Bug 798352 - Decimal Precision when Entering Mutual Fund Transaction

    Let xaccParseAmount parse up to 12 decimal places instead of 8.
    Bug 798382 - Typo corrections in comments

The following fixes and improvements were not associated with bug reports:

    More memory leak plugging and GList efficiency improvements.
    Fix build with cmake older than 3.18
      3.18 introduced 'cmake -E cat', but we still have to support versions
      starting from 3.14.
    [account-piecharts] don't use gnc:make-internal-option
    Several improvements to the experimental IFRS cost-basis report.
    Fix two minor issues with displaying links in the Document Link dialog.
    Fix obsolete links to bugzilla.gnome.org in translations.
    L18N: Fix glossary/es.po:906: duplicate message definition...

    901: ...this is the location of the first definition
    [budget-flow] delay creating exchange-fn until a valid budget exists

    because it needs budget period end-date.
    [options.scm] Addon previous commit... need to initialize with guid
    [options.scm] gnc:make-budget-option stores guid instead of object

    because object may become stale if UI is used to delete it, leading to stale pointer and segfault. storing guid is safer, and will return null if budget is deleted.
    Support SEPA internal transfer
    Fix issue with failure to run reconciliation with all accounts present in a multi-account OFX file. Do that by saving a GList of statements, rather than a pointer to a single one. Also freeing of info happens during the call to process_next_file.
    Import of OFX files with many securities opens too many matching dialogs Because ofx import is currently split per target account, and since each security has its own accounts, importing such OFX is a tedious process. The fix is to only split the transactions if we identify a potential transfer, currently based on amount, date and accounts. To do that, we insert transactions one by one into a list, making sure we have not already inserted one that has the same date, and the same absolute amount. If we have, we keep this potential transfer for a second phase. A naive approach would loop through added transactions for each new transaction by that ends up being O(N^2), which matters if we have many transactions. Instead, I'm using a hash to make this O(N log N).
    [date-utilities.scm] remove unused date selectors code & string
    Adjust GNC_DEFINE_TYPE_EXTENDED_END
      To match new signature of g_once_init_enter and g_once_init_leave.
    glib headers should not be included with 'extern "C"'.
    Add explanatory note for gnc_add_scheme_deprecated_module test code

    [test-engine-extras] don't create splits of opposing amt/value signs
      Splits' value and amount should never be of opposite signs. Fix
      test book which had created this incorrect value/amount
      pair. Strengthen env-create-multisplit-transaction to disallow
      test book errors.
    [trep-engine] new option "Date Filter" to specify dates to filter by
    [dialog-sx-since-last-run] don't allocate repeatedly to combine strings
    [test-transaction] "Report Currency" renamed to "Report's currency"
    Separate the warnings for "cut split/transaction" and
      "delete split/transaction"
    [gnc-autoclear] Move autoclear algorithm into gnome-utils
    GSettings - fix GObject warning when old prefs-version is unset
      This will happen when current master (future 5.0) has been run
      at least once.
    [utest-gnc-pricedb] add tests for gnc_price_list_equal
    [reconcile-view][bug] dr/cr query tests amount instead of value
      Because split->amount and split->value are generally of equal
      sign, EXCEPT in stock transactions whereby split->value can be
      zero and split->amount is non-zero. e.g. a stock split
      transaction. Querying the split amount instead will allow stock
      accounts with stock splits to be reconciled.
    [split-register-model.c] Hide price==1 for zero-amount splits
      mainly of interest in zero-amount zero-value splits in the stock
      register eg dividend splits.
    [core-utils] introduce gnc:format
    (gnc:format str [binding value]...) str will contain ${binding} which will
      be replaced to value.
    [test-core-utils] uses srfi-64
    [import-main-matcher] show destination acct for auto-match
    GSettings - add 'deprecate' and 'obsolete' conversions for user preferences
      'deprecate' is technically a noop. It serves to remind
      maintainers the 'deprecated' preference is to be obsoleted in
      the next major release. 'obsolete' goes one step further in that
      it will cause gnucash to reset the preference, effectively
      clearing the value stored in the preferences backend. This is
      the final phase of a preference. Following this it will be
      completely removed from the GSettings schema in the next major
      release. Notes * 'deprecate' and 'migrate' are related. Both are
      a reminder the preference is to be obsoleted in the next major
      release. 'deprecate' does only that though while 'migrate' will
      also trigger a copy of the old value to a new location in the
      databse. * This commit readds a couple of preferences that had
      been removed in the past to be able to properly obsolete them
      (and to test the obsoleting code)

(wiz)

py-multipart: not for python 2.7

(wiz)

py-mod_wsgi: add patch to distinfo

(wiz)

py-mod_wsgi: remove reference to non-existent header

To fix build

(wiz)

py-l18n: fix build with latest setuptools

(wiz)

gssdp12: remove unknown meson option to make this build with latest meson

(wiz)

doc: Updated multimedia/libtheora-docs to 1.1.1nb5

(wiz)

libtheora-docs: match PLIST with installed files

Bump PKGREVISION.

(wiz)

py-numexpr: mark as not for python 2.7

(wiz)

py-cxfreeze: fix build with python 3.10

Clean up some pkglint while here.

(wiz)

py-utils: mark as not for python 2.7

Also mark its users that way

(wiz)

py-pip: not for python 2.x

(Use py-pip20 instead)

(wiz)

doc: Updated devel/py-pbr to 5.8.0

(wiz)

py-pbr: update to 5.8.0.

5.8.0
-----

* Add python2 testing back to PBR
* Allow PEP517 without setup_requires
* Clarify the need for setup.py with PEP517

5.7.0
-----

* Add a PEP517 interface
* PBR package testing improvements
* Run python3.9 test jobs
* Retire django-openstack-auth
* Correct comment for 'D1_D2_SETUP_ARGS'

5.6.0
-----

* Reverse ordering of 'D1_D2_SETUP_ARGS'
* Add test for cfg -> py transformation
* Don't pass empty 'long_description'
* Move flake8 as a pre-commit local target
* Map requires-python to python-requires (attempt 2)
* Update dependencies to avoid failure with old pip
* Increase OS_TEST_TIMEOUT to 1200
* Prevent test failure due to use of setpref
* util: Convert 'D1_D2_SETUP_ARGS' to a list of tuples
* Revert "Map requires-python to python-requires"
* Dropping lower constraints testing
* Adding pre-commit

5.5.1
-----

* Run tempest-full for stable/train
* Remove use_2to3 backward compat for Setuptools
* More easy_install.ScriptWriter.get_header()

5.5.0
-----

* Remove bdist_wininst support
* Increase integration test timeout
* Add Release Notes to documentation
* Cleanup old legacy devstack-gate jobs
* Begin work to modernize pbr's integration testing
* Re-add ChangeLog
* Update some url to use opendev.org
* Support newer openstackdocstheme
* Use easy_install.ScriptWriter.get_header()
* Remove neutron-fwaas from the jobs' required project
* Update python requires packaging metadata for package
* trivial: Improve logging of run commands
* Map requires-python to python-requires
* Update hacking
* Add support for virtualenv 20.x

(wiz)

py-lazy-object-proxy: mark as not for python 2.7

(wiz)

py-gdbm: fix build with python 3.10

(wiz)

py-mpd2: not for python 2.x

(wiz)

doc/TODO: add some

+ ImageMagick-7.1.0.18, asciidoc-10.1.0, gopls-0.7.4, grpc-1.43.0,
  php74-7.4.27, php81-8.1.1, restish-0.9.2, rust-analyzer-0.0.2021.12.13.

(wiz)

accelerator3d: remove

Needs python 2.7, but py-OpenGL does not support this any longer.

(wiz)

doc: Updated devel/nspr to 4.33

(wiz)

nspr: update to 4.33.

NSPR 4.33 contains the following changes:
- build fixes for Firefox 96

(wiz)

doc: Updated net/tor to 0.4.6.9

(wiz)

tor: update to 0.4.6.9.

Changes in version 0.4.6.9 - 2021-12-15
  This version fixes several bugs from earlier versions of Tor. One important
  piece is the removal of DNS timeout metric from the overload general signal.
  See below for more details.

  o Major bugfixes (relay, overload):
    - Don't make Tor DNS timeout trigger an overload general state.
      These timeouts are different from DNS server timeout. They have to
      be seen as timeout related to UX and not because of a network
      problem. Fixes bug 40527; bugfix on 0.4.6.1-alpha.

  o Minor feature (reproducible build):
    - The repository can now build reproducible tarballs which adds the
      build command "make dist-reprod" for that purpose. Closes
      ticket 26299.

  o Minor features (compilation):
    - Give an error message if trying to build with a version of
      LibreSSL known not to work with Tor. (There's an incompatibility
      with LibreSSL versions 3.2.1 through 3.4.0 inclusive because of
      their incompatibility with OpenSSL 1.1.1's TLSv1.3 APIs.) Closes
      ticket 40511.

  o Minor features (fallbackdir):
    - Regenerate fallback directories generated on December 15, 2021.

  o Minor features (geoip data):
    - Update the geoip files to match the IPFire Location Database, as
      retrieved on 2021/12/15.

  o Minor bugfixes (compilation):
    - Fix our configuration logic to detect whether we had OpenSSL 3:
      previously, our logic was reversed. This has no other effect than
      to change whether we suppress deprecated API warnings. Fixes bug
      40429; bugfix on 0.3.5.13.

  o Minor bugfixes (relay):
    - Reject IPv6-only DirPorts. Our reachability self-test forces
      DirPorts to be IPv4, but our configuration parser allowed them to
      be IPv6-only, which led to an assertion failure. Fixes bug 40494;
      bugfix on 0.4.5.1-alpha.

  o Documentation (man, relay):
    - Missing "OverloadStatistics" in tor.1 manpage. Fixes bug 40504;
      bugfix on 0.4.6.1-alpha.

(wiz)

py-cyclonedx-*: mark as not for python 2.7

(wiz)

doc: Updated math/py-rpy to 3.4.5

(wiz)

py-rpy: update to 3.4.5.

Release 3.4.5
=============

Changes
-------

- The deprecation warning when using :func:`rpy2.robjects.lib.grid.activate`
  was missing (indirectly revealed through issue #804).

- The named argument `LINPACK` in :meth:`rpy2.robjects.vectors.Matrix.svd`
  is no longer present in R.

Bugs fixed
----------

- SIGPIPE sent to a process running Python+rpy2 could result in a segfault.
  This was caused by an incorrect setting of R signal handlers (issue #809).

Release 3.4.4
==============

Changes
-------

- `RRuntimeError` exceptions raised while evaluating R code
  an R magic (ipython/jupyter) are now propagated (issue #792).

Release 3.4.3
=============

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

- :mod:`rpy2.robjects.lib.ggplot2` maps more functions in the
  R package (issue #767)

- Utility function :func:`rpy2.robjects.lib.ggplot2.dict2rvec`
  to convert a Python `Dict[str, str]` into an R named vector
  of strings.

Bugs fixed
----------

- Calling mod:`rpy2.situation` to report on the environment no longer
  stops with an uncaught exception when no R home can be determined
  (issue #774)

- Converting pandas series with the older numpy types could result
  in an error (issue #781)

- Numpy converter was not properly turing R integer or float arrays
  into their numpy equivalent (issue #785)

- The HTML representation of R list without named element was
  incorrect (issue #787)

Release 3.4.2
=============

Bugs fixed
----------

- Multithreading during the initialization of the embedded R no longer
  triggers a fatal error (issue #729)

Changes
-------

- :mod:`pytest` is now an optional package. Optional sets of packages are
  `numpy`, `pandas`, `test`, and `all` (all optional packages). They
  can be specified during the installation. For example
  `pip install rpy2[test]`. (issue #670)

Release 3.4.1
=============

Bugs fixed
----------

- The file `requirements.txt` was missing from the source distribution
  on pypi (issue #764).

Release 3.4.0
=============

New Features
------------

- The mapping of the R C API now includes `Rf_isSymbol()`.

- Singleton class :class:`rpy2.rinterface_lib.sexp.RVersion` to report
  the R version for the embedded R.

- :func:`rpy2.rinterface.local_context` to create a context manager
  to evaluate R code within a local environment.

- The `staticmethod` :meth:`rpy2.robjects.vectors.DateVector.isrinstance`
  will tell whether an R objects is an R `Date` array.

Changes
-------

- The dynamic generation of docstrings for R man pages
  is now using R's `Rd2txt`.

- The :func:`rpy2.rinterface_lib._rinterface_capi._findVarInFrame`
  is replaced by the function
  :func:`rpy2.rinterface_lib._rinterface_capi._findvar_in_frame`
  (see fix to issue #710).

- The functions :func:`rpy2.robjects.numpy.activate()` and
  :func:`rpy2.robjects.pandas.activate()` are deprecated and will
  be removed in rpy2-3.5.0.

- :func:`rpy2.rinterface_lib.embedded.setinitialized` was renamed to
  :func:`rpy2.rinterface_lib.embedded._setinitialized` to indicate that
  one should not use it.

- :meth:`rpy2.robjects.lib.ggplot2.vars` to map the R function
  `ggplot2::vars` (issue #742).

- Report correctly the class of R matrix objects with R>=4.0: it is
  now `('matrix', 'array')`. With R<4.0 `('matrix')` is still reported.

- The conversion of R/rpy2 objects to python objects using R class name mapping
  is extended to more classes. The documentation about conversion covers the topic.

- If `R_NilValue` is not null when the initialization of the embedded R is attempted,
  it is now assumed that R was initialized through other means (e.g., an other C library in the
  same process) and the C-level initialization is be skipped.

- The conversion `rpy2py` is now working with any Python object inheriting
  from `_rinterface_capi.SupportsSEXP`.

Bugs fixed
----------

- The C function `Rf_findVarInFrame()` in the R API can trigger
  in an R-level error, and while this is rare, when it does
  when embedded in Python it creates a segfault. Calls are
  now wrapped in `R_ToplevelExec()` to limit the propagation
  of R exceptions. This solved issue #710.

- More complete and correct mapping of R class names in
  :func:`rpy2.rinterface_lib.sexp.rclass_get`.

- Initializing the embedded R caused the loss of ability to use Ctrl-C
  to send SIGINT to a Python process (issue #723)

- :mod:`rpy2.sitation` is now working when the environment variable
  `R_HOME` is set even though R is not in the `PATH` or in the Windows
  registry (issue #744).

- Handling an R language objects could result in a segfault when its
  R class was queried (issue #749).

- The conversion of R string arrays to `numpy` arrays was leaving
  R's `NA` value as R NA objects. NAs in this type of arrays are now
  turned to `None` in the resulting `numpy` array (issue #751).

- `rpy2.situation.get_rlib_path()` was returning an environment variable
  with an invalid separator on Windows (mentioned in issue #754).

- R strings encoded with something else than 'utf-8' could result in
  errors when trying to convert to Python strings (issue #754).

- Extracting documentation pages for R objects in packages could
  generate spurious warnings when several "section" tags are present.

- R `Date` arrays/vectors were not wrapped into
  :class:`rpy2.robjects.vectors.DateVector` objects but left as
  R arrays of floats (which they are at the C level).

- The HTML representation of short R lists without names could
  fail with an error.

- The :meth:`__repr__` of `robjects`-level objects was not displaying
  the rpy2 class the R object is mapped to.

Release 3.3.6
=============

Bugs fixed
----------

- The unit tests for importing R packages with `lib_loc` were
  broken (issue #720).

- Trying to create a memoryview for an R array with complex values
  was failing with an attribute error.

- Fix the constructor of metaclass
  :class:`rpy2.robjects.methods.RS4Auto_Type`.

- Fix call to end the embedded R in :class:`rpy2.robjects.R.__cleanup__`
  (issue #734).

Release 3.3.5
=============

Bugs fixed
----------

- The callback handler to read input to R returned an
  invalid result, leading to R asking for input
  without ever acknowledging it received it.

Release 3.3.4
=============

Bugs fixed
----------

- Creating an R vector object from a Python object implementing
  the buffer protocol could give incorrect results as C-level
  incompatibilities could be missed (issue #702).

- :func:`rpy2.robjects.packages.importr` could fail when `lib_loc`
  was specified (issue #705).

Release 3.3.3
=============

Bugs fixed
----------

- Fallback for when `str2lang` is missing (R < 3.6)

- Fix segfault with :meth:`PairListSexpVector.__getitem__` when
  elements of the R pairlist have a `NILSXP` name (issue #700)

Release 3.3.2
=============

Bugs fixed
----------

- Initial fixes to have rpy2 running in ABI mode on Windows.
  Few tests are not passing (many in callbacks for R's C API).

- System detection is now checking for FreeBSD.

Release 3.3.1
=============

Bugs fixed
-----------

- :meth:`rpy2.robjects.conversion.NameClassMap.update` can update
  the mapping (:class:`dict`) or the default class.

Changes
-------

- Adding local converters was overwriting the base `NameClassMap`.

Release 3.3.0
=============

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

- Trying to import an R package that is not installed will now raise an
  exception :class:`rpy2.robjects.packages.PackageNotInstalledError`.

- The R C API functions `void SET_FRAME(SEXP x, SEXP v)`,
  `void SET_ENCLOS(SEXP x, SEXP v)` and `void SET_HASHTAB(SEXP x, SEXP v)`
  are now accessible through rpy2.

- The module :mod:`rpy2.situation` can now return `LD_LIBRARY_PATH`
  information about R. For example with
  `python -m rpy2.situation LD_LIBRARY_PATH`

- :meth:`rpy2.robjects.methods.RS4.extends` lists the class names in the
  inheritance line.

- The conversion of R objects to Python allows much more flexibility
  and better allow the use of independent code converting different classes.
  This is currently limited to R objects that are lists, environments, or
  S4 objects. The Sphinx documentation contains an example. While this is
  still work in progress this should already address concerns
  at the origin of issue #539 about S4 classes.

- :class:`rpy2.robjects.language.LangVector` to map R language objects at
  the `robjects` level.

- :class:`rpy2.robjects.vectors.PairlistVector` to map R pairlist objects at
  the `robjects` level.

- An alternative function to display the output of R cells can be
  specified using `-d` or `--display` in the magic arguments
  (in :mod:`rpy2.ipython.rmagic`).

- Python classes representing underlying R objects no longer have to
  exclusively rely on inheritance from :mod:`rpy2.rinterface` objects`.
  An abstract class :class:`rpy2.rinterface_lib.sexp.SupportsSEXP` is added
  to identify objects supporting a `__sexp__` protocol, and that abstract
  class can also be used with type hints.
- :func:`rpy2.robjects.functions.wrap_r_functions` can create Python functions
  with matching signature from R functions

- :func:`rpy2.robjects.functions.wrap_r_functions` can create Python functions
  with matching signature from R functions.

- New class :class:`rpy2.rinterface_lib._rinterface_capi.UninitializedRCapsule`
  to allow the instanciation of "placeholder" rpy2 objects before the
  embedded R is initialized. This facilitate the use of static typing checks
  such as mypy, mocking for tests that do not involve the execution of R
  code, and allow cleaner implementations of module-level globals
  that are R objects.

- New class :class:`rpy2.robjects.vectors.DateVector` to represent R dates.

- :class:`pandas.Series` containing date objects can now be converted to R
  `Date` vectors.

Changes
-------

- When calling R C-API's `R_ParseVector` and a error occurs, the
  exception message now contains the parsing status.

- :mod:`rpy2.rinterface_lib.embedded` has a module-level "constant"
  `DEFAULT_C_STACK_LIMIT` used when initializing the embedded R.

- When creating a :mod:`rpy2.robjects.vectors.DataFrame` from (name, vector)
  pairs, the names are no longer transformed to syntactically valid R
  symbols (issue #660).

- The value `nan` in :mod:`pandas` Series with strings is now converted
  to R NA (issue #668).

- Initial support for :const:`pandas.NA` (still experimental in pandas
  at the time of writing, and rpy2 support is limited to arrays of strings).

- :mod:`pandas` series of dtype :class:`pandas.StringDType`, experimental in pandas 1.0,
  are now supported by the converted (in the pandas-to-R direction) (issue #669)

- Version checking for the mapping of R packages in :mod:`rpy2.robjects.lib` is
  now more permissive (check that version prefixes are matching).

Bugs fixed
-----------

- Building ABI only mode could require an API build environment (and fail
  with an error when not present).

- SVG output for the R magic were incorrectly bytes objects.

- :meth:`rpy2.rinterface_lib.sexp.StrSexpVector.__getitem__` was returning the string
  `'NA'` when an R NA value. Not it returns `rpy2.rinterface_lib.na_values.NA_Character`.

Release 3.2.7
=============

Bugs fixed
----------

- An f-string in `_rinterface_cffi_build.py` prevented installation
  on Python 3.5 (issue #654).

Release 3.2.6
=============

Bugs fixed
----------

- The conversion of date/time object with specified timezones
  was wrong when different than the local time zone (issue #634)

- Iterating over :mod:`rpy2.situation.iter_info()` could result
  in a error because of a typo in the code.

Changes
-------

- :mod:`pandas` 1.0.0 breaks the conversion layer. A warning
  is now emitted whenever trying to use `pandas` >= 1.0.

Release 3.2.5
=============

Bugs fixed
----------

- Latest release for R package `rlang` broke import through `importr()`.
  A workaround for :mod:`rpy2.robjects.lib.ggplot2` is to rename the
  offending R object (issue #631).

Changes
-------

- f-string requiring Python >= 3.6 removed.

Release 3.2.4
=============

Bugs fixed
----------

- An incomplete backport of the bug fixed in 3.2.3 broke the ABI mode.

Release 3.2.3
=============

Bugs fixed
-----------

- Error when parsing strings as R codes could result in a segfault.

Release 3.2.2
=============

Bugs fixed
----------

- Python format error when trying to report that the system is not reported
  on Windows (issue #597).

- The setup script would error on build if R is not installed. It is now
  printing an error message.

Release 3.2.1
=============

Bugs fixed
----------

- The wrapper for the R package `dbplyr` could not import the underlying
  package (refactoring elsewhere was not propagated there).

- Creating R objects called `names` `globalenv` caused the method
  :meth:`Sexp.names` to fail (issue #587).

- Whenever the pandas conversion was activated :class:`FloatSexpVector` instances
  with the R class `POSIXct` attached where not corrected mapped back to pandas
  datetime arrays. (issue #594).

- Fix installation when an installation when a prefix without write access is used
  (issue #588).

Release 3.2.0
=============

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

- rpy2 can built and used with :mod:`cffi`'s ABI or API modes (releases 3.0.x and
  3.1.x were using the ABI mode exclusively). At the time of writing the default
  is still the ABI mode but the choice can be controlled through the environment variable
  `RPY2_CFFI_MODE`. If set, possible values are `ABI` (default if the environment
  variable is not set), `API`, or `BOTH`. When the latter, both `API` and `ABI`
  modes are built, and the choice of which one to use can be made at run time.

Changes
-------

- The "consoleread" callback (reading input to the R console) is now assuming UTF-8
  (was previously assuming ASCII) and is no longer trying to add a "new line" character
  at the end of the input.

- Querying an R environment with an invalid key will generate a :class:`TypeError`
  or a :class:`ValueError` depending on the issue (rather than always :class:`ValueError`
  before.

Bugs fixed
----------

- `setup.py` is now again compatible with Python2 (issue #580).

- Unit tests were failing if numpy is not installed.

- :mod:`rpy2.situation` is no longer breaking when R is not the in path and
  there is no environment variable `R_HOME`.

- Build script for the cffi interface is now using the environment
  variable `R_HOME` whenever defined (rather that always infer it from the
  R in the PATH).

- Converting R strings back to Python was incorrectly using `Latin1` while `UTF-8` was
  intended (issue #537).

Release 3.1.0
=============

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

- Python matrix multiplication (`__matmul__` / `@`) added to
  R :class:`Matrix` objects.

- An :class:`threading.RLock` is added to :mod:`rpy2.rinterface_lib.openrlib` and is
  used by the context manager :func:`rpy2.rinterface_lib.memorymanagement.rmemory`
  to ensure that protect/unprotect cycles cannot be broken by thread switching, at least
  as long as the context manager is used to handle such cycles (see issue #571).

- The documentation covers the use of notebooks (mainly Jupyter/Jupyterlab).

- The PNG output in Jupyter notebooks R cells can now specify an argument `--type`
  (passed as the named argument `type` in the R function `png`).
  For example on some Linux systems and R installations, the type `cairo`
  can fix issues when alpha transparency is used.

Changes
-------

- Added callbacks for `ptr_R_Busy()` and `ptr_R_ProcessEvents()`.

- `rstart` now an objects in :mod:`rpy2.rinterface_lib.embedded`
  (set to `None` until R is initialized).

- Unit tests are included in a subpackage :mod:`rpy2.tests` as was the
  case before release 3.0.0 (issue #528).

- Experimental initialization for Microsoft Windows.

- :mod:`rpy2.situation` is now also reporting the rpy2 version.

- :func:`rpy2.robjecs.package_utils.default_symbol_check_after` was
  renamed  :func:`rpy2.robjecs.package_utils.default_symbol_resolve`.
  The named parameters `default_symbol_check_after` present in few methods
  in :mod:`rpy2.robjects.packages` and :mod:`rpy2.robjects.functions` were
  modified to keep a consistent naming.

- Trying to instantiate an :class:`rpy2.rlike.container.OrdDict` with a
  a :class:`dict` will result in a :class:`TypeError` rather than a
  :class:`ValueError`.

- Methods of :class:`rpy2.rlike.container.OrdDict` now raises a
  :class:`NotImplementedError` when not implemented.

- The creation of R vectors from Python sequences is now relying on a method
  :meth:`_populate_r_vector` that allows vectorized implementation to
  to improve speed.

- Continuous integration tests run against Python 3.6, 3.7, and 3.8. It is
  no longer checked against Python 3.5.

Bugs fixed
----------

- `aes` in :mod:`rpy2.robjects.lib.ggplot2` had stopped working with the
  R package ggplot2 reaching version 3.2.0. (issue #562).

- Better handling of recent :mod:`pandas` arrays with missing values
  (related to issue #544).

- The mapping of the R operator `%in%` reachable through the attribute `ro`
  of R vectors was always returning `True`. It is now working properly.

- R POSIXct vectors with `NA` dates were triggering an error when converted
  in a data frame converted to :mod:`pandas` (issue #561).

Release 3.0.5
=============

Bugs fixed
----------

- No longer allow installation if Python 3 but < 3.5.

- Fixed error `undefined symbol: DATAPTR` if R < 3.5 (issue #565).

Release 3.0.4
=============

Bugs fixed
----------

- Fixed conversion of `pandas` :class:`Series` of dtype `pandas.Int32Dtype`,
  or `pandas.Int64Dtype` (issue #544).

Release 3.0.3
=============

Bugs fixed
----------

- Fixed the evaluation of R code using the "R magic" was delaying all
  output to the end of the execution of that code, independently of
  whether the attribute `cache_display_data` was `True` or `False`
  (issue #543).

- Fixed conversion of :class:`pandas.Series` of `dtype` "object" when
  all items are either all of the same type or are :obj:`None` (issue #540).

Release 3.0.2
=============

Bugs fixed
----------

- Failing to import `pandas` or `numpy` when loading the "R magic" extension
  for jupyter/ipython was hiding the cause of the error in the `ImportError`
  exception.

- Fallback when an R `POSIXct` vector does not had an attribute `"tzone"`
  (issue #533).

- Callback for console reset was not set during R initialization.

- Fixed rternalized function returning rpy2 objects (issue #538).

- `--vanilla` is no longer among the default options used to initialize R
  (issue #534).

Release 3.0.1
=============

Bugs fixed
----------

- Script to install R packages for docker image never made it to version
  control.

- Conversion of R arrays/matrices into numpy object trigged a segfault
  during garbage collection (issue #524).

Release 3.0.0
=============

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

- rpy2 can be installed without a development environment.

- Unit tests are now relying on the Python module `pytest`.

- :attr:`rpy2.rinterface.NA_Integer` is now only defined when the embedded R
  is initialized.

Changes
-------

- complete rewrite of :mod:`rpy2.rinterface`.
  :mod:`cffi` is now used to interface with the R compiled shared library.
  This allows ABI calls and removes the need to compile binaries. However, if
  compilation is available (when installing or preparing pre-compiled binaries)
  faster implementations of performance bottlenecks will be available.

- calling :func:`rpy2.rinterface.endr` multiple times is now only ending R
  the first time it is called (note: an ended R cannot successfully be
  re-initialized).

- The conversion system in the mod:`rpy2.robjects.conversion` now has only
  two conversions `py2rpy` and rpy2py`. `py2rpy` tries to convert any
  Python object into an object rpy2 can use with R and `rpy2py` tries
  to convert any rpy2 object into a either a non-rpy2 Python object or
  a mod:`rpy2.robjects` level object.

- The method `get` for R environments is now called `find()` to avoid
  confusion with the method of the same name in Python (:meth:`dict.get`).

- :class:`rpy2.robjects.vectors.Vector`, :class:`rpy2.robjects.vectors.Matrix`,
  and :class:`rpy2.robjects.vectors.Array` can no longer be used to create
  R arrays of unspecified type. New type-specific classes (for example for
  vectors :class:`rpy2.robjects.vectors.IntVector`,
  :class:`rpy2.robjects.vectors.BoolVector`,
  :class:`rpy2.robjects.vectors.FloatVector`,
  :class:`rpy2.robjects.vectors.ComplexVector`, or
  :class:`rpy2.robjects.vectors.StrVector`) should be used instead.

- mod:`rpy2.rpy_classic`, an implementation of the `rpy` interface using
  :mod:`rpy2.rinterface` is no longer available.

- :class:`rpy2.robjects.ParsedCode` and
  :class:`rpy2.robjects.SourceCode` are moved to
  :class:`rpy2.robjects.packages.ParsedCode` and
  :class:`rpy2.robjects.packages.SourceCode`.

Bugs fixed
----------

- Row names in R data frames were lost when converting to pandas data frames
  (issue #484).

Known issues
------------

- Mismatch between R's POSIXlt `wday` and Python time struct_time's `tm_wday`
  (issue #523).

Release 2.9.6
=============

Bugs fixed
----------

- Latest release of :mod:`pandas` deprecated :meth:`DataFrame.from_items`.
  (issue #514).

- Latest release of :mod:`pandas` requires categories to be a list
  (not an other sequence).

Known issues
------------

- The numpy buffer implemented by R arrays is broken for complex numbers

Release 2.9.5
=============

Bugs fixed
----------

- Missing values in pandas :class:`Category` series were creating
  invalid R factors when converted (issue #493).

Release 2.9.4
=============

Bugs fixed
----------

- Fallback for failure to import numpy or pandas is now dissociated from
  failure to import :mod:`numpy2ri` or :mod:`pandas2ri` (issue #463).

- :func:`repr` for R POSIX date/time vectors is now showing a string
  representation of the date/time rather than the timestamp as a float
  (issue #467).

- The HTML representation of R data frame (the default representation in the
  Jupyter notebook) was displaying an inconsistent number of rows
  (found while workin on issue #466).

- Handle time zones in timezones in Pandas when converting to R data frames
  (issue #454).

- When exiting the Python process, the R cleanup is now explicitly request
  to happen before Python's exit. This is preventing possible segfaults
  the process is terminating (issue #471).

- dplyr method `ungroup()` was missing from
  :class:`rpy2.robjects.lib.dplyr.DataFrame` (issue #473).

Release 2.9.3
=============

Bugs fixed
----------

- Delegate finding where is local time zone file to either a user-specified
  module-level variable `default_timezone` or to the third-party
  module :mod:`tzlocal` (issue #448).

Release 2.9.2
=============

Changes
-------

- The pandas converter is converting :class:`pandas.Series` of `dtype` `"O"`
  to :class:`rpy2.robjects.vectors.StrVector` objects, issueing a warning
  about it (See issue #421).

- The conversion of pandas data frame is now working with columns rather
  than rows (introduce in bug fix for issue #442 below) and this is expected
  to result in more efficient conversions.

Bugs fixed
----------

- Allow floats in figure sizes for R magic (Pull request #63)

- Fixed pickling unpickling of robjects-level instances,
  regression introduced in fix for issue #432 with release 2.9.1 (issue #443).

- Fixed broken unit test for columns of `dtype` `"O"` in `pandas` data frames.

- Fixed incorrect conversion of R factors in data frames to columns of
  integers in pandas data frame (issue #442).

(wiz)

doc: Added databases/R-dbplyr version 2.1.1

(wiz)

databases/Makefile: + R-dbplyr

(wiz)

databases/R-dbplyr: import R-dbplyr-2.1.1

dbplyr is the database backend for dplyr. It allows you to use
remote database tables as if they are in-memory data frames by
automatically converting dplyr code into SQL.

(wiz)

py-lmfit: add build dependency on py-setuptools_scm

(wiz)

py-mysqldb: remove post-install rule that deletes two compiled objects

Breaks build with python 2.7 because the paths are different there.
Still packages with python 3.6, 3.9, 3.10 on NetBSD.

Bump PKGREVISION for PLIST change.

(wiz)

py-OpenGL*: mark as not for python 2.7

py-OpenGL-accelerate segfaults python2.7 during build;
all the users of the packages already disable python 2.7

(wiz)

py-trueskill: fix build with latest setuptools.

(wiz)

py-ranking: fix build with latest setuptools

(wiz)

py-eth-abi: does not support python 3.6

Also mark its user that way

(wiz)

py-xdis: does not build with python 2.7, mark it so

Also mark packages using it that way.

(wiz)

py-trytond: fix build with latest setuptools

(wiz)

py-pysvn: add missing dependency

(wiz)

py-pygit2: not for python 3.6

Move restriction higher so it works

(wiz)

py-line_profiler: remove incorrect check

(wiz)

altmeister-otf: fix some pkglint

(wiz)

py-iso3166: not for python 3.6

Also mark packages using it that way

(wiz)

py-hash: fix build with latest setuptools

(wiz)

neumeister-otf: fix some pkglint

(wiz)

py-gitinspector-devel: fix build

Using patch from py-gitinspector

(wiz)

doc: Updated devel/py-boost to 1.77.0nb1

(wiz)

py-boost: disable numpy

For some reason, the numpy so file is not created in the package as-is,
breaking the package in the packaging step.

Remove py-numpy dependency and matching entries in PLIST to fix package.

Please fix this and numpy support back if you want, but this solves
the immediate breakage.

Bump PKGREVISION.

(wiz)

py-behave: fix build with latest setuptools

(wiz)

py-bcolz: remove

Does not build, looks dead upstream, nothing in pkgsrc depends on it.

(wiz)

py-aenum: mark as incompatible with python 2.7

Also mark all packages using it.

(wiz)

py-peewee: mark as not for python 2.7

Uses sphinx features that last version compatible with python 2.7 does
not provide.

(wiz)

py-asyncpg: move incompatibility line earlier so it works

(wiz)

py-mutagen: add missing test dependency

(wiz)

py-PyRTL: mark as not for python 2.7

(wiz)

py-mutagen: mark as not for python 2.7

Also adapt the dependencies to not support python 2.7.

(wiz)

py-karaoke: remove, needs python 2.7 but its dependencies don't support it

(wiz)

py-funcparserlib: fix build with latest setuptools

(wiz)

doc: Updated x11/modular-xorg-xephyr to 1.20.14

(wiz)

doc: Updated x11/modular-xorg-xwayland to 1.20.14

(wiz)

doc: Updated x11/modular-xorg-server to 1.20.14

(wiz)

modular-xorg-*: update to 1.20.14

Mario Kleiner (2):
      Fix RandR leasing for more than 1 simultaneously active lease.
      modesetting: Allow Present flips with mismatched stride on atomic drivers.

Matt Turner (1):
      xserver 1.20.14

Povilas Kanapickas (4):
      record: Fix out of bounds access in SwapCreateRegister()
      xfixes: Fix out of bounds access in *ProcXFixesCreatePointerBarrier()
      Xext: Fix out of bounds access in SProcScreenSaverSuspend()
      render: Fix out of bounds access in SProcRenderCompositeGlyphs()

(wiz)

py-docutils: switch to versioned_dependencies.mk for py-pygments

(wiz)

doc: Updated editors/xvile to 9.8v

(wiz)

doc: Updated editors/vile to 9.8v

(wiz)

vile, xvile: update to 9.8v

Update provided by Thomas Dickey.

Changes:
20211212 (v)
> Lois Mansot
+ corrected range for &random function.
+ move the reframe_cursor_position() in update() to ensure that $cwline
  is shown correctly when viewing [Variables] while showvariables is
  set.
> Mark Robinson:
+ modify c-filt.c
  + include the quote chars in the literal colouring for C/C++/Java/etc
    to make it the same as most other syntax highlighting
  + add backquote "`" as a quoting character for JavaScript, e.g., for
    multi-line quotes.
  + highlight embedded "${xxx}" markers in JavaScript strings using the
    Ident2 color.
+ modify sh-filt.l
  + improve parsing of the beginning of here-document to allow for
    here-documents piped to programs, e.g.,
    cat <<EOF | pr
something
EOF
    or
    cat <<EOF && date
something
EOF
  + do not treat "#" as a comment marker if it is adjacent to nonblank
    text.
> Stephan Schulz:
+ add popup-choices mode (Savannah #58999)
> Tom Dickey:
+ eliminate filename conflict between vile/xvile in test-packages for
  FreeBSD ports, e.g., xvile's vileget becomes xvileget.
+ suppress some of the gcc warnings due to perl header-files.
+ updated plink.sh, from xterm, to work around poor performance of gcc
  on Fedora.
+ modify perl scripts to use /usr/bin/env to locate the perl program,
  and eliminate sinstall.sh
+ modify curses-driver to support italics, e.g., as in ncurses 6.
+ apply Xaw header-fix for XawPlus, neXtaw and Xaw3dxft.
+ use Xaw3d menu-headers for x11vile.h and x11menu.c (FreeBSD #186420).
+ replace Header keywords with Id, to make Git-snapshots match.
+ continue development of xftplain.c (--enable-freetype).  The current
  driver is able to display a single TrueType font using Xft.  The
  menu entries for the bitmap-fonts are replaced by a single menu with
  sizes.
  Some of the remaining issues:
  + provide conversion for X displays other than 24-bit TrueColor
  + implement fallback fonts, i.e., --enable-fontsets
+ modify configure script to check for ssp library, needed for some
  misconfigured libraries when cross-compiling to MinGW.
+ fix a case in vile-manfilt where cur_line may not have been allocated
  before first use.
+ update configure script to work with _Noreturn changes in ncurses
  20210320 development version.
+ modify makefile.in, wrapping cd commands in a subshell to work around
  "jobs" misfeature of pmake.
+ improve c-filt.c support for JavaScript
  + handle ranges, i.e., to avoid stopping on "/"
  + check for regex after ":"
  + add m,s,u,y to permissable flags ending regex
+ quoting-fixes in autoconf macros, per shellcheck warnings.
+ add check for 'U' in read_quoted() needed to make ^VUxxxx work.
+ modify lins_chars() to handle a case where a script inserts a UTF-8
  character (report by Thomas Dupond).
+ modify configure/makefiles to support ".PHONY" feature.
+ change configure/makefiles to use ARFLAGS rather than AR_OPTS
+ change configure --with-warnings to --enable-warnings for consistency
  with other configure scripts, using recent changes in that to filter
  gcc -Werror options to avoid breaking configure checks.
+ fix a few build problems with --disable-extensions
+ split-out x11plain.c, to start work on TrueType font support in xvile
+ modify configure script to allow clang to support loadable filters,
  since recent configurations work.
+ add/fix NetBSD pkgsrc-files for 9.8u
+ update FreeBSD package-files for 9.8u
+ update config.guess, config.sub

(wiz)

py-stsci.distutils: and do not break python 2

(wiz)

py-stsci.distutils: fix build with python 3

(wiz)

doc: Updated security/py-pip-audit to 1.1.1

(wiz)

py-pip-audit: update to 1.1.1.

## [1.1.1] - 2021-12-07

### Fixed

* Dependency sources: a crash caused by unexpected logging statements in `pip`'s
  JSON output was fixed

(wiz)

doc: Updated textproc/hgrep to 0.2.2

(wiz)

hgrep: update to 0.2.2.

Fix a build failure since new RC version of clap crate was released.
Fix a dynamic link error of pcre2 library by linking the library
statically. The error could happen when you installed Homebrew to
non-default location on macOS (#6).
Add --regex-size-limit option for built-in grep feature.
Add --dfa-size-limit option for built-in grep feature.
Use Rust compiler v1.57 to build binaries.

(wiz)

doc: Updated audio/musicpd to 0.23.5

(wiz)

musicpd: update to 0.23.5.

ver 0.23.5 (2021/12/01)
* protocol
  - support relative offsets for "searchadd"
  - fix "searchaddpl" bug (bogus error "Bad position")
* database
  - upnp: fix crash bug
* tags
  - fix MixRamp support
* migrate to PCRE2
* GCC 12 build fixes

(wiz)

doc: Updated mail/notmuch to 0.34.2

(wiz)

notmuch: update to 0.34.2.

Notmuch 0.34.2 (2021-12-09)
===========================

Library
-------

Fix a bug that wrongly resolved conflict between the `database_path`
parameter to `notmuch_database_open_with_config` and configuration
item `database.path` in favour of the latter.

Python Bindings (notmuch2)
--------------------------

When building the documentation for the `notmuch2` python module,
import from the built module, not a system wide installed one.

The notmuch2.Database constructor now uses the library function
`notmuch_database_open_with_config` to support the same configuration
and database location options as the library does.

Fix some unprintable exception objects.

(wiz)

p5-PerlMagick: reset PKGREVISION after ImageMagick update

(wiz)

doc: Updated graphics/ImageMagick to 7.1.0.17

(wiz)

ImageMagick: update to 7.1.0.17.

2021-12-04  7.1.0-17  <quetzlzacatenango@image...>
  * Release ImageMagick version 7.1.0-17 GIT revision 19425:e66261ffe:20211204

2021-11-27  7.1.0-17  <quetzlzacatenango@image...>
  * set threads to 1 when using ASAN to avoid false positive memory leaks.

(wiz)

doc: Updated devel/autoconf-archive to 2021.02.19

(wiz)

autoconf-archive: update to 2021.02.19.

* Noteworthy changes in release 2021.02.19 (2021-02-19) [stable]

  The macros AX_CHECK_PCRE2 and AX_PROG_ROBOT have been added. Many other
  macros have received updates.

(wiz)

doc/TODO: add some

+ ImageMagick-7.1.0.17, alsa-lib-1.2.6.1, hgrep-0.2.2, musicpd-0.23.5,
  notmuch-0.34.2, py-pip-audit-1.1.1.

(wiz)

sun-jdk7: add a BLAKE2s

(wiz)

sun-jre7: add two BLAKE2s

(wiz)

gcc6-aux: + BLAKE2s

(wiz)

gcc5-aux: + BLAKE2s

(wiz)

gcc-aux: + BLAKE2s

(wiz)

python310: add upstream bug report URL

(wiz)

python310: remove reference to non-existent file

(wiz)

python310: remove dragonfly patch that's not even in dports

(wiz)

doc: Updated lang/python310 to 3.10.1nb2

(wiz)

python310: remove patch that was carried forward from before 2.7 times

upstream says it should not be needed, and I don't see a difference
in building python310 and running its tests either.

Let me know if you do.

Bump PKGREVISION to be on the safe side.

(wiz)

doc: Updated x11/libX11 to 1.7.3.1

(wiz)

libX11: update to 1.7.3.1.

This release of libX11 corrects a packaging problem in 1.7.3 which caused
the m4 files needed for autoreconf to not be included in the tarballs.

(ProTip: if you have an ACLOCAL variable set in your environment, and it
doesn't have "-I m4" ahead of any other -I options, then automake may
assume that any m4 files found via those other -I options don't need to
be included in the tarballs when you run "make dist".)

As a bonus, this release also includes one tiny typo fix in the XIM specs.

(wiz)

doc: Updated lang/vala to 0.54.4

(wiz)

vala: update to 0.54.4.

Vala 0.54.4
===========
* Various improvements and bug fixes:
  - codegen:
    + Use CCodeConstant for member access of constant symbol
    + Emit constants without initializer list in defines section [#440]
    + Add and use CCodeConstantIdentifier for accessing constants
    + Check required length of enum type name for GType support
    + Add missing check while overriding virtual async interface methods [#852]
    + Drop inconsistent space for ObjectType parameters
    + Accept CCode.type attribute on parameters [#876]
    + Fix CCodeUnaryExpression.write() for PREFIX_INCREMENT/PREFIX_DECREMENT
  - vala:
    + Improve error message for invalid handler of dynamic signal
    + Using SignalHandler.disconnect() is required for dynamic signals
    + Check for unavailable value-type of variable initializer [#1253]
    + Add [Profile] as known attribute for methods
    + Report error on missing gio-2.0 package for async constructors
    + Fix signals with generic return
    + parser: Split out Parser.parse_switch_section_statement()
    + parser: Better handling of misplaced switch sections [#1246]
  - genie:
    + Amend text of indent and dedent for error messages [#497]
    + Accept INTERR token before type arguments when parsing type [#1245]
    + Properly handle plain "get" or "set" property accessors [#1248]
    + "exception" is the expected string for TokenType.ERRORDOMAIN
    + Accept accessibility/async modifiers on "construct" creation methods [#1235]
  - girparser: Accept "sealed" for transformed records and compact classes
  - gtkmodule: Improve error message for invalid signal element in ui-file
  - build: Add "test-update" which passed UPDATE_EXPECTED=1 to refresh c-expected
  - tests: Rename colliding test cases to avoid conflicts

* Bindings:
  - alsa: Add more API and fix a few things
  - gnome-desktop-3.0: Some parameter fixes
  - gstreamer: Cherry-pick bindings fixes from 0.56
  - gstreamer-base-1.0: Some parameter fixes [#1255]
  - gtk4: Update to 4.5.0~cd9b7307
  - pango: Cherry-pick bindings fixes from 0.56

(wiz)

doc/TODO: + py-django-4.0, tor-browser-11.0.2.

(wiz)

doc: Added devel/tomlplusplus version 2.5.0

(wiz)

devel/Makefile: + tomlplusplus

(wiz)

devel/tomlplusplus: import tomlplusplus-2.5.0

TOML library for C++

Features

* Supports the latest TOML release (v1.0.0), plus optional support
  for some unreleased TOML features
* Supports serializing to JSON
* Proper UTF-8 handling (incl. BOM)
* C++17 (plus some C++20 features where available, e.g. experimental
  support for char8_t strings)
* Header-only (optional!)
* Doesn't require RTTI
* Works with or without exceptions

(wiz)

py-pip-audit: add upstream patch fixing a test failure.

(wiz)

doc: Updated devel/SDL2 to 2.0.18

(wiz)

SDL2: update to 2.0.18.

---------------------------------------------------------------------------
2.0.18:
---------------------------------------------------------------------------

General:
* The SDL wiki documentation and development headers are automatically
kept in sync
* Each function has information about in which version of SDL it
was introduced
* Added the hint SDL_HINT_APP_NAME to let SDL know the name of your
application for various places it might show up in system information
* Added SDL_RenderGeometry() and SDL_RenderGeometryRaw() to allow
rendering of arbitrary shapes using the SDL 2D render API
* Added SDL_SetTextureUserData() and SDL_GetTextureUserData() to
associate application data with an SDL texture
* Added SDL_RenderWindowToLogical() and SDL_RenderLogicalToWindow()
to convert between window coordinates and logical render coordinates
* Added SDL_RenderSetVSync() to change whether a renderer present
is synchronized with vblank at runtime
* Added SDL_PremultiplyAlpha() to premultiply alpha on a block of
SDL_PIXELFORMAT_ARGB8888 pixels
* Added a window event SDL_WINDOWEVENT_DISPLAY_CHANGED which is
sent when a window changes what display it's centered on
* Added SDL_GetWindowICCProfile() to query a window's ICC profile,
and a window event SDL_WINDOWEVENT_ICCPROF_CHANGED that is sent
when it changes
* Added the hint SDL_HINT_VIDEO_EGL_ALLOW_TRANSPARENCY to allow
EGL windows to be transparent instead of opaque
* SDL_WaitEvent() has been redesigned to use less CPU in most cases
* Added SDL_SetWindowMouseRect() and SDL_GetWindowMouseRect() to
confine the mouse cursor to an area of a window
* You can now read precise mouse wheel motion using 'preciseX' and
'preciseY' event fields
* Added SDL_GameControllerHasRumble() and
SDL_GameControllerHasRumbleTriggers() to query whether a game
controller supports rumble
* Added SDL_JoystickHasRumble() and SDL_JoystickHasRumbleTriggers()
to query whether a joystick supports rumble
* SDL's hidapi implementation is now available as a public API in
SDL_hidapi.h

Linux:
* Added the hint SDL_HINT_SCREENSAVER_INHIBIT_ACTIVITY_NAME to set
the activity that's displayed by the system when the screensaver
is disabled
* Added the hint SDL_HINT_LINUX_JOYSTICK_CLASSIC to control whether
/dev/input/js* or /dev/input/event* are used as joystick devices
* Added the hint SDL_HINT_JOYSTICK_DEVICE to allow the user to
specify devices that will be opened in addition to the normal
joystick detection
* Added SDL_LinuxSetThreadPriorityAndPolicy() for more control over
a thread priority on Linux

(wiz)

doc: Updated security/py-pip-audit to 1.1.0

(wiz)

py-pip-audit: update to 1.1.0.

## [1.1.0]

### Added

* CLI: The `--path <PATH>` flag has been added, allowing users to limit
  dependency discovery to one or more paths (specified separately)
  when `pip-audit` is invoked in environment mode
  ([#148](https://github.com/trailofbits/pip-audit/pull/148))

* CLI: The `pip-audit` CLI can now be accessed through `python -m pip_audit`.
  All functionality is identical to the functionality provided by the
  `pip-audit` entrypoint
  ([#173](https://github.com/trailofbits/pip-audit/pull/173))

* CLI: The `--verbose` flag has been added, allowing users to receive more
  more verbose output from `pip-audit`. Supplying the `--verbose` flag
  overrides the `PIP_AUDIT_LOGLEVEL` environment variable and is equivalent to
  setting it to `debug`
  ([#185](https://github.com/trailofbits/pip-audit/pull/185))

### Changed

* CLI: `pip-audit` now clears its spinner bar from the terminal upon
  completion, preventing visual confusion
  ([#174](https://github.com/trailofbits/pip-audit/pull/174))

### Fixed

* Dependency sources: a crash caused by `platform.python_version` returning
  an version string that couldn't be parsed as a PEP-440 version was fixed
  ([#175](https://github.com/trailofbits/pip-audit/pull/175))

* Dependency sources: a crash caused by incorrect assumptions about
  the structure of source distributions was fixed
  ([#166](https://github.com/trailofbits/pip-audit/pull/166))

* Vulnerability sources: a performance issue on Windows caused by cache failures
  was fixed ([#178](https://github.com/trailofbits/pip-audit/pull/178))

## [1.0.1] - 2021-12-02

### Fixed

* CLI: The `--desc` flag no longer requires a following argument. If passed
  as a bare option, `--desc` is equivalent to `--desc on`
  ([#153](https://github.com/trailofbits/pip-audit/pull/153))

* Dependency resolution: The PyPI-based dependency resolver no longer throws
  an uncaught exception on package resolution errors; instead, the package
  is marked as skipped and an appropriate warning or fatal error (in
  `--strict` mode) is produced
  ([#162](https://github.com/trailofbits/pip-audit/pull/162))

* CLI: When providing the `--cache-dir` flag, the command to read the pip cache
  directory is no longer executed. Previously this was always executed and
  could result into failure when the command fails. In CI environments, the
  default `~/.cache` directory is typically not writable by the build user and
  this meant that the `python -m pip cache dir` would fail before this fix,
  even if the `--cache-dir` flag was provided.
  ([#161](https://github.com/trailofbits/pip-audit/pull/161))

## [1.0.0] - 2021-12-01

### Added

* This is the first stable release of `pip-audit`! The CLI is considered
  stable from this point on, and all changes will comply with
  [Semantic Versioning](https://semver.org/)

## [0.0.9] - 2021-12-01

### Added

* CLI: Skipped dependencies are now listed in the output of `pip-audit`,
  for supporting output formats
  ([#145](https://github.com/trailofbits/pip-audit/pull/145))
* CLI: `pip-audit` now supports a "strict" mode (enabled with `-S` or
  `--strict`) that fails if the audit if any individual dependency cannot be
  resolved or audited. The default behavior is still to skip any individual
  dependency errors ([#146](https://github.com/trailofbits/pip-audit/pull/146))

(wiz)

doc: Updated devel/py-pip-api to 0.0.26

(wiz)

py-pip-api: update to 0.0.26.

0.0.26
----------------------------------------------

- Pass PIP_DISABLE_PIP_VERSION_CHECK to all invocations (#114)

0.0.25
----------------------------------------------

- Support the --path parameter when calling pip list (#112)

0.0.24
----------------------------------------------

- Add support for parsing URL requirements (#109)
- Test against pip 21.3.1

(wiz)

doc: Updated lang/rust-analyzer to 0.0.2021.12.06

(wiz)

rust-analyzer: update to 0.0.2021.12.06.

10944: internal: Prefer resolution of inert attributes
10876: fix: Show parameter hints unconditionally for logical not expressions

(wiz)

doc: Updated textproc/enchant2 to 2.3.2

(wiz)

enchant2: update to 2.3.2.

2.3.2 (November 16, 2021)
-------------------------

Fix the generation of man pages, so that installation-dependent paths are
correctly inserted.

Improve error reporting when enchant-lsmod cannot open a dictionary for a
given language, in particular with the Aspell backend.

Fix compilation on MingW.

Some minor code cleanup.

(wiz)

doc: Updated audio/mpc to 0.34

(wiz)

mpc: update to 0.34.

0.34 (2021/11/30)
* add commands "albumart", "readpicture"
* don't print status after error
* custom status format
* support grouping "list" results
* meson: auto-build libmpdclient if not available
* require libmpdclient 2.16 or newer
* require MPD 0.21 or newer

(wiz)

doc: Updated math/gsl to 2.7.1

(wiz)

gsl: update to 2.7.1.

* What is new in gsl-2.7.1:

** update libtool version numbers

(wiz)

doc: Updated math/double-conversion to 3.1.6

(wiz)

double-conversion: update to 3.1.6.

2021-12-04:
  Update version number.

2021-10-04:
  Consistently use semicolons after DOUBLE_CONVERSION_ASSERT.

2021-07-16:
  Fix spelling.

2021-05-19:
  Loongarch is a RISC-style command system architecture.
  Add support for loongarch architecture.

(wiz)

doc: Added security/py-cyclonedx-bom version 1.5.3

(wiz)

doc: Added security/py-cyclonedx-python-lib version 0.11.1

(wiz)

security/Makefile: add some packages

(wiz)

security/py-cyclonedx-python-lib: import py-cyclonedx-python-lib-0.11.1

This CycloneDX module for Python can generate valid CycloneDX
bill-of-material document containing an aggregate of all project
dependencies.

This module is not designed for standalone use.

(wiz)

security/py-cyclonedx-bom: import py-cyclonedx-bom-1.5.3

This project provides a runnable Python-based application for
generating CycloneDX bill-of-material documents from either:

* Your current Python Environment

* Your project's manifest (e.g. Pipfile.lock, poetry.lock or
  requirements.txt)

* Conda as a Package Manager

The BOM will contain an aggregate of all your current project's
dependencies, or those defined by the manifest you supply.

CycloneDX is a lightweight BOM specification that is easily created,
human-readable, and simple to parse.

(wiz)

doc: Added devel/py-requirements-parser version 0.2.0

(wiz)

doc: Added devel/py-packageurl version 0.9.6

(wiz)

doc: Added devel/py-types-toml version 0.10.1

(wiz)

doc: Added devel/py-types-setuptools version 57.4.4

(wiz)

devel/Makefile: add some packages

(wiz)

devel/py-requirements-parser: import py-requirements-parser-0.2.0

This is a small Python module for parsing Pip requirement files.

The goal is to parse everything in the Pip requirement file format
spec.

(wiz)

devel/py-packageurl: import py-packageurl-0.9.6

Python library to parse and build "purl" aka. Package URLs.

(wiz)

devel/py-types-toml: import py-types-toml-0.10.1

This is a PEP 561 type stub package for the toml package. It can
be used by type-checking tools like mypy, PyCharm, pytype etc. to
check code that uses toml.

(wiz)

devel/py-types-setuptools: import py-types-setuptools-57.4.4

This is a PEP 561 type stub package for the setuptools package. It
can be used by type-checking tools like mypy, PyCharm, pytype etc.
to check code that uses setuptools.

(wiz)

doc: Updated x11/libX11 to 1.7.3

(wiz)

libX11: update to 1.7.3.

This release includes a number of bug fixes and adds support for the
_EVDEVK keysyms added in xorgproto 2021.2.

Alan Coopersmith (2):
      Build xz tarballs instead of bzip2
      libX11 1.7.3

Alex Richardson (2):
      Fix cross-compilation from macOS
      Avoid undefined behaviour after realloc()

Hodong (1):
      Fix a typo

Igor V. Kovalenko (1):
      libX11: do not crash in GetResReq() macro

Lev Korol (1):
      nls: Add en_IL locale

Liang Chang (1):
      Make Xutf8DrawString works correctly with ISO10646 fonts.

Matthieu Herrb (1):
      Avoid NULL pointer deref. Fixes issue #47.

Peter Hutterer (2):
      makekeys: handle the new _EVDEVK xorgproto symbols
      gitlab CI: update to use the ci-fairy image

Tatu Frisk (1):
      Fix hanging issue in _XReply

(wiz)

doc: Updated x11/xscope to 1.4.2

(wiz)

xscope: update to 1.4.2.

Alan Coopersmith (8):
      Fix up documentation of NAS command line flags
      Update README for gitlab migration
      Update configure.ac bug URL for gitlab migration
      Add support for process ID and names on Solarish systems
      gitlab CI: add a basic build test
      Build xz tarballs instead of bzip2
      Fix spelling/wording issues
      xscope 1.4.2

Emil Velikov (1):
      autogen.sh: use quoted string variables

Mihail Konev (1):
      autogen: add default patch prefix

Peter Hutterer (1):
      autogen.sh: use exec instead of waiting for configure to finish

Peter Wu (1):
      Add support for process ID and names

St辿phane Aulery (1):
      Improve help and manpage.

(wiz)

python39: add URL to upstream bug report

(wiz)

py-curses: bump PKGREVISION for python39 patch

(wiz)

python39: make curses module more amenable to NetBSD curses

(wiz)

py-cftime: not for python 2.x

While here, fix tests.

(wiz)

py-netCDF4: mark as not for python 2.x

This depends on numpy which does not support python 2.x.
Not sure why this doesn't break bulk builds.

(wiz)

py-xapian: mark as not for python 2.x

Does not package.

Leave 2.7 support code for now in case of highly motivated packagers
interested in fixing this.

(wiz)

py-tomli: not for python 2.x

(wiz)

doc: Updated textproc/py-pybtex-docutils to 1.0.1

(wiz)

py-pybtex-docutils: update to 1.0.1.

1.0.1 (in development)
----------------------

* Add py.typed so mypy finds the type annotations.

* Switch to github actions for regression testing.

1.0.0 (15 January 2021)
-----------------------

* Drop Python 2.7, 3.4, and 3.5 support.

* Add type annotations.

* Add support for sub and sup tags.

(wiz)

doc: Updated textproc/py-parse_type to 0.5.6

(wiz)

py-parse_type: update to 0.5.6.

Fix build with latest setuptools.

Changes not found.

(wiz)

py-feedparser5: fix build with latest setuptools.

(wiz)

py-demjson: fix build with latest setuptools.

(wiz)

py-anyjson: fix build with latest setuptools.

(wiz)

py-Tempita: fix build with latest setuptools.

(wiz)

py-collectd: fix package

(wiz)

py-m2crypto: remove, obsolete and does not build

(wiz)

py-gssapi: fix build

(wiz)

doc: Updated www/py-FormEncode to 2.0.1

(wiz)

py-FormEncode: update to 2.0.1.

2.0.1
-----
- Add support for 3.10
- use Pytest instead of Nose and Github Actions instead of Travis for tests
- Documentation updates
- Note this will be the last version to support Python 2.7. The next version
  will be 2.1 to signal this change. If you want to keep support for Python
  2.7 update your dependencies spec to be below 2.1

2.0.0
-----

- `FormEncode` can now run on Python 3.6 and higher without needing to run 2to3 first.
- `FormEncode` 2.0 is no longer compatible with Python 2.6 and 3.2 to 3.5.
  If you need Python 2.6 or 3.2 to 3.5 compatibility please use `FormEncode` 1.3.
  You might also try `FormEncode` 2.0.0a1 which supports Python 2.6 and Python 3.3-3.5.
- This will be the last major version to support Python 2.7
- Add strict flag to USPostalCode to raise error on postal codes that has too
  many digits instead of just truncating
- Various Python 3 fixes
- Serbian latin translation
- Changed License to MIT
- Dutch, UK, Greek and South Korean postal code format fixes
- Add postal code formats for Switzerland, Cyprus, Faroe Islands, San Marino, Ukraine and Vatican City.
- Add ISODateTimeConverter validator
- Add ability to target htmlfill to particular form or ignore a form
- Fix format errors in some translations
- The version of the library can be checked using formencode.__version__

(wiz)

py-pystache: fix build with latest setuptools

(wiz)

py-sparqlwrapper: fix build with latest setuptools

(wiz)

py-vatnumber: remove, unmaintained upstream.

(wiz)