sysutils/felix: update to 2.4.1

v2.4.1 (2023-06-21)
Changed
- Show status bar and registers even if current directory does not contain
  any item.

(pin)

doc: Updated lang/janet to 1.29.1

(charlotte)

lang/janet: Update to 1.29.1

Changes from 1.28.0:

## 1.29.1 - 2023-06-19

- Add support for passing booleans to PEGs for "always" and "never"
  matching.
- Allow dictionary types for `take` and `drop`
- Fix bug with closing channels while other fibers were waiting on them:
  `ev/take`, `ev/give`, and `ev/select`  will now return the correct
  (documented) value when another fiber closes the channel.
- Add `ffi/calling-conventions` to show all available calling conventions
  for FFI.
- Add `net/setsockopt`
- Add `signal` argument to `os/proc-kill` to send signals besides `SIGKILL`
  on Posix.
- Add `source` argument to `os/clock` to get different time sources.
- Various combinator functions now are variadic like `map`
- Add `file/lines` to iterate over lines in a file lazily.
- Reorganize test suite to be sorted by module rather than pseudo-randomly.
- Add `*task-id*`
- Add `env` argument to `fiber/new`.
- Add `JANET_NO_AMALG` flag to Makefile to properly incremental builds
- Optimize bytecode compiler to generate fewer instructions and improve
  loops.
- Fix bug with `ev/gather` and hung fibers.
- Add `os/isatty`
- Add `has-key?` and `has-value?`
- Make imperative arithmetic macros variadic
- `ev/connect` now yields to the event loop instead of blocking while
  waiting for an ACK.

(charlotte)

doc: Updated geography/qgis to 3.28.7

(gdt)

geography/qgis: Update to 3.28.7

update in freeze because: this is a leaf package, and micro updates
from qgis have an excellent track record of not causing problems.
(This commit is just version change and distinfo.)

(gdt)

Carry forward the SunOS sound setup from editors/xemacs (21.4).

I cannot test this right now, but it should at least unbreak the build
on SunOS.

(hauke)

doc: Updated editors/tp-note to 1.21.7

(pin)

editors/tp-note: update to 1.21.7

Maintenance release: improve the natural language detection speed

Maintenance release:

* Improve the natural language detection speed
* Reduce the binary size by 20%
* Improve the test harnesses

(pin)

doc: Updated devel/readline to 8.2nb2

(wiz)

readline: fix pkg-config files

Bump PKGREVISION.

(wiz)

Ticket #6762

(bsiegert)

Pullup ticket #6762 - requested by riastradh
security/heimdal: security fix

Revisions pulled up:
- security/heimdal/Makefile                                    1.160
- security/heimdal/distinfo                                    1.57
- security/heimdal/patches/patch-lib_krb5_store-int.c          1.1

---
  Module Name:    pkgsrc
  Committed By:  riastradh
  Date:          Mon Jun 19 19:13:03 UTC 2023

  Modified Files:
          pkgsrc/security/heimdal: Makefile distinfo
  Added Files:
          pkgsrc/security/heimdal/patches: patch-lib_krb5_store-int.c

  Log Message:
  security/heimdal: Patch CVE-2022-42898 away.

(bsiegert)

po4a: use a sufficiently capable diff program. Still broken on Solarish.

(schmonz)

py-cppy: not for Python 2.7

(adam)

deepstate: do not install binaries from py-future; forget a long gone patch

(adam)

py-claripy: relax z3-solver requirement

(adam)

freeradius: fix missing SOL_TCP

(adam)

py-argparse-manpage: fix post-install

(adam)

nickle: fix readline detection when using pkgsrc readline

(tnn)

nickle: put back patch-lex.c which got lost in recent rototil

Still needed on macOS.

(tnn)

shells/oh-my-posh: fix checksums

(pin)

libxslt: don't pass --version-script to Solaris linker.

Fixes "ld: fatal: option --version-script requires option -z
gnu-version-script-compat to be specified" on Solaris 11.4.

Tested to build as before on Tribblix, and therefore hopefully SmartOS.
NFCI on non-"SunOS" platforms.

(schmonz)

doc: Updated devel/gitui to 0.23.0

(pin)

devel/gitui: update to 0.23.0

[0.23.0] - 2022-06-19
Breaking Change
- focus_XYZ key bindings are merged into the move_XYZ set, so only one way to
  bind arrow-like keys from now on (#1539)

Added
- allow reset (soft,mixed,hard) from commit log (#1500)
- support reword of commit from log (#829)
- fuzzy find branch [@UUGTech] (#1350)
- list changes in commit message inside external editor [@bc-universe] (#1420)
- allow detaching HEAD and checking out specific commit from log view
  [@fralcow] (#1499)
- add no-verify option on commits to not run hooks [@dam5h] (#1374)
- allow fetch on status tab [@alensiljak] (#1471)
- allow copy file path on revision files and status tree [@yanganto] (#1516)
- print message of where log will be written if -l is set (#1472)
- show remote branches in log [@cruessler] (#1501)

Fixes
- fixed side effect of crossterm 0.26 on windows that caused double input of
  all keys [@pm100] (#1686)
- commit msg history ordered the wrong way (#1445)
- improve help documentation for amend cmd (#1448)
- lag issue when showing files tab (#1451)
- fix key binding shown in bottom bar for stash_open (#1454)
- --bugreport does not require param (#1466)
- edit-file command shown on commits msg (#1461)
- crash on branches popup in small terminal (#1470)
- edit command duplication (#1489)
- syntax errors in key_bindings.ron will be logged (#1491)
- Fix UI freeze when copying with xclip installed on Linux (#1497)
- Fix UI freeze when copying with wl-copy installed on Linux (#1497)
- commit hooks report "command not found" on Windows with wsl2 installed
  (#1528)
- crashes on entering submodules (#1510)
- fix race issue: revlog messages sometimes appear empty (#1473)
- default to tick-based updates [@cruessler] (#1444)
- add support for options handling in log and stashes views [@kamillo] (#1661)

Changed
- minimum supported rust version bumped to 1.65 (thank you time crate)

(pin)

doc: Updated shells/oh-my-posh to 17.4.0

(pin)

shells/oh-my-posh: update to 17.4.0

Features
- execution: lucky7 style (0616346)

(pin)

libcups: don't strip libraries on macOS. Fixes error seen in bulk build.

Still installs stripped (by default, anyway) on NetBSD.

(schmonz)

security/heimdal: Patch CVE-2022-42898 away.

(riastradh)

tiff: don't try to rebuild docs if sphinx-build is present.

Fixes PLIST mismatch (DESTDIR is missing _sources/multi_page.rst.txt and
multi_page.html) in that case, at least on my macOS system. No change on
my NetBSD system without sphinx.

(schmonz)

rust: fix build on Darwin if Homebrew is installed

The openssl-sys crate has a misguided piece of logic to prefer any
Homebrew tree to the usual place for finding OpenSSL. Remove this code
so that Rust actually uses pkgsrc OpenSSL as it should.

In particular, if you have Apple Game Porting toolkit installed, you get
a Homebrew installation for x86_64 on an ARM machine, which can never work.

OK during freeze adam@, wiz@

(bsiegert)

doc: mention freeze start

(wiz)

doc: Updated devel/py-codespell to 2.2.5

(wiz)

py-codespell: update to 2.2.5.

What's Changed

    Add fixing consciousely typo by @yarikoptic in #2777
    atalog typo (saw twice in a project) by @yarikoptic in #2782
    created fstrings by @marksmayo in #2784
    Replace bandit, isort, pylint, pyupgrade, and flake8 plus plugins with ruff by @cclauss in #2779
    Add Microsoft Mispelling by @Casey-Kiewit in #2789
    Use Latest Version In README Pre-Commit Example by @Casey-Kiewit in #2788
    ruff: Use format specifiers instead of percent format by @cclauss in #2787
    New typos by @gforcada in #2793
    Ruff: Ignore new rule PLC1901 by @cclauss in #2797
    Add various typo fixes by @luzpaz in #2795
    Another suggestion by @gforcada in #2798
    Fix AssertionError when opening with chardet by @n-thumann in #2785
    Fix typo (and variants) from xxHash by @DimitriPapadopoulos in #2761
    Add huld->hold Spelling Mistake by @Casey-Kiewit in #2792
    Fix CI issues by @DimitriPapadopoulos in #2796
    CI: fix new ruff error B028 by @DimitriPapadopoulos in #2802
    Skip subdirectories of hidden directories by @DimitriPapadopoulos in #2541
    Add tweek(s)->tweak(s) correction by @cfi-gb in #2804
    Add supoort->support by @thughes in #2806
    Add Suggestion For readded Mispelling by @Casey-Kiewit in #2805
    Add displaing->displaying by @peternewman in #2808
    Another suggestion for stati by @gforcada in #2810
    Add dimaond -> diamond by @Moiman in #2813
    Add vai->via, vie by @peternewman in #2815
    Remove invalid correction - Kwanza = currency of Angola by @DimitriPapadopoulos in #2803
    sinoid and friends by @DimitriPapadopoulos in #2807
    Add excludle->exclude by @peternewman in #2814
    Add gadjet->gadget and friend by @peternewman in #2818
    Add togueter->together by @luzpaz in #2821
    Add "mutabl->mutable" by @sirosen in #2822
    Add containin->containing by @janosh in #2824
    Add mandess->madness by @luzpaz in #2825
    some fixes by @marksmayo in #2820
    Fix new ruff errors reported by CI by @DimitriPapadopoulos in #2834
    Add fusipn->fusion by @luzpaz in #2836
    Add 'subcription->subscription' by @sirosen in #2832
    Add 'promtp->prompt' by @sirosen in #2833
    Add 'endpdoint' as a misspelling of 'endpoint' by @sirosen in #2830
    Add more misspellings of 'identities' by @sirosen in #2829
    Friends of existing typos by @DimitriPapadopoulos in #2837
    Fix 66ec667 by @DimitriPapadopoulos in #2838
    equivallent and friends by @DimitriPapadopoulos in #2827
    Add delpoys->deploys by @luzpaz in #2842
    Added variations on 'difference' by @luzpaz in #2841
    Add 'canoncical->canonical' by @sirosen in #2831
    Add additional typos by @kianmeng in #2835
    Add spelling corrections for inject and variants. by @cfi-gb in #2846
    Add 'credentail->credential' by @sirosen in #2844
    Add 'requestor->requester' by @sirosen in #2845
    Update README.rst by @andyholmes in #2466
    Add exrension(s)->extension(s) by @luzpaz in #2847
    Add facttory->factory and facttories-> factories by @nils-van-zuijlen in #2848
    Add several spelling corrections by @luzpaz in #2849
    Add comtaining->containing by @nils-van-zuijlen in #2855
    topoplogical and friends by @DimitriPapadopoulos in #2854
    Add placehoulder->placeholder by @sirosen in #2856
    More typos from GlobalProtect-openconnect by @DimitriPapadopoulos in #2850
    Add spelling correction for siver->silver/sliver/diver by @cfi-gb in #2858
    Add assertino->assertion by @sourcefrog in #2859
    add notority->notoriety by @EdwardBetts in #2861
    Misspellings from eLabFTW by @DimitriPapadopoulos in #2863
    Add invirant{,s}->invariant{,s} by @ojeda in #2870
    Add alloccat*->allocat* by @ojeda in #2869
    Add "multiply"/"multiplies" variations to instances of "multiple" by @jdufresne in #2872
    Add strin->string and optiona->optional by @peternewman in #2817
    Add wellplate -> well plate by @claydugo in #2873
    Add afile Misspelling by @Casey-Kiewit in #2876
    Ignore 'misspellings' due to string escapes by @Zac-HD in #2875
    Add pre-poulate Misspellings by @Casey-Kiewit in #2878

(wiz)

doc: Updated www/py-autobahn to 23.6.2

(wiz)

py-autobahn: update to 23.6.2.

No changelog found.

(wiz)

doc: Updated www/libproxy to 0.5.2

(wiz)

libproxy: update to 0.5.2.

Link with libsocket where needed (#234) by @amigadave in #235
Remove libsoup from doc and CI by @janbrummer in #238
Fix coverity findings by @janbrummer in #241
Add a comment that docs option needs introspection by @janbrummer in #242
Set pac data after download only by @janbrummer in #244
Fix race condition in px_manager_get_proxies_sync by @janbrummer in #245
Update gobject dependency in pkgconfig file by @floppym in #239
Reread env variables in each get_config call by @janbrummer in #246
Bump version to 0.5.2 by @janbrummer in #247

(wiz)

Updated devel/py-pyparsing, math/py-xgboost

(adam)

py-xgboost: updated to 1.7.6

1.7.6 Patch Release

Bug Fixes

Fix distributed training with mixed dense and sparse partitions.
Fix monotone constraints on CPU with large trees.
[spark] Make the spark model have the same UID as its estimator
Optimize prediction with QuantileDMatrix.

Document

Improve doxygen
Update the cuDF pip index URL.

Maintenance

Fix tests with pandas 2.0.

(adam)

py-pyparsing: updated to 3.1.0

Version 3.1.0 - June, 2023
--------------------------
- Added `tag_emitter.py` to examples. This example demonstrates how to insert
  tags into your parsed results that are not part of the original parsed text.

Version 3.1.0b2 - May, 2023
---------------------------
- Updated `create_diagram()` code to be compatible with railroad-diagrams package
  version 3.0.

- Fixed bug in `NotAny`, where parse actions on the negated expr were not being run.
  This could cause `NotAny` to incorrectly fail if the expr would normally match,
  but would fail to match if a condition used as a parse action returned False.

- Fixed `create_diagram()` to accept keyword args, to be passed through to the
  `template.render()` method to generate the output HTML (PR submitted by Aussie Schnore,
  good catch!)

- Fixed bug in `python_quoted_string` regex.

- Added `examples/bf.py` Brainf*ck parser/executor example. Illustrates using
  a pyparsing grammar to parse language syntax, and attach executable AST nodes to
  the parsed results.

Version 3.1.0b1 - April, 2023
-----------------------------
- Added support for Python 3.12.

- API CHANGE: A slight change has been implemented when unquoting a quoted string
  parsed using the `QuotedString` class. Formerly, when unquoting and processing
  whitespace markers such as \t and \n, these substitutions would occur first, and
  then any additional '\' escaping would be done on the resulting string. This would
  parse "\\n" as "\<newline>". Now escapes and whitespace markers are all processed
  in a single pass working left to right, so the quoted string "\\n" would get unquoted
  to "\n" (a backslash followed by "n").

- Added named field "url" to `pyparsing.common.url`, returning the entire
  parsed URL string.

- Fixed bug when parse actions returned an empty string for an expression that
  had a results name, that the results name was not saved. That is:

      expr = Literal("X").add_parse_action(lambda tokens: "")("value")
      result = expr.parse_string("X")
      print(result["value"])

  would raise a `KeyError`. Now empty strings will be saved with the associated
  results name.

- Fixed bug in `SkipTo` where ignore expressions were not properly handled while
  scanning for the target expression.

- Updated `ci.yml` permissions to limit default access to source - submitted by Joyce
  Brum of Google. Thanks so much!

- Updated the `lucene_grammar.py` example (better support for '*' and '?' wildcards)
  and corrected the test cases - brought to my attention by Elijah Nicol, good catch!

Version 3.1.0a1 - March, 2023
-----------------------------
- API ENHANCEMENT: `Optional(expr)` may now be written as `expr | ""`

  This will make this code:

      "{" + Optional(Literal("A") | Literal("a")) + "}"

  writable as:

      "{" + (Literal("A") | Literal("a") | "") + "}"

  Some related changes implemented as part of this work:
  - `Literal("")` now internally generates an `Empty()` (and no longer raises an exception)
  - `Empty` is now a subclass of `Literal`

- Added new class property `identifier` to all Unicode set classes in `pyparsing.unicode`,
  using the class's values for `cls.identchars` and `cls.identbodychars`. Now Unicode-aware
  parsers that formerly wrote:

      ppu = pyparsing.unicode
      ident = Word(ppu.Greek.identchars, ppu.Greek.identbodychars)

  can now write:

      ident = ppu.Greek.identifier
      # or
      # ident = ppu.Ελληνικά.identifier

- `ParseResults` now has a new method `deepcopy()`, in addition to the current
  `copy()` method. `copy()` only makes a shallow copy - any contained `ParseResults`
  are copied as references - changes in the copy will be seen as changes in the original.
  In many cases, a shallow copy is sufficient, but some applications require a deep copy.
  `deepcopy()` makes a deeper copy: any contained `ParseResults` or other mappings or
  containers are built with copies from the original, and do not get changed if the
  original is later changed.

- Reworked `delimited_list` function into the new `DelimitedList` class.
  `DelimitedList` has the same constructor interface as `delimited_list`, and
  in this release, `delimited_list` changes from a function to a synonym for
  `DelimitedList`. `delimited_list` and the older `delimitedList` method will be
  deprecated in a future release, in favor of `DelimitedList`.

- Error messages from `MatchFirst` and `Or` expressions will try to give more details
  if one of the alternatives matches better than the others, but still fails.

- Added new class method `ParserElement.using_each`, to simplify code
  that creates a sequence of `Literals`, `Keywords`, or other `ParserElement`
  subclasses.

  For instance, to define suppressible punctuation, you would previously
  write:

      LPAR, RPAR, LBRACE, RBRACE, SEMI = map(Suppress, "(){};")

  You can now write:

      LPAR, RPAR, LBRACE, RBRACE, SEMI = Suppress.using_each("(){};")

  `using_each` will also accept optional keyword args, which it will
  pass through to the class initializer. Here is an expression for
  single-letter variable names that might be used in an algebraic
  expression:

      algebra_var = MatchFirst(
          Char.using_each(string.ascii_lowercase, as_keyword=True)
      )

- Added new builtin `python_quoted_string`, which will match any form
  of single-line or multiline quoted strings defined in Python.

- Extended `expr[]` notation for repetition of `expr` to accept a
  slice, where the slice's stop value indicates a `stop_on`
  expression:

      test = "BEGIN aaa bbb ccc END"
      BEGIN, END = Keyword.using_each("BEGIN END".split())
      body_word = Word(alphas)

      expr = BEGIN + Group(body_word[...:END]) + END
      # equivalent to
      # expr = BEGIN + Group(ZeroOrMore(body_word, stop_on=END)) + END

      print(expr.parse_string(test))

  Prints:

      ['BEGIN', ['aaa', 'bbb', 'ccc'], 'END']

- `ParserElement.validate()` is deprecated. It predates the support for left-recursive
  parsers, and was prone to false positives (warning that a grammar was invalid when
  it was in fact valid).  It will be removed in a future pyparsing release. In its
  place, developers should use debugging and analytical tools, such as `ParserElement.set_debug()`
  and `ParserElement.create_diagram()`.

- Added bool `embed` argument to `ParserElement.create_diagram()`.
  When passed as True, the resulting diagram will omit the `<DOCTYPE>`,
  `<HEAD>`, and `<BODY>` tags so that it can be embedded in other
  HTML source. (Useful when embedding a call to `create_diagram()` in
  a PyScript HTML page.)

- Added `recurse` argument to `ParserElement.set_debug` to set the
  debug flag on an expression and all of its sub-expressions.

- Added '·' (Unicode MIDDLE DOT) to the set of Latin1.identbodychars.

- Fixed bug in `Word` when `max=2`. Also added performance enhancement
  when specifying `exact` argument.

- `Word` arguments are now validated if `min` and `max` are both
  given, that `min` <= `max`; raises `ValueError` if values are invalid.

- Fixed bug in srange, when parsing escaped '/' and '\' inside a
  range set.

- Fixed exception messages for some `ParserElements` with custom names,
  which instead showed their contained expression names.

- Fixed bug in pyparsing.common.url, when input URL is not alone
  on an input line.

- Multiple added and corrected type annotations. With much help from
  Stephen Rosen, thanks!

- Some documentation and error message clarifications on pyparsing's
  keyword logic, cited by Basil Peace.

- General docstring cleanup for Sphinx doc generation, PRs submitted
  by Devin J. Pohly. A dirty job, but someone has to do it - much
  appreciated!

- `invRegex.py` example renamed to `inv_regex.py` and updated to PEP-8
  variable and method naming. PR submitted by Ross J. Duff, thanks!

- Removed examples `sparser.py` and `pymicko.py`, since each included its
  own GPL license in the header. Since this conflicts with pyparsing's
  MIT license, they were removed from the distribution to avoid
  confusion among those making use of them in their own projects.

(adam)

doc: Updated sysutils/bup to 0.32.1

(gdt)

sysutils/bup: Update to 0.32.1

This is a bug-fix micro update.

(gdt)

doc: Updated math/nickle to 2.92

(wiz)

nickle: update to 2.92.

Version 2.92
debian: Build depend on a couple of packages
Use CC_BUILD to generate math test tables
Add ctype.h, remove explicit lex func prototypes from lex.l
Remove 'const' attribute when assigning to tm_zone field
Allow use of libedit in place of readline
Check for all readline helper functions separately
Define _GNU_SOURCE in edit.c to get asprintf declared
Remove redundant box value check in ThreadBoxSetDefault
test: Allow "UTC", "GMT", or "NONE" to be returned for GMT time zone
Make tm_zone and tm_gmtoff fields optional
Autodetect 'significand' function
Use base-10 for all float exponents. Use base 2 for base 2/8/16 powers
Support C-like hex float values in lexer
Add ChaCha implementation

(wiz)

doc: Updated graphics/oculante to 0.6.65

(pin)

graphics/oculante: update to 0.6.65

No ChangeLog provided, cherry pick some commits.

- remove unused code
- Improve animation sequence start
- Feat: Add Zen mode (o UI) default 'z' to toggle
- Enable support for Apple Silicon (aarch64)
- update deps
- Show zwn mode info on first start

(pin)

TODO: libtpms-0.9.6 [CVE-2023-1017, CVE-2023-1018]

(riastradh)

Updated net/py-lexicon

(adam)

py-lexicon: updated to 3.12.0

Lexicon v3.12.0

Added

Add duckdns provider (experimental support)
Add dnsservices provider
Add flexibleengine provider
Official support for Python 3.11

Modified

Upgrade API version used for azure provider
Various fixes for documentation
Fix check for extra dependencies

(adam)

doc: Updated security/acmesh to 3.0.6

(js)

Update security/acmesh to 3.0.6

This fixes a RCE vulnerability.

(js)

doc: Updated devel/R-fansi to 1.0.4

(mef)

(devel/R-fansi) Updated 0.5.0 to 1.0.4

# fansi Release Notes

## v1.0.4

CRAN compiled code warning suppression release.

* Fix void function declarations and definitions.
* Change `sprintf` to `snprintf`.

## v1.0.3

* Address problem uncovered by gcc-12 linters, although the issue itself could
  not manifest due to redundancy of checks in the code.

## v1.0.0-2

This is a major release and includes some behavior changes.

### Features

* New functions:
    * [#26](https://github.com/brodieG/fansi/issues/26) Replacement forms of
      `substr_cl` (i.e `substr_ctl<-`).
    * `state_at_end` to compute active state at end of a string.
    * `close_state` to generate a closing sequence given an active state.
    * [#31](https://github.com/brodieG/fansi/issues/31) `trimws_ctl` as an
      equivalent to `trimws`.
    * [#64](https://github.com/brodieG/fansi/issues/64) `normalize_sgr` converts
      compound _Control Sequences_ into normalized form (e.g. "ESC[44;31m"
      becomes "ESC[31mESC[44m") for better compatibility with
      [`crayon`](https://github.com/r-lib/crayon).  Additionally, most functions
      gain a `normalize` parameter so that they may return their output in
      normalized form (h/t @krlmlr).
* [#74](https://github.com/brodieG/fansi/issues/74)`substr_ctl` and related
  functions are now all-C instead of a combination of C offset computations and
  R level `substr` operations.  This greatly improves performance, particularly
  for vectors with many distinct strings.  Despite documentation claiming
  otherwise, `substr_ctl` was quite slow in that case.
* [#66](https://github.com/brodieG/fansi/issues/66) Improved grapheme support,
  including accounting for them in `type="width"` mode, as well as a
  `type="graphemes"` mode to measure in graphemes instead of characters.
  Implementation is based on heuristics designed to work in most common use
  cases.
* `html_esc` gains a `what` parameter to indicate which HTML special characters
  should be escaped.
* Many functions gain `carry` and `terminate` parameters to control how `fansi`
  generated substrings interact with surrounding formats.
* [#71](https://github.com/brodieG/fansi/issues/71) Functions that write SGR and
  OSC are now more parsimonious (see "Behavior Changes" below).
* [#73](https://github.com/brodieG/fansi/issues/73) Default parameter values
  retrieved with `getOption` now always have explicit fallback values defined
  (h/t @gadenbui).
* Better warnings and error messages, including more granular messages for
  `unhandled_ctl` for adjacent _Control Sequences_.
* `term.cap` parameter now accepts "all" as value, like the `ctl` parameter.

### Deprecated Functions

* All the "sgr" functions (e.g., `substr_sgr`, `strwrap_sgr`) are deprecated.
  They will likely live on indefinitely, but they are of limited usefulness and
  with the added support for OSC hyperlinks their name is misleading.
* `sgr_to_html` is now `to_html` with slight modifications to semantics; the old
  function remains and does not warn about unescaped "<" or ">" in the
  input string.

### Behavior Changes

The major intentional behavior change is to default `fansi` to always recognize
true color CSI SGR sequences (e.g. `"ESC[38;2;128;50;245m"`).  The prior
default was to match the active terminal capabilities, but it is unlikely that
the intent of a user manipulating a string with truecolor sequences is to
interpret them incorrectly, even if their terminal does.  `fansi` will continue
to warn in this case.  To keep the pre-1.0 behavior add `"old"` to the
`term.cap` parameter.

Additionally, `to_html` will now warn if it encounters unescaped HTML special
character "<" or ">" in the input string.

Finally, the 1.0 release is an extensive refactoring of many parts of the
SGR and OSC hyperlink controls (_Special Sequences_) intake and output
algorithms.  In some cases this means that some `fansi` functions will output
_Special Sequences_ slightly differently than they did before.  In almost all
cases the rendering of the output should remain unchanged, although there are
some corner cases with changes (e.g. in `strwrap_ctl` SGRs embedded in
whitespace sequences don't break the sequence).

The changes are a side effect of applying more consistent treatment of corner
cases around leading and trailing control sequences and (partially) invalid
control sequences.  Trailing _Special Sequences_ in the output is now omitted as
it would be immediately closed (assuming `terminate=TRUE`, the default).
Leading SGR is interpreted and re-output.

Normally output consistency alone would not be a reason to change behavior, but
in this case the changes should be almost always undetectable in the
**rendered** output, and maintaining old inconsistent behavior in the midst of a
complete refactoring of the internals was beyond my patience.  I apologize if
these behavior changes adversely affect your programs.

> WARNING: we will strive to keep rendered appearance of `fansi` outputs
> consistent across releases, but the exact bytes used in the output of _Special
> Sequences_ may change.

Other changes:

* Tests may no longer pass with R < 4.0 although the package should still
  function correctly.  This is primarily because of changes to the character
  width Unicode Database that ships with R, and many of the newly added grapheme
  tests touch parts of that database that changed (emoji).
* CSI sequences with more than one "intermediate" byte are now considered valid,
  even though they are likely to be very rare, and CSI sequences consume all
  subsequent bytes until a valid closing byte or end of string is encountered.
* `strip_ctl` only warns with malformed CSI and OSC if they are reported as
  supported via the `ctl` parameter.  If CSI and OSC are indicated as not
  supported, but two byte escapes are, the two initial bytes of CSI and OSCs
  will be stripped.
* "unknown" encoded strings are no longer translated to UTF-8 in UTF-8 locales
  (they are instead assumed to be UTF-8).
* `nchar_ctl` preserves `dim`, `dimnames`, and `names` as the base functions do.
* UTF-8 known to be invalid should not be output, even if present in input
  (UTF-8 validation is not complete, only sequences that are obviously wrong are
  detected).

### Bug Fixes

* Fix `tabs_as_spaces` to handle sequential tabs, and to perform better on very
  wide strings.
* Strings with invalid UTF-8 sequences with "unknown" declared encoding in UTF-8
  locales now cause errors instead of being silently translated into byte
  escaped versions (e.g. "\xf0\xc2" (2 bytes), used to be interpreted as
  "<f0><c2>" (four characters).  These now cause errors as they would have if
  they had had "UTF-8" declared encoding.
* In some cases true colors of form "38;2;x;x;x" and "48;2;x;x;x" would only be
  partially transcribed.

### Internal Changes

* More aggressive UTF-8 validation, also, invalid UTF-8 code points now advance
  only one byte instead of their putative width based on the initial byte.
* Reduce peak memory usage by making some intermediate buffers eligible for
  garbage collection prior to native code returning to R.
* Reworked internals to simplify buffer size computation and synchronization, in
  some cases this might cause slightly reduced performance.  Please report any
  significant performance regressions.
* `nchar_ctl(...)` is no longer a wrapper for `nchar(strip_ctl(...))` so that it
  may correctly support grapheme width calculations.

(mef)

doc: Updated devel/R-IRkernel to 1.3.2

(mef)

(devel/R-IRkernel) Updated 1.3.1 to 1.3.2, NEWS.md unknown

(mef)

doc: Updated net/R-pbdZMQ to 0.3.9

(mef)

(net/pbdZMQ) Updated 0.3.3 to 0.3.9

2022-12-24: Ver 0.3-9
  * Replace "sprintf()" in C.
  * Replace some ".pbd_env" from function arguments.
  * Add suppress messages to demo.

2022-10-16: Ver 0.3-8
  * Fix "strict-types" and "deprecated-non-prototype" warnings.
  * Fix "if() conditions comparing class() to string".

2022-02-04: Ver 0.3-7
  * Make a copy of './inst/zmq_copyright/*' to './src/zmqsrc'.
  * Add ZeroMQ authors to DESCRIPTION file.

2021-10-25: Ver 0.3-6
  * Change configure.ac for autoconf-2.71
  * Change tests to local in-process (inter-thread) communication transport
    "zmq_inproc".

2021-04-17: Ver 0.3-6
  * Check and add "libzmq>=4.3.0" and "-DENABLE_DRAFTS=ON" options.
  * Add more ZMQ socket options up to libzmq 4.3.4 (may not stable).

2021-02-27: Ver 0.3-6
  * Add more ZMQ socket options.
  * Add timeout for connection in tests.

2021-02-09: Ver 0.3-5
  * Update "conf.sub" and "conf.guess" from CRAN.

2020-12-13: Ver 0.3-4
  * Change "http://" to "https://".

2020-09-07: Ver 0.3-4
  * Fix warning 'char* strncpy(char*, const char*, size_t)' output truncated
    due to "-Werror=stringop-truncation" by gcc 8.3.1
  * Fix a "buf[1]" in zmq.send() call in "R_zmq_sendrecv.r"

2019-07-27: Ver 0.3-4
  * For osx, change "install.libs.R" and "zzz.r.in" for staged installation.

2019-07-10: Ver 0.3-4
  * Roll back to (R >= 3.5.0).
  * Change detection of ZeroMQ library version to "4.2.2" from "4.0.4".
  * Roll detection of ZeroMQ library version to "4.0.4".
  * Block ZeroMQ library version "4.1.6".

2019-05-03: Ver 0.3-4
  * Add "StagedInstall: FALSE" to DESCRIPTION to turn off WARNING in macos.

2019-04-26: Ver 0.3-4
  * Revmoe "^M" from "src/zmqsrc/src/condition_variable.hpp".

2019-04-01: Ver 0.3-4
  * Support REQ/REP sockets in sendfile/recvfile functions.

2019-02-18: Ver 0.3-4
  * Add "R/R_zmq_transfers.r" for transferfing files and directories.
  * Add importFrom utils for zip and unzip.
  * Remove "-Werror" from "src/zmqsrc/configure" to pass "R CMD check".

2019-02-17: Ver 0.3-4
  * Register "R_zmq_send_file" and "R_zmq_recv_file" in "src/zzz.c".
  * Fix Windows binary files transfer problems.

(mef)

doc: Updated devel/R-repr to 1.1.6

(mef)

(devel/R-repr) Updated 1.1.4 to 1.1.6

1.1.6 (release note unknown)

1.1.5 Latest

    [#157] Fix for dev stringr (@hadley)

(mef)

Added math/R-cellranger version 1.1.0

(mef)

(math/R-cellranger) import R-cellranger-1.1.0

Helper functions to work with spreadsheets and the "A1:D10" style of
cell range specification.

(mef)

doc: Updated devel/R-pak to 0.5.1

(mef)

(devel/R-pak) Updated 0.4.0 to 0.5.1

# pak 0.5.1

* No user visible changes.

# pak 0.5.0

* The `meta_*()` functions now consider the `platforms`, `cran_mirror` and
  `r_versions` config entries, see `?"pak-config"`.

* Better Posit Package Manager (PPM) support. New `+ppm_has_binaries()`,
  `ppm_r_versions()`, `ppm_repo_url()`, `ppm_snapshots()` and `ppm_platforms()`
  functions to help interacting with PPM. See the new 'pkgcache and Posit
  Package Manager on Linux' article at https://r-lib.github.io/pkgcache.
  (r-lib/pkgcache#47 and r-lib/pkgdepends#186).

* New `system_r_platform()` and `system_r_platform_data()` functions to query
  the current platform.

* pak now support git repositories as package references. E.g.
  `git::https://github.com/r-lib/pak.git`.

* pak now supports versioned CRAN packages, e.g. `dplyr@1.1.1` will
  always install dplyr 1.1.1. Note that only CRAN packages are supported,
  Bioconductor packages are not (yet).

* pak now has an alternative system requirements lookup
  implementation. It supports Fedora and Debian systems as well, in
  addition to Debian, Ubuntu, SUSE and RedHat derivatives.
  You can switch to this implementation by setting the
  `R_PKG_SYSREQS2` environment variable to `true`.

* pak now does a better job looking up dependencies for
  hand-selected dependency types. E.g. `dependencies = "LinkingTo"`.

* pak now removes `?ignore`-d packages from dependencies, and
  uses the correct version comparison for `?ignore-before.r`
  (https://github.com/r-lib/actions/issues/708).

* pak now does not fail for circular soft dependencies (#306).

* pak now reports dependency solver failures better in some cases
  (#305, https://github.com/r-lib/pak/issues/474).

* pak now uses locally built CRAN binaries from the cache. Use the
  `?nocache` parameter to opt out from this, or
  `cache_delete(package = ...)` to remove a package from the cache.

(mef)

doc: Updated devel/R-pkgcache to 2.1.1

(mef)

(devel/R-pkgcache) Updated 2.0.4 to 2.1.1

# pkgcache 2.1.1

* `parse_installed()` now has a `packages` argument, to list only a subset
  of all packages.

* `parse_packages()` can now parse `PACKAGES` files with trailing
  whitespace (#93).

* The Bioconductor repositories now include the 'books' repository,
  available since Bioconductor 3.12.

# pkgcache 2.1.0

* pkgcache now supports binary packages on x86_64 macOS and R 4.3.0 and
  later (#89).

* Better Posit Package Manager (PPM) support. New `ppm_has_binaries()`,
  `ppm_r_versions()`, `ppm_repo_url()`, `ppm_snapshots()` and `ppm_platforms()`
  functions to help interacting with PPM. See the new 'pkgcache and Posit
  Package Manager on Linux' article at https://r-lib.github.io/pkgcache.
  (#47 and r-lib/pkgdepends#186).

(mef)

doc: Updated devel/R-renv to 0.17.3

(mef)

(devel/R-renv) Updated 0.16.0 to 0.17.3

# renv 0.17.3

* Fixed an issue where `renv::install("bioc::<package>")` could fail if
  `BiocManager` was not already installed. (#1184)

* Fixed an issue where package names were not included in the output
  of `renv::diagnostics()`. (#1182)

* The clarity of the message printed by `renv::status()` has been improved;
  in particular, `renv` should better report the recommended actions when
  a package required by the project is not installed.

* `renv::snapshot()` gains the `exclude` argument, for explicitly
  excluding certain packages from the generated lockfile.

* Fixed an issue where `renv` was passing the wrong argument name to
  `installed.packages()`, causing usages of `renv` to fail with
  R (<= 3.4.0). (#1173)

* `renv` now sets the `SDKROOT` environment variable on macOS if it detects
  that R was built using an LLVM build of `clang` on macOS.

* `renv::install()` now parses the remotes included within, for example,
  a `DESCRIPTION` file's `Config/Needs/...` field.

* `renv` now checks that the index directory is writable before attempting to
  use it, e.g. for the `R` available packages index maintained by `renv`. (#1171)

* `renv` now checks that the version of `curl` used for downloads appears to
  be functional, and reports a warning if it does not (for example, because
  a requisite system library is missing). The version of `curl` used for
  downloads can also be configured via the `RENV_CURL_EXECUTABLE` environment
  variable.

# renv 0.17.2

* Fixed a regression that caused package hashes to be computed incorrectly
  in some cases. This was a regression in the 0.17.1 release. (#1168)

# renv 0.17.1

* `renv` gains the configuration option `install.remotes`, which can be used
  to control whether `renv` attempts to read and use the `Remotes:` field
  included with installed packages. This can be set to `FALSE` if you'd
  like to ignore this field; e.g. because you know they will not be
  accessible. (#1133)

* General performance optimizations. In particular, `renv::update(check = TRUE)`
  should now execute much faster.

* `renv` now stores project settings within `renv/settings.json` rather than
  `renv/settings.dcf`. Old settings will be automatically migrated.

* The `renv` sandbox is now placed within the `renv` cache directory. (#1158)

* Fixed an issue where `renv::status()` could erroneously report a project was
  out-of-sync when using explicit snapshots. (#1159)

* Fixed an issue where `renv::hydrate()` would print spurious warnings. (#1160)

* `renv::status()` now suggests running `renv::restore()` if there are no
  packages installed in the project library. (#1060)

* Fixed an issue where `renv` would fail to query [r-universe](https://r-universe.dev/)
  repositories. (#1156)

* `renv` no longer tries to set the `SDKROOT` environment variable on
  macOS for R (>= 4.0.0).

* Fixed an issue where installation of Bioconductor packages could fail
  when `BiocManager` was not installed. (#1156, #1155)

* Fixed an issue where the amount of time elapsed reported by `renv::install()`
  failed to include the time spent retrieving packages.

# renv 0.17.0

* The performance of `renv::snapshot()` has improved.

* `renv` now maintains an index of available packages, as retrieved from the
  active package repositories, that is shared across `R` sessions. This should
  improve `renv`'s performance when querying for available packages across
  multiple different `R` sessions.

* `renv::hydrate()` gains the `prompt` parameter. When `TRUE` (the default),
  `renv` will prompt for confirmation before attempting to hydrate the active
  library.

* Improved handling of package installation via SSH URLs with `git`. (#667)

* Improved handling of R version validation when using projects with
  Bioconductor. If you find `renv` is erroneously reporting that your version
  of R is incompatible with the version of Bioconductor you are using, you can
  set `options(renv.bioconductor.validate = FALSE)` to disable this validation
  behaviour. (#1148)

* Package names can now be associated with remotes in `renv::install()`; for
  example, you can use `renv::install("igraph=igraph/rigraph")` to install
  the `igraph` package. This is mainly important when using the `renv.auth`
  authentication tools, where the package name of a remote needs to be
  declared explicitly. (#667)

* Fixed an issue that could prevent `renv::restore()` from functioning when
  attempting to install packages which had been archived on CRAN. (#1141)

* `renv::install()` gains the `dependencies` argument, which behaves similarly
  to the equivalent argument in `remotes::install_*()`. In particular, this can
  be set to fields like `Config/Needs/dev` to tell `renv` to use custom
  DESCRIPTION fields for dependency resolution in installation.

* Fixed an issue where the function variant of the `renv.auth` option was not
  resolved correctly. (#667)

* `renv::install()` now supports remotes with a trailing slash -- such slashes
  are now removed. (#1135)

* Integration with the RStudio (Posit) Package Manager is now disabled
  by default on aarch64 Linux instances.

* The `RENV_CACHE_MODE` environment variable can now also be used
  to adjust the permissions set on downloaded package tarballs / binaries.
  (#988)

* Fixed an issue where fields of the form `Remotes.1` could
  enter lockfile records for older versions of R. (#736)

* Fixed the performance of `renv::update()` in cases where
  integration with MRAN is enabled.

* Fixed an issue where package installation using `pak` could fail
  in some cases.

* `renv_file_find()` can now scan up to the root directory in Docker
  containers. (#1115)

* `renv` no longer uses the R temporary directory on Windows for the
  sandbox. The sandbox directory can be customized via the
  `RENV_PATHS_SANDBOX` environment variable if required. (#835)

* `renv` now reports the elapsed time when installing packages. (#1104)

* For projects using "explicit" snapshots, `renv` now reports if
  a package is required by the project, but the package is not
  currently installed. (#949)

* Fixed an issue where `renv::snapshot()` could fail to detect when
  no changes had been made to the lockfile.

* Fixed an issue where `renv` could emit JSON lockfiles which could not
  be parsed by external JSON readers. (#1102)

* `renv` now marks the sandbox as non-writable, which should hopefully
  alleviate issues where attempts to update installed packages would
  inadvertently install the updated package into the sandbox. (#1090)

* `renv::use()` gains the `sandbox` argument, which allows one to control
  whether the system library is sandboxed after a call to `renv::use()`.
  (#1083)

* The path to the Conda `environment.yml` file created by `renv` can
  now be customized via the `RENV_PATHS_CONDA_EXPORT` environment
  variable. We recommend setting this within your project-local
  `.Renviron` file as appropriate. (#1089)

* Fixed an issue where the `renv` sandbox location did not respect the
  active `renv` profile. (#1088)

(mef)

doc: Updated devel/R-pkgbuild to 1.4.1

(mef)

(devel/R-pkgbuild) Updated 1.4.0 to 1.4.1

# pkgbuild 1.4.1

* New `Config/build/extra-sources` `DESCRIPTION` option to make pkgbuild aware
  of extra source files to consider in `needs_compile()`.

* New `Config/build/bootstrap` `DESCRIPTION` option. Set it to `TRUE` to run
  `Rscript bootstrap.R` in the package root prior to building the source
  package (#157, @paleolimbot).

* pkgbuild now supports Rtools43.

* pkgbuild now always _appends_ its extra compiler flags to the ones that
  already exist in the system and/or user `Makevars` files (#156).

(mef)

doc: Updated devel/R-checkmate to 2.2.0

(mef)

(devel/R-checkmate) Updated 2.1.0 to 2.2.0

# Version 2.2.0
* Fixed C compiler warnings for windows
* Added `checkPermutation` (#230).
* Better handling of long vectors (#231).
* Fixed a bounds check for numeric vectors and infinite bounds.

(mef)

doc: Updated devel/R-BH to 1.81.0.1

(mef)

(devel/R-BH) Updated 1.78.0.0 to 1.81.0.1

2023-01-17  Dirk Eddelbuettel  <edd@debian.org>

* DESCRIPTION (Version, Date): Release 1.81.0-1

* inst/include/boost/math/tools/config.hpp: Comment-out 'warning'
message as CRAN Policy would prefer not to see these
* inst/include/boost/multiprecision/detail/standalone_config.hpp: Idem
* inst/include/boost/spirit/home/x3.hpp: Idem (for four lines)

2022-12-29  Dirk Eddelbuettel  <edd@debian.org>

* inst/include/boost/core/snprintf.hpp: Added
* .../lexical_cast/detail/converter_lexical_streams.hpp: Used here
* .../core/lightweight_test.hpp: Use snprintf
* .../core/type_name.hpp: Idem
* .../interprocess/detail/win32_api.hpp: Idem
* .../numeric/odeint/integrate/max_step_checker.hpp: Idem
* .../test/imp/debug.ipp: Idem
* .../utils/setcolor.hpp: Idem

2022-12-15  Dirk Eddelbuettel  <edd@debian.org>

* inst/include/boost/*: Upgraded to Boost 1.81, as usual with two
patch sets from CRAN demands on file path length and diagnostics

* README.md: Updated package list for 1.81
* DESCRIPTION (Description): Idem

* local/scripts/CreateBoost.sh: Update for 1.81.0, added url

2022-10-31  Dirk Eddelbuettel  <edd@debian.org>

* .github/workflows/ci.yaml (jobs): Update to actions/checkout@v3

(mef)

doc: Updated devel/R-testthat to 3.1.9

(mef)

(devel/R-testthat) Updated 3.1.8 to 3.1.9

# testthat 3.1.9

* New `expect_contains()` and `expect_in()` that works similarly to
  `expect_true(all(expected %in% object))` or
  `expect_true(all(object %in% expected))` but give more informative failure
  messages (#1346).

* New `is_snapshot()` returns `TRUE` if code is running inside a snapshot test
  (#1796) and `is_checking()` returns `TRUE` if test is running inside of
  `R CMD check` (#1795)

* `ProgressReporter` only reports the run time of test files that take longer
  than 1s, rather than 0.1s. (#1806) and re-displays all failures at the end
  of the results. Skips are now only shown at the end of reporter summaries,
  not as tests are run. This makes them less intrusive in interactive tests
  while still allowing you to verify that the correct tests are skipped (#1801).
  When using parallel tests, links to failed tests (#1787) and links to
  accept/review snapshot (#1802) now work.

* `set_state_inspector()` allows to to register a function that's called
  before and after every test, reporting on any differences. This
  is very useful for detecting if any of your tests have made changes to
  global state (like options, env vars, or connections) (#1674). This
  function was inspired by renv's testing infrastructure.

* `skip_on_cran()` no longer skips (errors) when run interactively.

* `teardown_env()` works in more cases.

* All packages, regardless of whether or not they use rlang, now
  use the new snapshot display for errors, warnings, and messages.

* testthat no longer truncates tracebacks and uses rlang's default tree
  display.

(mef)

doc: Updated devel/R-cli to 3.6.1

(mef)

(devel/R-cli) Updated 3.6.0 to 3.6.1

# cli 3.6.1

* ANSI hyperlinks are now turned off on the RStudio render plane (#581).

(mef)

doc: Updated devel/R-git2r to 0.32.0

(mef)

(devel/R-git2r) Updated 0.30.1 to 0.32.0

# git2r 0.32.0 (2023-04-12)

## CHANGES

* Updated the bundled libgit2 source code to version 1.6.3.

* Fix build failure of bundled libgit2 when structure for nanosecond
  type is mssing.

# git2r 0.31.0 (2023-01-26)

## CHANGES

* Removed unused macro 'AC_PROG_CC' from the build configuration
  script.

(mef)

Updated devel/py-zope.deprecation, www/py-zope.proxy, devel/py-zope.testing, textproc/py-sphinx-autoapi

(adam)

py-sphinx-autoapi: updated to 2.1.1

v2.1.1 (2023-06-10)
-------------------

Bugfixes
- Fix "document isn't included" warning when using autoapi_add_toctree_entry
- Types used in PEP-604 union syntax can be linked with intersphinx
- Fix class overrides not rendering correctly.
- Fix separated type comments for arguments not merging correctly in Python 3.7
- Fixed viewcode being unable to find the source code for imported objects

Improved Documentation
- Made it clearer how to customise what objects AutoAPI will document.

(adam)

py-zope.testing: updated to 5.0.1

5.0.1 (2022-12-20)
==================
- Make wheels no longer universal.

5.0 (2022-12-20)
================

Backwards incompatible changes
------------------------------
- Drop support for Python 2.7, 3.5, 3.6.
- Drop modules which do not seem to be Python compatible:
    + ``zope.testing.loghandler``
    + ``zope.testing.server``
- Drop doctest option ``IGNORE_EXCEPTION_MODULE_IN_PYTHON2``.
- Remove functions:
    + ``zope.testing.renormalizing.strip_dottedname_from_traceback``
    + ``zope.testing.renormalizing.is_dotted_name``

Features
--------
- Add support for Python 3.11.

(adam)

py-zope.proxy: updated to 5.0.0

5.0.0 (2023-01-18)
- Drop support for Python 2.7, 3.5, 3.6.
- Remove proxying code for names that no longer exist in Python 3
  like ``__long__`` and some others.

(adam)

py-zope.deprecation: updated to 5.0

5.0 (2023-03-29)
- Drop support for Python 2.7, 3.4, 3.5, 3.6.
- Drop support for deprecated ``python setup.py test``.
- Add support for Python 3.8, 3.9, 3.10, 3.11.

(adam)

Updated devel/py-zope.configuration

(adam)

doc/TODO: add some

+ gettext-lib-0.22, grpc-1.56.0, libcups-2.4.5, libproxy-0.5.2,
  lua-language-server-3.6.22, mupdf-1.22.2, protobuf-23.3,
  py-autobahn-23.6.2, py-codespell-2.2.5, py-contourpy-1.1.0,
  py-google-api-python-client-2.89.0, py-numpy-1.25.0,
  py-platformdirs-3.6.0, py-project-api-1.5.2, py-protobuf-4.23.3,
  py-service_identity-23.1.0, py-sphinx-autodoc-typehints-1.23.2,
  py-tox-4.6.2, py-virtualenv-20.23.1, py-yapf-0.40.0,
  xkeyboard-config-2.39.

(wiz)

py-zope.configuration: updated to 5.0; renamed from py-ZopeConfiguration

5.0 (2023-05-04)
- Drop support for Python 2.7, 3.5, 3.6.
- Add support for Python 3.11.

(adam)

Updated devel/py-zope.schema

(adam)

py-zope.schema: updated to 7.0.1; renamed from py-ZopeSchema

7.0.1 (2023-01-02)
- Fix fallback when ``zope.i18nmessageid`` is not installed (regression
  introduced in 7.0.0).

7.0.0 (2023-01-01)
- Add support for Python 3.11.
- Drop support for Python 2.7, 3.5, 3.6.
- Drop ``zope.schema._compat`` module.
- Fix test deprecation warning on Python 3.11.

(adam)

geeqie: fix missing library issue using patch submitted upstream

(wiz)

Updated devel/py-zope.i18nmessageid

(adam)