Backport of #31358 to release-0.10

In order to run unittests with a release build, we need the test functions to be accessible when NDEBUG is defined. Moving the functions into the test fixture ensures they are available and only available for use by the unit tests.

Pulling this out of #31358 because it won't backport to rele...

Breaking: Python 3.7 is the minimum supported Python version.

Changes since 0.4.3:

• 1696737 feat: Ex command ":py=" evaluate and print expression
• 86cc50e test: always use the same python regardless of $PATH
• 71d2d65 packaging: Add pynvim.version attribute
• 056f6f9 fix: ignore flaky OSE...

16 bytes is not enough room for existing usage of the buffer by the tests, so the name may get truncated and cause log_spec test to fail:

FAILED   test/functional/core/log_spec.lua @ 30: log messages are formatted with name or test id
test/helpers.lua:146: retry() attempts: 51
test/h...

After finally getting v0.9.4 uploaded to Debian, some [tests are failing] due to endianness issues.

I'm collecting big-endian fixes here and backporting the big-endian CI to validate them in the PR.

Add loong64 architecture

GHA's ubuntu-latest runner has updated to Ubuntu 22.04, which doesn't have libfuse installed by default.

Debian-based distributions provide a standard interface to launch a terminal via the x-terminal-emulator name. In order for a terminal emualtor to satisfy that interface, it must

• Be VT100 compatiable
• Support the "-e <command> <args>" CLI option
• Support the "-T <title>" CLI option

Adjust ...

Similar to #6242, this will let us ship the changelog in Debian's alacritty package.

The test is supposed to generate the pixmap in memory and then compare it to what's present in tests/images/. However, it's first overwriting the "expected" file, likely left over from generating the expected data in the first place.

This test would sometimes fail in the StopVimInTerminal() because the terminal state was "running", rather than "finished". Adding a debug dump of the terminal lines to the "finished" assert showed that the child vim was "recording @a".

This implies that part of the \<C-o>:\<C-u>qa!\<CR> was ...

Verified on Debian's hppa porterbox.

Closes #5802

Running the tests with “cargo test --target-dir <path>” fails because "target/" doesn't exist.

Rust 1.57.0 added $CARGO_TARGET_TMPDIR as a scratch directory for integration tests to use, which Cargo itself won't touch.

Therefore, use $CARGO_TARGET_TMPDIR if it's available, `$CARGO_TARGET_DI...

This allows highlighting the number column for a sign.

Command-line completion for ":sign define" is also updated to handle "culhl" and "numhl".

Although this can already be done using BUSTED_ARGS, it complements our existing shortcut of TEST_FILTER.

This filetype is a super-set of the diff filetype, as described at https://dep-team.pages.debian.net/deps/dep3/

This is already skipped in all CI environments, so it should also be skipped in environments that don't like fragile tests. Since there's no convenient way to express these concisely, add the explicit fragile skip.

This is already skipped in all CI environments, so it should also be skipped in environments that don't like fragile tests. Since there's no convenient way to express these concisely, add the explicit fragile skip.

[skip ci]

vim-patch:8.2.3664: cannot adjust sign highlighting for 'cursorline'

Problem: Cannot adjust sign highlighting for 'cursorline'. Solution: Add CursorLineSign and CursorLineFold highlight groups. (Gregory Anders, closes vim/vim#9201) https://github.com/vim/vim/commit/e413ea04...

vim-patch:8.2.3874: cannot highlight the number column for a sign

Problem: Cannot highlight the number column for a sign. Solution: Add the "numhl" argument. (James McCoy, closes vim/vim#9381) https://github.com/vim/vim/commit/a80aad717464760a5a50ac2201ce35b24a0cf7a5

Added tests to an outstanding PR -- https://github.com/neovim/neovim/pull/9077 -- to finish getting it ready.

Uploaded a new version of the neovim package:

neovim (0.3.1-3) unstable; urgency=medium .

• rules: Use non-verbose build when DEB_BUILD_OPTIONS=terse
• Use Release build type, not RelWithBuildInfo, to avoid -Og and asserts (Closes: #916338)

As of CMake 3.12, check_include_files() also link the check executable against the libraries listed in CMAKE_REQUIRED_LIBRARIES. Therefore we should unset the CMAKE_REQUIRED_* variables after each respective use to avoid them unnecessarily bleeding into other checks.

https://bitbucket.org/blais/beancount/pull-requests/90/resolve-some-deprecationwarnings

Unix's typical locale-related environment variables aren't always set appropriately on a Mac. Instead of relying on them, query the locale information using Mac specific APIs and then set $LANG appropriately for the rest of nvim.

Closes #5873

It's an empty string, so there's no reason to try to translate it.

Closes #7717

vim-patch:8.0.0590: cannot add a context to locations

Problem: Cannot add a context to locations. Solution: Add the "context" entry in location entries. (Yegappan Lakshmanan, closes vim/vim#1012)

https://github.com/vim/vim/commit/8f77c5a4ec756f3f866bd6b18feb6fca6f2a2e91

...

vim-patch:8.0.0420: text garbled when the system encoding differs from 'encoding'

Problem: When running :make the output may be in the system encoding, different from 'encoding'. Solution: Add the 'makeencoding' option. (Ken Takata)

https://github.com/vim/vim/commit/2c7292...

vim-patch:8.0.0404: not enough testing for quickfix

Problem: Not enough testing for quickfix. Solution: Add some more tests. (Yegappan Lakshmanan)

https://github.com/vim/vim/commit/391b1dd040af204b150d43c5a1c97477ee450a28

vim-patch:8.0.0484: :lhelpgrep does not fail after a suc...

vim-patch:8.0.0517: there is no way to remove quickfix lists

Problem: There is no way to remove quickfix lists (for testing). Solution: Add the 'f' action to setqflist(). Add tests. (Yegappan Lakshmanan)

https://github.com/vim/vim/commit/b6fa30ccc39cdb7f1d07b99fe2f4c6b6167...

vim_memcmp exists to abstract how the memory comparison is performed on various systems and to fallback to Vim's own memcmp implementation, if needed. However, there have been various direct memcmp calls in Vim since at least 2010 without issue.

Converting the remaining vim_memcmp calls to memc...

vim-patch:7.4.1748

Problem: "gD" does not find match in first column of first line. (Gary Johnson) Solution: Accept match at the cursor.

https://github.com/vim/vim/commit/1538fc34fae3fae39773ca43f6ff52401fce61d8

vim-patch:f9660b5

Add missing test file.

https://git...

Gpg4win expects native Windows paths even when it's being run from a cygwin environment. In order to satisfy that, update s:shellescape to know whether it should convert the paths to native Windows paths using cygpath -am.

In order to detect whether cygpath should be invoked, look for ...

vim-patch:7.4.1708

Problem: New regexp engine does not work properly with EBCDIC. Solution: Define equivalence class characters. (Owen Leibman)

https://github.com/vim/vim/commit/2a6fa564a3b5061c14ff63b8b0f12801df0b0ac2

Although nvim doesn't support EBCDIC systems, this keeps us inlin...

vim-patch:7.4.1751

Problem: Crash when 'tagstack' is off. (Dominique Pelle) Solution: Fix it. (Hirohito Higashi)

https://github.com/vim/vim/commit/def5abe0a2727041ecee69afdccfca405333bd24

vim-patch:8d8aa0a

Add missing test file.

https://github.com/vim/vim/commit/8d8aa0a367380...

vim-patch:7.4.1754

Problem: When 'filetype' was set and reloading a buffer which does not cause it to be set, the syntax isn't loaded. (KillTheMule) Solution: Remember whether the FileType event was fired and fire it if not. (Anton Lindqvist, closes vim/vim#747)...

• 1727: garbagecollect_for_testing()
• 1761: NA Coverity issue
• 1769: Adding attributes to if_pyth's sys.stdout
• 1781: synIDattr() support for mode=g in terminal
• 1812: Athena/Motif GUI fixes
• 1838: Rename of test functions (e.g., garbagecollect_for_testing()) to start with "test_"
• 1844: Mo...

vim-patch:7.4.1867

Problem: Memory leak in test_matchstrpos. Solution: Free the string before overwriting. (Yegappan Lakshmanan)

https://github.com/vim/vim/commit/3c809343c72d9964475f421fd03bb892bc584a51

Previously, free_oldval was set immediately on entering did_set_string_option. However, opt_idx hadn't been adjusted to account for diverting 'viminfo' manipulation to 'shada'.

Therefore, the code which determines whether to free the old value was looking at the flags for 'viminfo' while the co...

Closes #5759

vim-patch:7.4.1752

Problem: When adding to the quickfix list the current position is reset. Solution: Do not reset the position when not needed. (Yegappan Lakshmanan)

https://github.com/vim/vim/commit/c1808d5822ed9534ef7f0fe509b15bee92a5cc28

vim-patch:7.4.1768

Problem: Argu...

I commonly work on a patch series from different computers. Being able to get the preprocessed patch without performing the branch prep and empty commit would ease that workflow.

neovim/neovim#5529 merged support for Vim's partial functions, which made nvim more strict about dictionary functions. This means the implicit self variable isn't available unless a function is annoted as a dict function or defined as part of a dictionary.

Closes #392

shada_get_buflist finds all the relevant buffers and generates the kSDItemBufferList ShadaEntry.

This cleans up a lint issue introduced in #5760

In 7.4.1896, guards were added around specific mark_adjust() calls to avoid updating marks if a new line was being added to the end of the buffer. However, this breaks updating the state of windows in 'diffmode' when a buffer is changed in that manner, leading to incorrect diff views.

It's not atypical to have extra, non-"vim-patch" commits in a PR merging a bigger or set of related patches from Vim. Rather than immediately aborting, display the patch header to the user and let them decide whether to continue reviewing.

There have been various breakages of “make shadow” when new tests are added. I rely on this to parallelize the build of Debian's packages, so prefer for it not to get broken.

Note that this is currently failing CI due to an existing test failure, which has a fix in #515.

There is an ongoing effort[0] to make FOSS software reproducibly buildable. In order to make NeoVim build reproducibly, it is necessary to allow defining the date/time used in longVersionWithDate as part of the build process.

This commit enables that by adding support for the SOURCE_DATE_E...

There is an ongoing effort[0] to make FOSS software reproducibly buildable. In order to make Vim build reproducibly, it is necessary to allow defining the date/time that is part of VIM_VERSION_LONG as part of the build process.

This commit enables that by adding support for the SOURCE_DATE_EPOC...

Also, update the URLs for the canonical sources of the files.

Various test expressions in the configure script were using unquoted variables which reference pathnames. If these paths contain whitespace the test expression will be malformed.

Although the autoconf related files don't change often, it's useful to ensure they can be rebuilt. This will help catch possible breakage and make it easier for pull requests which modify src/configure.in to get sanity tested.

In 8e46f7264cc303ae3fd723b2f27badccb1d68fbf test86 and test89 were adjusted to work correctly when using a shadowdir. This commit applies a similar fix to test87.

Using a Vim that is built to be able to access X's PRIMARY and CLIPBOARD selections (i.e., "* and "+), performing “:let &term=&term” causes Vim to "forget" that it knows how to access those registers.

This is because set_termname() ends up calling clip_init(), which sets owned = FALSE on both Vi...

This should address #74.

This addresses #417