Git 2.35.0

Git 2.35 Release Notes
======================

Updates since Git 2.34
----------------------

Backward compatibility warts

 * "_" is now treated as any other URL-valid characters in an URL when
   matching the per-URL configuration variable names.

 * The color palette used by "git grep" has been updated to match that
   of GNU grep.


Note to those who build from the source

 * You may need to define NO_UNCOMPRESS2 Makefile macro if you build
   with zlib older than 1.2.9.

 * If your compiler cannot grok C99, the build will fail.  See the
   instruction at the beginning of git-compat-util.h if this happens
   to you.


UI, Workflows & Features

 * "git status --porcelain=v2" now show the number of stash entries
   with --show-stash like the normal output does.

 * "git stash" learned the "--staged" option to stash away what has
   been added to the index (and nothing else).

 * "git var GIT_DEFAULT_BRANCH" is a way to see what name is used for
   the newly created branch if "git init" is run.

 * Various operating modes of "git reset" have been made to work
   better with the sparse index.

 * "git submodule deinit" for a submodule whose .git metadata
   directory is embedded in its working tree refused to work, until
   the submodule gets converted to use the "absorbed" form where the
   metadata directory is stored in superproject, and a gitfile at the
   top-level of the working tree of the submodule points at it.  The
   command is taught to convert such submodules to the absorbed form
   as needed.

 * The completion script (in contrib/) learns that the "--date"
   option of commands from the "git log" family takes "human" and
   "auto" as valid values.

 * "Zealous diff3" style of merge conflict presentation has been added.

 * The "git log --format=%(describe)" placeholder has been extended to
   allow passing selected command-line options to the underlying "git
   describe" command.

 * "default" and "reset" have been added to our color palette.

 * The cryptographic signing using ssh keys can specify literal keys
   for keytypes whose name do not begin with the "ssh-" prefix by
   using the "key::" prefix mechanism (e.g. "key::ecdsa-sha2-nistp256").

 * "git fetch" without the "--update-head-ok" option ought to protect
   a checked out branch from getting updated, to prevent the working
   tree that checks it out to go out of sync.  The code was written
   before the use of "git worktree" got widespread, and only checked
   the branch that was checked out in the current worktree, which has
   been updated.

 * "git name-rev" has been tweaked to give output that is shorter and
   easier to understand.

 * "git apply" has been taught to ignore a message without a patch
   with the "--allow-empty" option.  It also learned to honor the
   "--quiet" option given from the command line.

 * The "init" and "set" subcommands in "git sparse-checkout" have been
   unified for a better user experience and performance.

 * Many git commands that deal with working tree files try to remove a
   directory that becomes empty (i.e. "git switch" from a branch that
   has the directory to another branch that does not would attempt
   remove all files in the directory and the directory itself).  This
   drops users into an unfamiliar situation if the command was run in
   a subdirectory that becomes subject to removal due to the command.
   The commands have been taught to keep an empty directory if it is
   the directory they were started in to avoid surprising users.

 * "git am" learns "--empty=(stop|drop|keep)" option to tweak what is
   done to a piece of e-mail without a patch in it.

 * The default merge message prepared by "git merge" records the name
   of the current branch; the name can be overridden with a new option
   to allow users to pretend a merge is made on a different branch.

 * The way "git p4" shows file sizes in its output has been updated to
   use human-readable units.

 * "git -c branch.autosetupmerge=inherit branch new old" makes "new"
   to have the same upstream as the "old" branch, instead of marking
   "old" itself as its upstream.


Performance, Internal Implementation, Development Support etc.

 * The use of errno as a means to carry the nature of error in the ref
   API implementation has been reworked and reduced.

 * Teach and encourage first-time contributors to this project to
   state the base commit when they submit their topic.

 * The command line completion for "git send-email" options have been
   tweaked to make it easier to keep it in sync with the command itself.

 * Ensure that the sparseness of the in-core index matches the
   index.sparse configuration specified by the repository immediately
   after the on-disk index file is read.

 * Code clean-up to eventually allow information on remotes defined
   for an arbitrary repository to be read.

 * Build optimization.

 * Tighten code for testing pack-bitmap.

 * Weather balloon to break people with compilers that do not support
   C99.

 * The "reftable" backend for the refs API, without integrating into
   the refs subsystem, has been added.

 * More tests are marked as leak-free.

 * The test framework learns to list unsatisfied test prerequisites,
   and optionally error out when prerequisites that are expected to be
   satisfied are not.

 * The default setting for trace2 event nesting was too low to cause
   test failures, which is worked around by bumping it up in the test
   framework.

 * Drop support for TravisCI and update test workflows at GitHub.

 * Many tests that used to need GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME
   mechanism to force "git" to use 'master' as the default name for
   the initial branch no longer need it; the use of the mechanism from
   them have been removed.

 * Allow running our tests while disabling fsync.

 * Document the parameters given to the reflog entry iterator callback
   functions.
   (merge e6e94f34b2 jc/reflog-iterator-callback-doc later to maint).

 * The test helper for refs subsystem learned to write bogus and/or
   nonexistent object name to refs to simulate error situations we
   want to test Git in.

 * "diff --histogram" optimization.

 * Weather balloon to find compilers that do not grok variable
   declaration in the for() loop.

 * diff and blame commands have been taught to work better with sparse
   index.

 * The chainlint test script linter in the test suite has been updated.

 * The DEVELOPER=yes build uses -std=gnu99 now.

 * "git format-patch" uses a single rev_info instance and then exits.
   Mark the structure with UNLEAK() macro to squelch leak sanitizer.

 * New interface into the tmp-objdir API to help in-core use of the
   quarantine feature.

 * Broken &&-chains in the test scripts have been corrected.

 * The RCS keyword substitution in "git p4" used to be done assuming
   that the contents are UTF-8 text, which can trigger decoding
   errors.  We now treat the contents as a bytestring for robustness
   and correctness.

 * The conditions to choose different definitions of the FLEX_ARRAY
   macro for vendor compilers has been simplified to make it easier to
   maintain.

 * Correctness and performance update to "diff --color-moved" feature.

 * "git upload-pack" (the other side of "git fetch") used a 8kB buffer
   but most of its payload came on 64kB "packets".  The buffer size
   has been enlarged so that such a packet fits.

 * "git fetch" and "git pull" are now declared sparse-index clean.
   Also "git ls-files" learns the "--sparse" option to help debugging.

 * Similar message templates have been consolidated so that
   translators need to work on fewer number of messages.


Fixes since v2.34
-----------------

 * "git grep" looking in a blob that has non-UTF8 payload was
   completely broken when linked with certain versions of PCREv2
   library in the latest release.

 * Other code cleanup, docfix, build fix, etc.

 * "git pull" with any strategy when the other side is behind us
   should succeed as it is a no-op, but doesn't.

 * An earlier change in 2.34.0 caused JGit application (that abused
   GIT_EDITOR mechanism when invoking "git config") to get stuck with
   a SIGTTOU signal; it has been reverted.

 * An earlier change that broke .gitignore matching has been reverted.

 * Things like "git -c branch.sort=bogus branch new HEAD", i.e. the
   operation modes of the "git branch" command that do not need the
   sort key information, no longer errors out by seeing a bogus sort
   key.
   (merge 98e7ab6d42 jc/fix-ref-sorting-parse later to maint).

 * The compatibility implementation for unsetenv(3) were written to
   mimic ancient, non-POSIX, variant seen in an old glibc; it has been
   changed to return an integer to match the more modern era.
   (merge a38989bd5b jc/unsetenv-returns-an-int later to maint).

 * The clean/smudge conversion code path has been prepared to better
   work on platforms where ulong is narrower than size_t.
   (merge 596b5e77c9 mc/clean-smudge-with-llp64 later to maint).

 * Redact the path part of packfile URI that appears in the trace output.
   (merge 0ba558ffb1 if/redact-packfile-uri later to maint).

 * CI has been taught to catch some Unicode directional formatting
   sequence that can be used in certain mischief.
   (merge 0e7696c64d js/ci-no-directional-formatting later to maint).

 * The "--date=format:" gained a workaround for the lack of
   system support for a non-local timezone to handle "%s" placeholder.
   (merge 9b591b9403 jk/strbuf-addftime-seconds-since-epoch later to maint).

 * The "merge" subcommand of "git jump" (in contrib/) silently ignored
   pathspec and other parameters.
   (merge 67ba13e5a4 jk/jump-merge-with-pathspec later to maint).

 * The code to decode the length of packed object size has been
   corrected.
   (merge 34de5b8eac jt/pack-header-lshift-overflow later to maint).

 * The advice message given by "git pull" when the user hasn't made a
   choice between merge and rebase still said that the merge is the
   default, which no longer is the case.  This has been corrected.
   (merge 71076d0edd ah/advice-pull-has-no-preference-between-rebase-and-merge later to maint).

 * "git fetch", when received a bad packfile, can fail with SIGPIPE.
   This wasn't wrong per-se, but we now detect the situation and fail
   in a more predictable way.
   (merge 2a4aed42ec jk/fetch-pack-avoid-sigpipe-to-index-pack later to maint).

 * The function to cull a child process and determine the exit status
   had two separate code paths for normal callers and callers in a
   signal handler, and the latter did not yield correct value when the
   child has caught a signal.  The handling of the exit status has
   been unified for these two code paths.  An existing test with
   flakiness has also been corrected.
   (merge 5263e22cba jk/t7006-sigpipe-tests-fix later to maint).

 * When a non-existent program is given as the pager, we tried to
   reuse an uninitialized child_process structure and crashed, which
   has been fixed.
   (merge f917f57f40 em/missing-pager later to maint).

 * The single-key-input mode in "git add -p" had some code to handle
   keys that generate a sequence of input via ReadKey(), which did not
   handle end-of-file correctly, which has been fixed.
   (merge fc8a8126df cb/add-p-single-key-fix later to maint).

 * "git rebase -x" added an unnecessary 'exec' instructions before
   'noop', which has been corrected.
   (merge cc9dcdee61 en/rebase-x-fix later to maint).

 * When the "git push" command is killed while the receiving end is
   trying to report what happened to the ref update proposals, the
   latter used to die, due to SIGPIPE.  The code now ignores SIGPIPE
   to increase our chances to run the post-receive hook after it
   happens.
   (merge d34182b9e3 rj/receive-pack-avoid-sigpipe-during-status-reporting later to maint).

 * "git worktree add" showed "Preparing worktree" message to the
   standard output stream, but when it failed, the message from die()
   went to the standard error stream.  Depending on the order the
   stdio streams are flushed at the program end, this resulted in
   confusing output.  It has been corrected by sending all the chatty
   messages to the standard error stream.
   (merge b50252484f es/worktree-chatty-to-stderr later to maint).

 * Coding guideline document has been updated to clarify what goes to
   standard error in our system.
   (merge e258eb4800 es/doc-stdout-vs-stderr later to maint).

 * The sparse-index/sparse-checkout feature had a bug in its use of
   the matching code to determine which path is in or outside the
   sparse checkout patterns.
   (merge 8c5de0d265 ds/sparse-deep-pattern-checkout-fix later to maint).

 * "git rebase -x" by mistake started exporting the GIT_DIR and
   GIT_WORK_TREE environment variables when the command was rewritten
   in C, which has been corrected.
   (merge 434e0636db en/rebase-x-wo-git-dir-env later to maint).

 * When "git log" implicitly enabled the "decoration" processing
   without being explicitly asked with "--decorate" option, it failed
   to read and honor the settings given by the "--decorate-refs"
   option.

 * "git fetch --set-upstream" did not check if there is a current
   branch, leading to a segfault when it is run on a detached HEAD,
   which has been corrected.
   (merge 17baeaf82d ab/fetch-set-upstream-while-detached later to maint).

 * Among some code paths that ask an yes/no question, only one place
   gave a prompt that looked different from the others, which has been
   updated to match what the others create.
   (merge 0fc8ed154c km/help-prompt-fix later to maint).

 * "git log --invert-grep --author=" used to exclude commits
   written by the given author, but now "--invert-grep" only affects
   the matches made by the "--grep=" option.
   (merge 794c000267 rs/log-invert-grep-with-headers later to maint).

 * "git grep --perl-regexp" failed to match UTF-8 characters with
   wildcard when the pattern consists only of ASCII letters, which has
   been corrected.
   (merge 32e3e8bc55 rs/pcre2-utf later to maint).

 * Certain sparse-checkout patterns that are valid in non-cone mode
   led to segfault in cone mode, which has been corrected.

 * Use of certain "git rev-list" options with "git fast-export"
   created nonsense results (the worst two of which being "--reverse"
   and "--invert-grep --grep=").  The use of "--first-parent" is
   made to behave a bit more sensible than before.
   (merge 726a228dfb ws/fast-export-with-revision-options later to maint).

 * Perf tests were run with end-user's shell, but it has been
   corrected to use the shell specified by $TEST_SHELL_PATH.
   (merge 9ccab75608 ja/perf-use-specified-shell later to maint).

 * Fix dependency rules to generate hook-list.h header file.
   (merge d3fd1a6667 ab/makefile-hook-list-dependency-fix later to maint).

 * "git stash" by default triggers its "push" action, but its
   implementation also made "git stash -h" to show short help only for
   "git stash push", which has been corrected.
   (merge ca7990cea5 ab/do-not-limit-stash-help-to-push later to maint).

 * "git apply --3way" bypasses the attempt to do a three-way
   application in more cases to address the regression caused by the
   recent change to use direct application as a fallback.
   (merge 34d607032c jz/apply-3-corner-cases later to maint).

 * Fix performance-releated bug in "git subtree" (in contrib/).
   (merge 3ce8888fb4 jl/subtree-check-parents-argument-passing-fix later to maint).

 * Extend the guidance to choose the base commit to build your work
   on, and hint/nudge contributors to read others' changes.
   (merge fdfae830f8 jc/doc-submitting-patches-choice-of-base later to maint).

 * A corner case bug in the ort merge strategy has been corrected.
   (merge d30126c20d en/merge-ort-renorm-with-rename-delete-conflict-fix later to maint).

 * "git stash apply" forgot to attempt restoring untracked files when
   it failed to restore changes to tracked ones.
   (merge 71cade5a0b en/stash-df-fix later to maint).

 * Calling dynamically loaded functions on Windows has been corrected.
   (merge 4a9b204920 ma/windows-dynload-fix later to maint).

 * Some lockfile code called free() in signal-death code path, which
   has been corrected.
   (merge 58d4d7f1c5 ps/lockfile-cleanup-fix later to maint).

 * Other code cleanup, docfix, build fix, etc.
   (merge 74db416c9c cw/protocol-v2-doc-fix later to maint).
   (merge f9b2b6684d ja/doc-cleanup later to maint).
   (merge 7d1b866778 jc/fix-first-object-walk later to maint).
   (merge 538ac74604 js/trace2-avoid-recursive-errors later to maint).
   (merge 152923b132 jk/t5319-midx-corruption-test-deflake later to maint).
   (merge 9081a421a6 ab/checkout-branch-info-leakfix later to maint).
   (merge 42c456ff81 rs/mergesort later to maint).
   (merge ad506e6780 tl/midx-docfix later to maint).
   (merge bf5b83fd8a hk/ci-checkwhitespace-commentfix later to maint).
   (merge 49f1eb3b34 jk/refs-g11-workaround later to maint).
   (merge 7d3fc7df70 jt/midx-doc-fix later to maint).
   (merge 7b089120d9 hn/create-reflog-simplify later to maint).
   (merge 9e12400da8 cb/mingw-gmtime-r later to maint).
   (merge 0bf0de6cc7 tb/pack-revindex-on-disk-cleanup later to maint).
   (merge 2c68f577fc ew/cbtree-remove-unused-and-broken-cb-unlink later to maint).
   (merge eafd6e7e55 ab/die-with-bug later to maint).
   (merge 91028f7659 jc/grep-patterntype-default-doc later to maint).
   (merge 47ca93d071 ds/repack-fixlets later to maint).
   (merge e6a9bc0c60 rs/t4202-invert-grep-test-fix later to maint).
   (merge deb5407a42 gh/gpg-doc-markup-fix later to maint).
   (merge 999bba3e0b rs/daemon-plug-leak later to maint).
   (merge 786eb1ba39 js/l10n-mention-ngettext-early-in-readme later to maint).
   (merge 2f12b31b74 ab/makefile-msgfmt-wo-stats later to maint).
   (merge 0517f591ca fs/gpg-unknown-key-test-fix later to maint).
   (merge 97d6fb5a1f ma/header-dup-cleanup later to maint).

FUD and Security Leftovers

• Linux Servers at Risk of RCE Due to Critical CWP Bugs [Ed: It's not a Linux issue but a program that can run on top of it; FUD pattern?]

• CWP bugs allow code execution as root on Linux servers, patch now [Ed: Microsoft boosters make a bug in CentOS Web Panel sound like it's an issue with Linux]

  Two security vulnerabilities that impact the Control Web Panel (CWP) software can be chained by unauthenticated attackers to gain remote code execution (RCE) as root on vulnerable Linux servers. CWP, previously known as CentOS Web Panel, is a free Linux control panel for managing dedicated web hosting servers and virtual private servers.

• CISA Publishes Infographic on Layering Network Security Through Segmentation

  CISA has published an infographic to emphasize the importance of implementing network segmentation—a physical or virtual architectural approach that divides a network into multiple segments, each acting as its own subnetwork, to provide additional security and control that can help prevent or minimize the impact of a cyberattack.

• LHS Episode #449: Insecurities Everywhere

  Hello and welcome to the 449th installment of Linux in the Ham Shack. In this short-topics episode, the hosts discuss results from the 2021 QSO parties, the FCC tech advisory council, the proliferation of Linux malware, SDR++, programming in Python and much more. Thank you for listening and have a fantastic week.

Today in Techrights

• Links 25/1/2022: GPL Settlement With Patrick McHardy, Godot 4.0 Alpha 1, and DXVK 1.9.4 Released
• Proprietary Software is Pollution
• Microsoft GitHub Exposé — Part XV — Cover-Up and Defamation

today's leftovers

• The Apache Weekly News Round-up: week ending 21 January 2022

  We're wrapping up another great week with the following activities from the Apache community...

• How the debuginfod project evolved in 2021

  Debuginfod is a web service for distributing debugging resources over HTTP. The project began in 2019 and has seen a lot of growth in 2021.

• Am getting the hang of it…. [Ed: archbang-winter-2201-x86_64.iso has been released]

  With a bit of spare time on my hands between decorating and cooking, been slowly setting up dk tiling window manager. Did try lemonbar as a panel but I soon lost interest and went back to tint2, more the fact I am used to it more. One thing I have added is super + a which fires up rofi showing open windows, not only that it tells you which workspace window is on. Was trying out floating windows and just moved them to workspace 10 to test them out only they would not return to full screen. Found it very odd so tried them on 9 and they worked fine. So thought I would take a look at dkrc file and sure enough there is a line that sets windows, gaps and borders on workspace 10.

• Brazilian telecoms giant accelerates innovation with SUSE | SUSE Communities

  “SUSE Rancher makes it easier for our IT staff to manage and scale a large container environment. As a result, we can spend less time on repetitive manual tasks and more time on value-added development.” Marcos Borges, Telco/DC Application Manager, Algar Telecom.

• Linux Foundation launches Open Source Software Development, Linux, and Git certification [Ed: Paid-for LF spam by SJVN at ZDNet. 'Obligatory' articles for sponsors.]

  Want a good-paying programming job? By ZipRecruiter's count, the average annual pay for an open-source developer in the United States is $123,411. That's not bad. There's also a lot of demand for Linux and open-source pros. The Linux Foundation and edX, the leading massive open online course (MOOC) provider, reported in their 2021 Open Source Jobs Report that the pair found more demand for top open-source workers than ever.