* [ANNOUNCE] Git v2.45.0
@ 2024-04-29 17:12 Junio C Hamano
2024-05-01 17:39 ` rsbecker
2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen
0 siblings, 2 replies; 12+ messages in thread
From: Junio C Hamano @ 2024-04-29 17:12 UTC (permalink / raw)
To: git; +Cc: Linux Kernel, git-packagers
The latest feature release Git v2.45.0 is now available at the
usual places. It is comprised of 540 non-merge commits since
v2.44.0, contributed by 96 people, 38 of which are new faces [*].
The tarballs are found at:
https://www.kernel.org/pub/software/scm/git/
The following public repositories all have a copy of the 'v2.45.0'
tag and the 'master' branch that the tag points at:
url = https://git.kernel.org/pub/scm/git/git
url = https://kernel.googlesource.com/pub/scm/git/git
url = git://repo.or.cz/alt-git.git
url = https://github.com/gitster/git
New contributors whose contributions weren't in v2.44.0 are as follows.
Welcome to the Git development community!
Ahelenia Ziemiańska, Angelo Dureghello, Aryan Gupta, Benjamin
Flesch, Bo Anderson, Brian C Tracy, Brian Lyles, Bruno Haible,
Chuck Lever, Dario Gjorgjevski, Dirk Gouders, Eugenio Gigante,
Florian Schmidt, Haritha D, Harmen Stoppels, Jean-Rémy
Falleri, Jiamu Sun, Jonas Wunderlich, Jonathan Davies, Julio
Bacellari, Kipras Melnikovas, Kisaragi Hiu, Marcel Röthke,
Matthew Rollings, Max Gautier, mirth hickford, Paweł Dominiak,
Pi Fisher, Ralph Seichter, Richard Macklin, shejialuo, Steven
Jeuris, Thalia Archibald, Tiago Pascoal, Vincenzo Mezzela,
Vũ Tiến Hưng, Xing Xin, and Yehezkel Bernat.
Returning contributors who helped this release are as follows.
Thanks for your continued support.
Alexander Shopov, Arkadii Yakovets, Bagas Sanjaya, Beat Bolli,
brian m. carlson, Chandra Pratap, Christian Couder, Derrick
Stolee, Đoàn Trần Công Danh, Dragan Simic, Elijah Newren,
Emir SARI, Eric Sunshine, Eric W. Biederman, Ghanshyam Thakkar,
Han Young, Jakub Wilk, Jean-Noël Avila, Jeff Hostetler, Jeff
King, Jiang Xin, Johannes Schindelin, Johannes Sixt, John Cai,
Josh Steadmon, Josh Triplett, Junio C Hamano, Justin Tobler,
Karthik Nayak, Kate Golovanova, Kristoffer Haugsbakk, Kyle
Lippincott, Kyle Meyer, Linus Arver, Lumynous, Manlio Perillo,
Matthias Aßhauer, Matthias Rüster, M Hickford, Michael Lohmann,
Michael Osipov, Mike Hommey, Orgad Shaneh, Patrick Steinhardt,
Peter Hutterer, Peter Krefting, Philippe Blain, Phillip Wood,
Ralf Thielow, René Scharfe, Rubén Justo, Sergey Organov,
SZEDER Gábor, Taylor Blau, Teng Long, Ville Skyttä, Yasushi
SHOJI, and Yi-Jyun Pan.
[*] We are counting not just the authorship contribution but issue
reporting, mentoring, helping and reviewing that are recorded in
the commit trailers.
----------------------------------------------------------------
Git v2.45 Release Notes
=======================
Backward Compatibility Notes
UI, Workflows & Features
* Integrate the reftable code into the refs framework as a backend.
With "git init --ref-format=reftable", hopefully it would be a lot
more efficient to manage a repository with many references.
* "git checkout -p" and friends learned that that "@" is a synonym
for "HEAD".
* Variants of vimdiff learned to honor mergetool.<variant>.layout
settings.
* "git reflog" learned a "list" subcommand that enumerates known reflogs.
* When a merge conflicted at a submodule, merge-ort backend used to
unconditionally give a lengthy message to suggest how to resolve
it. Now the message can be squelched as an advice message.
* "git for-each-ref" learned "--include-root-refs" option to show
even the stuff outside the 'refs/' hierarchy.
* "git rev-list --missing=print" has learned to optionally take
"--allow-missing-tips", which allows the objects at the starting
points to be missing.
* "git merge-tree" has learned that the three trees involved in the
3-way merge only need to be trees, not necessarily commits.
* "git log --merge" learned to pay attention to CHERRY_PICK_HEAD and
other kinds of *_HEAD pseudorefs.
* Platform specific tweaks for OS/390 has been added to
config.mak.uname.
* Users with safe.bareRepository=explicit can still work from within
$GIT_DIR of a seconary worktree (which resides at .git/worktrees/$name/)
of the primary worktree without explicitly specifying the $GIT_DIR
environment variable or the --git-dir=<path> option.
* The output format for dates "iso-strict" has been tweaked to show
a time in the Zulu timezone with "Z" suffix, instead of "+00:00".
* "git diff" and friends learned two extra configuration variables,
diff.srcPrefix and diff.dstPrefix.
* The status.showUntrackedFiles configuration variable had a name
that tempts users to set a Boolean value expressed in our usual
"false", "off", and "0", but it only took "no". This has been
corrected so "true" and its synonyms are taken as "normal", while
"false" and its synonyms are taken as "no".
* Remove an ancient and not well maintained Hg-to-git migration
script from contrib/.
* Hints that suggest what to do after resolving conflicts can now be
squelched by disabling advice.mergeConflict.
* Allow git-cherry-pick(1) to automatically drop redundant commits via
a new `--empty` option, similar to the `--empty` options for
git-rebase(1) and git-am(1). Includes a soft deprecation of
`--keep-redundant-commits` as well as some related docs changes and
sequencer code cleanup.
* "git config" learned "--comment=<message>" option to leave a
comment immediately after the "variable = value" on the same line
in the configuration file.
* core.commentChar used to be limited to a single byte, but has been
updated to allow an arbitrary multi-byte sequence.
* "git add -p" and other "interactive hunk selection" UI has learned to
skip showing the hunk immediately after it has already been shown, and
an additional action to explicitly ask to reshow the current hunk.
* "git pack-refs" learned the "--auto" option, which defers the decision of
whether and how to pack to the ref backend. This is used by the reftable
backend to avoid repacking of an already-optimal ref database. The new mode
is triggered from "git gc --auto".
* "git add -u <pathspec>" and "git commit [-i] <pathspec>" did not
diagnose a pathspec element that did not match any files in certain
situations, unlike "git add <pathspec>" did.
* The userdiff patterns for C# has been updated.
* Git writes a "waiting for your editor" message on an incomplete
line after launching an editor, and then append another error
message on the same line if the editor errors out. It now clears
the "waiting for..." line before giving the error message.
* The filename used for rejected hunks "git apply --reject" creates
was limited to PATH_MAX, which has been lifted.
* When "git bisect" reports the commit it determined to be the
culprit, we used to show it in a format that does not honor common
UI tweaks, like log.date and log.decorate. The code has been
taught to use "git show" to follow more customizations.
Performance, Internal Implementation, Development Support etc.
* The code to iterate over refs with the reftable backend has seen
some optimization.
* More tests that are marked as "ref-files only" have been updated to
improve test coverage of reftable backend.
* Some parts of command line completion script (in contrib/) have
been micro-optimized.
* The way placeholders are to be marked-up in documentation have been
specified; use "_<placeholder>_" to typeset the word inside a pair
of <angle-brackets> emphasized.
* "git --no-lazy-fetch cmd" allows to run "cmd" while disabling lazy
fetching of objects from the promisor remote, which may be handy
for debugging.
* The implementation in "git clean" that makes "-n" and "-i" ignore
clean.requireForce has been simplified, together with the
documentation.
* Uses of xwrite() helper have been audited and updated for better
error checking and simpler code.
* Some trace2 events that lacked def_param have learned to show it,
enriching the output.
* The parse-options code that deals with abbreviated long option
names have been cleaned up.
* The code in reftable backend that creates new table files works
better with the tempfile framework to avoid leaving cruft after a
failure.
* The reftable code has its own custom binary search function whose
comparison callback has an unusual interface, which caused the
binary search to degenerate into a linear search, which has been
corrected.
* The code to iterate over reflogs in the reftable has been optimized
to reduce memory allocation and deallocation.
* Work to support a repository that work with both SHA-1 and SHA-256
hash algorithms has started.
* A new fuzz target that exercises config parsing code has been
added.
* Fix the way recently added tests interpolate variables defined
outside them, and document the best practice to help future
developers.
* Introduce an experimental protocol for contributors to propose the
topic description to be used in the "What's cooking" report, the
merge commit message for the topic, and in the release notes and
document it in the SubmittingPatches document.
* The t/README file now gives a hint on running individual tests in
the "t/" directory with "make t<num>-*.sh t<num>-*.sh".
(merge 8d383806fc pb/test-scripts-are-build-targets later to maint).
* The "hint:" messages given by the advice mechanism, when given a
message with a blank line, left a line with trailing whitespace,
which has been cleansed.
* Documentation rules has been explicitly described how to mark-up
literal parts and a few manual pages have been updated as examples.
* The .editorconfig file has been taught that a Makefile uses HT
indentation.
* t-prio-queue test has been cleaned up by using C99 compound
literals; this is meant to also serve as a weather-balloon to smoke
out folks with compilers who have trouble compiling code that uses
the feature.
* Windows binary used to decide the use of unix-domain socket at
build time, but it learned to make the decision at runtime instead.
* The "shared repository" test in the t0610 reftable test failed
under restrictive umask setting (e.g. 007), which has been
corrected.
* Document and apply workaround for a buggy version of dash that
mishandles "local var=val" construct.
* The codepaths that reach date_mode_from_type() have been updated to
pass "struct date_mode" by value to make them thread safe.
* The strategy to compact multiple tables of reftables after many
operations accumulate many entries has been improved to avoid
accumulating too many tables uncollected.
* The code to iterate over reftable blocks has seen some optimization
to reduce memory allocation and deallocation.
* The way "git fast-import" handles paths described in its input has
been tightened up and more clearly documented.
* The cvsimport tests required that the platform understands
traditional timezone notations like CST6CDT, which has been
updated to work on those systems as long as they understand
POSIX notation with explicit tz transition dates.
* The code to format trailers have been cleaned up.
Fixes since v2.44
-----------------
* "git apply" on a filesystem without filemode support have learned
to take a hint from what is in the index for the path, even when
not working with the "--index" or "--cached" option, when checking
the executable bit match what is required by the preimage in the
patch.
(merge 45b625142d cp/apply-core-filemode later to maint).
* "git column" has been taught to reject negative padding value, as
it would lead to nonsense behaviour including division by zero.
(merge 76fb807faa kh/column-reject-negative-padding later to maint).
* "git am --help" now tells readers what actions are available in
"git am --whitespace=<action>", in addition to saying that the
option is passed through to the underlying "git apply".
(merge a171dac734 jc/am-whitespace-doc later to maint).
* "git tag --column" failed to check the exit status of its "git
column" invocation, which has been corrected.
(merge 92e66478fc rj/tag-column-fix later to maint).
* Credential helper based on libsecret (in contrib/) has been updated
to handle an empty password correctly.
(merge 8f1f2023b7 mh/libsecret-empty-password-fix later to maint).
* "git difftool --dir-diff" learned to honor the "--trust-exit-code"
option; it used to always exit with 0 and signalled success.
(merge eb84c8b6ce ps/difftool-dir-diff-exit-code later to maint).
* The code incorrectly attempted to use textconv cache when asked,
even when we are not running in a repository, which has been
corrected.
(merge affe355fe7 jk/textconv-cache-outside-repo-fix later to maint).
* Remove an empty file that shouldn't have been added in the first
place.
(merge 4f66942215 js/remove-cruft-files later to maint).
* The logic to access reflog entries by date and number had ugly
corner cases at the boundaries, which have been cleaned up.
(merge 5edd126720 jk/reflog-special-cases-fix later to maint).
* An error message from "git upload-pack", which responds to "git
fetch" requests, had a trailing NUL in it, which has been
corrected.
(merge 3f4c7a0805 sg/upload-pack-error-message-fix later to maint).
* Clarify wording in the CodingGuidelines that requires <git-compat-util.h>
to be the first header file.
(merge 4e89f0e07c jc/doc-compat-util later to maint).
* "git commit -v --cleanup=scissors" used to add the scissors line
twice in the log message buffer, which has been corrected.
(merge e90cc075cc jt/commit-redundant-scissors-fix later to maint).
* A custom remote helper no longer cannot access the newly created
repository during "git clone", which is a regression in Git 2.44.
This has been corrected.
(merge 199f44cb2e ps/remote-helper-repo-initialization-fix later to maint).
* Various parts of upload-pack have been updated to bound the resource
consumption relative to the size of the repository to protect from
abusive clients.
(merge 6cd05e768b jk/upload-pack-bounded-resources later to maint).
* The upload-pack program, when talking over v2, accepted the
packfile-uris protocol extension from the client, even if it did
not advertise the capability, which has been corrected.
(merge a922bfa3b5 jk/upload-pack-v2-capability-cleanup later to maint).
* Make sure failure return from merge_bases_many() is properly caught.
(merge 25fd20eb44 js/merge-base-with-missing-commit later to maint).
* FSMonitor client code was confused when FSEvents were given in a
different case on a case-insensitive filesystem, which has been
corrected.
(merge 29c139ce78 jh/fsmonitor-icase-corner-case-fix later to maint).
* The "core.commentChar" configuration variable only allows an ASCII
character, which was not clearly documented, which has been
corrected.
(merge fb7c556f58 kh/doc-commentchar-is-a-byte later to maint).
* With release 2.44 we got rid of all uses of test_i18ngrep and there
is no in-flight topic that adds a new use of it. Make a call to
test_i18ngrep a hard failure, so that we can remove it at the end
of this release cycle.
(merge 381a83dfa3 jc/test-i18ngrep later to maint).
* The command line completion script (in contrib/) learned to
complete "git reflog" better.
(merge 1284f9cc11 rj/complete-reflog later to maint).
* The logic to complete the command line arguments to "git worktree"
subcommand (in contrib/) has been updated to correctly honor things
like "git -C dir" etc.
(merge 3574816d98 rj/complete-worktree-paths-fix later to maint).
* When git refuses to create a branch because the proposed branch
name is not a valid refname, an advice message is given to refer
the user to exact naming rules.
(merge 8fbd903e58 kh/branch-ref-syntax-advice later to maint).
* Code simplification by getting rid of code that sets an environment
variable that is no longer used.
(merge 72a8d3f027 pw/rebase-i-ignore-cherry-pick-help-environment later to maint).
* The code to find the effective end of log messages can fall into an
endless loop, which has been corrected.
(merge 2541cba2d6 fs/find-end-of-log-message-fix later to maint).
* Mark-up used in the documentation has been improved for
consistency.
(merge 45d5ed3e50 ja/doc-markup-fixes later to maint).
* The status.showUntrackedFiles configuration variable was
incorrectly documented to accept "false", which has been corrected.
* Leaks from "git restore" have been plugged.
(merge 2f64da0790 rj/restore-plug-leaks later to maint).
* "git bugreport --no-suffix" was not supported and instead
segfaulted, which has been corrected.
(merge b3b57c69da js/bugreport-no-suffix-fix later to maint).
* The documentation for "%(trailers[:options])" placeholder in the
"--pretty" option of commands in the "git log" family has been
updated.
(merge bff85a338c bl/doc-key-val-sep-fix later to maint).
* "git checkout --conflict=bad" reported a bad conflictStyle as if it
were given to a configuration variable; it has been corrected to
report that the command line option is bad.
(merge 5a99c1ac1a pw/checkout-conflict-errorfix later to maint).
* Code clean-up in the "git log" machinery that implements custom log
message formatting.
(merge 1c10b8e5b0 jk/pretty-subject-cleanup later to maint).
* "git config" corrupted literal HT characters written in the
configuration file as part of a value, which has been corrected.
(merge e6895c3f97 ds/config-internal-whitespace-fix later to maint).
* A unit test for reftable code tried to enumerate all files in a
directory after reftable operations and expected to see nothing but
the files it wanted to leave there, but was fooled by .nfs* cruft
files left, which has been corrected.
(merge 0068aa7946 ps/reftable-unit-test-nfs-workaround later to maint).
* The implementation and documentation of "object-format" option
exchange between the Git itself and its remote helpers did not
quite match, which has been corrected.
* The "--pretty=<shortHand>" option of the commands in the "git log"
family, defined as "[pretty] shortHand = <expansion>" should have
been looked up case insensitively, but was not, which has been
corrected.
(merge f999d5188b bl/pretty-shorthand-config-fix later to maint).
* "git apply" failed to extract the filename the patch applied to,
when the change was about an empty file created in or deleted from
a directory whose name ends with a SP, which has been corrected.
(merge 776ffd1a30 jc/apply-parse-diff-git-header-names-fix later to maint).
* Update a more recent tutorial doc.
(merge 95ab557b4b dg/myfirstobjectwalk-updates later to maint).
* The test script had an incomplete and ineffective attempt to avoid
clobbering the testing user's real crontab (and its equivalents),
which has been completed.
(merge 73cb87773b es/test-cron-safety later to maint).
* Use advice_if_enabled() API to rewrite a simple pattern to
call advise() after checking advice_enabled().
(merge 6412d01527 rj/use-adv-if-enabled later to maint).
* Another "set -u" fix for the bash prompt (in contrib/) script.
(merge d7805bc743 vs/complete-with-set-u-fix later to maint).
* "git checkout/switch --detach foo", after switching to the detached
HEAD state, gave the tracking information for the 'foo' branch,
which was pointless.
* "git apply" has been updated to lift the hardcoded pathname length
limit, which in turn allowed a mksnpath() function that is no
longer used.
(merge 708f7e0590 rs/apply-lift-path-length-limit later to maint).
* A file descriptor leak in an error codepath, used when "git apply
--reject" fails to create the *.rej file, has been corrected.
(merge 2b1f456adf rs/apply-reject-fd-leakfix later to maint).
* A config parser callback function fell through instead of returning
after recognising and processing a variable, wasting cycles, which
has been corrected.
(merge a816ccd642 ds/fetch-config-parse-microfix later to maint).
* Fix was added to work around a regression in libcURL 8.7.0 (which has
already been fixed in their tip of the tree).
(merge 92a209bf24 jk/libcurl-8.7-regression-workaround later to maint).
* The variable that holds the value read from the core.excludefile
configuration variable used to leak, which has been corrected.
(merge 0e0fefb29f jc/unleak-core-excludesfile later to maint).
* vreportf(), which is used by error() and friends, has been taught
to give the error message printf-format string when its vsnprintf()
call fails, instead of showing nothing useful to identify the
nature of the error.
(merge c63adab961 rs/usage-fallback-to-show-message-format later to maint).
* Adjust to an upcoming changes to GNU make that breaks our Makefiles.
(merge 227b8fd902 tb/make-indent-conditional-with-non-spaces later to maint).
* Git 2.44 introduced a regression that makes the updated code to
barf in repositories with multi-pack index written by older
versions of Git, which has been corrected.
* When .git/rr-cache/ rerere database gets corrupted or rerere is fed to
work on a file with conflicted hunks resolved incompletely, the rerere
machinery got confused and segfaulted, which has been corrected.
(merge 167395bb47 mr/rerere-crash-fix later to maint).
* The "receive-pack" program (which responds to "git push") was not
converted to run "git maintenance --auto" when other codepaths that
used to run "git gc --auto" were updated, which has been corrected.
(merge 7bf3057d9c ps/run-auto-maintenance-in-receive-pack later to maint).
* Other code cleanup, docfix, build fix, etc.
(merge f0e578c69c rs/use-xstrncmpz later to maint).
(merge 83e6eb7d7a ba/credential-test-clean-fix later to maint).
(merge 64562d784d jb/doc-interactive-singlekey-do-not-need-perl later to maint).
(merge c431a235e2 cp/t9146-use-test-path-helpers later to maint).
(merge 82d75402d5 ds/doc-send-email-capitalization later to maint).
(merge 41bff66e35 jc/doc-add-placeholder-fix later to maint).
(merge 6835f0efe9 jw/remote-doc-typofix later to maint).
(merge 244001aa20 hs/rebase-not-in-progress later to maint).
(merge 2ca6c07db2 jc/no-include-of-compat-util-from-headers later to maint).
(merge 87bd7fbb9c rs/fetch-simplify-with-starts-with later to maint).
(merge f39addd0d9 rs/name-rev-with-mempool later to maint).
(merge 9a97b43e03 rs/submodule-prefix-simplify later to maint).
(merge 40b8076462 ak/rebase-autosquash later to maint).
(merge 3223204456 eg/add-uflags later to maint).
(merge 5f78d52dce es/config-doc-sort-sections later to maint).
(merge 781fb7b4c2 as/option-names-in-messages later to maint).
(merge 51d41dc243 jk/doc-remote-helpers-markup-fix later to maint).
(merge e1aaf309db pb/ci-win-artifact-names-fix later to maint).
(merge ad538c61da jc/index-pack-fsck-levels later to maint).
(merge 67471bc704 ja/doc-formatting-fix later to maint).
(merge 86f9ce7dd6 bl/doc-config-fixes later to maint).
(merge 0d527842b7 az/grep-group-error-message-update later to maint).
(merge 7c43bdf07b rs/strbuf-expand-bad-format later to maint).
(merge 8b68b48d5c ds/typofix-core-config-doc later to maint).
(merge 39bb692152 rs/imap-send-use-xsnprintf later to maint).
(merge 8d320cec60 jc/t2104-style-fixes later to maint).
(merge b4454d5a7b pw/t3428-cleanup later to maint).
(merge 84a7c33a4b pf/commitish-committish later to maint).
(merge 8882ee9d68 la/mailmap-entry later to maint).
(merge 44bdba2fa6 rs/no-openssl-compilation-fix-on-macos later to maint).
(merge f412d72c19 yb/replay-doc-linkfix later to maint).
(merge 5da40be8d7 xx/rfc2822-date-format-in-doc later to maint).
----------------------------------------------------------------
Changes since v2.44.0 are as follows:
Ahelenia Ziemiańska (1):
grep: improve errors for unmatched ( and )
Alexander Shopov (5):
transport-helper.c: trivial fix of error message
builtin/remote.c: trivial fix of error message
builtin/clone.c: trivial fix of message
revision.c: trivial fix to message
l10n: bg.po: Updated Bulgarian translation (5652t)
Arkadii Yakovets (1):
l10n: uk: v2.45 update
Aryan Gupta (1):
tests: modernize the test script t0010-racy-git.sh
Bagas Sanjaya (1):
l10n: po-id for 2.45
Beat Bolli (25):
completion: use awk for filtering the config entries
date: make "iso-strict" conforming for the UTC timezone
t0006: add more tests with a negative TZ offset
doc: avoid redundant use of cat
contrib/subtree/t: avoid redundant use of cat
t/lib-cvs.sh: avoid redundant use of cat
t/annotate-tests.sh: avoid redundant use of cat
t/perf: avoid redundant use of cat
t/t0*: avoid redundant uses of cat
t/t1*: avoid redundant uses of cat
t/t3*: avoid redundant uses of cat
t/t4*: avoid redundant uses of cat
t/t5*: avoid redundant uses of cat
t/t6*: avoid redundant uses of cat
t/t7*: avoid redundant use of cat
t/t8*: avoid redundant use of cat
t/t9*: avoid redundant uses of cat
t/t1*: merge a "grep | sed" pipeline
t/t3*: merge a "grep | awk" pipeline
t/t4*: merge a "grep | sed" pipeline
t/t5*: merge a "grep | sed" pipeline
t/t8*: merge "grep | sed" pipelines
t/t9*: merge "grep | sed" pipelines
contrib/coverage-diff: avoid redundant pipelines
git-quiltimport: avoid an unnecessary subshell
Bo Anderson (5):
t/lib-credential: clean additional credential
osxkeychain: replace deprecated SecKeychain API
osxkeychain: erase all matching credentials
osxkeychain: erase matching passwords only
osxkeychain: store new attributes
Brian C Tracy (1):
fuzz: add fuzzer for config parsing
Brian Lyles (13):
docs: clarify file options in git-config `--edit`
docs: fix typo in git-config `--default`
docs: correct trailer `key_value_separator` description
docs: adjust trailer `separator` and `key_value_separator` language
pretty: update tests to use `test_config`
pretty: find pretty formats case-insensitively
docs: address inaccurate `--empty` default with `--exec`
docs: clean up `--empty` formatting in git-rebase(1) and git-am(1)
rebase: update `--empty=ask` to `--empty=stop`
sequencer: handle unborn branch with `--allow-empty`
sequencer: do not require `allow_empty` for redundant commit options
cherry-pick: enforce `--keep-redundant-commits` incompatibility
cherry-pick: add `--empty` for more robust redundant commit handling
Chandra Pratap (2):
apply: ignore working tree filemode when !core.filemode
t9146: replace test -d/-e/-f with appropriate test_path_is_* function
Christian Couder (5):
revision: clarify a 'return NULL' in get_reference()
oidset: refactor oidset_insert_from_set()
t6022: fix 'test' style and 'even though' typo
rev-list: allow missing tips with --missing=[print|allow*]
revision: fix --missing=[print|allow*] for annotated tags
Derrick Stolee (1):
fetch: return when parsing submodule.recurse
Dirk Gouders (6):
Documentation/user-manual.txt: example for generating object hashes
MyFirstObjectWalk: use additional arg in config_fn_t
MyFirstObjectWalk: fix misspelled "builtins/"
MyFirstObjectWalk: fix filtered object walk
MyFirstObjectWalk: fix description for counting omitted objects
MyFirstObjectWalk: add stderr to pipe processing
Dragan Simic (8):
documentation: send-email: use camel case consistently
config: minor addition of whitespace
config: really keep value-internal whitespace verbatim
t1300: add more tests for whitespace and inline comments
config.txt: describe handling of whitespace further
grep docs: describe --recurse-submodules further and improve formatting a bit
grep docs: describe --no-index further and improve formatting a bit
config: fix some small capitalization issues, as spotted
Emir SARI (1):
l10n: tr: Update Turkish translations
Eric Sunshine (2):
docs: sort configuration variable groupings alphabetically
test-lib: fix non-functioning GIT_TEST_MAINT_SCHEDULER fallback
Eric W. Biederman (23):
object-file-convert: stubs for converting from one object format to another
oid-array: teach oid-array to handle multiple kinds of oids
object-names: support input of oids in any supported hash
repository: add a compatibility hash algorithm
loose: compatibilty short name support
object-file: update the loose object map when writing loose objects
object-file: add a compat_oid_in parameter to write_object_file_flags
commit: convert mergetag before computing the signature of a commit
commit: export add_header_signature to support handling signatures on tags
tag: sign both hashes
object: factor out parse_mode out of fast-import and tree-walk into in object.h
object-file-convert: don't leak when converting tag objects
object-file-convert: convert commits that embed signed tags
object-file: update object_info_extended to reencode objects
rev-parse: add an --output-object-format parameter
builtin/cat-file: let the oid determine the output algorithm
tree-walk: init_tree_desc take an oid to get the hash algorithm
object-file: handle compat objects in check_object_signature
builtin/ls-tree: let the oid determine the output algorithm
test-lib: compute the compatibility hash so tests may use it
t1006: rename sha1 to oid
t1006: test oid compatibility with cat-file
t1016-compatObjectFormat: add tests to verify the conversion between objects
Eugenio Gigante (1):
add: use unsigned type for collection of bits
Florian Schmidt (1):
wt-status: don't find scissors line beyond buf len
Ghanshyam Thakkar (5):
add-patch: classify '@' as a synonym for 'HEAD'
add -p tests: remove PERL prerequisites
setup: remove unnecessary variable
builtin/commit: error out when passing untracked path with -i
builtin/add: error out when passing untracked path with -u
Haritha D (1):
build: support z/OS (OS/390).
Harmen Stoppels (1):
rebase: make warning less passive aggressive
Jakub Wilk (1):
git-remote.txt: fix typo
Jean-Noël Avila (18):
doc: git-rev-parse: enforce command-line description syntax
doc: close unclosed angle-bracket of a placeholder in git-clone doc
doc: end sentences with full-stop
doc: clarify the format of placeholders
doc: git-init: format verbatim parts
doc: git-init: format placeholders
doc: git-init: rework definition lists
doc: git-init: rework config item init.templateDir
doc: git-clone: format verbatim words
doc: git-clone: format placeholders
doc: format alternatives in synopsis
doc: fix some placeholders formating
doc: rework CodingGuidelines with new formatting rules
doc: allow literal and emphasis format in doc vs help tests
doc: git-init: apply new documentation formatting guidelines
doc: git-clone: apply new documentation formatting guidelines
doc: git-clone: do not autoreference the manpage in itself
l10n: fr: v2.45.0
Jeff Hostetler (17):
name-hash: add index_dir_find()
t7527: add case-insensitve test for FSMonitor
fsmonitor: refactor refresh callback on directory events
fsmonitor: clarify handling of directory events in callback helper
fsmonitor: refactor refresh callback for non-directory events
dir: create untracked_cache_invalidate_trimmed_path()
fsmonitor: refactor untracked-cache invalidation
fsmonitor: move untracked-cache invalidation into helper functions
fsmonitor: return invalidated cache-entry count on directory event
fsmonitor: remove custom loop from non-directory path handler
fsmonitor: return invalidated cache-entry count on non-directory event
fsmonitor: trace the new invalidated cache-entry count
fsmonitor: refactor bit invalidation in refresh callback
fsmonitor: support case-insensitive events
t0211: demonstrate missing 'def_param' events for certain commands
trace2: avoid emitting 'def_param' set more than once
trace2: emit 'def_param' set with 'cmd_name' event
Jeff King (51):
t0303: check that helper_test_clean removes all credentials
userdiff: skip textconv caching when not in a repository
Revert "refs: allow @{n} to work with n-sized reflog"
get_oid_basic(): special-case ref@{n} for oldest reflog entry
read_ref_at(): special-case ref@{0} for an empty reflog
upload-pack: drop separate v2 "haves" array
upload-pack: switch deepen-not list to an oid_array
upload-pack: use oidset for deepen_not list
upload-pack: use a strmap for want-ref lines
upload-pack: accept only a single packfile-uri line
upload-pack: always turn off save_commit_buffer
upload-pack: use PARSE_OBJECT_SKIP_HASH_CHECK in more places
upload-pack: free tree buffers after parsing
upload-pack: use repository struct to get config
upload-pack: centralize setup of sideband-all config
upload-pack: use existing config mechanism for advertisement
upload-pack: only accept packfile-uris if we advertised it
doc/gitremote-helpers: fix missing single-quote
config: forbid newline as core.commentChar
strbuf: simplify comment-handling in add_lines() helper
strbuf: avoid static variables in strbuf_add_commented_lines()
commit: refactor base-case of adjust_comment_line_char()
strbuf: avoid shadowing global comment_line_char name
environment: store comment_line_char as a string
strbuf: accept a comment string for strbuf_stripspace()
strbuf: accept a comment string for strbuf_commented_addf()
strbuf: accept a comment string for strbuf_add_commented_lines()
prefer comment_line_str to comment_line_char for printing
find multi-byte comment chars in NUL-terminated strings
find multi-byte comment chars in unterminated buffers
sequencer: handle multi-byte comment characters when writing todo list
wt-status: drop custom comment-char stringification
environment: drop comment_line_char compatibility macro
config: allow multi-byte core.commentChar
shortlog: stop setting pp.print_email_subject
pretty: split oneline and email subject printing
pretty: drop print_email_subject flag
log: do not set up extra_headers for non-email formats
format-patch: return an allocated string from log_write_email_headers()
format-patch: simplify after-subject MIME header handling
doc/gitremote-helpers: fix more missing single-quotes
transport-helper: use write helpers more consistently
transport-helper: drop "object-format <algo>" option
transport-helper: send "true" value for object-format option
contrib: drop hg-to-git script
format-patch: fix leak of empty header string
rebase: use child_process_clear() to clean
config: add core.commentString
http: reset POSTFIELDSIZE when clearing curl handle
INSTALL: bump libcurl version to 7.21.3
remote-curl: add Transfer-Encoding header only for older curl
Jiamu Sun (1):
bugreport.c: fix a crash in `git bugreport` with `--no-suffix` option
Jiang Xin (1):
l10n: TEAMS: retire l10n teams no update in 1 year
Johannes Schindelin (22):
merge-tree: accept 3 trees as arguments
merge-tree: fail with a non-zero exit code on missing tree objects
merge-ort: do check `parse_tree()`'s return value
t4301: verify that merge-tree fails on missing blob objects
Always check `parse_tree*()`'s return value
cache-tree: avoid an unnecessary check
fill_tree_descriptor(): mark error message for translation
neue: remove a bogus empty file
commit-reach(paint_down_to_common): plug two memory leaks
commit-reach(repo_in_merge_bases_many): optionally expect missing commits
commit-reach(repo_in_merge_bases_many): report missing commits
commit-reach(paint_down_to_common): prepare for handling shallow commits
commit-reach(paint_down_to_common): start reporting errors
commit-reach(merge_bases_many): pass on "missing commits" errors
commit-reach(get_merge_bases_many_0): pass on "missing commits" errors
commit-reach(repo_get_merge_bases): pass on "missing commits" errors
commit-reach(get_octopus_merge_bases): pass on "missing commits" errors
commit-reach(repo_get_merge_bases_many): pass on "missing commits" errors
commit-reach(repo_get_merge_bases_many_dirty): pass on errors
merge-recursive: prepare for `merge_submodule()` to report errors
merge-ort/merge-recursive: do report errors in `merge_submodule()`
merge-tree: fix argument type of the `--merge-base` option
John Cai (1):
t5300: fix test_with_bad_commit()
Jonas Wunderlich (1):
doc: status.showUntrackedFiles does not take "false"
Josh Triplett (2):
commit: avoid redundant scissor line with --cleanup=scissors -v
commit: unify logic to avoid multiple scissors lines when merging
Julio Bacellari (1):
doc: remove outdated information about interactive.singleKey
Junio C Hamano (64):
apply: correctly reverse patch's pre- and post-image mode bits
apply: code simplification
t9210: do not rely on lazy fetching to fail
git: --no-lazy-fetch option
doc: add shortcut to "am --whitespace=<action>"
doc: apply the new placeholder rules to git-add documentation
compat: drop inclusion of <git-compat-util.h>
Start the 2.45 cycle
git: document GIT_NO_REPLACE_OBJECTS environment variable
doc: clarify the wording on <git-compat-util.h> requirement
git: extend --no-lazy-fetch to work across subprocesses
The second batch
The third batch
test_i18ngrep: hard deprecate and forbid its use
unpack: replace xwrite() loop with write_in_full()
sideband: avoid short write(2)
repack: check error writing to pack-objects subprocess
clean: further clean-up of implementation around "--force"
The fourth batch
The fifth batch
setup: notice more types of implicit bare repositories
The sixth batch
status: unify parsing of --untracked= and status.showUntrackedFiles
status: allow --untracked=false and friends
The seventh batch
The eighth batch
config: fix --comment formatting
config: allow tweaking whitespace between value and comment
diff.*Prefix: use camelCase in the doc and test titles
The ninth batch
apply: parse names out of "diff --git" more carefully
The tenth batch
The eleventh batch
SubmittingPatches: release-notes entry experiment
The twelfth batch
t4126: make sure a directory with SP at the end is usable
t4126: fix "funny directory name" test on Windows (again)
advice: omit trailing whitespace
checkout: omit "tracking" information on a detached HEAD
The thirteenth batch
t2104: style fixes
The fourteenth batch
revision: optionally record matches with pathspec elements
The fifteenth batch
CodingGuidelines: describe "export VAR=VAL" rule
CodingGuidelines: quote assigned value in 'local var=$val'
t: local VAR="VAL" (quote positional parameters)
t: local VAR="VAL" (quote command substitution)
t: local VAR="VAL" (quote ${magic-reference})
t: teach lint that RHS of 'local VAR=VAL' needs to be quoted
t0610: local VAR="VAL" fix
t1016: local VAR="VAL" fix
config: do not leak excludes_file
Makefile(s): do not enforce "all indents must be done with tab"
The sixteenth batch
t2104: style fixes
The seventeenth batch
The eighteenth batch
The ninteenth batch
The twentieth batch
Git 2.45-rc0
A bit more topics before -rc1
Git 2.45-rc1
Git 2.45
Justin Tobler (3):
reftable/stack: expose option to disable auto-compaction
reftable/stack: add env to disable autocompaction
reftable/stack: use geometric table compaction
Karthik Nayak (7):
refs: introduce `is_pseudoref()` and `is_headref()`
refs: extract out `loose_fill_ref_dir_regular_file()`
refs: introduce `refs_for_each_include_root_refs()`
ref-filter: rename 'FILTER_REFS_ALL' to 'FILTER_REFS_REGULAR'
for-each-ref: add new option to include root refs
update-ref: use {old,new}-oid instead of {old,new}value
githooks: use {old,new}-oid instead of {old,new}-value
Kipras Melnikovas (1):
mergetools: vimdiff: use correct tool's name when reading mergetool config
Kristoffer Haugsbakk (9):
column: disallow negative padding
column: guard against negative padding
gitcli: drop mention of “non-dashed form”
config: document `core.commentChar` as ASCII-only
t3200: improve test style
advice: make all entries stylistically consistent
advice: use backticks for verbatim
advice: use double quotes for regular quoting
branch: advise about ref syntax rules
Linus Arver (15):
trailer: free trailer_info _after_ all related usage
shortlog: add test for de-duplicating folded trailers
trailer: rename functions to use 'trailer'
trailer: reorder format_trailers_from_commit() parameters
trailer: move interpret_trailers() to interpret-trailers.c
trailer_info_get(): reorder parameters
format_trailers(): use strbuf instead of FILE
format_trailer_info(): move "fast path" to caller
format_trailers_from_commit(): indirectly call trailer_info_get()
format_trailer_info(): use trailer_item objects
format_trailer_info(): drop redundant unfold_value()
format_trailer_info(): append newline for non-trailer lines
trailer: begin formatting unification
trailer: finish formatting unification
mailmap: change primary address for Linus Arver
M Hickford (1):
libsecret: retrieve empty password
Marcel Röthke (1):
rerere: fix crashes due to unmatched opening conflict markers
Matthias Aßhauer (1):
Win32: detect unix socket support at runtime
Max Gautier (1):
editorconfig: add Makefiles to "text files"
Michael Lohmann (2):
revision: ensure MERGE_HEAD is a ref in prepare_show_merge
revision: implement `git log --merge` also for rebase/cherry-pick/revert
Orgad Shaneh (1):
docs: remove duplicate entry and fix typo in 2.45 changelog
Patrick Steinhardt (99):
refs: introduce reftable backend
ci: add jobs to test with the reftable backend
refs/reftable: fix leak when copying reflog fails
reftable/record: introduce function to compare records by key
reftable/merged: allocation-less dropping of shadowed records
reftable/merged: skip comparison for records of the same subiter
reftable/pq: allocation-less comparison of entry keys
reftable/block: swap buffers instead of copying
reftable/record: don't try to reallocate ref record name
reftable/reader: add comments to `table_iter_next()`
t: move tests exercising the "files" backend
t0410: convert tests to use DEFAULT_REPO_FORMAT prereq
t1400: exercise reflog with gaps with reftable backend
t1404: make D/F conflict tests compatible with reftable backend
t1405: remove unneeded cleanup step
t2011: exercise D/F conflicts with HEAD with the reftable backend
t7003: ensure filter-branch prunes reflogs with the reftable backend
git-difftool--helper: honor `--trust-exit-code` with `--dir-diff`
dir-iterator: pass name to `prepare_next_entry_data()` directly
dir-iterator: support iteration in sorted order
refs/files: sort reflogs returned by the reflog iterator
refs/files: sort merged worktree and common reflogs
refs: always treat iterators as ordered
refs: drop unused params from the reflog iterator callback
refs: stop resolving ref corresponding to reflogs
builtin/reflog: introduce subcommand to list reflogs
builtin/clone: allow remote helpers to detect repo
refs/reftable: don't fail empty transactions in repo without HEAD
reftable/pq: use `size_t` to track iterator index
reftable/merged: make `merged_iter` structure private
reftable/merged: advance subiter on subsequent iteration
reftable/merged: make subiters own their records
reftable/merged: remove unnecessary null check for subiters
reftable/merged: handle subiter cleanup on close only
reftable/merged: circumvent pqueue with single subiter
reftable/merged: avoid duplicate pqueue emptiness check
reftable/record: reuse refname when decoding
reftable/record: reuse refname when copying
reftable/record: decode keys in place
reftable: allow inlining of a few functions
refs/reftable: precompute prefix length
refs/reftable: reload correct stack when creating reflog iter
reftable/record: convert old and new object IDs to arrays
reftable/record: avoid copying author info
reftable/record: reuse refnames when decoding log records
reftable/record: reuse message when decoding log records
reftable/record: use scratch buffer when decoding records
refs/reftable: track last log record name via strbuf
t0610: remove unused variable assignment
lockfile: report when rollback fails
reftable/stack: register new tables as tempfiles
reftable/stack: register lockfiles during compaction
reftable/stack: register compacted tables as tempfiles
reftable/record: fix memory leak when decoding object records
reftable/block: fix binary search over restart counter
t5601: exercise clones with "includeIf.*.onbranch"
reftable: fix tests being broken by NFS' delete-after-close semantics
t7800: improve test descriptions with empty arguments
t7800: use single quotes for test bodies
t/README: document how to loop around test cases
reftable/stack: fix error handling in `reftable_stack_init_addition()`
reftable/error: discern locked/outdated errors
reftable/stack: use error codes when locking fails during compaction
reftable/stack: gracefully handle failed auto-compaction due to locks
refs/reftable: print errors on compaction failure
t/helper: drop pack-refs wrapper
refs: move `struct pack_refs_opts` to where it's used
refs: remove `PACK_REFS_ALL` flag
refs/reftable: expose auto compaction via new flag
builtin/pack-refs: release allocated memory
builtin/pack-refs: introduce new "--auto" flag
builtin/gc: move `struct maintenance_run_opts`
t6500: extract objects with "17" prefix
builtin/gc: forward git-gc(1)'s `--auto` flag when packing refs
builtin/gc: pack refs when using `git maintenance run --auto`
reftable/basics: fix return type of `binsearch()` to be `size_t`
reftable/basics: improve `binsearch()` test
reftable/refname: refactor binary search over refnames
reftable/block: refactor binary search over restart points
reftable/block: fix error handling when searching restart points
reftable/record: extract function to decode key lengths
reftable/block: avoid decoding keys when searching restart points
t0610: make `--shared=` tests reusable
t0610: execute git-pack-refs(1) with specified umask
reftable/block: rename `block_reader_start()`
reftable/block: merge `block_iter_seek()` and `block_reader_seek()`
reftable/block: better grouping of functions
reftable/block: introduce `block_reader_release()`
reftable/block: move ownership of block reader into `struct table_iter`
reftable/reader: iterate to next block in place
reftable/block: reuse uncompressed blocks
reftable/block: open-code call to `uncompress2()`
reftable/block: reuse `zstream` state on inflation
reftable/block: avoid copying block iterators on seek
pack-bitmap: gracefully handle missing BTMP chunks
run-command: introduce function to prepare auto-maintenance process
builtin/receive-pack: convert to use git-maintenance(1)
docs: improve changelog entry for `git pack-refs --auto`
docs: address typos in Git v2.45 changelog
Peter Hutterer (1):
diff: add diff.srcPrefix and diff.dstPrefix configuration variables
Peter Krefting (2):
bisect: report the found commit with "show"
l10n: sv.po: Update Swedish translation
Philippe Blain (5):
merge-ort: turn submodule conflict suggestions into an advice
ci(github): make Windows test artifacts name unique
sequencer: allow disabling conflict advice
builtin/am: allow disabling conflict advice
t/README: mention test files are make targets
Phillip Wood (9):
rebase -i: stop setting GIT_CHERRY_PICK_HELP
xdiff-interface: refactor parsing of merge.conflictstyle
merge-ll: introduce LL_MERGE_OPTIONS_INIT
merge options: add a conflict style member
checkout: cleanup --conflict=<style> parsing
checkout: fix interaction between --conflict and --merge
t3428: modernize test setup
t3428: use test_commit_message
t3428: restore coverage for "apply" backend
Pi Fisher (1):
typo: replace 'commitish' with 'committish'
Ralf Thielow (1):
l10n: Update German translation
Ralph Seichter (1):
config: add --comment option to add a comment
René Scharfe (32):
use xstrncmpz()
fetch: convert strncmp() with strlen() to starts_with()
mem-pool: add mem_pool_strfmt()
name-rev: use mem_pool_strfmt()
submodule: use strvec_pushf() for --submodule-prefix
t-ctype: allow NUL anywhere in the specification string
t-ctype: simplify EOF check
t-ctype: align output of i
t-ctype: avoid duplicating class names
parse-options: recognize abbreviated negated option with arg
parse-options: set arg of abbreviated option lazily
parse-options: factor out register_abbrev() and struct parsed_option
parse-options: detect ambiguous self-negation
parse-options: normalize arg and long_name before comparison
parse-options: rearrange long_name matching code
t-prio-queue: shorten array index message
t-prio-queue: check result array bounds
factor out strbuf_expand_bad_format()
cat-file: use strbuf_expand_bad_format()
midx: use strvec_pushf() for pack-objects base name
mem-pool: use st_add() in mem_pool_strvfmt()
imap-send: use xsnprintf to format command
t-prio-queue: simplify using compound literals
apply: avoid fixed-size buffer in create_one_file()
path: remove mksnpath()
apply: don't leak fd on fdopen() error
usage: report vsnprintf(3) failure
date: make DATE_MODE thread-safe
git-compat-util: fix NO_OPENSSL on current macOS
imap-send: increase command size limit
apply: avoid using fixed-size buffer in write_out_one_reject()
don't report vsnprintf(3) error as bug
Richard Macklin (1):
rebase: fix typo in autosquash documentation
Rubén Justo (18):
tag: error when git-column fails
completion: fix __git_complete_worktree_paths
completion: reflog with implicit "show"
completion: reflog show <log-options>
completion: introduce __git_find_subcommand
completion: factor out __git_resolve_builtins
completion: reflog subcommands and options
checkout: plug some leaks in git-restore
add-patch: introduce 'p' in interactive-patch
add-patch: do not print hunks repeatedly
add: use advise_if_enabled for ADVICE_ADD_IGNORED_FILE
add: use advise_if_enabled for ADVICE_ADD_EMPTY_PATHSPEC
add: use advise_if_enabled for ADVICE_ADD_EMBEDDED_REPO
launch_editor: waiting message on error
apply: plug a leak in apply_data
add-interactive: plug a leak in get_untracked_files
add-patch: plug a leak handling the '/' command
add: plug a leak on interactive_add
SZEDER Gábor (1):
upload-pack: don't send null character in abort message to the client
Sergey Organov (1):
clean: improve -n and -f implementation and documentation
Steven Jeuris (1):
userdiff: better method/property matching for C#
Taylor Blau (9):
Documentation/config/pack.txt: fix broken AsciiDoc mark-up
upload-pack: disallow object-info capability by default
midx-write: move writing-related functions from midx.c
midx-write.c: factor out common want_included_pack() routine
midx-write.c: check count of packs to repack after grouping
midx-write.c: use `--stdin-packs` when repacking
t/t7700-repack.sh: fix test breakages with `GIT_TEST_MULTI_PACK_INDEX=1 `
Makefile(s): avoid recipe prefix in conditional statements
Documentation/RelNotes/2.45.0.txt: fix typo
Teng Long (1):
l10n: zh_CN: for git 2.45 rounds
Thalia Archibald (8):
fast-import: tighten path unquoting
fast-import: directly use strbufs for paths
fast-import: allow unquoted empty path for root
fast-import: remove dead strbuf
fast-import: improve documentation for path quoting
fast-import: document C-style escapes for paths
fast-import: forbid escaped NUL in paths
fast-import: make comments more precise
Ville Skyttä (2):
completion: fix prompt with unset SHOWCONFLICTSTATE in nounset mode
completion: protect prompt against unset SHOWUPSTREAM in nounset mode
Vincenzo Mezzela (1):
t7301: use test_path_is_(missing|file)
Vũ Tiến Hưng (2):
l10n: Update Vietnamese team contact
l10n: vi: Updated translation for 2.45
Xing Xin (1):
Documentation: fix typos describing date format
Yehezkel Bernat (1):
Documentation: fix linkgit reference
Yi-Jyun Pan (1):
l10n: zh-TW: Git 2.45
brian m. carlson (7):
loose: add a mapping between SHA-1 and SHA-256 for loose objects
commit: write commits for both hashes
cache: add a function to read an OID of a specific algorithm
object-file-convert: add a function to convert trees between algorithms
object-file-convert: convert tag objects when writing
object-file-convert: convert commit objects when writing
repository: implement extensions.compatObjectFormat
shejialuo (1):
t9117: prefer test_path_* helper functions
Đoàn Trần Công Danh (1):
t9604: Fix test for musl libc and new Debian
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [ANNOUNCE] Git v2.45.0
2024-04-29 17:12 [ANNOUNCE] Git v2.45.0 Junio C Hamano
@ 2024-05-01 17:39 ` rsbecker
2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen
1 sibling, 0 replies; 12+ messages in thread
From: rsbecker @ 2024-05-01 17:39 UTC (permalink / raw)
To: 'Junio C Hamano', git; +Cc: 'Linux Kernel', git-packagers
On Monday, April 29, 2024 1:12 PM, Junio C Hamano wrote:
>The latest feature release Git v2.45.0 is now available at the
>usual places. It is comprised of 540 non-merge commits since
>v2.44.0, contributed by 96 people, 38 of which are new faces [*].
>
>The tarballs are found at:
>
> https://www.kernel.org/pub/software/scm/git/
>
>The following public repositories all have a copy of the 'v2.45.0'
>tag and the 'master' branch that the tag points at:
>
> url = https://git.kernel.org/pub/scm/git/git
> url = https://kernel.googlesource.com/pub/scm/git/git
> url = git://repo.or.cz/alt-git.git
> url = https://github.com/gitster/git
The HPE NonStop builds for git 2.45.0 are available at the ITUGLIB website at https://ituglib.connect-community.org/apps/Ituglib/SrchOpenSrcLib.xhtml.
Thanks to the entire git and ITUGLIB teams for making these happen.
Regards,
Randall
--
Brief whoami: NonStop&UNIX developer since approximately
UNIX(421664400)
NonStop(211288444200000000)
-- In real life, I talk too much.
^ permalink raw reply [flat|nested] 12+ messages in thread
* t4216-log-bloom.sh broken ?
2024-04-29 17:12 [ANNOUNCE] Git v2.45.0 Junio C Hamano
2024-05-01 17:39 ` rsbecker
@ 2024-05-02 5:56 ` Torsten Bögershausen
2024-05-02 16:06 ` Junio C Hamano
2024-05-02 16:26 ` Junio C Hamano
1 sibling, 2 replies; 12+ messages in thread
From: Torsten Bögershausen @ 2024-05-02 5:56 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git, Linux Kernel, git-packagers
[]
Highjacking the Git v2.45.0 announcement:
There are 4 test cases in t4216-log-bloom.sh, that do not pass on one
Mac here (they pass on another machine)
I haven't digged further, but here are some log files,
hopefully helpful to somebody.
not ok 141 - Bloom reader notices too-small data chunk
#
# check_corrupt_graph BDAT clear 00000000 &&
# echo "warning: ignoring too-small changed-path chunk" \
# "(4 < 12) in commit-graph file" >expect.err &&
# test_cmp expect.err err
#
not ok 142 - Bloom reader notices out-of-bounds filter offsets
#
# check_corrupt_graph BIDX 12 FFFFFFFF &&
# # use grep to avoid depending on exact chunk size
# grep "warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph" err
#
not ok 143 - Bloom reader notices too-small index chunk
#
# # replace the index with a single entry, making most
# # lookups out-of-bounds
# check_corrupt_graph BIDX clear 00000000 &&
# echo "warning: commit-graph changed-path index chunk" \
# "is too small" >expect.err &&
# test_cmp expect.err err
#
not ok 144 - Bloom reader notices out-of-order index offsets
#
# # we do not know any real offsets, but we can pick
# # something plausible; we should not get to the point of
# # actually reading from the bogus offsets anyway.
# corrupt_graph BIDX 4 0000000c00000005 &&
# echo "warning: ignoring decreasing changed-path index offsets" \
# "(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph" >expect.err &&
# git -c core.commitGraph=false log -- A/B/file2 >expect.out &&
# git -c core.commitGraph=true log -- A/B/file2 >out 2>err &&
# test_cmp expect.out out &&
# test_cmp expect.err err
#
# failed 4 among 144 test(s)
1..144
ok 140 - Bloom generation backfills empty commits
expecting success of 4216.141 'Bloom reader notices too-small data chunk':
check_corrupt_graph BDAT clear 00000000 &&
echo "warning: ignoring too-small changed-path chunk" \
"(4 < 12) in commit-graph file" >expect.err &&
test_cmp expect.err err
++ check_corrupt_graph BDAT clear 00000000
++ corrupt_graph BDAT clear 00000000
++ graph=.git/objects/info/commit-graph
++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
++ test 0 = 0
++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
} && (exit "$eval_ret"); eval_ret=$?; :'
++ git commit-graph write --reachable --changed-paths
++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000
++ fn=.git/objects/info/commit-graph
++ shift
++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
++ git -c core.commitGraph=false log -- A/B/file2
++ git -c core.commitGraph=true log -- A/B/file2
++ test_cmp expect.out out
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.out out
++ echo 'warning: ignoring too-small changed-path chunk' '(4 < 12) in commit-graph file'
++ test_cmp expect.err err
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.err err
--- expect.err 2024-05-02 05:51:10.000000000 +0000
+++ err 2024-05-02 05:51:10.000000000 +0000
@@ -1 +0,0 @@
-warning: ignoring too-small changed-path chunk (4 < 12) in commit-graph file
error: last command exited with $?=1
++ rm -rf .git/objects/info/commit-graph
++ exit 1
++ eval_ret=1
++ :
not ok 141 - Bloom reader notices too-small data chunk
expecting success of 4216.142 'Bloom reader notices out-of-bounds filter offsets':
check_corrupt_graph BIDX 12 FFFFFFFF &&
# use grep to avoid depending on exact chunk size
grep "warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph" err
++ check_corrupt_graph BIDX 12 FFFFFFFF
++ corrupt_graph BIDX 12 FFFFFFFF
++ graph=.git/objects/info/commit-graph
++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
++ test 0 = 0
++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
} && (exit "$eval_ret"); eval_ret=$?; :'
++ git commit-graph write --reachable --changed-paths
++ corrupt_chunk_file .git/objects/info/commit-graph BIDX 12 FFFFFFFF
++ fn=.git/objects/info/commit-graph
++ shift
++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF
++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF
++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF
++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
++ git -c core.commitGraph=false log -- A/B/file2
++ git -c core.commitGraph=true log -- A/B/file2
++ test_cmp expect.out out
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.out out
++ grep 'warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph' err
error: last command exited with $?=1
++ rm -rf .git/objects/info/commit-graph
++ exit 1
++ eval_ret=1
++ :
not ok 142 - Bloom reader notices out-of-bounds filter offsets
expecting success of 4216.143 'Bloom reader notices too-small index chunk':
# replace the index with a single entry, making most
# lookups out-of-bounds
check_corrupt_graph BIDX clear 00000000 &&
echo "warning: commit-graph changed-path index chunk" \
"is too small" >expect.err &&
test_cmp expect.err err
++ check_corrupt_graph BIDX clear 00000000
++ corrupt_graph BIDX clear 00000000
++ graph=.git/objects/info/commit-graph
++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
++ test 0 = 0
++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
} && (exit "$eval_ret"); eval_ret=$?; :'
++ git commit-graph write --reachable --changed-paths
++ corrupt_chunk_file .git/objects/info/commit-graph BIDX clear 00000000
++ fn=.git/objects/info/commit-graph
++ shift
++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000
++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000
++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000
++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
++ git -c core.commitGraph=false log -- A/B/file2
++ git -c core.commitGraph=true log -- A/B/file2
++ test_cmp expect.out out
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.out out
++ echo 'warning: commit-graph changed-path index chunk' 'is too small'
++ test_cmp expect.err err
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.err err
--- expect.err 2024-05-02 05:51:11.000000000 +0000
+++ err 2024-05-02 05:51:11.000000000 +0000
@@ -1 +0,0 @@
-warning: commit-graph changed-path index chunk is too small
error: last command exited with $?=1
++ rm -rf .git/objects/info/commit-graph
++ exit 1
++ eval_ret=1
++ :
not ok 143 - Bloom reader notices too-small index chunk
expecting success of 4216.144 'Bloom reader notices out-of-order index offsets':
# we do not know any real offsets, but we can pick
# something plausible; we should not get to the point of
# actually reading from the bogus offsets anyway.
corrupt_graph BIDX 4 0000000c00000005 &&
echo "warning: ignoring decreasing changed-path index offsets" \
"(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph" >expect.err &&
git -c core.commitGraph=false log -- A/B/file2 >expect.out &&
git -c core.commitGraph=true log -- A/B/file2 >out 2>err &&
test_cmp expect.out out &&
test_cmp expect.err err
++ corrupt_graph BIDX 4 0000000c00000005
++ graph=.git/objects/info/commit-graph
++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
++ test 0 = 0
++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
} && (exit "$eval_ret"); eval_ret=$?; :'
++ git commit-graph write --reachable --changed-paths
++ corrupt_chunk_file .git/objects/info/commit-graph BIDX 4 0000000c00000005
++ fn=.git/objects/info/commit-graph
++ shift
++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005
++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005
++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005
++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
++ echo 'warning: ignoring decreasing changed-path index offsets' '(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph'
++ git -c core.commitGraph=false log -- A/B/file2
++ git -c core.commitGraph=true log -- A/B/file2
++ test_cmp expect.out out
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.out out
++ test_cmp expect.err err
++ test 2 -ne 2
++ eval 'diff -u' '"$@"'
+++ diff -u expect.err err
--- expect.err 2024-05-02 05:51:11.000000000 +0000
+++ err 2024-05-02 05:51:11.000000000 +0000
@@ -1 +0,0 @@
-warning: ignoring decreasing changed-path index offsets (12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph
error: last command exited with $?=1
++ rm -rf .git/objects/info/commit-graph
++ exit 1
++ eval_ret=1
++ :
not ok 144 - Bloom reader notices out-of-order index offsets
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen
@ 2024-05-02 16:06 ` Junio C Hamano
2024-05-02 18:59 ` Torsten Bögershausen
2024-05-02 16:26 ` Junio C Hamano
1 sibling, 1 reply; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 16:06 UTC (permalink / raw)
To: Torsten Bögershausen; +Cc: git
Torsten Bögershausen <[email protected]> writes:
> []
>
> Highjacking the Git v2.45.0 announcement:
I am not sure why you want to do so. You are perfectly capable of
sending a new message to the list (and I know you have done so in
the past). Why would linux-kernel@ list want to hear about the
macOS issues?
> There are 4 test cases in t4216-log-bloom.sh, that do not pass on one
> Mac here (they pass on another machine)
Another machine being another mac?
> expecting success of 4216.141 'Bloom reader notices too-small data chunk':
> check_corrupt_graph BDAT clear 00000000 &&
> echo "warning: ignoring too-small changed-path chunk" \
> "(4 < 12) in commit-graph file" >expect.err &&
> test_cmp expect.err err
>
> ++ check_corrupt_graph BDAT clear 00000000
> ++ corrupt_graph BDAT clear 00000000
> ++ graph=.git/objects/info/commit-graph
> ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
> ++ test 0 = 0
> ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
> } && (exit "$eval_ret"); eval_ret=$?; :'
> ++ git commit-graph write --reachable --changed-paths
> ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000
> ++ fn=.git/objects/info/commit-graph
> ++ shift
> ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
> override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
Is this failure preventing the later steps of the test work as
expected, I wonder. Is there something curious with the permission
bits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/
subdirectory? Is there something curious with the "umask" of the
user running the test? Are they different from what you see on your
other mac that does not exhibit the problems?
Thanks.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen
2024-05-02 16:06 ` Junio C Hamano
@ 2024-05-02 16:26 ` Junio C Hamano
1 sibling, 0 replies; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 16:26 UTC (permalink / raw)
To: Torsten Bögershausen; +Cc: git
Torsten Bögershausen <[email protected]> writes:
> ++ check_corrupt_graph BDAT clear 00000000
> ++ corrupt_graph BDAT clear 00000000
> ++ graph=.git/objects/info/commit-graph
> ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
> ++ test 0 = 0
> ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
> } && (exit "$eval_ret"); eval_ret=$?; :'
> ++ git commit-graph write --reachable --changed-paths
> ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000
> ++ fn=.git/objects/info/commit-graph
> ++ shift
> ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
> override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
Another probing question. Do t5318, t5319, t5324, and t5328 pass?
The "mv" is part of the corrupt_chunk_file helper is used by this:
corrupt_graph () {
graph=.git/objects/info/commit-graph &&
test_when_finished "rm -rf $graph" &&
git commit-graph write --reachable --changed-paths &&
corrupt_chunk_file $graph "$@"
}
and reads like this:
corrupt_chunk_file () {
fn=$1; shift
perl "$TEST_DIRECTORY"/lib-chunk/corrupt-chunk-file.pl \
"$@" <"$fn" >"$fn.tmp" &&
mv "$fn.tmp" "$fn"
}
If the final "mv" is what is failing in the above trace, the test
tried to corrupt the objects/info/commit-graph file, expecting that
later steps will notice breakage. But if the broken file written in
commit-graph.tmp failed to become the final commit-graph, later
steps will see healthy graph file and it is understandable that we
do not see any breakage that we expect, thus failing the test.
Is "mv" somehow configured to always go interactive on that box and
on that box alone? For example if you have ~/bin/mv that does
something silly like
#!/bin/bash
exec /bin/mv -i "$@
and have ~/bin early in $PATH, it may be sufficient to explain the
error we see (I am not claiming that is the only way to cause this
test failure; it is just one possible one).
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 16:06 ` Junio C Hamano
@ 2024-05-02 18:59 ` Torsten Bögershausen
2024-05-02 19:26 ` SZEDER Gábor
0 siblings, 1 reply; 12+ messages in thread
From: Torsten Bögershausen @ 2024-05-02 18:59 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Thu, May 02, 2024 at 09:06:41AM -0700, Junio C Hamano wrote:
> Torsten Bögershausen <[email protected]> writes:
>
> > There are 4 test cases in t4216-log-bloom.sh, that do not pass on one
> > Mac here (they pass on another machine)
>
> Another machine being another mac?
Yes, different mac, different MacOs, different $PATH probably.
>
> > expecting success of 4216.141 'Bloom reader notices too-small data chunk':
> > check_corrupt_graph BDAT clear 00000000 &&
> > echo "warning: ignoring too-small changed-path chunk" \
> > "(4 < 12) in commit-graph file" >expect.err &&
> > test_cmp expect.err err
> >
> > ++ check_corrupt_graph BDAT clear 00000000
> > ++ corrupt_graph BDAT clear 00000000
> > ++ graph=.git/objects/info/commit-graph
> > ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
> > ++ test 0 = 0
> > ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
> > } && (exit "$eval_ret"); eval_ret=$?; :'
> > ++ git commit-graph write --reachable --changed-paths
> > ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000
> > ++ fn=.git/objects/info/commit-graph
> > ++ shift
> > ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
> > override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
>
> Is this failure preventing the later steps of the test work as
> expected, I wonder. Is there something curious with the permission
> bits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/
> subdirectory? Is there something curious with the "umask" of the
> user running the test? Are they different from what you see on your
> other mac that does not exhibit the problems?
>
> Thanks.
>
>
mv is /bin/mv, that seems to be good:
$ type mv
mv is /bin/mv
$ alias | grep mv
$ which mv
/bin/mv
$ umask
0022
I don't know, why we see
r--r--r-- tb/staff for .git/objects/info/commit-graph
But, the "-r--r--r--" may be part of the problem, here is another one:
$ find . -name commit-graph -print0 | xargs -0 ls -l
-r--r--r-- 1 tb staff 1792 May 2 12:12 ./trash directory.t5318-commit-graph/bare/objects/info/commit-graph
(And some more 6 in total. All with -r--r--r--)
Which means, yes, t5318 does not pass either:
t5318-commit-graph.sh not ok 101 - reader notices too-small oid fanout chunk
t5318-commit-graph.sh not ok 102 - reader notices fanout/lookup table mismatch
t5318-commit-graph.sh not ok 103 - reader notices out-of-bounds fanout
t5318-commit-graph.sh not ok 104 - reader notices too-small commit data chunk
t5318-commit-graph.sh not ok 105 - reader notices out-of-bounds extra edge
t5318-commit-graph.sh not ok 106 - reader notices too-small generations chunk
Same problem here:
++ mv full/.git/objects/info/commit-graph.tmp full/.git/objects/info/commit-graph
override r--r--r-- tb/staff for full/.git/objects/info/commit-graph? (y/n [n]) not overwritten
The rest of the test suite passes, I see the same failures even under a
fresh cloned repo.
Any hints are appreciated.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 18:59 ` Torsten Bögershausen
@ 2024-05-02 19:26 ` SZEDER Gábor
2024-05-02 19:55 ` Junio C Hamano
0 siblings, 1 reply; 12+ messages in thread
From: SZEDER Gábor @ 2024-05-02 19:26 UTC (permalink / raw)
To: Torsten Bögershausen; +Cc: Junio C Hamano, git
On Thu, May 02, 2024 at 08:59:03PM +0200, Torsten Bögershausen wrote:
> On Thu, May 02, 2024 at 09:06:41AM -0700, Junio C Hamano wrote:
> > Torsten Bögershausen <[email protected]> writes:
> >
> > > There are 4 test cases in t4216-log-bloom.sh, that do not pass on one
> > > Mac here (they pass on another machine)
> >
> > Another machine being another mac?
>
> Yes, different mac, different MacOs, different $PATH probably.
>
> >
> > > expecting success of 4216.141 'Bloom reader notices too-small data chunk':
> > > check_corrupt_graph BDAT clear 00000000 &&
> > > echo "warning: ignoring too-small changed-path chunk" \
> > > "(4 < 12) in commit-graph file" >expect.err &&
> > > test_cmp expect.err err
> > >
> > > ++ check_corrupt_graph BDAT clear 00000000
> > > ++ corrupt_graph BDAT clear 00000000
> > > ++ graph=.git/objects/info/commit-graph
> > > ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'
> > > ++ test 0 = 0
> > > ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph
> > > } && (exit "$eval_ret"); eval_ret=$?; :'
> > > ++ git commit-graph write --reachable --changed-paths
> > > ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000
> > > ++ fn=.git/objects/info/commit-graph
> > > ++ shift
> > > ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > > ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > > ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000
> > > ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph
> > > override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten
> >
> > Is this failure preventing the later steps of the test work as
> > expected, I wonder. Is there something curious with the permission
> > bits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/
> > subdirectory? Is there something curious with the "umask" of the
> > user running the test? Are they different from what you see on your
> > other mac that does not exhibit the problems?
> >
> > Thanks.
> >
> >
>
> mv is /bin/mv, that seems to be good:
>
> $ type mv
> mv is /bin/mv
>
> $ alias | grep mv
>
> $ which mv
> /bin/mv
>
> $ umask
> 0022
>
> I don't know, why we see
> r--r--r-- tb/staff for .git/objects/info/commit-graph
>
> But, the "-r--r--r--" may be part of the problem, here is another one:
> $ find . -name commit-graph -print0 | xargs -0 ls -l
> -r--r--r-- 1 tb staff 1792 May 2 12:12 ./trash directory.t5318-commit-graph/bare/objects/info/commit-graph
> (And some more 6 in total. All with -r--r--r--)
>
> Which means, yes, t5318 does not pass either:
> t5318-commit-graph.sh not ok 101 - reader notices too-small oid fanout chunk
> t5318-commit-graph.sh not ok 102 - reader notices fanout/lookup table mismatch
> t5318-commit-graph.sh not ok 103 - reader notices out-of-bounds fanout
> t5318-commit-graph.sh not ok 104 - reader notices too-small commit data chunk
> t5318-commit-graph.sh not ok 105 - reader notices out-of-bounds extra edge
> t5318-commit-graph.sh not ok 106 - reader notices too-small generations chunk
>
> Same problem here:
> ++ mv full/.git/objects/info/commit-graph.tmp full/.git/objects/info/commit-graph
> override r--r--r-- tb/staff for full/.git/objects/info/commit-graph? (y/n [n]) not overwritten
'mv' in macOS doesn't conform to POSIX, and asks for confirmation when
the destination exists (is read-only?) even without '-i' and even when
its stdin is not a terminal, which it won't get as its stdin in our
test suite is redirected from /dev/null.
This is a recurring issue, see e.g.:
https://public-inbox.org/git/[email protected]/
c20d4d702f (t1450: use "mv -f" within loose object directory, 2017-01-24)
'mv -f' did the trick in the past.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 19:26 ` SZEDER Gábor
@ 2024-05-02 19:55 ` Junio C Hamano
2024-05-02 20:05 ` Junio C Hamano
2024-05-02 20:15 ` Junio C Hamano
0 siblings, 2 replies; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 19:55 UTC (permalink / raw)
To: SZEDER Gábor; +Cc: Torsten Bögershausen, git
SZEDER Gábor <[email protected]> writes:
>> Yes, different mac, different MacOs, different $PATH probably.
>> ...
>
> 'mv' in macOS doesn't conform to POSIX, and asks for confirmation when
> the destination exists (is read-only?) even without '-i' and even when
> its stdin is not a terminal, which it won't get as its stdin in our
> test suite is redirected from /dev/null.
>
> This is a recurring issue, see e.g.:
>
> https://public-inbox.org/git/[email protected]/
> c20d4d702f (t1450: use "mv -f" within loose object directory, 2017-01-24)
>
> 'mv -f' did the trick in the past.
Hmph, while that is a good data point to explain why this one fails,
it is a very unsatisfactory workaround, compared to a world where we
do not have to worry about such a broken mv (perhaps by noticing a
broken macOS /bin/mv and automatically doing mv () { mv -f "$@" }
for them).
I am curious what other differences Torsten will find out between
good macs and bad ones. Perhaps we can narrow down the bad apples?
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 19:55 ` Junio C Hamano
@ 2024-05-02 20:05 ` Junio C Hamano
2024-05-02 20:15 ` Junio C Hamano
1 sibling, 0 replies; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 20:05 UTC (permalink / raw)
To: SZEDER Gábor; +Cc: Torsten Bögershausen, git
Junio C Hamano <[email protected]> writes:
> ... it is a very unsatisfactory workaround, compared to a world where we
> do not have to worry about such a broken mv (perhaps by noticing a
> broken macOS /bin/mv and automatically doing mv () { mv -f "$@" }
> for them).
>
> I am curious what other differences Torsten will find out between
> good macs and bad ones. Perhaps we can narrow down the bad apples?
In any case, while we are waiting, I did a few "grep":
$ git grep 'mv \(.*\)\.tmp \1' t
t/lib-t6000.sh: mv sed.script.tmp sed.script
t/t7508-status.sh: rm "$1" && mv "$1".tmp "$1"
t/t8011-blame-split-file.sh: mv one.tmp one &&
t/t8011-blame-split-file.sh: mv two.tmp two &&
t/t9400-git-cvsserver-server.sh: mv merge.tmp merge &&
t/t9400-git-cvsserver-server.sh: mv merge.tmp merge &&
t/t9802-git-p4-filetype.sh: mv empty-symlink,v.tmp empty-symlink,v
$ git grep 'mv "\(.*\)\.tmp" "\1"' t
t/lib-chunk.sh: mv "$fn.tmp" "$fn"
t/t3404-rebase-interactive.sh: mv "$1.tmp" "$1"
t/t5515-fetch-merge-logic.sh: mv "$file.tmp" "$file"
t/t7600-merge.sh:) >"$1.tmp" && mv "$1.tmp" "$1"
t/t9001-send-email.sh: mv "$1.tmp" "$1"
t/t9001-send-email.sh: mv "$1.tmp" "$1"
$ git grep 'mv \(.*\)+ \1' t
t/t4025-hunk-header.sh: mv file+ file
t/t6402-merge-rename.sh: mv A+ A &&
t/t6402-merge-rename.sh: mv A+ A &&
t/t6432-merge-recursive-space-options.sh: mv text.txt+ text.txt &&
t/t6432-merge-recursive-space-options.sh: mv text.txt+ text.txt &&
There is nothing other than the one we found in lib-chunk.sh that
work on files under .git/objects/ that are made read-only, so we
should be OK with a workaround like the attached.
t/lib-chunk.sh | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git c/t/lib-chunk.sh w/t/lib-chunk.sh
index a7cd9c3c6d..9f01df190b 100644
--- c/t/lib-chunk.sh
+++ w/t/lib-chunk.sh
@@ -13,5 +13,6 @@ corrupt_chunk_file () {
fn=$1; shift
perl "$TEST_DIRECTORY"/lib-chunk/corrupt-chunk-file.pl \
"$@" <"$fn" >"$fn.tmp" &&
- mv "$fn.tmp" "$fn"
+ # some vintages of macOS 'mv' fails to overwrite a read-only file.
+ mv -f "$fn.tmp" "$fn"
}
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 19:55 ` Junio C Hamano
2024-05-02 20:05 ` Junio C Hamano
@ 2024-05-02 20:15 ` Junio C Hamano
2024-05-02 20:51 ` Torsten Bögershausen
1 sibling, 1 reply; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 20:15 UTC (permalink / raw)
To: SZEDER Gábor; +Cc: Torsten Bögershausen, git
Junio C Hamano <[email protected]> writes:
> I am curious what other differences Torsten will find out between
> good macs and bad ones. Perhaps we can narrow down the bad apples?
So, "no, your 'mv' is broken" seems to be the answer to the question
on the Subject line, and it is rather well-known, it seems.
----- >8 --------- >8 --------- >8 --------- >8 -----
Subject: [PATCH] t/lib-chunk: work around broken "mv" on some vintage of macOS
When the destination is read-only, "mv" on some version of macOS
asks whether to replace the destination even though in the test its
stdin is not a terminal (and thus doesn't conform to POSIX[1]).
The helper to corrupt a chunk-file is designed to work on the
files like commit-graph and multi-pack-index files that are
generally read-only, so use "mv -f" to work around this issue.
Helped-by: SZEDER Gábor <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>
---
t/lib-chunk.sh | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/t/lib-chunk.sh b/t/lib-chunk.sh
index a7cd9c3c6d..9f01df190b 100644
--- a/t/lib-chunk.sh
+++ b/t/lib-chunk.sh
@@ -13,5 +13,6 @@ corrupt_chunk_file () {
fn=$1; shift
perl "$TEST_DIRECTORY"/lib-chunk/corrupt-chunk-file.pl \
"$@" <"$fn" >"$fn.tmp" &&
- mv "$fn.tmp" "$fn"
+ # some vintages of macOS 'mv' fails to overwrite a read-only file.
+ mv -f "$fn.tmp" "$fn"
}
--
2.45.0-31-gd4cc1ec35f
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 20:15 ` Junio C Hamano
@ 2024-05-02 20:51 ` Torsten Bögershausen
2024-05-02 21:20 ` Junio C Hamano
0 siblings, 1 reply; 12+ messages in thread
From: Torsten Bögershausen @ 2024-05-02 20:51 UTC (permalink / raw)
To: Junio C Hamano; +Cc: SZEDER Gábor, git
On Thu, May 02, 2024 at 01:15:57PM -0700, Junio C Hamano wrote:
> Junio C Hamano <[email protected]> writes:
>
> > I am curious what other differences Torsten will find out between
> > good macs and bad ones. Perhaps we can narrow down the bad apples?
>
> So, "no, your 'mv' is broken" seems to be the answer to the question
> on the Subject line, and it is rather well-known, it seems.
[]
> "$@" <"$fn" >"$fn.tmp" &&
> - mv "$fn.tmp" "$fn"
> + # some vintages of macOS 'mv' fails to overwrite a read-only file.
> + mv -f "$fn.tmp" "$fn"
Thanks folks for digging - the patch makes both test cases pass.
Some official information:
$ man mv
on the both machines says
[snip]
If the destination path does not have a mode which permits writing,
mv prompts the user for confirmation as specified for the -i option.
To summarize what I have at hand right now:
MacOs 11.7.10 fails t4216 and t5318, both pass under MacOs 13.6.6
But why doesn't `mv` not hang then ?
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: t4216-log-bloom.sh broken ?
2024-05-02 20:51 ` Torsten Bögershausen
@ 2024-05-02 21:20 ` Junio C Hamano
0 siblings, 0 replies; 12+ messages in thread
From: Junio C Hamano @ 2024-05-02 21:20 UTC (permalink / raw)
To: Torsten Bögershausen; +Cc: SZEDER Gábor, git
Torsten Bögershausen <[email protected]> writes:
> Thanks folks for digging - the patch makes both test cases pass.
> Some official information:
>
> $ man mv
>
> on the both machines says
> [snip]
> If the destination path does not have a mode which permits writing,
> mv prompts the user for confirmation as specified for the -i option.
>
> To summarize what I have at hand right now:
> MacOs 11.7.10 fails t4216 and t5318, both pass under MacOs 13.6.6
>
> But why doesn't `mv` not hang then ?
It is a sensible choice for the command to assume a "safer" no and
continuing rather than hanging, so that part is understandable.
So presumably 13.6.6 has a working "mv" but their "man mv" is not
updated to how the command actually works?
In any case, let me declare victory and queue the patch I sent
earlier.
Thanks.
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2024-05-02 21:21 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-04-29 17:12 [ANNOUNCE] Git v2.45.0 Junio C Hamano
2024-05-01 17:39 ` rsbecker
2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen
2024-05-02 16:06 ` Junio C Hamano
2024-05-02 18:59 ` Torsten Bögershausen
2024-05-02 19:26 ` SZEDER Gábor
2024-05-02 19:55 ` Junio C Hamano
2024-05-02 20:05 ` Junio C Hamano
2024-05-02 20:15 ` Junio C Hamano
2024-05-02 20:51 ` Torsten Bögershausen
2024-05-02 21:20 ` Junio C Hamano
2024-05-02 16:26 ` Junio C Hamano
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).
{
"title": "[ANNOUNCE] Git v2.45.0",
"description": "* [ANNOUNCE] Git v2.45.0 @ 2024-04-29 17:12 Junio C Hamano 2024-05-01 17:39 ` rsbecker 2024-05-02 5:56 ` t4216-log-bloom.sh broken ? Torsten Bögershausen 0 siblings, 2 replies; 12+ messages in...",
"links": [
],
"image": "",
"content": "<pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#ee0af79d15d5071ef42f089f0ea4b9ced4f206d45\">*</a> <u><b>[ANNOUNCE] Git v2.45.0</b></u>\n<b>@ 2024-04-29 17:12 Junio C Hamano</b>\n 2024-05-01 17:39 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">rsbecker</a>\n 2024-05-02 5:56 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">t4216-log-bloom.sh broken ?</a> Torsten Bögershausen\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#re0af79d15d5071ef42f089f0ea4b9ced4f206d45\">0 siblings, 2 replies; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-04-29 17:12 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240429171233\">git</a>; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/lkml/?t=20240429171233\">Linux Kernel</a>, git-packagers\nThe latest feature release Git v2.45.0 is now available at the\nusual places. It is comprised of 540 non-merge commits since\nv2.44.0, contributed by 96 people, 38 of which are new faces [*].\nThe tarballs are found at:\n <a target=\"_blank\" href=\"https://www.kernel.org/pub/software/scm/git/\">https://www.kernel.org/pub/software/scm/git/</a>\nThe following public repositories all have a copy of the 'v2.45.0'\ntag and the 'master' branch that the tag points at:\n url = <a target=\"_blank\" href=\"https://git.kernel.org/pub/scm/git/git\">https://git.kernel.org/pub/scm/git/git</a>\n url = <a target=\"_blank\" href=\"https://kernel.googlesource.com/pub/scm/git/git\">https://kernel.googlesource.com/pub/scm/git/git</a>\n url = git://repo.or.cz/alt-git.git\n url = <a target=\"_blank\" href=\"https://github.com/gitster/git\">https://github.com/gitster/git</a>\nNew contributors whose contributions weren't in v2.44.0 are as follows.\nWelcome to the Git development community!\n Ahelenia Ziemiańska, Angelo Dureghello, Aryan Gupta, Benjamin\n Flesch, Bo Anderson, Brian C Tracy, Brian Lyles, Bruno Haible,\n Chuck Lever, Dario Gjorgjevski, Dirk Gouders, Eugenio Gigante,\n Florian Schmidt, Haritha D, Harmen Stoppels, Jean-Rémy\n Falleri, Jiamu Sun, Jonas Wunderlich, Jonathan Davies, Julio\n Bacellari, Kipras Melnikovas, Kisaragi Hiu, Marcel Röthke,\n Matthew Rollings, Max Gautier, mirth hickford, Paweł Dominiak,\n Pi Fisher, Ralph Seichter, Richard Macklin, shejialuo, Steven\n Jeuris, Thalia Archibald, Tiago Pascoal, Vincenzo Mezzela,\n Vũ Tiến Hưng, Xing Xin, and Yehezkel Bernat.\nReturning contributors who helped this release are as follows.\nThanks for your continued support.\n Alexander Shopov, Arkadii Yakovets, Bagas Sanjaya, Beat Bolli,\n brian m. carlson, Chandra Pratap, Christian Couder, Derrick\n Stolee, Đoàn Trần Công Danh, Dragan Simic, Elijah Newren,\n Emir SARI, Eric Sunshine, Eric W. Biederman, Ghanshyam Thakkar,\n Han Young, Jakub Wilk, Jean-Noël Avila, Jeff Hostetler, Jeff\n King, Jiang Xin, Johannes Schindelin, Johannes Sixt, John Cai,\n Josh Steadmon, Josh Triplett, Junio C Hamano, Justin Tobler,\n Karthik Nayak, Kate Golovanova, Kristoffer Haugsbakk, Kyle\n Lippincott, Kyle Meyer, Linus Arver, Lumynous, Manlio Perillo,\n Matthias Aßhauer, Matthias Rüster, M Hickford, Michael Lohmann,\n Michael Osipov, Mike Hommey, Orgad Shaneh, Patrick Steinhardt,\n Peter Hutterer, Peter Krefting, Philippe Blain, Phillip Wood,\n Ralf Thielow, René Scharfe, Rubén Justo, Sergey Organov,\n SZEDER Gábor, Taylor Blau, Teng Long, Ville Skyttä, Yasushi\n SHOJI, and Yi-Jyun Pan.\n[*] We are counting not just the authorship contribution but issue\n reporting, mentoring, helping and reviewing that are recorded in\n the commit trailers.\n----------------------------------------------------------------\nGit v2.45 Release Notes\n=======================\nBackward Compatibility Notes\nUI, Workflows & Features\n * Integrate the reftable code into the refs framework as a backend.\n With \"git init --ref-format=reftable\", hopefully it would be a lot\n more efficient to manage a repository with many references.\n * \"git checkout -p\" and friends learned that that \"@\" is a synonym\n for \"HEAD\".\n * Variants of vimdiff learned to honor mergetool.<variant>.layout\n settings.\n * \"git reflog\" learned a \"list\" subcommand that enumerates known reflogs.\n * When a merge conflicted at a submodule, merge-ort backend used to\n unconditionally give a lengthy message to suggest how to resolve\n it. Now the message can be squelched as an advice message.\n * \"git for-each-ref\" learned \"--include-root-refs\" option to show\n even the stuff outside the 'refs/' hierarchy.\n * \"git rev-list --missing=print\" has learned to optionally take\n \"--allow-missing-tips\", which allows the objects at the starting\n points to be missing.\n * \"git merge-tree\" has learned that the three trees involved in the\n 3-way merge only need to be trees, not necessarily commits.\n * \"git log --merge\" learned to pay attention to CHERRY_PICK_HEAD and\n other kinds of *_HEAD pseudorefs.\n * Platform specific tweaks for OS/390 has been added to\n config.mak.uname.\n * Users with safe.bareRepository=explicit can still work from within\n $GIT_DIR of a seconary worktree (which resides at .git/worktrees/$name/)\n of the primary worktree without explicitly specifying the $GIT_DIR\n environment variable or the --git-dir=<path> option.\n * The output format for dates \"iso-strict\" has been tweaked to show\n a time in the Zulu timezone with \"Z\" suffix, instead of \"+00:00\".\n * \"git diff\" and friends learned two extra configuration variables,\n diff.srcPrefix and diff.dstPrefix.\n * The status.showUntrackedFiles configuration variable had a name\n that tempts users to set a Boolean value expressed in our usual\n \"false\", \"off\", and \"0\", but it only took \"no\". This has been\n corrected so \"true\" and its synonyms are taken as \"normal\", while\n \"false\" and its synonyms are taken as \"no\".\n * Remove an ancient and not well maintained Hg-to-git migration\n script from contrib/.\n * Hints that suggest what to do after resolving conflicts can now be\n squelched by disabling advice.mergeConflict.\n * Allow git-cherry-pick(1) to automatically drop redundant commits via\n a new `--empty` option, similar to the `--empty` options for\n git-rebase(1) and git-am(1). Includes a soft deprecation of\n `--keep-redundant-commits` as well as some related docs changes and\n sequencer code cleanup.\n * \"git config\" learned \"--comment=<message>\" option to leave a\n comment immediately after the \"variable = value\" on the same line\n in the configuration file.\n * core.commentChar used to be limited to a single byte, but has been\n updated to allow an arbitrary multi-byte sequence.\n * \"git add -p\" and other \"interactive hunk selection\" UI has learned to\n skip showing the hunk immediately after it has already been shown, and\n an additional action to explicitly ask to reshow the current hunk.\n * \"git pack-refs\" learned the \"--auto\" option, which defers the decision of\n whether and how to pack to the ref backend. This is used by the reftable\n backend to avoid repacking of an already-optimal ref database. The new mode\n is triggered from \"git gc --auto\".\n * \"git add -u <pathspec>\" and \"git commit [-i] <pathspec>\" did not\n diagnose a pathspec element that did not match any files in certain\n situations, unlike \"git add <pathspec>\" did.\n * The userdiff patterns for C# has been updated.\n * Git writes a \"waiting for your editor\" message on an incomplete\n line after launching an editor, and then append another error\n message on the same line if the editor errors out. It now clears\n the \"waiting for...\" line before giving the error message.\n * The filename used for rejected hunks \"git apply --reject\" creates\n was limited to PATH_MAX, which has been lifted.\n * When \"git bisect\" reports the commit it determined to be the\n culprit, we used to show it in a format that does not honor common\n UI tweaks, like log.date and log.decorate. The code has been\n taught to use \"git show\" to follow more customizations.\nPerformance, Internal Implementation, Development Support etc.\n * The code to iterate over refs with the reftable backend has seen\n some optimization.\n * More tests that are marked as \"ref-files only\" have been updated to\n improve test coverage of reftable backend.\n * Some parts of command line completion script (in contrib/) have\n been micro-optimized.\n * The way placeholders are to be marked-up in documentation have been\n specified; use \"_<placeholder>_\" to typeset the word inside a pair\n of <angle-brackets> emphasized.\n * \"git --no-lazy-fetch cmd\" allows to run \"cmd\" while disabling lazy\n fetching of objects from the promisor remote, which may be handy\n for debugging.\n * The implementation in \"git clean\" that makes \"-n\" and \"-i\" ignore\n clean.requireForce has been simplified, together with the\n documentation.\n * Uses of xwrite() helper have been audited and updated for better\n error checking and simpler code.\n * Some trace2 events that lacked def_param have learned to show it,\n enriching the output.\n * The parse-options code that deals with abbreviated long option\n names have been cleaned up.\n * The code in reftable backend that creates new table files works\n better with the tempfile framework to avoid leaving cruft after a\n failure.\n * The reftable code has its own custom binary search function whose\n comparison callback has an unusual interface, which caused the\n binary search to degenerate into a linear search, which has been\n corrected.\n * The code to iterate over reflogs in the reftable has been optimized\n to reduce memory allocation and deallocation.\n * Work to support a repository that work with both SHA-1 and SHA-256\n hash algorithms has started.\n * A new fuzz target that exercises config parsing code has been\n added.\n * Fix the way recently added tests interpolate variables defined\n outside them, and document the best practice to help future\n developers.\n * Introduce an experimental protocol for contributors to propose the\n topic description to be used in the \"What's cooking\" report, the\n merge commit message for the topic, and in the release notes and\n document it in the SubmittingPatches document.\n * The t/README file now gives a hint on running individual tests in\n the \"t/\" directory with \"make t<num>-*.sh t<num>-*.sh\".\n (merge 8d383806fc pb/test-scripts-are-build-targets later to maint).\n * The \"hint:\" messages given by the advice mechanism, when given a\n message with a blank line, left a line with trailing whitespace,\n which has been cleansed.\n * Documentation rules has been explicitly described how to mark-up\n literal parts and a few manual pages have been updated as examples.\n * The .editorconfig file has been taught that a Makefile uses HT\n indentation.\n * t-prio-queue test has been cleaned up by using C99 compound\n literals; this is meant to also serve as a weather-balloon to smoke\n out folks with compilers who have trouble compiling code that uses\n the feature.\n * Windows binary used to decide the use of unix-domain socket at\n build time, but it learned to make the decision at runtime instead.\n * The \"shared repository\" test in the t0610 reftable test failed\n under restrictive umask setting (e.g. 007), which has been\n corrected.\n * Document and apply workaround for a buggy version of dash that\n mishandles \"local var=val\" construct.\n * The codepaths that reach date_mode_from_type() have been updated to\n pass \"struct date_mode\" by value to make them thread safe.\n * The strategy to compact multiple tables of reftables after many\n operations accumulate many entries has been improved to avoid\n accumulating too many tables uncollected.\n * The code to iterate over reftable blocks has seen some optimization\n to reduce memory allocation and deallocation.\n * The way \"git fast-import\" handles paths described in its input has\n been tightened up and more clearly documented.\n * The cvsimport tests required that the platform understands\n traditional timezone notations like CST6CDT, which has been\n updated to work on those systems as long as they understand\n POSIX notation with explicit tz transition dates.\n * The code to format trailers have been cleaned up.\nFixes since v2.44\n-----------------\n * \"git apply\" on a filesystem without filemode support have learned\n to take a hint from what is in the index for the path, even when\n not working with the \"--index\" or \"--cached\" option, when checking\n the executable bit match what is required by the preimage in the\n patch.\n (merge 45b625142d cp/apply-core-filemode later to maint).\n * \"git column\" has been taught to reject negative padding value, as\n it would lead to nonsense behaviour including division by zero.\n (merge 76fb807faa kh/column-reject-negative-padding later to maint).\n * \"git am --help\" now tells readers what actions are available in\n \"git am --whitespace=<action>\", in addition to saying that the\n option is passed through to the underlying \"git apply\".\n (merge a171dac734 jc/am-whitespace-doc later to maint).\n * \"git tag --column\" failed to check the exit status of its \"git\n column\" invocation, which has been corrected.\n (merge 92e66478fc rj/tag-column-fix later to maint).\n * Credential helper based on libsecret (in contrib/) has been updated\n to handle an empty password correctly.\n (merge 8f1f2023b7 mh/libsecret-empty-password-fix later to maint).\n * \"git difftool --dir-diff\" learned to honor the \"--trust-exit-code\"\n option; it used to always exit with 0 and signalled success.\n (merge eb84c8b6ce ps/difftool-dir-diff-exit-code later to maint).\n * The code incorrectly attempted to use textconv cache when asked,\n even when we are not running in a repository, which has been\n corrected.\n (merge affe355fe7 jk/textconv-cache-outside-repo-fix later to maint).\n * Remove an empty file that shouldn't have been added in the first\n place.\n (merge 4f66942215 js/remove-cruft-files later to maint).\n * The logic to access reflog entries by date and number had ugly\n corner cases at the boundaries, which have been cleaned up.\n (merge 5edd126720 jk/reflog-special-cases-fix later to maint).\n * An error message from \"git upload-pack\", which responds to \"git\n fetch\" requests, had a trailing NUL in it, which has been\n corrected.\n (merge 3f4c7a0805 sg/upload-pack-error-message-fix later to maint).\n * Clarify wording in the CodingGuidelines that requires <git-compat-util.h>\n to be the first header file.\n (merge 4e89f0e07c jc/doc-compat-util later to maint).\n * \"git commit -v --cleanup=scissors\" used to add the scissors line\n twice in the log message buffer, which has been corrected.\n (merge e90cc075cc jt/commit-redundant-scissors-fix later to maint).\n * A custom remote helper no longer cannot access the newly created\n repository during \"git clone\", which is a regression in Git 2.44.\n This has been corrected.\n (merge 199f44cb2e ps/remote-helper-repo-initialization-fix later to maint).\n * Various parts of upload-pack have been updated to bound the resource\n consumption relative to the size of the repository to protect from\n abusive clients.\n (merge 6cd05e768b jk/upload-pack-bounded-resources later to maint).\n * The upload-pack program, when talking over v2, accepted the\n packfile-uris protocol extension from the client, even if it did\n not advertise the capability, which has been corrected.\n (merge a922bfa3b5 jk/upload-pack-v2-capability-cleanup later to maint).\n * Make sure failure return from merge_bases_many() is properly caught.\n (merge 25fd20eb44 js/merge-base-with-missing-commit later to maint).\n * FSMonitor client code was confused when FSEvents were given in a\n different case on a case-insensitive filesystem, which has been\n corrected.\n (merge 29c139ce78 jh/fsmonitor-icase-corner-case-fix later to maint).\n * The \"core.commentChar\" configuration variable only allows an ASCII\n character, which was not clearly documented, which has been\n corrected.\n (merge fb7c556f58 kh/doc-commentchar-is-a-byte later to maint).\n * With release 2.44 we got rid of all uses of test_i18ngrep and there\n is no in-flight topic that adds a new use of it. Make a call to\n test_i18ngrep a hard failure, so that we can remove it at the end\n of this release cycle.\n (merge 381a83dfa3 jc/test-i18ngrep later to maint).\n * The command line completion script (in contrib/) learned to\n complete \"git reflog\" better.\n (merge 1284f9cc11 rj/complete-reflog later to maint).\n * The logic to complete the command line arguments to \"git worktree\"\n subcommand (in contrib/) has been updated to correctly honor things\n like \"git -C dir\" etc.\n (merge 3574816d98 rj/complete-worktree-paths-fix later to maint).\n * When git refuses to create a branch because the proposed branch\n name is not a valid refname, an advice message is given to refer\n the user to exact naming rules.\n (merge 8fbd903e58 kh/branch-ref-syntax-advice later to maint).\n * Code simplification by getting rid of code that sets an environment\n variable that is no longer used.\n (merge 72a8d3f027 pw/rebase-i-ignore-cherry-pick-help-environment later to maint).\n * The code to find the effective end of log messages can fall into an\n endless loop, which has been corrected.\n (merge 2541cba2d6 fs/find-end-of-log-message-fix later to maint).\n * Mark-up used in the documentation has been improved for\n consistency.\n (merge 45d5ed3e50 ja/doc-markup-fixes later to maint).\n * The status.showUntrackedFiles configuration variable was\n incorrectly documented to accept \"false\", which has been corrected.\n * Leaks from \"git restore\" have been plugged.\n (merge 2f64da0790 rj/restore-plug-leaks later to maint).\n * \"git bugreport --no-suffix\" was not supported and instead\n segfaulted, which has been corrected.\n (merge b3b57c69da js/bugreport-no-suffix-fix later to maint).\n * The documentation for \"%(trailers[:options])\" placeholder in the\n \"--pretty\" option of commands in the \"git log\" family has been\n updated.\n (merge bff85a338c bl/doc-key-val-sep-fix later to maint).\n * \"git checkout --conflict=bad\" reported a bad conflictStyle as if it\n were given to a configuration variable; it has been corrected to\n report that the command line option is bad.\n (merge 5a99c1ac1a pw/checkout-conflict-errorfix later to maint).\n * Code clean-up in the \"git log\" machinery that implements custom log\n message formatting.\n (merge 1c10b8e5b0 jk/pretty-subject-cleanup later to maint).\n * \"git config\" corrupted literal HT characters written in the\n configuration file as part of a value, which has been corrected.\n (merge e6895c3f97 ds/config-internal-whitespace-fix later to maint).\n * A unit test for reftable code tried to enumerate all files in a\n directory after reftable operations and expected to see nothing but\n the files it wanted to leave there, but was fooled by .nfs* cruft\n files left, which has been corrected.\n (merge 0068aa7946 ps/reftable-unit-test-nfs-workaround later to maint).\n * The implementation and documentation of \"object-format\" option\n exchange between the Git itself and its remote helpers did not\n quite match, which has been corrected.\n * The \"--pretty=<shortHand>\" option of the commands in the \"git log\"\n family, defined as \"[pretty] shortHand = <expansion>\" should have\n been looked up case insensitively, but was not, which has been\n corrected.\n (merge f999d5188b bl/pretty-shorthand-config-fix later to maint).\n * \"git apply\" failed to extract the filename the patch applied to,\n when the change was about an empty file created in or deleted from\n a directory whose name ends with a SP, which has been corrected.\n (merge 776ffd1a30 jc/apply-parse-diff-git-header-names-fix later to maint).\n * Update a more recent tutorial doc.\n (merge 95ab557b4b dg/myfirstobjectwalk-updates later to maint).\n * The test script had an incomplete and ineffective attempt to avoid\n clobbering the testing user's real crontab (and its equivalents),\n which has been completed.\n (merge 73cb87773b es/test-cron-safety later to maint).\n * Use advice_if_enabled() API to rewrite a simple pattern to\n call advise() after checking advice_enabled().\n (merge 6412d01527 rj/use-adv-if-enabled later to maint).\n * Another \"set -u\" fix for the bash prompt (in contrib/) script.\n (merge d7805bc743 vs/complete-with-set-u-fix later to maint).\n * \"git checkout/switch --detach foo\", after switching to the detached\n HEAD state, gave the tracking information for the 'foo' branch,\n which was pointless.\n * \"git apply\" has been updated to lift the hardcoded pathname length\n limit, which in turn allowed a mksnpath() function that is no\n longer used.\n (merge 708f7e0590 rs/apply-lift-path-length-limit later to maint).\n * A file descriptor leak in an error codepath, used when \"git apply\n --reject\" fails to create the *.rej file, has been corrected.\n (merge 2b1f456adf rs/apply-reject-fd-leakfix later to maint).\n * A config parser callback function fell through instead of returning\n after recognising and processing a variable, wasting cycles, which\n has been corrected.\n (merge a816ccd642 ds/fetch-config-parse-microfix later to maint).\n * Fix was added to work around a regression in libcURL 8.7.0 (which has\n already been fixed in their tip of the tree).\n (merge 92a209bf24 jk/libcurl-8.7-regression-workaround later to maint).\n * The variable that holds the value read from the core.excludefile\n configuration variable used to leak, which has been corrected.\n (merge 0e0fefb29f jc/unleak-core-excludesfile later to maint).\n * vreportf(), which is used by error() and friends, has been taught\n to give the error message printf-format string when its vsnprintf()\n call fails, instead of showing nothing useful to identify the\n nature of the error.\n (merge c63adab961 rs/usage-fallback-to-show-message-format later to maint).\n * Adjust to an upcoming changes to GNU make that breaks our Makefiles.\n (merge 227b8fd902 tb/make-indent-conditional-with-non-spaces later to maint).\n * Git 2.44 introduced a regression that makes the updated code to\n barf in repositories with multi-pack index written by older\n versions of Git, which has been corrected.\n * When .git/rr-cache/ rerere database gets corrupted or rerere is fed to\n work on a file with conflicted hunks resolved incompletely, the rerere\n machinery got confused and segfaulted, which has been corrected.\n (merge 167395bb47 mr/rerere-crash-fix later to maint).\n * The \"receive-pack\" program (which responds to \"git push\") was not\n converted to run \"git maintenance --auto\" when other codepaths that\n used to run \"git gc --auto\" were updated, which has been corrected.\n (merge 7bf3057d9c ps/run-auto-maintenance-in-receive-pack later to maint).\n * Other code cleanup, docfix, build fix, etc.\n (merge f0e578c69c rs/use-xstrncmpz later to maint).\n (merge 83e6eb7d7a ba/credential-test-clean-fix later to maint).\n (merge 64562d784d jb/doc-interactive-singlekey-do-not-need-perl later to maint).\n (merge c431a235e2 cp/t9146-use-test-path-helpers later to maint).\n (merge 82d75402d5 ds/doc-send-email-capitalization later to maint).\n (merge 41bff66e35 jc/doc-add-placeholder-fix later to maint).\n (merge 6835f0efe9 jw/remote-doc-typofix later to maint).\n (merge 244001aa20 hs/rebase-not-in-progress later to maint).\n (merge 2ca6c07db2 jc/no-include-of-compat-util-from-headers later to maint).\n (merge 87bd7fbb9c rs/fetch-simplify-with-starts-with later to maint).\n (merge f39addd0d9 rs/name-rev-with-mempool later to maint).\n (merge 9a97b43e03 rs/submodule-prefix-simplify later to maint).\n (merge 40b8076462 ak/rebase-autosquash later to maint).\n (merge 3223204456 eg/add-uflags later to maint).\n (merge 5f78d52dce es/config-doc-sort-sections later to maint).\n (merge 781fb7b4c2 as/option-names-in-messages later to maint).\n (merge 51d41dc243 jk/doc-remote-helpers-markup-fix later to maint).\n (merge e1aaf309db pb/ci-win-artifact-names-fix later to maint).\n (merge ad538c61da jc/index-pack-fsck-levels later to maint).\n (merge 67471bc704 ja/doc-formatting-fix later to maint).\n (merge 86f9ce7dd6 bl/doc-config-fixes later to maint).\n (merge 0d527842b7 az/grep-group-error-message-update later to maint).\n (merge 7c43bdf07b rs/strbuf-expand-bad-format later to maint).\n (merge 8b68b48d5c ds/typofix-core-config-doc later to maint).\n (merge 39bb692152 rs/imap-send-use-xsnprintf later to maint).\n (merge 8d320cec60 jc/t2104-style-fixes later to maint).\n (merge b4454d5a7b pw/t3428-cleanup later to maint).\n (merge 84a7c33a4b pf/commitish-committish later to maint).\n (merge 8882ee9d68 la/mailmap-entry later to maint).\n (merge 44bdba2fa6 rs/no-openssl-compilation-fix-on-macos later to maint).\n (merge f412d72c19 yb/replay-doc-linkfix later to maint).\n (merge 5da40be8d7 xx/rfc2822-date-format-in-doc later to maint).\n----------------------------------------------------------------\nChanges since v2.44.0 are as follows:\nAhelenia Ziemiańska (1):\n grep: improve errors for unmatched ( and )\nAlexander Shopov (5):\n transport-helper.c: trivial fix of error message\n builtin/remote.c: trivial fix of error message\n builtin/clone.c: trivial fix of message\n revision.c: trivial fix to message\n l10n: bg.po: Updated Bulgarian translation (5652t)\nArkadii Yakovets (1):\n l10n: uk: v2.45 update\nAryan Gupta (1):\n tests: modernize the test script t0010-racy-git.sh\nBagas Sanjaya (1):\n l10n: po-id for 2.45\nBeat Bolli (25):\n completion: use awk for filtering the config entries\n date: make \"iso-strict\" conforming for the UTC timezone\n t0006: add more tests with a negative TZ offset\n doc: avoid redundant use of cat\n contrib/subtree/t: avoid redundant use of cat\n t/lib-cvs.sh: avoid redundant use of cat\n t/annotate-tests.sh: avoid redundant use of cat\n t/perf: avoid redundant use of cat\n t/t0*: avoid redundant uses of cat\n t/t1*: avoid redundant uses of cat\n t/t3*: avoid redundant uses of cat\n t/t4*: avoid redundant uses of cat\n t/t5*: avoid redundant uses of cat\n t/t6*: avoid redundant uses of cat\n t/t7*: avoid redundant use of cat\n t/t8*: avoid redundant use of cat\n t/t9*: avoid redundant uses of cat\n t/t1*: merge a \"grep | sed\" pipeline\n t/t3*: merge a \"grep | awk\" pipeline\n t/t4*: merge a \"grep | sed\" pipeline\n t/t5*: merge a \"grep | sed\" pipeline\n t/t8*: merge \"grep | sed\" pipelines\n t/t9*: merge \"grep | sed\" pipelines\n contrib/coverage-diff: avoid redundant pipelines\n git-quiltimport: avoid an unnecessary subshell\nBo Anderson (5):\n t/lib-credential: clean additional credential\n osxkeychain: replace deprecated SecKeychain API\n osxkeychain: erase all matching credentials\n osxkeychain: erase matching passwords only\n osxkeychain: store new attributes\nBrian C Tracy (1):\n fuzz: add fuzzer for config parsing\nBrian Lyles (13):\n docs: clarify file options in git-config `--edit`\n docs: fix typo in git-config `--default`\n docs: correct trailer `key_value_separator` description\n docs: adjust trailer `separator` and `key_value_separator` language\n pretty: update tests to use `test_config`\n pretty: find pretty formats case-insensitively\n docs: address inaccurate `--empty` default with `--exec`\n docs: clean up `--empty` formatting in git-rebase(1) and git-am(1)\n rebase: update `--empty=ask` to `--empty=stop`\n sequencer: handle unborn branch with `--allow-empty`\n sequencer: do not require `allow_empty` for redundant commit options\n cherry-pick: enforce `--keep-redundant-commits` incompatibility\n cherry-pick: add `--empty` for more robust redundant commit handling\nChandra Pratap (2):\n apply: ignore working tree filemode when !core.filemode\n t9146: replace test -d/-e/-f with appropriate test_path_is_* function\nChristian Couder (5):\n revision: clarify a 'return NULL' in get_reference()\n oidset: refactor oidset_insert_from_set()\n t6022: fix 'test' style and 'even though' typo\n rev-list: allow missing tips with --missing=[print|allow*]\n revision: fix --missing=[print|allow*] for annotated tags\nDerrick Stolee (1):\n fetch: return when parsing submodule.recurse\nDirk Gouders (6):\n Documentation/user-manual.txt: example for generating object hashes\n MyFirstObjectWalk: use additional arg in config_fn_t\n MyFirstObjectWalk: fix misspelled \"builtins/\"\n MyFirstObjectWalk: fix filtered object walk\n MyFirstObjectWalk: fix description for counting omitted objects\n MyFirstObjectWalk: add stderr to pipe processing\nDragan Simic (8):\n documentation: send-email: use camel case consistently\n config: minor addition of whitespace\n config: really keep value-internal whitespace verbatim\n t1300: add more tests for whitespace and inline comments\n config.txt: describe handling of whitespace further\n grep docs: describe --recurse-submodules further and improve formatting a bit\n grep docs: describe --no-index further and improve formatting a bit\n config: fix some small capitalization issues, as spotted\nEmir SARI (1):\n l10n: tr: Update Turkish translations\nEric Sunshine (2):\n docs: sort configuration variable groupings alphabetically\n test-lib: fix non-functioning GIT_TEST_MAINT_SCHEDULER fallback\nEric W. Biederman (23):\n object-file-convert: stubs for converting from one object format to another\n oid-array: teach oid-array to handle multiple kinds of oids\n object-names: support input of oids in any supported hash\n repository: add a compatibility hash algorithm\n loose: compatibilty short name support\n object-file: update the loose object map when writing loose objects\n object-file: add a compat_oid_in parameter to write_object_file_flags\n commit: convert mergetag before computing the signature of a commit\n commit: export add_header_signature to support handling signatures on tags\n tag: sign both hashes\n object: factor out parse_mode out of fast-import and tree-walk into in object.h\n object-file-convert: don't leak when converting tag objects\n object-file-convert: convert commits that embed signed tags\n object-file: update object_info_extended to reencode objects\n rev-parse: add an --output-object-format parameter\n builtin/cat-file: let the oid determine the output algorithm\n tree-walk: init_tree_desc take an oid to get the hash algorithm\n object-file: handle compat objects in check_object_signature\n builtin/ls-tree: let the oid determine the output algorithm\n test-lib: compute the compatibility hash so tests may use it\n t1006: rename sha1 to oid\n t1006: test oid compatibility with cat-file\n t1016-compatObjectFormat: add tests to verify the conversion between objects\nEugenio Gigante (1):\n add: use unsigned type for collection of bits\nFlorian Schmidt (1):\n wt-status: don't find scissors line beyond buf len\nGhanshyam Thakkar (5):\n add-patch: classify '@' as a synonym for 'HEAD'\n add -p tests: remove PERL prerequisites\n setup: remove unnecessary variable\n builtin/commit: error out when passing untracked path with -i\n builtin/add: error out when passing untracked path with -u\nHaritha D (1):\n build: support z/OS (OS/390).\nHarmen Stoppels (1):\n rebase: make warning less passive aggressive\nJakub Wilk (1):\n git-remote.txt: fix typo\nJean-Noël Avila (18):\n doc: git-rev-parse: enforce command-line description syntax\n doc: close unclosed angle-bracket of a placeholder in git-clone doc\n doc: end sentences with full-stop\n doc: clarify the format of placeholders\n doc: git-init: format verbatim parts\n doc: git-init: format placeholders\n doc: git-init: rework definition lists\n doc: git-init: rework config item init.templateDir\n doc: git-clone: format verbatim words\n doc: git-clone: format placeholders\n doc: format alternatives in synopsis\n doc: fix some placeholders formating\n doc: rework CodingGuidelines with new formatting rules\n doc: allow literal and emphasis format in doc vs help tests\n doc: git-init: apply new documentation formatting guidelines\n doc: git-clone: apply new documentation formatting guidelines\n doc: git-clone: do not autoreference the manpage in itself\n l10n: fr: v2.45.0\nJeff Hostetler (17):\n name-hash: add index_dir_find()\n t7527: add case-insensitve test for FSMonitor\n fsmonitor: refactor refresh callback on directory events\n fsmonitor: clarify handling of directory events in callback helper\n fsmonitor: refactor refresh callback for non-directory events\n dir: create untracked_cache_invalidate_trimmed_path()\n fsmonitor: refactor untracked-cache invalidation\n fsmonitor: move untracked-cache invalidation into helper functions\n fsmonitor: return invalidated cache-entry count on directory event\n fsmonitor: remove custom loop from non-directory path handler\n fsmonitor: return invalidated cache-entry count on non-directory event\n fsmonitor: trace the new invalidated cache-entry count\n fsmonitor: refactor bit invalidation in refresh callback\n fsmonitor: support case-insensitive events\n t0211: demonstrate missing 'def_param' events for certain commands\n trace2: avoid emitting 'def_param' set more than once\n trace2: emit 'def_param' set with 'cmd_name' event\nJeff King (51):\n t0303: check that helper_test_clean removes all credentials\n userdiff: skip textconv caching when not in a repository\n Revert \"refs: allow @{n} to work with n-sized reflog\"\n get_oid_basic(): special-case ref@{n} for oldest reflog entry\n read_ref_at(): special-case ref@{0} for an empty reflog\n upload-pack: drop separate v2 \"haves\" array\n upload-pack: switch deepen-not list to an oid_array\n upload-pack: use oidset for deepen_not list\n upload-pack: use a strmap for want-ref lines\n upload-pack: accept only a single packfile-uri line\n upload-pack: always turn off save_commit_buffer\n upload-pack: use PARSE_OBJECT_SKIP_HASH_CHECK in more places\n upload-pack: free tree buffers after parsing\n upload-pack: use repository struct to get config\n upload-pack: centralize setup of sideband-all config\n upload-pack: use existing config mechanism for advertisement\n upload-pack: only accept packfile-uris if we advertised it\n doc/gitremote-helpers: fix missing single-quote\n config: forbid newline as core.commentChar\n strbuf: simplify comment-handling in add_lines() helper\n strbuf: avoid static variables in strbuf_add_commented_lines()\n commit: refactor base-case of adjust_comment_line_char()\n strbuf: avoid shadowing global comment_line_char name\n environment: store comment_line_char as a string\n strbuf: accept a comment string for strbuf_stripspace()\n strbuf: accept a comment string for strbuf_commented_addf()\n strbuf: accept a comment string for strbuf_add_commented_lines()\n prefer comment_line_str to comment_line_char for printing\n find multi-byte comment chars in NUL-terminated strings\n find multi-byte comment chars in unterminated buffers\n sequencer: handle multi-byte comment characters when writing todo list\n wt-status: drop custom comment-char stringification\n environment: drop comment_line_char compatibility macro\n config: allow multi-byte core.commentChar\n shortlog: stop setting pp.print_email_subject\n pretty: split oneline and email subject printing\n pretty: drop print_email_subject flag\n log: do not set up extra_headers for non-email formats\n format-patch: return an allocated string from log_write_email_headers()\n format-patch: simplify after-subject MIME header handling\n doc/gitremote-helpers: fix more missing single-quotes\n transport-helper: use write helpers more consistently\n transport-helper: drop \"object-format <algo>\" option\n transport-helper: send \"true\" value for object-format option\n contrib: drop hg-to-git script\n format-patch: fix leak of empty header string\n rebase: use child_process_clear() to clean\n config: add core.commentString\n http: reset POSTFIELDSIZE when clearing curl handle\n INSTALL: bump libcurl version to 7.21.3\n remote-curl: add Transfer-Encoding header only for older curl\nJiamu Sun (1):\n bugreport.c: fix a crash in `git bugreport` with `--no-suffix` option\nJiang Xin (1):\n l10n: TEAMS: retire l10n teams no update in 1 year\nJohannes Schindelin (22):\n merge-tree: accept 3 trees as arguments\n merge-tree: fail with a non-zero exit code on missing tree objects\n merge-ort: do check `parse_tree()`'s return value\n t4301: verify that merge-tree fails on missing blob objects\n Always check `parse_tree*()`'s return value\n cache-tree: avoid an unnecessary check\n fill_tree_descriptor(): mark error message for translation\n neue: remove a bogus empty file\n commit-reach(paint_down_to_common): plug two memory leaks\n commit-reach(repo_in_merge_bases_many): optionally expect missing commits\n commit-reach(repo_in_merge_bases_many): report missing commits\n commit-reach(paint_down_to_common): prepare for handling shallow commits\n commit-reach(paint_down_to_common): start reporting errors\n commit-reach(merge_bases_many): pass on \"missing commits\" errors\n commit-reach(get_merge_bases_many_0): pass on \"missing commits\" errors\n commit-reach(repo_get_merge_bases): pass on \"missing commits\" errors\n commit-reach(get_octopus_merge_bases): pass on \"missing commits\" errors\n commit-reach(repo_get_merge_bases_many): pass on \"missing commits\" errors\n commit-reach(repo_get_merge_bases_many_dirty): pass on errors\n merge-recursive: prepare for `merge_submodule()` to report errors\n merge-ort/merge-recursive: do report errors in `merge_submodule()`\n merge-tree: fix argument type of the `--merge-base` option\nJohn Cai (1):\n t5300: fix test_with_bad_commit()\nJonas Wunderlich (1):\n doc: status.showUntrackedFiles does not take \"false\"\nJosh Triplett (2):\n commit: avoid redundant scissor line with --cleanup=scissors -v\n commit: unify logic to avoid multiple scissors lines when merging\nJulio Bacellari (1):\n doc: remove outdated information about interactive.singleKey\nJunio C Hamano (64):\n apply: correctly reverse patch's pre- and post-image mode bits\n apply: code simplification\n t9210: do not rely on lazy fetching to fail\n git: --no-lazy-fetch option\n doc: add shortcut to \"am --whitespace=<action>\"\n doc: apply the new placeholder rules to git-add documentation\n compat: drop inclusion of <git-compat-util.h>\n Start the 2.45 cycle\n git: document GIT_NO_REPLACE_OBJECTS environment variable\n doc: clarify the wording on <git-compat-util.h> requirement\n git: extend --no-lazy-fetch to work across subprocesses\n The second batch\n The third batch\n test_i18ngrep: hard deprecate and forbid its use\n unpack: replace xwrite() loop with write_in_full()\n sideband: avoid short write(2)\n repack: check error writing to pack-objects subprocess\n clean: further clean-up of implementation around \"--force\"\n The fourth batch\n The fifth batch\n setup: notice more types of implicit bare repositories\n The sixth batch\n status: unify parsing of --untracked= and status.showUntrackedFiles\n status: allow --untracked=false and friends\n The seventh batch\n The eighth batch\n config: fix --comment formatting\n config: allow tweaking whitespace between value and comment\n diff.*Prefix: use camelCase in the doc and test titles\n The ninth batch\n apply: parse names out of \"diff --git\" more carefully\n The tenth batch\n The eleventh batch\n SubmittingPatches: release-notes entry experiment\n The twelfth batch\n t4126: make sure a directory with SP at the end is usable\n t4126: fix \"funny directory name\" test on Windows (again)\n advice: omit trailing whitespace\n checkout: omit \"tracking\" information on a detached HEAD\n The thirteenth batch\n t2104: style fixes\n The fourteenth batch\n revision: optionally record matches with pathspec elements\n The fifteenth batch\n CodingGuidelines: describe \"export VAR=VAL\" rule\n CodingGuidelines: quote assigned value in 'local var=$val'\n t: local VAR=\"VAL\" (quote positional parameters)\n t: local VAR=\"VAL\" (quote command substitution)\n t: local VAR=\"VAL\" (quote ${magic-reference})\n t: teach lint that RHS of 'local VAR=VAL' needs to be quoted\n t0610: local VAR=\"VAL\" fix\n t1016: local VAR=\"VAL\" fix\n config: do not leak excludes_file\n Makefile(s): do not enforce \"all indents must be done with tab\"\n The sixteenth batch\n t2104: style fixes\n The seventeenth batch\n The eighteenth batch\n The ninteenth batch\n The twentieth batch\n Git 2.45-rc0\n A bit more topics before -rc1\n Git 2.45-rc1\n Git 2.45\nJustin Tobler (3):\n reftable/stack: expose option to disable auto-compaction\n reftable/stack: add env to disable autocompaction\n reftable/stack: use geometric table compaction\nKarthik Nayak (7):\n refs: introduce `is_pseudoref()` and `is_headref()`\n refs: extract out `loose_fill_ref_dir_regular_file()`\n refs: introduce `refs_for_each_include_root_refs()`\n ref-filter: rename 'FILTER_REFS_ALL' to 'FILTER_REFS_REGULAR'\n for-each-ref: add new option to include root refs\n update-ref: use {old,new}-oid instead of {old,new}value\n githooks: use {old,new}-oid instead of {old,new}-value\nKipras Melnikovas (1):\n mergetools: vimdiff: use correct tool's name when reading mergetool config\nKristoffer Haugsbakk (9):\n column: disallow negative padding\n column: guard against negative padding\n gitcli: drop mention of “non-dashed form”\n config: document `core.commentChar` as ASCII-only\n t3200: improve test style\n advice: make all entries stylistically consistent\n advice: use backticks for verbatim\n advice: use double quotes for regular quoting\n branch: advise about ref syntax rules\nLinus Arver (15):\n trailer: free trailer_info _after_ all related usage\n shortlog: add test for de-duplicating folded trailers\n trailer: rename functions to use 'trailer'\n trailer: reorder format_trailers_from_commit() parameters\n trailer: move interpret_trailers() to interpret-trailers.c\n trailer_info_get(): reorder parameters\n format_trailers(): use strbuf instead of FILE\n format_trailer_info(): move \"fast path\" to caller\n format_trailers_from_commit(): indirectly call trailer_info_get()\n format_trailer_info(): use trailer_item objects\n format_trailer_info(): drop redundant unfold_value()\n format_trailer_info(): append newline for non-trailer lines\n trailer: begin formatting unification\n trailer: finish formatting unification\n mailmap: change primary address for Linus Arver\nM Hickford (1):\n libsecret: retrieve empty password\nMarcel Röthke (1):\n rerere: fix crashes due to unmatched opening conflict markers\nMatthias Aßhauer (1):\n Win32: detect unix socket support at runtime\nMax Gautier (1):\n editorconfig: add Makefiles to \"text files\"\nMichael Lohmann (2):\n revision: ensure MERGE_HEAD is a ref in prepare_show_merge\n revision: implement `git log --merge` also for rebase/cherry-pick/revert\nOrgad Shaneh (1):\n docs: remove duplicate entry and fix typo in 2.45 changelog\nPatrick Steinhardt (99):\n refs: introduce reftable backend\n ci: add jobs to test with the reftable backend\n refs/reftable: fix leak when copying reflog fails\n reftable/record: introduce function to compare records by key\n reftable/merged: allocation-less dropping of shadowed records\n reftable/merged: skip comparison for records of the same subiter\n reftable/pq: allocation-less comparison of entry keys\n reftable/block: swap buffers instead of copying\n reftable/record: don't try to reallocate ref record name\n reftable/reader: add comments to `table_iter_next()`\n t: move tests exercising the \"files\" backend\n t0410: convert tests to use DEFAULT_REPO_FORMAT prereq\n t1400: exercise reflog with gaps with reftable backend\n t1404: make D/F conflict tests compatible with reftable backend\n t1405: remove unneeded cleanup step\n t2011: exercise D/F conflicts with HEAD with the reftable backend\n t7003: ensure filter-branch prunes reflogs with the reftable backend\n git-difftool--helper: honor `--trust-exit-code` with `--dir-diff`\n dir-iterator: pass name to `prepare_next_entry_data()` directly\n dir-iterator: support iteration in sorted order\n refs/files: sort reflogs returned by the reflog iterator\n refs/files: sort merged worktree and common reflogs\n refs: always treat iterators as ordered\n refs: drop unused params from the reflog iterator callback\n refs: stop resolving ref corresponding to reflogs\n builtin/reflog: introduce subcommand to list reflogs\n builtin/clone: allow remote helpers to detect repo\n refs/reftable: don't fail empty transactions in repo without HEAD\n reftable/pq: use `size_t` to track iterator index\n reftable/merged: make `merged_iter` structure private\n reftable/merged: advance subiter on subsequent iteration\n reftable/merged: make subiters own their records\n reftable/merged: remove unnecessary null check for subiters\n reftable/merged: handle subiter cleanup on close only\n reftable/merged: circumvent pqueue with single subiter\n reftable/merged: avoid duplicate pqueue emptiness check\n reftable/record: reuse refname when decoding\n reftable/record: reuse refname when copying\n reftable/record: decode keys in place\n reftable: allow inlining of a few functions\n refs/reftable: precompute prefix length\n refs/reftable: reload correct stack when creating reflog iter\n reftable/record: convert old and new object IDs to arrays\n reftable/record: avoid copying author info\n reftable/record: reuse refnames when decoding log records\n reftable/record: reuse message when decoding log records\n reftable/record: use scratch buffer when decoding records\n refs/reftable: track last log record name via strbuf\n t0610: remove unused variable assignment\n lockfile: report when rollback fails\n reftable/stack: register new tables as tempfiles\n reftable/stack: register lockfiles during compaction\n reftable/stack: register compacted tables as tempfiles\n reftable/record: fix memory leak when decoding object records\n reftable/block: fix binary search over restart counter\n t5601: exercise clones with \"includeIf.*.onbranch\"\n reftable: fix tests being broken by NFS' delete-after-close semantics\n t7800: improve test descriptions with empty arguments\n t7800: use single quotes for test bodies\n t/README: document how to loop around test cases\n reftable/stack: fix error handling in `reftable_stack_init_addition()`\n reftable/error: discern locked/outdated errors\n reftable/stack: use error codes when locking fails during compaction\n reftable/stack: gracefully handle failed auto-compaction due to locks\n refs/reftable: print errors on compaction failure\n t/helper: drop pack-refs wrapper\n refs: move `struct pack_refs_opts` to where it's used\n refs: remove `PACK_REFS_ALL` flag\n refs/reftable: expose auto compaction via new flag\n builtin/pack-refs: release allocated memory\n builtin/pack-refs: introduce new \"--auto\" flag\n builtin/gc: move `struct maintenance_run_opts`\n t6500: extract objects with \"17\" prefix\n builtin/gc: forward git-gc(1)'s `--auto` flag when packing refs\n builtin/gc: pack refs when using `git maintenance run --auto`\n reftable/basics: fix return type of `binsearch()` to be `size_t`\n reftable/basics: improve `binsearch()` test\n reftable/refname: refactor binary search over refnames\n reftable/block: refactor binary search over restart points\n reftable/block: fix error handling when searching restart points\n reftable/record: extract function to decode key lengths\n reftable/block: avoid decoding keys when searching restart points\n t0610: make `--shared=` tests reusable\n t0610: execute git-pack-refs(1) with specified umask\n reftable/block: rename `block_reader_start()`\n reftable/block: merge `block_iter_seek()` and `block_reader_seek()`\n reftable/block: better grouping of functions\n reftable/block: introduce `block_reader_release()`\n reftable/block: move ownership of block reader into `struct table_iter`\n reftable/reader: iterate to next block in place\n reftable/block: reuse uncompressed blocks\n reftable/block: open-code call to `uncompress2()`\n reftable/block: reuse `zstream` state on inflation\n reftable/block: avoid copying block iterators on seek\n pack-bitmap: gracefully handle missing BTMP chunks\n run-command: introduce function to prepare auto-maintenance process\n builtin/receive-pack: convert to use git-maintenance(1)\n docs: improve changelog entry for `git pack-refs --auto`\n docs: address typos in Git v2.45 changelog\nPeter Hutterer (1):\n diff: add diff.srcPrefix and diff.dstPrefix configuration variables\nPeter Krefting (2):\n bisect: report the found commit with \"show\"\n l10n: sv.po: Update Swedish translation\nPhilippe Blain (5):\n merge-ort: turn submodule conflict suggestions into an advice\n ci(github): make Windows test artifacts name unique\n sequencer: allow disabling conflict advice\n builtin/am: allow disabling conflict advice\n t/README: mention test files are make targets\nPhillip Wood (9):\n rebase -i: stop setting GIT_CHERRY_PICK_HELP\n xdiff-interface: refactor parsing of merge.conflictstyle\n merge-ll: introduce LL_MERGE_OPTIONS_INIT\n merge options: add a conflict style member\n checkout: cleanup --conflict=<style> parsing\n checkout: fix interaction between --conflict and --merge\n t3428: modernize test setup\n t3428: use test_commit_message\n t3428: restore coverage for \"apply\" backend\nPi Fisher (1):\n typo: replace 'commitish' with 'committish'\nRalf Thielow (1):\n l10n: Update German translation\nRalph Seichter (1):\n config: add --comment option to add a comment\nRené Scharfe (32):\n use xstrncmpz()\n fetch: convert strncmp() with strlen() to starts_with()\n mem-pool: add mem_pool_strfmt()\n name-rev: use mem_pool_strfmt()\n submodule: use strvec_pushf() for --submodule-prefix\n t-ctype: allow NUL anywhere in the specification string\n t-ctype: simplify EOF check\n t-ctype: align output of i\n t-ctype: avoid duplicating class names\n parse-options: recognize abbreviated negated option with arg\n parse-options: set arg of abbreviated option lazily\n parse-options: factor out register_abbrev() and struct parsed_option\n parse-options: detect ambiguous self-negation\n parse-options: normalize arg and long_name before comparison\n parse-options: rearrange long_name matching code\n t-prio-queue: shorten array index message\n t-prio-queue: check result array bounds\n factor out strbuf_expand_bad_format()\n cat-file: use strbuf_expand_bad_format()\n midx: use strvec_pushf() for pack-objects base name\n mem-pool: use st_add() in mem_pool_strvfmt()\n imap-send: use xsnprintf to format command\n t-prio-queue: simplify using compound literals\n apply: avoid fixed-size buffer in create_one_file()\n path: remove mksnpath()\n apply: don't leak fd on fdopen() error\n usage: report vsnprintf(3) failure\n date: make DATE_MODE thread-safe\n git-compat-util: fix NO_OPENSSL on current macOS\n imap-send: increase command size limit\n apply: avoid using fixed-size buffer in write_out_one_reject()\n don't report vsnprintf(3) error as bug\nRichard Macklin (1):\n rebase: fix typo in autosquash documentation\nRubén Justo (18):\n tag: error when git-column fails\n completion: fix __git_complete_worktree_paths\n completion: reflog with implicit \"show\"\n completion: reflog show <log-options>\n completion: introduce __git_find_subcommand\n completion: factor out __git_resolve_builtins\n completion: reflog subcommands and options\n checkout: plug some leaks in git-restore\n add-patch: introduce 'p' in interactive-patch\n add-patch: do not print hunks repeatedly\n add: use advise_if_enabled for ADVICE_ADD_IGNORED_FILE\n add: use advise_if_enabled for ADVICE_ADD_EMPTY_PATHSPEC\n add: use advise_if_enabled for ADVICE_ADD_EMBEDDED_REPO\n launch_editor: waiting message on error\n apply: plug a leak in apply_data\n add-interactive: plug a leak in get_untracked_files\n add-patch: plug a leak handling the '/' command\n add: plug a leak on interactive_add\nSZEDER Gábor (1):\n upload-pack: don't send null character in abort message to the client\nSergey Organov (1):\n clean: improve -n and -f implementation and documentation\nSteven Jeuris (1):\n userdiff: better method/property matching for C#\nTaylor Blau (9):\n Documentation/config/pack.txt: fix broken AsciiDoc mark-up\n upload-pack: disallow object-info capability by default\n midx-write: move writing-related functions from midx.c\n midx-write.c: factor out common want_included_pack() routine\n midx-write.c: check count of packs to repack after grouping\n midx-write.c: use `--stdin-packs` when repacking\n t/t7700-repack.sh: fix test breakages with `GIT_TEST_MULTI_PACK_INDEX=1 `\n Makefile(s): avoid recipe prefix in conditional statements\n Documentation/RelNotes/2.45.0.txt: fix typo\nTeng Long (1):\n l10n: zh_CN: for git 2.45 rounds\nThalia Archibald (8):\n fast-import: tighten path unquoting\n fast-import: directly use strbufs for paths\n fast-import: allow unquoted empty path for root\n fast-import: remove dead strbuf\n fast-import: improve documentation for path quoting\n fast-import: document C-style escapes for paths\n fast-import: forbid escaped NUL in paths\n fast-import: make comments more precise\nVille Skyttä (2):\n completion: fix prompt with unset SHOWCONFLICTSTATE in nounset mode\n completion: protect prompt against unset SHOWUPSTREAM in nounset mode\nVincenzo Mezzela (1):\n t7301: use test_path_is_(missing|file)\nVũ Tiến Hưng (2):\n l10n: Update Vietnamese team contact\n l10n: vi: Updated translation for 2.45\nXing Xin (1):\n Documentation: fix typos describing date format\nYehezkel Bernat (1):\n Documentation: fix linkgit reference\nYi-Jyun Pan (1):\n l10n: zh-TW: Git 2.45\nbrian m. carlson (7):\n loose: add a mapping between SHA-1 and SHA-256 for loose objects\n commit: write commits for both hashes\n cache: add a function to read an OID of a specific algorithm\n object-file-convert: add a function to convert trees between algorithms\n object-file-convert: convert tag objects when writing\n object-file-convert: convert commit objects when writing\n repository: implement extensions.compatObjectFormat\nshejialuo (1):\n t9117: prefer test_path_* helper functions\nĐoàn Trần Công Danh (1):\n t9604: Fix test for musl libc and new Debian\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#me0af79d15d5071ef42f089f0ea4b9ced4f206d45\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#re0af79d15d5071ef42f089f0ea4b9ced4f206d45\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#eb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">*</a> <b>RE: [ANNOUNCE] Git v2.45.0</b>\n 2024-04-29 17:12 <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#me0af79d15d5071ef42f089f0ea4b9ced4f206d45\">[ANNOUNCE] Git v2.45.0</a> Junio C Hamano\n<b>@ 2024-05-01 17:39 ` rsbecker</b>\n 2024-05-02 5:56 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">t4216-log-bloom.sh broken ?</a> Torsten Bögershausen\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">1 sibling, 0 replies; 12+ messages in thread</a>\nFrom: rsbecker @ 2024-05-01 17:39 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: 'Junio C Hamano', <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240501174008\">git</a>; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/lkml/?t=20240501174008\">'Linux Kernel'</a>, git-packagers\nOn Monday, April 29, 2024 1:12 PM, Junio C Hamano wrote:\n<span>>The latest feature release Git v2.45.0 is now available at the\n>usual places. It is comprised of 540 non-merge commits since\n>v2.44.0, contributed by 96 people, 38 of which are new faces [*].\n>\n>The tarballs are found at:\n>\n> <a target=\"_blank\" href=\"https://www.kernel.org/pub/software/scm/git/\">https://www.kernel.org/pub/software/scm/git/</a>\n>\n>The following public repositories all have a copy of the 'v2.45.0'\n>tag and the 'master' branch that the tag points at:\n>\n> url = <a target=\"_blank\" href=\"https://git.kernel.org/pub/scm/git/git\">https://git.kernel.org/pub/scm/git/git</a>\n> url = <a target=\"_blank\" href=\"https://kernel.googlesource.com/pub/scm/git/git\">https://kernel.googlesource.com/pub/scm/git/git</a>\n> url = git://repo.or.cz/alt-git.git\n> url = <a target=\"_blank\" href=\"https://github.com/gitster/git\">https://github.com/gitster/git</a>\n</span>\nThe HPE NonStop builds for git 2.45.0 are available at the ITUGLIB website at <a target=\"_blank\" href=\"https://ituglib.connect-community.org/apps/Ituglib/SrchOpenSrcLib.xhtml\">https://ituglib.connect-community.org/apps/Ituglib/SrchOpenSrcLib.xhtml</a>.\nThanks to the entire git and ITUGLIB teams for making these happen.\nRegards,\nRandall\n--\nBrief whoami: NonStop&UNIX developer since approximately\nUNIX(421664400)\nNonStop(211288444200000000)\n-- In real life, I talk too much.\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e9413de4a25e0a9b9018ea1dbac13cb390d66d152\">*</a> <b>t4216-log-bloom.sh broken ?</b>\n 2024-04-29 17:12 <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#me0af79d15d5071ef42f089f0ea4b9ced4f206d45\">[ANNOUNCE] Git v2.45.0</a> Junio C Hamano\n 2024-05-01 17:39 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">rsbecker</a>\n<b>@ 2024-05-02 5:56 ` Torsten Bögershausen</b>\n 2024-05-02 16:06 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9347d051b559e6d8fe6d2db94432f5828f2be60a\">Junio C Hamano</a>\n 2024-05-02 16:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r9413de4a25e0a9b9018ea1dbac13cb390d66d152\">1 sibling, 2 replies; 12+ messages in thread</a>\nFrom: Torsten Bögershausen @ 2024-05-02 5:56 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/raw\">raw</a>)\n To: Junio C Hamano; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502055628\">git</a>, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/lkml/?t=20240502055628\">Linux Kernel</a>, git-packagers\n[]\nHighjacking the Git v2.45.0 announcement:\nThere are 4 test cases in t4216-log-bloom.sh, that do not pass on one\nMac here (they pass on another machine)\nI haven't digged further, but here are some log files,\nhopefully helpful to somebody.\nnot ok 141 - Bloom reader notices too-small data chunk\n#\n# check_corrupt_graph BDAT clear 00000000 &&\n# echo \"warning: ignoring too-small changed-path chunk\" \\\n# \"(4 < 12) in commit-graph file\" >expect.err &&\n# test_cmp expect.err err\n#\nnot ok 142 - Bloom reader notices out-of-bounds filter offsets\n#\n# check_corrupt_graph BIDX 12 FFFFFFFF &&\n# # use grep to avoid depending on exact chunk size\n# grep \"warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph\" err\n#\nnot ok 143 - Bloom reader notices too-small index chunk\n#\n# # replace the index with a single entry, making most\n# # lookups out-of-bounds\n# check_corrupt_graph BIDX clear 00000000 &&\n# echo \"warning: commit-graph changed-path index chunk\" \\\n# \"is too small\" >expect.err &&\n# test_cmp expect.err err\n#\nnot ok 144 - Bloom reader notices out-of-order index offsets\n#\n# # we do not know any real offsets, but we can pick\n# # something plausible; we should not get to the point of\n# # actually reading from the bogus offsets anyway.\n# corrupt_graph BIDX 4 0000000c00000005 &&\n# echo \"warning: ignoring decreasing changed-path index offsets\" \\\n# \"(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph\" >expect.err &&\n# git -c core.commitGraph=false log -- A/B/file2 >expect.out &&\n# git -c core.commitGraph=true log -- A/B/file2 >out 2>err &&\n# test_cmp expect.out out &&\n# test_cmp expect.err err\n#\n# failed 4 among 144 test(s)\n1..144\nok 140 - Bloom generation backfills empty commits\nexpecting success of 4216.141 'Bloom reader notices too-small data chunk':\n\tcheck_corrupt_graph BDAT clear 00000000 &&\n\techo \"warning: ignoring too-small changed-path chunk\" \\\n\t\t\"(4 < 12) in commit-graph file\" >expect.err &&\n\ttest_cmp expect.err err\n++ check_corrupt_graph BDAT clear 00000000\n++ corrupt_graph BDAT clear 00000000\n++ graph=.git/objects/info/commit-graph\n++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n++ test 0 = 0\n++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n\t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n++ git commit-graph write --reachable --changed-paths\n++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000\n++ fn=.git/objects/info/commit-graph\n++ shift\n++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\noverride r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n++ git -c core.commitGraph=false log -- A/B/file2\n++ git -c core.commitGraph=true log -- A/B/file2\n++ test_cmp expect.out out\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.out out\n++ echo 'warning: ignoring too-small changed-path chunk' '(4 < 12) in commit-graph file'\n++ test_cmp expect.err err\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.err err\n--- expect.err\t2024-05-02 05:51:10.000000000 +0000\n+++ err\t2024-05-02 05:51:10.000000000 +0000\n@@ -1 +0,0 @@\n-warning: ignoring too-small changed-path chunk (4 < 12) in commit-graph file\nerror: last command exited with $?=1\n++ rm -rf .git/objects/info/commit-graph\n++ exit 1\n++ eval_ret=1\n++ :\nnot ok 141 - Bloom reader notices too-small data chunk\nexpecting success of 4216.142 'Bloom reader notices out-of-bounds filter offsets':\n\tcheck_corrupt_graph BIDX 12 FFFFFFFF &&\n\t# use grep to avoid depending on exact chunk size\n\tgrep \"warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph\" err\n++ check_corrupt_graph BIDX 12 FFFFFFFF\n++ corrupt_graph BIDX 12 FFFFFFFF\n++ graph=.git/objects/info/commit-graph\n++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n++ test 0 = 0\n++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n\t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n++ git commit-graph write --reachable --changed-paths\n++ corrupt_chunk_file .git/objects/info/commit-graph BIDX 12 FFFFFFFF\n++ fn=.git/objects/info/commit-graph\n++ shift\n++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF\n++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF\n++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 12 FFFFFFFF\n++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\noverride r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n++ git -c core.commitGraph=false log -- A/B/file2\n++ git -c core.commitGraph=true log -- A/B/file2\n++ test_cmp expect.out out\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.out out\n++ grep 'warning: ignoring out-of-range offset (4294967295) for changed-path filter at pos 3 of .git/objects/info/commit-graph' err\nerror: last command exited with $?=1\n++ rm -rf .git/objects/info/commit-graph\n++ exit 1\n++ eval_ret=1\n++ :\nnot ok 142 - Bloom reader notices out-of-bounds filter offsets\nexpecting success of 4216.143 'Bloom reader notices too-small index chunk':\n\t# replace the index with a single entry, making most\n\t# lookups out-of-bounds\n\tcheck_corrupt_graph BIDX clear 00000000 &&\n\techo \"warning: commit-graph changed-path index chunk\" \\\n\t\t\"is too small\" >expect.err &&\n\ttest_cmp expect.err err\n++ check_corrupt_graph BIDX clear 00000000\n++ corrupt_graph BIDX clear 00000000\n++ graph=.git/objects/info/commit-graph\n++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n++ test 0 = 0\n++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n\t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n++ git commit-graph write --reachable --changed-paths\n++ corrupt_chunk_file .git/objects/info/commit-graph BIDX clear 00000000\n++ fn=.git/objects/info/commit-graph\n++ shift\n++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000\n++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000\n++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX clear 00000000\n++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\noverride r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n++ git -c core.commitGraph=false log -- A/B/file2\n++ git -c core.commitGraph=true log -- A/B/file2\n++ test_cmp expect.out out\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.out out\n++ echo 'warning: commit-graph changed-path index chunk' 'is too small'\n++ test_cmp expect.err err\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.err err\n--- expect.err\t2024-05-02 05:51:11.000000000 +0000\n+++ err\t2024-05-02 05:51:11.000000000 +0000\n@@ -1 +0,0 @@\n-warning: commit-graph changed-path index chunk is too small\nerror: last command exited with $?=1\n++ rm -rf .git/objects/info/commit-graph\n++ exit 1\n++ eval_ret=1\n++ :\nnot ok 143 - Bloom reader notices too-small index chunk\nexpecting success of 4216.144 'Bloom reader notices out-of-order index offsets':\n\t# we do not know any real offsets, but we can pick\n\t# something plausible; we should not get to the point of\n\t# actually reading from the bogus offsets anyway.\n\tcorrupt_graph BIDX 4 0000000c00000005 &&\n\techo \"warning: ignoring decreasing changed-path index offsets\" \\\n\t\t\"(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph\" >expect.err &&\n\tgit -c core.commitGraph=false log -- A/B/file2 >expect.out &&\n\tgit -c core.commitGraph=true log -- A/B/file2 >out 2>err &&\n\ttest_cmp expect.out out &&\n\ttest_cmp expect.err err\n++ corrupt_graph BIDX 4 0000000c00000005\n++ graph=.git/objects/info/commit-graph\n++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n++ test 0 = 0\n++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n\t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n++ git commit-graph write --reachable --changed-paths\n++ corrupt_chunk_file .git/objects/info/commit-graph BIDX 4 0000000c00000005\n++ fn=.git/objects/info/commit-graph\n++ shift\n++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005\n++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005\n++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BIDX 4 0000000c00000005\n++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\noverride r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n++ echo 'warning: ignoring decreasing changed-path index offsets' '(12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph'\n++ git -c core.commitGraph=false log -- A/B/file2\n++ git -c core.commitGraph=true log -- A/B/file2\n++ test_cmp expect.out out\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.out out\n++ test_cmp expect.err err\n++ test 2 -ne 2\n++ eval 'diff -u' '\"$@\"'\n+++ diff -u expect.err err\n--- expect.err\t2024-05-02 05:51:11.000000000 +0000\n+++ err\t2024-05-02 05:51:11.000000000 +0000\n@@ -1 +0,0 @@\n-warning: ignoring decreasing changed-path index offsets (12 > 5) for positions 1 and 2 of .git/objects/info/commit-graph\nerror: last command exited with $?=1\n++ rm -rf .git/objects/info/commit-graph\n++ exit 1\n++ eval_ret=1\n++ :\nnot ok 144 - Bloom reader notices out-of-order index offsets\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502055621.GA29945@tb-raspi4/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r9413de4a25e0a9b9018ea1dbac13cb390d66d152\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e9347d051b559e6d8fe6d2db94432f5828f2be60a\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 5:56 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">t4216-log-bloom.sh broken ?</a> Torsten Bögershausen\n<b>@ 2024-05-02 16:06 ` Junio C Hamano</b>\n 2024-05-02 18:59 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4d5da250c83d37cecf4709dd92741dee72eba18c\">Torsten Bögershausen</a>\n 2024-05-02 16:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r9347d051b559e6d8fe6d2db94432f5828f2be60a\">1 sibling, 1 reply; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 16:06 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: Torsten Bögershausen; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502160644\">git</a>\nTorsten Bögershausen <[email protected]> writes:\n<span>> []\n>\n> Highjacking the Git v2.45.0 announcement:\n</span>\nI am not sure why you want to do so. You are perfectly capable of\nsending a new message to the list (and I know you have done so in\nthe past). Why would linux-kernel@ list want to hear about the\nmacOS issues?\n<span>> There are 4 test cases in t4216-log-bloom.sh, that do not pass on one\n> Mac here (they pass on another machine)\n</span>\nAnother machine being another mac?\n<span>> expecting success of 4216.141 'Bloom reader notices too-small data chunk':\n> \tcheck_corrupt_graph BDAT clear 00000000 &&\n> \techo \"warning: ignoring too-small changed-path chunk\" \\\n> \t\t\"(4 < 12) in commit-graph file\" >expect.err &&\n> \ttest_cmp expect.err err\n>\n> ++ check_corrupt_graph BDAT clear 00000000\n> ++ corrupt_graph BDAT clear 00000000\n> ++ graph=.git/objects/info/commit-graph\n> ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n> ++ test 0 = 0\n> ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n> \t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n> ++ git commit-graph write --reachable --changed-paths\n> ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000\n> ++ fn=.git/objects/info/commit-graph\n> ++ shift\n> ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\n> override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n</span>\nIs this failure preventing the later steps of the test work as\nexpected, I wonder. Is there something curious with the permission\nbits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/\nsubdirectory? Is there something curious with the \"umask\" of the\nuser running the test? Are they different from what you see on your\nother mac that does not exhibit the problems?\nThanks.\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9347d051b559e6d8fe6d2db94432f5828f2be60a\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r9347d051b559e6d8fe6d2db94432f5828f2be60a\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#efcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 5:56 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">t4216-log-bloom.sh broken ?</a> Torsten Bögershausen\n 2024-05-02 16:06 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9347d051b559e6d8fe6d2db94432f5828f2be60a\">Junio C Hamano</a>\n<b>@ 2024-05-02 16:26 ` Junio C Hamano</b>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">1 sibling, 0 replies; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 16:26 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: Torsten Bögershausen; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502162648\">git</a>\nTorsten Bögershausen <[email protected]> writes:\n<span>> ++ check_corrupt_graph BDAT clear 00000000\n> ++ corrupt_graph BDAT clear 00000000\n> ++ graph=.git/objects/info/commit-graph\n> ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n> ++ test 0 = 0\n> ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n> \t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n> ++ git commit-graph write --reachable --changed-paths\n> ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000\n> ++ fn=.git/objects/info/commit-graph\n> ++ shift\n> ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\n> override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n</span>\nAnother probing question. Do t5318, t5319, t5324, and t5328 pass?\nThe \"mv\" is part of the corrupt_chunk_file helper is used by this:\n corrupt_graph () {\n graph=.git/objects/info/commit-graph &&\n test_when_finished \"rm -rf $graph\" &&\n git commit-graph write --reachable --changed-paths &&\n corrupt_chunk_file $graph \"$@\"\n }\nand reads like this:\n corrupt_chunk_file () {\n fn=$1; shift\n perl \"$TEST_DIRECTORY\"/lib-chunk/corrupt-chunk-file.pl \\\n \"$@\" <\"$fn\" >\"$fn.tmp\" &&\n mv \"$fn.tmp\" \"$fn\"\n }\nIf the final \"mv\" is what is failing in the above trace, the test\ntried to corrupt the objects/info/commit-graph file, expecting that\nlater steps will notice breakage. But if the broken file written in\ncommit-graph.tmp failed to become the final commit-graph, later\nsteps will see healthy graph file and it is understandable that we\ndo not see any breakage that we expect, thus failing the test.\nIs \"mv\" somehow configured to always go interactive on that box and\non that box alone? For example if you have ~/bin/mv that does\nsomething silly like\n\t#!/bin/bash\n\texec /bin/mv -i \"$@\nand have ~/bin early in $PATH, it may be sufficient to explain the\nerror we see (I am not claiming that is the only way to cause this\ntest failure; it is just one possible one).\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e4d5da250c83d37cecf4709dd92741dee72eba18c\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 16:06 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9347d051b559e6d8fe6d2db94432f5828f2be60a\">Junio C Hamano</a>\n<b>@ 2024-05-02 18:59 ` Torsten Bögershausen</b>\n 2024-05-02 19:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4af68eff8b6328b24dbc99e697d6bf854aa02166\">SZEDER Gábor</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r4d5da250c83d37cecf4709dd92741dee72eba18c\">0 siblings, 1 reply; 12+ messages in thread</a>\nFrom: Torsten Bögershausen @ 2024-05-02 18:59 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/raw\">raw</a>)\n To: Junio C Hamano; <b>+Cc:</b> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502185915\">git</a>\nOn Thu, May 02, 2024 at 09:06:41AM -0700, Junio C Hamano wrote:\n<span>> Torsten Bögershausen <[email protected]> writes:\n>\n> > There are 4 test cases in t4216-log-bloom.sh, that do not pass on one\n> > Mac here (they pass on another machine)\n>\n> Another machine being another mac?\n</span>\nYes, different mac, different MacOs, different $PATH probably.\n<span>>\n> > expecting success of 4216.141 'Bloom reader notices too-small data chunk':\n> > \tcheck_corrupt_graph BDAT clear 00000000 &&\n> > \techo \"warning: ignoring too-small changed-path chunk\" \\\n> > \t\t\"(4 < 12) in commit-graph file\" >expect.err &&\n> > \ttest_cmp expect.err err\n> >\n> > ++ check_corrupt_graph BDAT clear 00000000\n> > ++ corrupt_graph BDAT clear 00000000\n> > ++ graph=.git/objects/info/commit-graph\n> > ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n> > ++ test 0 = 0\n> > ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n> > \t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n> > ++ git commit-graph write --reachable --changed-paths\n> > ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000\n> > ++ fn=.git/objects/info/commit-graph\n> > ++ shift\n> > ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\n> > override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n>\n> Is this failure preventing the later steps of the test work as\n> expected, I wonder. Is there something curious with the permission\n> bits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/\n> subdirectory? Is there something curious with the \"umask\" of the\n> user running the test? Are they different from what you see on your\n> other mac that does not exhibit the problems?\n>\n> Thanks.\n>\n>\n</span>\nmv is /bin/mv, that seems to be good:\n$ type mv\nmv is /bin/mv\n$ alias | grep mv\n$ which mv\n/bin/mv\n$ umask\n0022\nI don't know, why we see\nr--r--r-- tb/staff for .git/objects/info/commit-graph\nBut, the \"-r--r--r--\" may be part of the problem, here is another one:\n$ find . -name commit-graph -print0 | xargs -0 ls -l\n-r--r--r-- 1 tb staff 1792 May 2 12:12 ./trash directory.t5318-commit-graph/bare/objects/info/commit-graph\n(And some more 6 in total. All with -r--r--r--)\nWhich means, yes, t5318 does not pass either:\nt5318-commit-graph.sh not ok 101 - reader notices too-small oid fanout chunk\nt5318-commit-graph.sh not ok 102 - reader notices fanout/lookup table mismatch\nt5318-commit-graph.sh not ok 103 - reader notices out-of-bounds fanout\nt5318-commit-graph.sh not ok 104 - reader notices too-small commit data chunk\nt5318-commit-graph.sh not ok 105 - reader notices out-of-bounds extra edge\nt5318-commit-graph.sh not ok 106 - reader notices too-small generations chunk\nSame problem here:\n++ mv full/.git/objects/info/commit-graph.tmp full/.git/objects/info/commit-graph\noverride r--r--r-- tb/staff for full/.git/objects/info/commit-graph? (y/n [n]) not overwritten\nThe rest of the test suite passes, I see the same failures even under a\nfresh cloned repo.\nAny hints are appreciated.\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4d5da250c83d37cecf4709dd92741dee72eba18c\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502185903.GA11895@tb-raspi4/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r4d5da250c83d37cecf4709dd92741dee72eba18c\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e4af68eff8b6328b24dbc99e697d6bf854aa02166\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 18:59 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4d5da250c83d37cecf4709dd92741dee72eba18c\">Torsten Bögershausen</a>\n<b>@ 2024-05-02 19:26 ` SZEDER Gábor</b>\n 2024-05-02 19:55 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m05c9a5a4f71490b5234ecf285f32ad262ced1acb\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r4af68eff8b6328b24dbc99e697d6bf854aa02166\">0 siblings, 1 reply; 12+ messages in thread</a>\nFrom: SZEDER Gábor @ 2024-05-02 19:26 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: Torsten Bögershausen; <b>+Cc:</b> Junio C Hamano, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502192625\">git</a>\nOn Thu, May 02, 2024 at 08:59:03PM +0200, Torsten Bögershausen wrote:\n<span>> On Thu, May 02, 2024 at 09:06:41AM -0700, Junio C Hamano wrote:\n> > Torsten Bögershausen <[email protected]> writes:\n> >\n> > > There are 4 test cases in t4216-log-bloom.sh, that do not pass on one\n> > > Mac here (they pass on another machine)\n> >\n> > Another machine being another mac?\n> \n> Yes, different mac, different MacOs, different $PATH probably.\n> \n> >\n> > > expecting success of 4216.141 'Bloom reader notices too-small data chunk':\n> > > \tcheck_corrupt_graph BDAT clear 00000000 &&\n> > > \techo \"warning: ignoring too-small changed-path chunk\" \\\n> > > \t\t\"(4 < 12) in commit-graph file\" >expect.err &&\n> > > \ttest_cmp expect.err err\n> > >\n> > > ++ check_corrupt_graph BDAT clear 00000000\n> > > ++ corrupt_graph BDAT clear 00000000\n> > > ++ graph=.git/objects/info/commit-graph\n> > > ++ test_when_finished 'rm -rf .git/objects/info/commit-graph'\n> > > ++ test 0 = 0\n> > > ++ test_cleanup='{ rm -rf .git/objects/info/commit-graph\n> > > \t\t} && (exit \"$eval_ret\"); eval_ret=$?; :'\n> > > ++ git commit-graph write --reachable --changed-paths\n> > > ++ corrupt_chunk_file .git/objects/info/commit-graph BDAT clear 00000000\n> > > ++ fn=.git/objects/info/commit-graph\n> > > ++ shift\n> > > ++ perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > > ++ command /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > > ++ /usr/bin/perl /Users/tb/NoBackup/projects/git/git.pu/t/lib-chunk/corrupt-chunk-file.pl BDAT clear 00000000\n> > > ++ mv .git/objects/info/commit-graph.tmp .git/objects/info/commit-graph\n> > > override r--r--r-- tb/staff for .git/objects/info/commit-graph? (y/n [n]) not overwritten\n> >\n> > Is this failure preventing the later steps of the test work as\n> > expected, I wonder. Is there something curious with the permission\n> > bits of /Users/tb/NoBackup/projects/git/git.pu/ directory or its t/\n> > subdirectory? Is there something curious with the \"umask\" of the\n> > user running the test? Are they different from what you see on your\n> > other mac that does not exhibit the problems?\n> >\n> > Thanks.\n> >\n> >\n> \n> mv is /bin/mv, that seems to be good:\n> \n> $ type mv\n> mv is /bin/mv\n> \n> $ alias | grep mv\n> \n> $ which mv\n> /bin/mv\n> \n> $ umask\n> 0022\n> \n> I don't know, why we see\n> r--r--r-- tb/staff for .git/objects/info/commit-graph\n> \n> But, the \"-r--r--r--\" may be part of the problem, here is another one:\n> $ find . -name commit-graph -print0 | xargs -0 ls -l\n> -r--r--r-- 1 tb staff 1792 May 2 12:12 ./trash directory.t5318-commit-graph/bare/objects/info/commit-graph\n> (And some more 6 in total. All with -r--r--r--)\n> \n> Which means, yes, t5318 does not pass either:\n> t5318-commit-graph.sh not ok 101 - reader notices too-small oid fanout chunk\n> t5318-commit-graph.sh not ok 102 - reader notices fanout/lookup table mismatch\n> t5318-commit-graph.sh not ok 103 - reader notices out-of-bounds fanout\n> t5318-commit-graph.sh not ok 104 - reader notices too-small commit data chunk\n> t5318-commit-graph.sh not ok 105 - reader notices out-of-bounds extra edge\n> t5318-commit-graph.sh not ok 106 - reader notices too-small generations chunk\n> \n> Same problem here:\n> ++ mv full/.git/objects/info/commit-graph.tmp full/.git/objects/info/commit-graph\n> override r--r--r-- tb/staff for full/.git/objects/info/commit-graph? (y/n [n]) not overwritten\n</span>\n'mv' in macOS doesn't conform to POSIX, and asks for confirmation when\nthe destination exists (is read-only?) even without '-i' and even when\nits stdin is not a terminal, which it won't get as its stdin in our\ntest suite is redirected from /dev/null.\nThis is a recurring issue, see e.g.:\n <a target=\"_blank\" href=\"https://public-inbox.org/git/[email protected]/\">https://public-inbox.org/git/[email protected]/</a>\n c20d4d702f (t1450: use \"mv -f\" within loose object directory, 2017-01-24)\n'mv -f' did the trick in the past.\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4af68eff8b6328b24dbc99e697d6bf854aa02166\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r4af68eff8b6328b24dbc99e697d6bf854aa02166\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e05c9a5a4f71490b5234ecf285f32ad262ced1acb\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 19:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4af68eff8b6328b24dbc99e697d6bf854aa02166\">SZEDER Gábor</a>\n<b>@ 2024-05-02 19:55 ` Junio C Hamano</b>\n 2024-05-02 20:05 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">Junio C Hamano</a>\n 2024-05-02 20:15 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#md36a80faa93acbc8bc00ed855c37b75c72fba0a6\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r05c9a5a4f71490b5234ecf285f32ad262ced1acb\">0 siblings, 2 replies; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 19:55 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: SZEDER Gábor; <b>+Cc:</b> Torsten Bögershausen, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502195556\">git</a>\nSZEDER Gábor <[email protected]> writes:\n<span>>> Yes, different mac, different MacOs, different $PATH probably.\n>> ...\n>\n> 'mv' in macOS doesn't conform to POSIX, and asks for confirmation when\n> the destination exists (is read-only?) even without '-i' and even when\n> its stdin is not a terminal, which it won't get as its stdin in our\n> test suite is redirected from /dev/null.\n>\n> This is a recurring issue, see e.g.:\n>\n> <a target=\"_blank\" href=\"https://public-inbox.org/git/[email protected]/\">https://public-inbox.org/git/[email protected]/</a>\n> c20d4d702f (t1450: use \"mv -f\" within loose object directory, 2017-01-24)\n>\n> 'mv -f' did the trick in the past.\n</span>\nHmph, while that is a good data point to explain why this one fails,\nit is a very unsatisfactory workaround, compared to a world where we\ndo not have to worry about such a broken mv (perhaps by noticing a\nbroken macOS /bin/mv and automatically doing mv () { mv -f \"$@\" }\nfor them).\nI am curious what other differences Torsten will find out between\ngood macs and bad ones. Perhaps we can narrow down the bad apples?\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m05c9a5a4f71490b5234ecf285f32ad262ced1acb\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r05c9a5a4f71490b5234ecf285f32ad262ced1acb\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#ec14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 19:55 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m05c9a5a4f71490b5234ecf285f32ad262ced1acb\">Junio C Hamano</a>\n<b>@ 2024-05-02 20:05 ` Junio C Hamano</b>\n 2024-05-02 20:15 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#md36a80faa93acbc8bc00ed855c37b75c72fba0a6\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">1 sibling, 0 replies; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 20:05 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: SZEDER Gábor; <b>+Cc:</b> Torsten Bögershausen, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502200545\">git</a>\nJunio C Hamano <[email protected]> writes:\n<span>> ... it is a very unsatisfactory workaround, compared to a world where we\n> do not have to worry about such a broken mv (perhaps by noticing a\n> broken macOS /bin/mv and automatically doing mv () { mv -f \"$@\" }\n> for them).\n>\n> I am curious what other differences Torsten will find out between\n> good macs and bad ones. Perhaps we can narrow down the bad apples?\n</span>\nIn any case, while we are waiting, I did a few \"grep\":\n $ git grep 'mv \\(.*\\)\\.tmp \\1' t\n t/lib-t6000.sh:\tmv sed.script.tmp sed.script\n t/t7508-status.sh:\trm \"$1\" && mv \"$1\".tmp \"$1\"\n t/t8011-blame-split-file.sh:\tmv one.tmp one &&\n t/t8011-blame-split-file.sh:\tmv two.tmp two &&\n t/t9400-git-cvsserver-server.sh:\tmv merge.tmp merge &&\n t/t9400-git-cvsserver-server.sh:\tmv merge.tmp merge &&\n t/t9802-git-p4-filetype.sh:\t\tmv empty-symlink,v.tmp empty-symlink,v\n $ git grep 'mv \"\\(.*\\)\\.tmp\" \"\\1\"' t\n t/lib-chunk.sh:\tmv \"$fn.tmp\" \"$fn\"\n t/t3404-rebase-interactive.sh:\tmv \"$1.tmp\" \"$1\"\n t/t5515-fetch-merge-logic.sh:\tmv \"$file.tmp\" \"$file\"\n t/t7600-merge.sh:) >\"$1.tmp\" && mv \"$1.tmp\" \"$1\"\n t/t9001-send-email.sh:\tmv \"$1.tmp\" \"$1\"\n t/t9001-send-email.sh:\tmv \"$1.tmp\" \"$1\"\n $ git grep 'mv \\(.*\\)+ \\1' t\n t/t4025-hunk-header.sh:\tmv file+ file\n t/t6402-merge-rename.sh:\tmv A+ A &&\n t/t6402-merge-rename.sh:\tmv A+ A &&\n t/t6432-merge-recursive-space-options.sh:\tmv text.txt+ text.txt &&\n t/t6432-merge-recursive-space-options.sh:\tmv text.txt+ text.txt &&\nThere is nothing other than the one we found in lib-chunk.sh that\nwork on files under .git/objects/ that are made read-only, so we\nshould be OK with a workaround like the attached.\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#Z2e.:..:xmqqwmocc73f.fsf::40gitster.g:1t:lib-chunk.sh\">t/lib-chunk.sh</a> | 3 ++-\n 1 file <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#ec14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">changed</a>, 2 insertions(+), 1 deletion(-)\n<span><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#iZ2e.:..:xmqqwmocc73f.fsf::40gitster.g:1t:lib-chunk.sh\">diff</a> --git c/t/lib-chunk.sh w/t/lib-chunk.sh\nindex a7cd9c3c6d..9f01df190b 100644\n--- c/t/lib-chunk.sh\n+++ w/t/lib-chunk.sh\n</span><span>@@ -13,5 +13,6 @@ corrupt_chunk_file () {\n</span> \tfn=$1; shift\n \tperl \"$TEST_DIRECTORY\"/lib-chunk/corrupt-chunk-file.pl \\\n \t\t\"$@\" <\"$fn\" >\"$fn.tmp\" &&\n<span>-\tmv \"$fn.tmp\" \"$fn\"\n</span><span>+\t# some vintages of macOS 'mv' fails to overwrite a read-only file.\n+\tmv -f \"$fn.tmp\" \"$fn\"\n</span> }\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#related\">related</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#ed36a80faa93acbc8bc00ed855c37b75c72fba0a6\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 19:55 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m05c9a5a4f71490b5234ecf285f32ad262ced1acb\">Junio C Hamano</a>\n 2024-05-02 20:05 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">Junio C Hamano</a>\n<b>@ 2024-05-02 20:15 ` Junio C Hamano</b>\n 2024-05-02 20:51 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m5f0b708412c9a092e342407cb60ce4ba0cf95962\">Torsten Bögershausen</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rd36a80faa93acbc8bc00ed855c37b75c72fba0a6\">1 sibling, 1 reply; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 20:15 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: SZEDER Gábor; <b>+Cc:</b> Torsten Bögershausen, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502201600\">git</a>\nJunio C Hamano <[email protected]> writes:\n<span>> I am curious what other differences Torsten will find out between\n> good macs and bad ones. Perhaps we can narrow down the bad apples?\n</span>\nSo, \"no, your 'mv' is broken\" seems to be the answer to the question\non the Subject line, and it is rather well-known, it seems.\n----- >8 --------- >8 --------- >8 --------- >8 -----\nSubject: [PATCH] t/lib-chunk: work around broken \"mv\" on some vintage of macOS\nWhen the destination is read-only, \"mv\" on some version of macOS\nasks whether to replace the destination even though in the test its\nstdin is not a terminal (and thus doesn't conform to POSIX[1]).\nThe helper to corrupt a chunk-file is designed to work on the\nfiles like commit-graph and multi-pack-index files that are\ngenerally read-only, so use \"mv -f\" to work around this issue.\nHelped-by: SZEDER Gábor <[email protected]>\nSigned-off-by: Junio C Hamano <[email protected]>\n---\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#Z2e.:..:xmqqsez0c6ma.fsf::40gitster.g:1t:lib-chunk.sh\">t/lib-chunk.sh</a> | 3 ++-\n 1 file <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#ed36a80faa93acbc8bc00ed855c37b75c72fba0a6\">changed</a>, 2 insertions(+), 1 deletion(-)\n<span><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#iZ2e.:..:xmqqsez0c6ma.fsf::40gitster.g:1t:lib-chunk.sh\">diff</a> --git a/t/lib-chunk.sh b/t/lib-chunk.sh\nindex a7cd9c3c6d..9f01df190b 100644\n--- a/t/lib-chunk.sh\n+++ b/t/lib-chunk.sh\n</span><span>@@ -13,5 +13,6 @@ corrupt_chunk_file () {\n</span> \tfn=$1; shift\n \tperl \"$TEST_DIRECTORY\"/lib-chunk/corrupt-chunk-file.pl \\\n \t\t\"$@\" <\"$fn\" >\"$fn.tmp\" &&\n<span>-\tmv \"$fn.tmp\" \"$fn\"\n</span><span>+\t# some vintages of macOS 'mv' fails to overwrite a read-only file.\n+\tmv -f \"$fn.tmp\" \"$fn\"\n</span> }\n-- \n2.45.0-31-gd4cc1ec35f\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#md36a80faa93acbc8bc00ed855c37b75c72fba0a6\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#related\">related</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#rd36a80faa93acbc8bc00ed855c37b75c72fba0a6\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e5f0b708412c9a092e342407cb60ce4ba0cf95962\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 20:15 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#md36a80faa93acbc8bc00ed855c37b75c72fba0a6\">Junio C Hamano</a>\n<b>@ 2024-05-02 20:51 ` Torsten Bögershausen</b>\n 2024-05-02 21:20 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m99d41c5069ccc95f69c227be88108d54fe7d04f9\">Junio C Hamano</a>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r5f0b708412c9a092e342407cb60ce4ba0cf95962\">0 siblings, 1 reply; 12+ messages in thread</a>\nFrom: Torsten Bögershausen @ 2024-05-02 20:51 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/raw\">raw</a>)\n To: Junio C Hamano; <b>+Cc:</b> SZEDER Gábor, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502205128\">git</a>\nOn Thu, May 02, 2024 at 01:15:57PM -0700, Junio C Hamano wrote:\n<span>> Junio C Hamano <[email protected]> writes:\n>\n> > I am curious what other differences Torsten will find out between\n> > good macs and bad ones. Perhaps we can narrow down the bad apples?\n>\n> So, \"no, your 'mv' is broken\" seems to be the answer to the question\n> on the Subject line, and it is rather well-known, it seems.\n</span>\n[]\n<span>> \t\t\"$@\" <\"$fn\" >\"$fn.tmp\" &&\n> -\tmv \"$fn.tmp\" \"$fn\"\n> +\t# some vintages of macOS 'mv' fails to overwrite a read-only file.\n> +\tmv -f \"$fn.tmp\" \"$fn\"\n</span>\nThanks folks for digging - the patch makes both test cases pass.\nSome official information:\n$ man mv\non the both machines says\n[snip]\nIf the destination path does not have a mode which permits writing,\nmv prompts the user for confirmation as specified for the -i option.\nTo summarize what I have at hand right now:\nMacOs 11.7.10 fails t4216 and t5318, both pass under MacOs 13.6.6\nBut why doesn't `mv` not hang then ?\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m5f0b708412c9a092e342407cb60ce4ba0cf95962\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/20240502205121.GB11895@tb-raspi4/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r5f0b708412c9a092e342407cb60ce4ba0cf95962\">12+ messages in thread</a></pre><hr /><pre><a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#e99d41c5069ccc95f69c227be88108d54fe7d04f9\">*</a> <b>Re: t4216-log-bloom.sh broken ?</b>\n 2024-05-02 20:51 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m5f0b708412c9a092e342407cb60ce4ba0cf95962\">Torsten Bögershausen</a>\n<b>@ 2024-05-02 21:20 ` Junio C Hamano</b>\n <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r99d41c5069ccc95f69c227be88108d54fe7d04f9\">0 siblings, 0 replies; 12+ messages in thread</a>\nFrom: Junio C Hamano @ 2024-05-02 21:20 UTC (<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> / <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a>)\n To: Torsten Bögershausen; <b>+Cc:</b> SZEDER Gábor, <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/git/?t=20240502212106\">git</a>\nTorsten Bögershausen <[email protected]> writes:\n<span>> Thanks folks for digging - the patch makes both test cases pass.\n> Some official information:\n>\n> $ man mv\n>\n> on the both machines says\n> [snip]\n> If the destination path does not have a mode which permits writing,\n> mv prompts the user for confirmation as specified for the -i option.\n>\n> To summarize what I have at hand right now:\n> MacOs 11.7.10 fails t4216 and t5318, both pass under MacOs 13.6.6\n>\n> But why doesn't `mv` not hang then ?\n</span>\nIt is a sensible choice for the command to assume a \"safer\" no and\ncontinuing rather than hanging, so that part is understandable.\nSo presumably 13.6.6 has a working \"mv\" but their \"man mv\" is not\nupdated to how the command actually works?\nIn any case, let me declare victory and queue the patch I sent\nearlier.\nThanks.\n<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m99d41c5069ccc95f69c227be88108d54fe7d04f9\">^</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/\">permalink</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/raw\">raw</a> <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/#R\">reply</a>\t[<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#u\"><b>flat</b></a>|<a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t/#u\">nested</a>] <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#r99d41c5069ccc95f69c227be88108d54fe7d04f9\">12+ messages in thread</a></pre><hr /><pre>end of thread, other threads:[<a target=\"_blank\" href=\"https://lore.kernel.org/git/?t=20240502212106\">~2024-05-02 21:21 UTC</a> | <a target=\"_blank\" href=\"https://lore.kernel.org/git/\">newest</a>]\n<b>Thread overview:</b> 12+ messages (download: <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t.mbox.gz\">mbox.gz</a> follow: <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/t.atom\">Atom feed</a>\n-- links below jump to the message on this page --\n2024-04-29 17:12 <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#me0af79d15d5071ef42f089f0ea4b9ced4f206d45\">[ANNOUNCE] Git v2.45.0</a> Junio C Hamano\n2024-05-01 17:39 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mb7fb7b6ad6c9c7896dc1bf465fe2d9dd6395d19b\">rsbecker</a>\n2024-05-02 5:56 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9413de4a25e0a9b9018ea1dbac13cb390d66d152\">t4216-log-bloom.sh broken ?</a> Torsten Bögershausen\n2024-05-02 16:06 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m9347d051b559e6d8fe6d2db94432f5828f2be60a\">Junio C Hamano</a>\n2024-05-02 18:59 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4d5da250c83d37cecf4709dd92741dee72eba18c\">Torsten Bögershausen</a>\n2024-05-02 19:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m4af68eff8b6328b24dbc99e697d6bf854aa02166\">SZEDER Gábor</a>\n2024-05-02 19:55 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m05c9a5a4f71490b5234ecf285f32ad262ced1acb\">Junio C Hamano</a>\n2024-05-02 20:05 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mc14b0d7f0e4c0e50f13820ac1e13bf66a53fde6f\">Junio C Hamano</a>\n2024-05-02 20:15 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#md36a80faa93acbc8bc00ed855c37b75c72fba0a6\">Junio C Hamano</a>\n2024-05-02 20:51 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m5f0b708412c9a092e342407cb60ce4ba0cf95962\">Torsten Bögershausen</a>\n2024-05-02 21:20 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#m99d41c5069ccc95f69c227be88108d54fe7d04f9\">Junio C Hamano</a>\n2024-05-02 16:26 ` <a target=\"_blank\" href=\"https://lore.kernel.org/git/[email protected]/T/#mfcda7704ffa747f47ae7eae06dd4d0ebd8b4d397\">Junio C Hamano</a>\n</pre><hr /><pre>This is a public inbox, see <a target=\"_blank\" href=\"https://lore.kernel.org/git/_/text/mirror/\">mirroring instructions</a>\nfor how to clone and mirror all data and code used for this inbox;\nas well as URLs for NNTP newsgroup(s).</pre>",
"author": "",
"favicon": "",
"source": "lore.kernel.org",
"published": "",
"ttr": 2340,
"type": ""
}