Skip to contentSkip to conversation

What's cooking in git.git (May 2025, #07; Fri, 23)

DISCUSSION Junio C Hamano 3 participants 7 messages May 24, 2025 – Jun 2, 2025

lore
Share & tools
lore
#0 lore raw

Here are the topics that have been cooking in my tree. Commits prefixed with '+' are in 'next' (being in 'next' is a sign that a topic is stable enough to be used and are candidate to be in a future release). Commits prefixed with '-' are only in 'seen', and aren't considered "accepted" at all and may be annotated with an URL to a message that raises issues but they are no means exhaustive. A topic without enough support may be discarded after a long period of no activity (of course they can be resubmit when new interests arise).

Copies of the source code to Git live in many repositories, and the following is a list of the ones I push into or their mirrors. Some repositories have only a subset of branches.

With maint, master, next, seen, todo:

	git://git.kernel.org/pub/scm/git/git.git/
	git://repo.or.cz/alt-git.git/
	https://kernel.googlesource.com/pub/scm/git/git/
	https://github.com/git/git/
	https://gitlab.com/git-scm/git/

With all the integration branches and topics broken out:

https://github.com/gitster/git/

Even though the preformatted documentation in HTML and man format are not sources, they are published in these repositories for convenience (replace "htmldocs" with "manpages" for the manual pages):

git://git.kernel.org/pub/scm/git/git-htmldocs.git/
https://github.com/gitster/git-htmldocs.git/

Release tarballs are available at:

https://www.kernel.org/pub/software/scm/git/

--------------------------------------------------
[Graduated to 'master']

* ag/send-email-hostname-f (2025-05-13) 1 commit
(merged to 'next' on 2025-05-16 at d5aa175e7d)
+ send-email: try to get fqdn by running hostname -f on Linux and macOS

Teach "git send-email" to also consult hostname -f for mail
domain to compute the identity given to SMTP servers.
source: <PN3PR01MB959701F40F805351472EA4CCB897A@PN3PR01MB9597.INDPRD01.PROD.OUTLOOK.COM>

* ds/scalar-no-maintenance (2025-05-14) 5 commits (merged to 'next' on 2025-05-16 at f9364331c0)
+ scalar reconfigure: improve --maintenance docs (merged to 'next' on 2025-05-08 at 1006cdd399)
+ scalar reconfigure: add --maintenance=<mode> option
+ scalar clone: add --no-maintenance option
+ scalar register: add --no-maintenance option
+ scalar: customize register_dir()'s behavior

Two "scalar" subcommands that adds a repository that hasn't been under "scalar"'s control are taught an option not to enable the scheduled maintenance on it.
source: <[email protected]>

* en/replay-wo-the-repository (2025-05-14) 1 commit
(merged to 'next' on 2025-05-16 at 23608dbaab)
+ replay: replace the_repository with repo parameter passed to cmd_replay ()

The dependency on the_repository variable has been reduced from the code paths in "git replay".
source: <[email protected]>

* js/ci-build-win-in-release-mode (2025-05-05) 1 commit (merged to 'next' on 2025-05-22 at 213a3f0262)
+ ci(win+Meson): build in Release mode

win+Meson CI pipeline, unlike other pipelines for Windows, used to build artifacts in develper mode, which has been changed to build them in release mode for consistency.
source: <[email protected]>

* lo/json-writer-docs (2025-05-16) 2 commits (merged to 'next' on 2025-05-16 at ab81da1b16)
+ json-writer: describe the usage of jw_* functions
+ json-writer: add docstrings to jw_* functions

In-code docstring updates.
source: <[email protected]>

* ly/pack-bitmap-load-leakfix (2025-05-12) 1 commit (merged to 'next' on 2025-05-14 at 0be31eac6b)
+ pack-bitmap: fix memory leak if load_bitmap_entries_v1 failed

Leakfix.
source: <[email protected]>

* ps/ci-gitlab-enable-msvc-meson-job (2025-05-13) 1 commit (merged to 'next' on 2025-05-16 at 81da2fbef9)
+ gitlab-ci: always run MSVC-based Meson job

CI settings at GitLab has been updated to run MSVC based Meson job automatically (as opposed to be done only upon manual request).
source: <20250428-pks-gitlab-ci-execute-win-meson-v1-1-f68683552b9e@pks.im>

--------------------------------------------------
[New Topics]

* op/cvsserver-perl-warning (2025-05-22) 1 commit
- cvsserver: avoid precedence problem between ! and %s

Recent versions of Perl started warning against "! A =~ /pattern/" which does not negate the result of the matching.
source: <[email protected]>

* ps/meson-tap-parse (2025-05-21) 4 commits
- meson: parse TAP output generated by our tests
- meson: introduce kwargs variable for tests
- t/test-lib: don't print shell traces to stdout
- t: fix cases where output breaks TAP format

Meson-based build/test framework now understands TAP output generated by our tests.

Expecting a reroll.
cf. <[email protected]>
source: <[email protected]>

* am/sparse-index-name-hash-fix (2025-05-21) 1 commit
- name-hash: don't add sparse directories in threaded lazy init

Avoid adding directory path to a sparse-index tree entries to the name-hash, since they would bloat the hashtable without anybody querying for them. This was done already for a single threaded part of the code, but now the multi-threaded code also does the same.

Will merge to 'next'.
source: <[email protected]>

* jw/doc-txt-to-adoc-refs (2025-05-21) 2 commits
- SQUASH???
- doc: update references to renamed AsciiDoc files

Some leftover references to documentation source files that no longer exist, due to recent ".txt" -> ".adoc" renaming, have been corrected.

Waiting for review response.
source: <[email protected]>

* jk/diff-no-index-with-pathspec (2025-05-22) 3 commits
- diff --no-index: support limiting by pathspec
- pathspec: add flag to indicate operation without repository
- pathspec: add match_leading_pathspec variant

"git diff --no-index dirA dirB" can limit the comparison with pathspec at the end of the command line, just like normal "git diff".

Comments?
source: <[email protected]>

* mm/apply-reverse-mode-of-deleted-path (2025-05-23) 2 commits
- apply: set file mode when --reverse creates a deleted file
- t4129: test that git apply warns for unexpected mode changes

"git apply --index/--cached" when applying a deletion patch in reverse failed to give the mode bits of the path "removed" by the patch to the file it creates, which has been corrected.

Will merge to 'next'?
source: <[email protected]>

--------------------------------------------------
[Cooking]

* ag/doc-send-email-update-2 (2025-05-19) 4 commits
- docs: remove credential helper links for emails from gitcredentials
- docs: improve formatting in git-send-email documentation
- docs: add credential helper for yahoo and link Google's sendgmail tool
- Merge branch 'ag/doc-send-email' into ag/doc-send-email-update-2

Documentation for "git send-email" has been updated with a bit more credential helper and OAuth information.

Comments?
source: <[email protected]>

* es/meson-configure-build-options-fix (2025-05-19) 1 commit (merged to 'next' on 2025-05-21 at b468031e13)
+ meson: reformat default options to workaround bug in meson configure

Build procedure updates.

Will merge to 'master'.
source: <[email protected]>

* jt/receive-pack-skip-connectivity-check (2025-05-20) 2 commits (merged to 'next' on 2025-05-22 at 3ced8c5d65)
+ builtin/receive-pack: add option to skip connectivity check
+ t5410: test receive-pack connectivity check

"git receive-pack" optionally learns not to care about connectivity check, which can be useful when the repository arranges to ensure connectivity by some other means.

Will merge to 'master'.
source: <[email protected]>

* kh/notes-doc-fixes (2025-05-23) 9 commits
- doc: notes: use stuck form throughout
- doc: notes: treat --stdin equally between copy/remove
- doc: notes: point out copy --stdin use with argv
- doc: notes: clearly state that --stripspace is the default
- doc: notes: remove stripspace discussion from other options
- doc: notes: rework --[no-]stripspace
- doc: notes: split out options with negated forms
- doc: config: mention core.commentChar on commit.cleanup
- doc: stripspace: mention where the default comes from

"git notes --help" documentation updates.

Comments?
Getting much better.
source: <[email protected]>

* kn/passing-leak-tests (2025-05-20) 1 commit
(merged to 'next' on 2025-05-22 at bc0d708c5c)
+ t: remove unexpected SANITIZE_LEAK variables

Remove the leftover hints to the test framework to mark tests that do not pass the leak checker tests, as they should no longer be needed.

Will merge to 'master'.
source: <20250520-kn-remove-unexpected-exported-v1-1-bb60cec57e84@gmail.com>

* ps/midx-negative-packfile-cache (2025-05-20) 2 commits
- midx: stop repeatedly looking up nonexistent packfiles
- packfile: explain ordering of how we look up auxiliary pack files

When a stale .midx file refers to .pack files that no longer exist, we ended up checking for these non-existent files repeatedly, which has been optimized by memoizing the non-existence.

Will merge to 'next'?
source: <20250520-pks-pack-avoid-stats-on-missing-v2-0-333c5217fb05@pks.im>

* pw/midx-repack-overflow-fix (2025-05-22) 4 commits
- midx docs: clarify tie breaking
- midx: avoid negative array index
- midx repack: avoid potential integer overflow on 64 bit systems
- midx repack: avoid integer overflow on 32 bit systems

Integer overflow fix around code paths for "git multi-pack-index repack"..

Will merge to 'next'.
cf. <aC/C9oQrcx/[email protected]>
source: <[email protected]>

* pw/stash-p-pathspec-fixes (2025-05-20) 2 commits
- stash: allow "git stash [<options>] --patch <pathspec>" to assume push
- stash: allow "git stash -p <pathspec>" to assume push again

"git stash -p <pathspec>" improvements.

Comments?
source: <[email protected]>

* kn/fetch-push-bulk-ref-update (2025-05-19) 4 commits (merged to 'next' on 2025-05-22 at 7ab014070f)
+ receive-pack: use batched reference updates
+ send-pack: fix memory leak around duplicate refs
+ fetch: use batched reference updates
+ refs: add function to translate errors to strings

"git push" and "git fetch" are taught to update refs in batches to gain performance.

Will cook in 'next'.
source: <20250519-501-update-git-fetch-1-to-use-partial-transactions-v3-0-6cdfd4f769b9@gmail.com>

* js/misc-defensive (2025-05-15) 14 commits
- shallow: handle missing shallow commits gracefully
- test-tool repository: check return value of lookup_commit()
- submodule: check return value of submodule_from_path()
- inherit_tracking(): defensive programming
- describe: defensive programming
- fetch: defensive programming
- push: defensive programming
- stash: defensive programming
- stash: defensive programming
- verify_commit_graph(): defensive programming
- unparse_commit(): defensive programming
- fetch-pack: defensive programming
- get_parent(): defensive programming
- revision: defensive programming

Assorted changes that please CodeQL.

Comments?
source: <[email protected]>

* js/misc-fixes (2025-05-15) 11 commits
(merged to 'next' on 2025-05-21 at e803806107)
+ sequencer: stop pretending that an assignment is a condition
+ bundle-uri: avoid using undefined output of sscanf()
+ commit-graph: avoid using stale stack addresses
+ trace2: avoid "futile conditional"
+ Avoid redundant conditions
+ fetch: avoid unnecessary work when there is no current branch
+ has_dir_name(): make code more obvious
+ upload-pack: rename enum to reflect the operation
+ commit-graph: avoid malloc'ing a local variable
+ fetch: carefully clear local variable's address after use
+ commit: simplify code

Assorted fixes for issues found with CodeQL.

Will merge to 'master'.
source: <[email protected]>

* ly/commit-graph-fill-oids-leakfix (2025-05-15) 1 commit (merged to 'next' on 2025-05-19 at 972bbc7c11)
+ commit-graph: fix memory leak when fill_oids_from_packs() fails

Leakfix.

Will merge to 'master'.
source: <[email protected]>

* ly/mailinfo-decode-header-leakfix (2025-05-15) 1 commit (merged to 'next' on 2025-05-19 at 87066488fc)
+ mailinfo: fix pointential memory leak if decode_header failed

Leakfix.

Will merge to 'master'.
source: <[email protected]>

* ly/sequencer-rearrange-leakfix (2025-05-15) 1 commit (merged to 'next' on 2025-05-19 at f0ad6cfe21)
+ sequencer: fix memory leak if todo_list_rearrange_squash() failed

Leakfix.

Will merge to 'master'.
source: <[email protected]>

* en/sequencer-comment-messages (2025-05-16) 1 commit (merged to 'next' on 2025-05-21 at b6516794fb)
+ sequencer: make it clearer that commit descriptions are just comments

Prefix '#' to the commit title in the "rebase -i" todo file, just like a merge commit being replayed.

Will merge to 'master'.
source: <[email protected]>

* jk/no-funny-object-types (2025-05-16) 13 commits (merged to 'next' on 2025-05-19 at 4c995dbd23)
+ object-file: drop support for writing objects with unknown types
+ hash-object: handle --literally with OPT_NEGBIT
+ hash-object: merge HASH_* and INDEX_* flags
+ hash-object: stop allowing unknown types
+ t: add lib-loose.sh
+ t/helper: add zlib test-tool
+ oid_object_info(): drop type_name strbuf
+ fsck: stop using object_info->type_name strbuf
+ oid_object_info_convert(): stop using string for object type
+ cat-file: use type enum instead of buffer for -t option
+ object-file: drop OBJECT_INFO_ALLOW_UNKNOWN_TYPE flag
+ cat-file: make --allow-unknown-type a noop
+ object-file.h: fix typo in variable declaration

Support to create a loose object file with unknown object type has been dropped.

Will merge to 'master'.
source: <[email protected]>

* kj/my-first-contribution-updates (2025-05-19) 3 commits (merged to 'next' on 2025-05-21 at f8c92423fb)
+ docs: replace git_config to repo_config
+ docs: clarify cmd_psuh signature and explain UNUSED macro
+ docs: remove unused mentoring mailing list reference

Doc updates.

Will merge to 'master'.
source: <[email protected]>

* kj/renamed-submodule (2025-05-16) 1 commit
- submodule: prevent overwriting .gitmodules entry on path reuse

The case where a new submodule takes a path where used to be a completely different subproject is now dealt a bit better than before.

Comments?
source: <[email protected]>

* pw/update-thunderbird-patch-inline (2025-05-16) 1 commit
- contrib: update thunderbird-patch-inline

Update bitrotten instruction for sending patches via Thunderbird (in contrib/).

Comments?
source: <[email protected]>

* bc/stash-export-import (2025-05-22) 5 commits
- builtin/stash: provide a way to import stashes from a ref
- builtin/stash: provide a way to export stashes to a ref
- builtin/stash: factor out revision parsing into a function
- reflog-walk: expose read_complete_reflog
- object-name: make get_oid quietly return an error

An interchange format for stash entries is defined, and subcommand of "git stash" to import/export has been added.

Will merge to 'next'?
source: <[email protected]>

* ds/sparse-apply-add-p (2025-05-16) 4 commits (merged to 'next' on 2025-05-21 at 933f316786)
+ p2000: add performance test for patch-mode commands
+ reset: integrate sparse index with --patch
+ git add: make -p/-i aware of sparse index
+ apply: integrate with the sparse index

"git apply" and "git add -i/-p" code paths no longer unnecessarily expand sparse-index while working.

Will merge to 'master'.
Kicked out of next and then is about to come back.
source: <[email protected]>

* en/merge-tree-check (2025-05-16) 2 commits
(merged to 'next' on 2025-05-19 at c3278b91fa)
+ merge-tree: add a new --quiet flag
+ merge-ort: add a new mergeability_only option

"git merge-tree" learned an option to see if it resolves cleanly without actually creating a result.

Will merge to 'master'.
source: <[email protected]>

* lm/add-p-context (2025-05-12) 4 commits
- add-patch: add diff.context command line overrides
- add-patch: respect diff.context configuration
- test: refactor to use "test_config"
- test: refactor to use "test_grep"

"git add/etc -p" now honors diff.context configuration variable, and learns to honor -U<n> option.

Comments?
source: <[email protected]>

* ps/contrib-sweep (2025-05-16) 12 commits
- Revert "contrib: remove "thunderbird-patch-inline""
- contrib: remove some scripts in "stats" directory
- contrib: remove "git-new-workdir"
- contrib: remove "emacs" directory
- contrib: remove "git-resurrect.sh"
- contrib: remove "persistent-https" remote helper
- contrib: remove "mw-to-git"
- contrib: remove "hooks" directory
- contrib: remove "thunderbird-patch-inline"
- contrib: remove remote-helper stubs
- contrib: remove "examples" directory
- contrib: remove "remotes2config.sh"

Remove bunch of stuff from contrib/ hierarchy.

I've reverted the thunderbird thing for now.
source: <[email protected]>

* rj/build-tweaks-part2 (2025-05-19) 5 commits (merged to 'next' on 2025-05-19 at fea40b8fb1)
+ configure.ac: upgrade to a compilation check for sysinfo
+ meson.build: correct setting of GIT_EXEC_PATH
+ meson: correct path to system config/attribute files
+ meson: correct install location of YAML.pm
+ meson.build: quote the GITWEBDIR build configuration

Updates to meson-based build procedure.

Will merge to 'master'.
source: <[email protected]>

* ps/object-store (2025-05-14) 18 commits
- odb: rename read_object_with_reference()
- odb: rename pretend_object_file()
- odb: rename has_object()
- odb: rename repo_read_object_file()
- odb: rename oid_object_info()
- odb: trivial refactorings to get rid of the_repository
- odb: get rid of the_repository when handling submodule alternates
- odb: get rid of the_repository when handling the primary alternate
- odb: get rid of the_repository in for_each() functions
- odb: get rid of the_repository when handling alternates
- odb: get rid of the_repository in odb_mkstemp()
- odb: get rid of the_repository in assert_oid_type()
- odb: get rid of the_repository in find_odb()
- odb: introduce parent pointers
- object-store: rename files to "odb.{c,h}"
- object-store: rename object_directory to odb_alternate
- object-store: rename raw_object_store to object_database
- Merge branch 'ps/object-store-cleanup' into ps/object-store

Code clean-up around object access API.

Comments?
source: <20250514-pks-object-store-wo-the-repository-v3-0-47df1d4ead22@pks.im>

* sj/use-mmap-to-check-packed-refs (2025-05-14) 3 commits (merged to 'next' on 2025-05-21 at a0ed4fdf95)
+ packed-backend: mmap large "packed-refs" file during fsck
+ packed-backend: extract snapshot allocation in load_contents
+ packed-backend: fsck should warn when "packed-refs" file is empty

The code path to access the "packed-refs" file while "fsck" is taught to mmap the file, instead of reading the whole file in the memory.

Will merge to 'master'.
source: <aCS7O8tNekg_u9Wp@ArchLinux>

* cc/promisor-remote-capability (2025-05-19) 5 commits
- promisor-remote: use string constants for 'name' and 'url' too
- promisor-remote: allow a client to check fields
- promisor-remote: refactor how we parse advertised fields
- promisor-remote: allow a server to advertise more fields
- promisor-remote: refactor to get rid of 'struct strvec'

Comments?
source: <[email protected]>

* jc/doc-synopsis-option-markup (2025-05-12) 4 commits
(merged to 'next' on 2025-05-21 at cb897d1302)
+ git-var doc: fix usage of $ENV_VAR vs ENV_VAR
+ git-verify-* doc: update mark-up of synopsis option descriptions
+ git-{var,write-tree} docs: update mark-up of synopsis option descriptions
+ git-daemon doc: update mark-up of synopsis option descriptions

Doc mark-up fixes.

Will merge to 'master'.
source: <[email protected]>

* jc/you-still-use-whatchanged (2025-05-12) 6 commits (merged to 'next' on 2025-05-22 at e79dc9090e)
+ whatschanged: list it in BreakingChanges document
+ whatchanged: remove when built with WITH_BREAKING_CHANGES
+ whatchanged: require --i-still-use-this
+ tests: prepare for a world without whatchanged
+ doc: prepare for a world without whatchanged
+ you-still-use-that??: help deprecating commands for removal

"git whatchanged" that is longer to type than "git log --raw" which is its modern rough equivalent has outlived its usefulness more than 10 years ago. Plan to deprecate and remove it.

Will cook in 'next'.
source: <[email protected]>

* cc/fast-import-export-signature-names (2025-04-24) 1 commit . fast-(import|export): improve on the signature algorithm name

Clean up the way how signature on commit objects are exported to and imported from fast-import stream.

Expecting a reroll.
cf. <[email protected]>
cf. https://github.com/git/git/actions/runs/14671270673/job/41178138711
source: <[email protected]>

* sj/string-list-typefix (2025-04-22) 5 commits
- u-string-list: move "remove duplicates" test to "u-string-list.c"
- u-string-list: move "filter string" test to "u-string-list.c"
- u-string-list: move "test_split_in_place" to "u-string-list.c"
- u-string-list: move "test_split" into "u-string-list.c"
- string-list: fix sign compare warnings

Code and test clean-up around string-list API.

Expecting a reroll.
cf. <[email protected]>
source: <aAetW0dan8S3Fljq@ArchLinux>

* tb/midx-avoid-cruft-packs (2025-04-15) 9 commits
- repack: exclude cruft pack(s) from the MIDX where possible
- pack-objects: introduce '--stdin-packs=follow'
- pack-objects: swap 'show_{object,commit}_pack_hint'
- pack-objects: fix typo in 'show_object_pack_hint()'
- pack-objects: perform name-hash traversal for unpacked objects
- pack-objects: declare 'rev_info' for '--stdin-packs' earlier
- pack-objects: factor out handling '--stdin-packs'
- pack-objects: limit scope in 'add_object_entry_from_pack()'
- pack-objects: use standard option incompatibility functions

"pack-objects" has been taught to avoid pointing into objects in cruft packs from midx.

Expecting a (hopefully small and final) reroll? cf.<CABPp-BEukTWwsuC7MMR8D5_UAhyw-LgT=DsPKAWeR_ZmVVhjzQ@mail.gmail.com>
source: <[email protected]>

* tb/pack-bitmap-lookup-tables (2025-04-17) 4 commits
- t/perf/lib-bitmap.sh: avoid test_perf during setup
- t/perf: avoid testing bitmaps without lookup table
- p5312: removed duplicate performance test script
- pack-bitmap: write lookup table extension by default

Enable lookup tables extension in pack bitmap (and midx bitmap) by default.

Comments?
source: <[email protected]>

* pb/status-rebase-fixes (2025-03-28) 4 commits
- wt-status: suggest 'git rebase --continue' to conclude 'merge' instruction
- wt-status: also abbreviate 'merge' and 'fixup -C' lines during rebase
- SQUASH??? - <CAPig+cS92W_gYuNsaTvQxiP3xBK7Wpg0__uVkgAU1x0OFJUZgQ@mail.gmail.com>
- rebase -r: do create merge commit after empty resolution

A few fixes around "git status" while "git rebase" is running, plus a corner case bug fix for "git rebase -r".

Expecting a (small and hopefully final) clarifying reroll. cf. <[email protected]> cf. <[email protected]>
source: <[email protected]>

* md/userdiff-bash-shell-function (2025-05-16) 1 commit (merged to 'next' on 2025-05-16 at 1fe8b68a72)
+ userdiff: extend Bash pattern to cover more shell function forms

The userdiff pattern for shell scripts has been updated to cope with more bash-isms.

Will merge to 'master'.
cf. <[email protected]>
source: <[email protected]>

* ds/path-walk-2 (2025-05-16) 13 commits
- pack-objects: allow --shallow and --path-walk
- path-walk: add new 'edge_aggressive' option
- pack-objects: thread the path-based compression
- pack-objects: refactor path-walk delta phase
- scalar: enable path-walk during push via config
- pack-objects: enable --path-walk via config
- repack: add --path-walk option
- t5538: add tests to confirm deltas in shallow pushes
- pack-objects: introduce GIT_TEST_PACK_PATH_WALK
- p5313: add performance tests for --path-walk
- pack-objects: update usage to match docs
- pack-objects: add --path-walk option
- pack-objects: extract should_attempt_deltas()

"git pack-objects" learns to find delta bases from blobs at the same path, using the --path-walk API.

Comments?
source: <[email protected]>

--------------------------------------------------
[Discarded]

* ib/diff-S-G-with-longhand (2025-02-12) 10 commits
. diff: docs: Use --patch-{grep,modifies} over -G/-S . diff: --pickaxe-{all,regex} help: Add --patch-{grep,modifies} . diff: test: Use --patch-{grep,modifies} over -G/-S . completion: Support --patch-{grep,modifies}
. diff: --patch-{grep,modifies} arg names for -G and -S . docs: gitdiffcore: -G and -S: Use regex/string placeholders . diff: short help: Add -G and --pickaxe-grep
. diff: short help: Correct -S description
. diff: -G description: Correct copy/paste error
. t/t4209-log-pickaxe: Naming typo: -G takes a regex

The commands in the "diff" family learned longhands for "-S" and "-G" options.

Has been in "Expecting a reroll" state for too long.
source: <[email protected]>

* ps/pack-check-pack-first (2025-05-16) 1 commit . packfile: avoid access(3p) calls for missing packs

The packfile registration code used to check ".pack" file the last after checking ".keep" and other files; the ordering is reversed.

Superseded by the ps/midx-negative-packfile-cache topic
source: <20250516-pks-pack-avoid-stats-on-missing-v1-1-e2ef4d8798a3@pks.im>

#1 lore raw
Quoted 10 lines · Junio C Hamano

* ps/midx-negative-packfile-cache (2025-05-20) 2 commits
- midx: stop repeatedly looking up nonexistent packfiles
- packfile: explain ordering of how we look up auxiliary pack files

When a stale .midx file refers to .pack files that no longer exist, we ended up checking for these non-existent files repeatedly, which has been optimized by memoizing the non-existence.

Will merge to 'next'?
source: <20250520-pks-pack-avoid-stats-on-missing-v2-0-333c5217fb05@pks.im>

I wanted to send one more iteration of this where I hide the ugliness of (void *)(intptr_t)-1 behind a macro, as suggested. But I saw that Taylor has built on top of these patches, so I don't want to make his life harder. Cc'd him.

Quoted 20 lines

* ps/object-store (2025-05-14) 18 commits
- odb: rename read_object_with_reference()
- odb: rename pretend_object_file()
- odb: rename has_object()
- odb: rename repo_read_object_file()
- odb: rename oid_object_info()
- odb: trivial refactorings to get rid of the_repository
- odb: get rid of the_repository when handling submodule alternates
- odb: get rid of the_repository when handling the primary alternate
- odb: get rid of the_repository in for_each() functions
- odb: get rid of the_repository when handling alternates
- odb: get rid of the_repository in odb_mkstemp()
- odb: get rid of the_repository in assert_oid_type()
- odb: get rid of the_repository in find_odb()
- odb: introduce parent pointers
- object-store: rename files to "odb.{c,h}"
- object-store: rename object_directory to odb_alternate
- object-store: rename raw_object_store to object_database
- Merge branch 'ps/object-store-cleanup' into ps/object-store

Quoted 4 lines

Code clean-up around object access API.

Comments?
source: <20250514-pks-object-store-wo-the-repository-v3-0-47df1d4ead22@pks.im>

I think the only outstanding discussion is whether to name things odb_alternate or odb_source [1]. In case others agree that odb_source is a better name I'm happy to revise, but if not I'd rather keep it as-is.

Other than that I think the patch series is in a good shape.

Thanks!

Patrick

[1]: <tjsbotrnrffykmi3letktpb3bly4nqw4wxzyrszgbln7pznem4@3kwiq4zvaebw>

#2 lore raw
Quoted 9 lines · Patrick Steinhardt
Quoted 8 lines · Junio C Hamano

* ps/midx-negative-packfile-cache (2025-05-20) 2 commits
- midx: stop repeatedly looking up nonexistent packfiles
- packfile: explain ordering of how we look up auxiliary pack files

When a stale .midx file refers to .pack files that no longer exist, we ended up checking for these non-existent files repeatedly, which has been optimized by memoizing the non-existence.

Quoted 7 lines
Quoted 2 lines

Will merge to 'next'?
source: <20250520-pks-pack-avoid-stats-on-missing-v2-0-333c5217fb05@pks.im>

I wanted to send one more iteration of this where I hide the ugliness of (void *)(intptr_t)-1 behind a macro, as suggested. But I saw that Taylor has built on top of these patches, so I don't want to make his life harder. Cc'd him.

Yeah, I do not think I've merged Taylor's work anywhere yet, but I did notice that it was built on top of this topic. I think it is perfectly OK to clean it up before your topic hits 'next'; adjusting to such a change is a responsibility of the author of a topic that depends on anything not in 'next' yet.

Quoted 4 lines

I think the only outstanding discussion is whether to name things odb_alternate or odb_source [1]. In case others agree that odb_source is a better name I'm happy to revise, but if not I'd rather keep it as-is.

The model in which the term "alternates" was born is "A repository has its own object directory, the primary one, and in addition it can borrow from zero or more alternate object directories that are used by other repositories". The presence of the primary makes the word "alternate" meaningful.

Is the model now "A repository has one object store, which consists of one or more X, all of which are equals"? If there is no primary that is more special than others, then calling X an "alternate" may indeed sound funny, although (1) I do not find it terribly confusing and (2) I do not find "source" much better, either.

The names we use to call the collection and the underlying implementations of the collection in the reference world unfortunately does not quite help to guide us, as we do not take two implementations and compose into one unified view, which is what we are doing in the object store. Hmmm...

We call pathspec elements given on the command line collectively a pathspec. "Object store elements like loose object directories and packfiles form the object store"? That may be a mouthful. I dunno.

#3 lore raw
Quoted 12 lines · Junio C Hamano
Quoted 4 lines · Patrick Steinhardt

I think the only outstanding discussion is whether to name things odb_alternate or odb_source [1]. In case others agree that odb_source is a better name I'm happy to revise, but if not I'd rather keep it as-is.

The model in which the term "alternates" was born is "A repository has its own object directory, the primary one, and in addition it can borrow from zero or more alternate object directories that are used by other repositories". The presence of the primary makes the word "alternate" meaningful.

Quoted 5 lines

Is the model now "A repository has one object store, which consists of one or more X, all of which are equals"? If there is no primary that is more special than others, then calling X an "alternate" may indeed sound funny, although (1) I do not find it terribly confusing and (2) I do not find "source" much better, either.

My understanding is that the object store still has a primary X and zero or more alternative X. The idea is that eventually, with pluggable ODBs, X can be a different backend/provider instead of just being "files". If this is the case, calling X an "alternate" would mean we have a primary "alternate" and potentially a set of "alternate" alternates.

This sounds a bit odd and doesn't quite match what I would intuitively expect. But, I also don't find it super confusing either.

Quoted 5 lines

The names we use to call the collection and the underlying implementations of the collection in the reference world unfortunately does not quite help to guide us, as we do not take two implementations and compose into one unified view, which is what we are doing in the object store. Hmmm...

Similar to references, I still think of a pluggable ODB as a "backend". The main difference being that with references there is only a single backend active ("file" or "reftables") at a time, while for the object store there could be multiple.

-Justin

Quoted 3 lines

We call pathspec elements given on the command line collectively a pathspec. "Object store elements like loose object directories and packfiles form the object store"? That may be a mouthful. I dunno.

#4 lore raw
Quoted 19 lines · Justin Tobler
Quoted 12 lines · Junio C Hamano
Quoted 4 lines · Patrick Steinhardt

I think the only outstanding discussion is whether to name things odb_alternate or odb_source [1]. In case others agree that odb_source is a better name I'm happy to revise, but if not I'd rather keep it as-is.

The model in which the term "alternates" was born is "A repository has its own object directory, the primary one, and in addition it can borrow from zero or more alternate object directories that are used by other repositories". The presence of the primary makes the word "alternate" meaningful.

Quoted 5 lines

Is the model now "A repository has one object store, which consists of one or more X, all of which are equals"? If there is no primary that is more special than others, then calling X an "alternate" may indeed sound funny, although (1) I do not find it terribly confusing and (2) I do not find "source" much better, either.

Quoted 8 lines

My understanding is that the object store still has a primary X and zero or more alternative X. The idea is that eventually, with pluggable ODBs, X can be a different backend/provider instead of just being "files". If this is the case, calling X an "alternate" would mean we have a primary "alternate" and potentially a set of "alternate" alternates.

This sounds a bit odd and doesn't quite match what I would intuitively expect. But, I also don't find it super confusing either.

Yeah, I understand that confusion indeed. I don't think that the other proposals we've got are a lot better, either:

- odb_backend was shot down because it may cause the association that one object database has one backend. But backends are per alternate, so there's a mismatch in expectations.

- odb_source is better, but we now have the problem that we use "alternate" interchangably in most cases where we also use odb_source. This will likely lead to somewhat awkward interfaces.

The problem with odb_source might eventually go away once we clearly distinguish the "alternates" concept from the low-level mechanism to access objects. But I'm just not certain at all whether it won't cause more confusion when in most cases "alternates" and "sources" can be used somewhat interchangably.

I dunno. The more I think about this the more I start to like the odb_source name.

Quoted 10 lines
Quoted 5 lines

The names we use to call the collection and the underlying implementations of the collection in the reference world unfortunately does not quite help to guide us, as we do not take two implementations and compose into one unified view, which is what we are doing in the object store. Hmmm...

Similar to references, I still think of a pluggable ODB as a "backend". The main difference being that with references there is only a single backend active ("file" or "reftables") at a time, while for the object store there could be multiple.

Yeah, that's basically the major source of confusion I want to avoid with the "backend" terminology. I really do want to make it explicit that there is not only one backend, but multiple ones.

Patrick

#5 lore raw
Quoted 6 lines · Patrick Steinhardt

Yeah, I understand that confusion indeed. I don't think that the other proposals we've got are a lot better, either:

- odb_backend was shot down because it may cause the association that one object database has one backend. But backends are per alternate, so there's a mismatch in expectations.

I do not see where that association would come from, though. But I agree with the other objection that the word "backend" is more about implementation and less about an instance that is realized by that implementation, i.e. two such components that runs the code for a single backend may be part of a single object database.

Quoted 4 lines

- odb_source is better, but we now have the problem that we use "alternate" interchangably in most cases where we also use odb_source. This will likely lead to somewhat awkward interfaces.

Quoted 8 lines

The problem with odb_source might eventually go away once we clearly distinguish the "alternates" concept from the low-level mechanism to access objects. But I'm just not certain at all whether it won't cause more confusion when in most cases "alternates" and "sources" can be used somewhat interchangably.

I dunno. The more I think about this the more I start to like the odb_source name.

Yeah, I do not mind calling the instantiation backed by a backend implementation a odb_source.

In any case, when deciding the terminology, we should look at what we currently have in the glossary and imagine how they should look like in the updated world. Currently,

- "alternate object database" is described as inheriting the entirety of another "object database" (we deliberately do not say that this other object database belongs to another repository, as a standalone object database is a valid option).

- "object database" is described as what holds a set of "objects". There is no complication here ;-)

When treating the set of objects stored in $GIT_DIR/objects/??/ directories (i.e. "loose objects") and the set of objects stored in $GIT_DIR/objects/pack/ directory (i.e. "packed objects") as two separate odb_something, with a vision that we may add different implementations of such collection later, it would be very confusing to call each of them "an alternate". "source" may not be ideal, but it is the best among what we collectively have come up with, I think.

#6 lore raw
Quoted 43 lines · Junio C Hamano
Quoted 6 lines · Patrick Steinhardt

Yeah, I understand that confusion indeed. I don't think that the other proposals we've got are a lot better, either:

- odb_backend was shot down because it may cause the association that one object database has one backend. But backends are per alternate, so there's a mismatch in expectations.

I do not see where that association would come from, though. But I agree with the other objection that the word "backend" is more about implementation and less about an instance that is realized by that implementation, i.e. two such components that runs the code for a single backend may be part of a single object database.

Quoted 4 lines

- odb_source is better, but we now have the problem that we use "alternate" interchangably in most cases where we also use odb_source. This will likely lead to somewhat awkward interfaces.

Quoted 8 lines

The problem with odb_source might eventually go away once we clearly distinguish the "alternates" concept from the low-level mechanism to access objects. But I'm just not certain at all whether it won't cause more confusion when in most cases "alternates" and "sources" can be used somewhat interchangably.

I dunno. The more I think about this the more I start to like the odb_source name.

Yeah, I do not mind calling the instantiation backed by a backend implementation a odb_source.

In any case, when deciding the terminology, we should look at what we currently have in the glossary and imagine how they should look like in the updated world. Currently,

- "alternate object database" is described as inheriting the entirety of another "object database" (we deliberately do not say that this other object database belongs to another repository, as a standalone object database is a valid option).

- "object database" is described as what holds a set of "objects". There is no complication here ;-)

Quoted 7 lines

When treating the set of objects stored in $GIT_DIR/objects/??/ directories (i.e. "loose objects") and the set of objects stored in $GIT_DIR/objects/pack/ directory (i.e. "packed objects") as two separate odb_something, with a vision that we may add different implementations of such collection later, it would be very confusing to call each of them "an alternate". "source" may not be ideal, but it is the best among what we collectively have come up with, I think.

Okay, let's settle on odb_source then. I'll send a new version once I have wrangled all the conflicts :) Thanks for the discussion, all!

Patrick