7 UI, Workflows & Features
9 * "git help log" has been enhanced by sharing more material from the
10 documentation for the underlying "git rev-list" command.
12 * "git for-each-ref --format=<>" learned %(contents:size).
14 * "git merge" learned to selectively omit " into <branch>" at the end
15 of the title of default merge message with merge.suppressDest
18 * The component to respond to "git fetch" request is made more
19 configurable to selectively allow or reject object filtering
20 specification used for partial cloning.
22 * Stop when "sendmail.*" configuration variables are defined, which
23 could be a mistaken attempt to define "sendemail.*" variables.
25 * The existing backends for "git mergetool" based on variants of vim
26 have been refactored and then support for "nvim" has been added.
28 * "git bisect" learns the "--first-parent" option to find the first
29 breakage along the first-parent chain.
31 * "git log --first-parent -p" showed patches only for single-parent
32 commits on the first-parent chain; the "--first-parent" option has
33 been made to imply "-m". Use "--no-diff-merges" to restore the
34 previous behaviour to omit patches for merge commits.
36 * The commit labels used to explain each side of conflicted hunks
37 placed by the sequencer machinery have been made more readable by
40 * The "--batch-size" option of "git multi-pack-index repack" command
41 is now used to specify that very small packfiles are collected into
42 one until the total size roughly exceeds it.
44 * The recent addition of SHA-256 support is marked as experimental in
47 * "git fetch" learned --no-write-fetch-head option to avoid writing
50 * Command line completion (in contrib/) usually omits redundant,
51 deprecated and/or dangerous options from its output; it learned to
52 optionally include all of them.
54 * The output from the "diff" family of the commands had abbreviated
55 object names of blobs involved in the patch, but its length was not
56 affected by the --abbrev option. Now it is.
58 * "git worktree" gained a "repair" subcommand to help users recover
59 after moving the worktrees or repository manually without telling
60 Git. Also, "git init --separate-git-dir" no longer corrupts
61 administrative data related to linked worktrees.
63 * The "--format=" option to the "for-each-ref" command and friends
64 learned a few more tricks, e.g. the ":short" suffix that applies to
65 "objectname" now also can be used for "parent", "tree", etc.
67 * "git worktree add" learns that the "-d" is a synonym to "--detach"
68 option to create a new worktree without being on a branch.
70 * "format-patch --range-diff=<prev> <origin>..HEAD" has been taught
71 not to ignore <origin> when <prev> is a single version.
73 * "add -p" now allows editing paths that were only added in intent.
75 * The 'meld' backend of the "git mergetool" learned to give the
76 underlying 'meld' the '--auto-merge' option, which would help
77 reduce the amount of text that requires manual merging.
79 * "git for-each-ref" and friends that list refs used to allow only
80 one --merged or --no-merged to filter them; they learned to take
81 combination of both kind of filtering.
84 Performance, Internal Implementation, Development Support etc.
86 * The changed-path Bloom filter is improved using ideas from an
87 independent implementation.
89 * Updates to the changed-paths bloom filter.
91 * The test framework has been updated so that most tests will run
92 with predictable (artificial) timestamps.
94 * Preliminary clean-up of the refs API in preparation for adding a
95 new refs backend "reftable".
97 * Dev support to limit the use of test_must_fail to only git commands.
99 * While packing many objects in a repository with a promissor remote,
100 lazily fetching missing objects from the promissor remote one by
101 one may be inefficient---the code now attempts to fetch all the
102 missing objects in batch (obviously this won't work for a lazy
103 clone that lazily fetches tree objects as you cannot even enumerate
104 what blobs are missing until you learn which trees are missing).
106 * The pretend-object mechanism checks if the given object already
107 exists in the object store before deciding to keep the data
108 in-core, but the check would have triggered lazy fetching of such
109 an object from a promissor remote.
111 * The argv_array API is useful for not just managing argv but any
112 "vector" (NULL-terminated array) of strings, and has seen adoption
113 to a certain degree. It has been renamed to "strvec" to reduce the
116 * The final leg of SHA-256 transition plus doc updates. Note that
117 there is no inter-operability between SHA-1 and SHA-256
120 * CMake support to build with MSVC for Windows bypassing the Makefile.
122 * A new helper function has_object() has been introduced to make it
123 easier to mark object existence checks that do and don't want to
124 trigger lazy fetches, and a few such checks are converted using it.
126 * A no-op replacement function implemented as a C preprocessor macro
127 does not perform as good a job as one implemented as a "static
128 inline" function in catching errors in parameters; replace the
129 former with the latter in <git-compat-util.h> header.
131 * Test framework update.
132 (merge d572f52a64 es/test-cmp-typocatcher later to maint).
134 * Updates to "git merge" tests, in preparation for a new merge
137 * midx and commit-graph files now use the byte defined in their file
138 format specification for identifying the hash function used for
141 * The FETCH_HEAD is now always read from the filesystem regardless of
142 the ref backend in use, as its format is much richer than the
143 normal refs, and written directly by "git fetch" as a plain file..
145 * A handful of places in in-tree code still relied on being able to
146 execute the git subcommands, especially built-ins, in "git-foo"
147 form, which have been corrected.
149 * An unused binary has been discarded, and and a bunch of commands
150 have been turned into into built-in.
152 * A handful of places in in-tree code still relied on being able to
153 execute the git subcommands, especially built-ins, in "git-foo"
154 form, which have been corrected.
156 * When a packfile is removed by "git repack", multi-pack-index gets
157 cleared; the code was taught to do so less aggressively by first
158 checking if the midx actually refers to a pack that no longer
161 * Internal API clean-up to handle two options "diff-index" and "log"
162 have, which happen to share the same short form, more sensibly.
164 * The "add -i/-p" machinery has been written in C but it is not used
165 by default yet. It is made default to those who are participating
166 in feature.experimental experiment.
168 * Allow maintainers to tweak $(TAR) invocations done while making
169 distribution tarballs.
171 * "git index-pack" learned to resolve deltified objects with greater
178 * "git clone --separate-git-dir=$elsewhere" used to stomp on the
179 contents of the existing directory $elsewhere, which has been
180 taught to fail when $elsewhere is not an empty directory.
181 (merge dfaa209a79 bw/fail-cloning-into-non-empty later to maint).
183 * With the base fix to 2.27 regresion, any new extensions in a v0
184 repository would still be silently honored, which is not quite
185 right. Instead, complain and die loudly.
186 (merge ec91ffca04 jk/reject-newer-extensions-in-v0 later to maint).
188 * Fetching from a lazily cloned repository resulted at the server
189 side in attempts to lazy fetch objects that the client side has,
190 many of which will not be available from the third-party anyway.
191 (merge 77aa0941ce jt/avoid-lazy-fetching-upon-have-check later to maint).
193 * Fix to an ancient bug caused by an over-eager attempt for
195 (merge a98f7fb366 rs/add-index-entry-optim-fix later to maint).
197 * Pushing a ref whose name contains non-ASCII character with the
198 "--force-with-lease" option did not work over smart HTTP protocol,
199 which has been corrected.
200 (merge cd85b447bf bc/push-cas-cquoted-refname later to maint).
202 * "git mv src dst", when src is an unmerged path, errored out
203 correctly but with an incorrect error message to claim that src is
204 not tracked, which has been clarified.
205 (merge 9b906af657 ct/mv-unmerged-path-error later to maint).
207 * Fix to a regression introduced during 2.27 cycle.
208 (merge cada7308ad en/fill-directory-exponential later to maint).
210 * Command line completion (in contrib/) update.
211 (merge 688b87c81b mp/complete-show-color-moved later to maint).
213 * All "mergy" operations that internally use the merge-recursive
214 machinery should honor the merge.renormalize configuration, but
217 * Doc cleanup around "worktree".
218 (merge dc9c144be5 es/worktree-doc-cleanups later to maint).
220 * The "git blame --first-parent" option was not documented, but now
222 (merge 11bc12ae1e rp/blame-first-parent-doc later to maint).
224 * The logic to find the ref transaction hook script attempted to
225 cache the path to the found hook without realizing that it needed
226 to keep a copied value, as the API it used returned a transitory
227 buffer space. This has been corrected.
228 (merge 09b2aa30c9 ps/ref-transaction-hook later to maint).
230 * Recent versions of "git diff-files" shows a diff between the index
231 and the working tree for "intent-to-add" paths as a "new file"
232 patch; "git apply --cached" should be able to take "git diff-files"
233 and should act as an equivalent to "git add" for the path, but the
234 command failed to do so for such a path.
235 (merge 4c025c667e rp/apply-cached-with-i-t-a later to maint).
237 * "git diff [<tree-ish>] $path" for a $path that is marked with i-t-a
238 bit was not showing the mode bits from the working tree.
239 (merge cb0dd22b82 rp/ita-diff-modefix later to maint).
241 * Ring buffer with size 4 used for bin-hex translation resulted in a
242 wrong object name in the sequencer's todo output, which has been
244 (merge 5da69c0dac ak/sequencer-fix-find-uniq-abbrev later to maint).
246 * When given more than one target line ranges, "git blame -La,b
247 -Lc,d" was over-eager to coalesce groups of original lines and
248 showed incorrect results, which has been corrected.
249 (merge c2ebaa27d6 jk/blame-coalesce-fix later to maint).
251 * The regexp to identify the function boundary for FORTRAN programs
253 (merge 75c3b6b2e8 pb/userdiff-fortran-update later to maint).
255 * A few end-user facing messages have been updated to be
256 hash-algorithm agnostic.
257 (merge 4279000d3e jc/object-names-are-not-sha-1 later to maint).
259 * "unlink" emulation on MinGW has been optimized.
260 (merge 680e0b4524 jh/mingw-unlink later to maint).
262 * The purpose of "git init --separate-git-dir" is to initialize a
263 new project with the repository separate from the working tree,
264 or, in the case of an existing project, to move the repository
265 (the .git/ directory) out of the working tree. It does not make
266 sense to use --separate-git-dir with a bare repository for which
267 there is no working tree, so disallow its use with bare
269 (merge ccf236a23a es/init-no-separate-git-dir-in-bare later to maint).
271 * "ls-files -o" mishandled the top-level directory of another git
272 working tree that hangs in the current git working tree.
273 (merge ab282aa548 en/dir-nonbare-embedded later to maint).
275 * Fix some incorrect UNLEAK() annotations.
276 (merge 3e19816dc0 jk/unleak-fixes later to maint).
278 * Use more buffered I/O where we used to call many small write(2)s.
279 (merge a698d67b08 rs/more-buffered-io later to maint).
281 * The patch-id computation did not ignore the "incomplete last line"
282 marker like whitespaces.
283 (merge 82a62015a7 rs/patch-id-with-incomplete-line later to maint).
285 * Updates into a lazy/partial clone with a submodule did not work
286 well with transfer.fsckobjects set.
288 * The parser for "git for-each-ref --format=..." was too loose when
289 parsing the "%(trailers...)" atom, and forgot that "trailers" and
290 "trailers:<modifiers>" are the only two allowed forms, which has
292 (merge 2c22e102f8 hv/ref-filter-trailers-atom-parsing-fix later to maint).
294 * Long ago, we decided to use 3 threads by default when running the
295 index-pack task in parallel, which has been adjusted a bit upwards.
296 (merge fbff95b67f jk/index-pack-w-more-threads later to maint).
298 * "git restore/checkout --no-overlay" with wildcarded pathspec
299 mistakenly removed matching paths in subdirectories, which has been
301 (merge bfda204ade rs/checkout-no-overlay-pathspec-fix later to maint).
303 * The description of --cached/--index options in "git apply --help"
305 (merge d064702be3 rp/apply-cached-doc later to maint).
307 * Feeding "$ZERO_OID" to "git log --ignore-missing --stdin", and
308 running "git log --ignore-missing $ZERO_OID" fell back to start
309 digging from HEAD; it has been corrected to become a no-op, like
310 "git log --tags=no-tag-matches-this-pattern" does.
311 (merge 04a0e98515 jk/rev-input-given-fix later to maint).
313 * Various callers of run_command API has been modernized.
314 (merge afbdba391e jc/run-command-use-embedded-args later to maint).
316 * List of options offered and accepted by "git add -i/-p" were
317 inconsistent, which have been corrected.
318 (merge ce910287e7 pw/add-p-allowed-options-fix later to maint).
320 * Various callers of run_command API has been modernized.
321 (merge afbdba391e jc/run-command-use-embedded-args later to maint).
323 * "git diff --stat -w" showed 0-line changes for paths whose changes
324 were only whitespaces, which was not intuitive. We now omit such
325 paths from the stat output.
326 (merge 1cf3d5db9b mr/diff-hide-stat-wo-textual-change later to maint).
328 * It was possible for xrealloc() to send a non-NULL pointer that has
329 been freed, which has been fixed.
330 (merge 6479ea4a8a jk/xrealloc-avoid-use-after-free later to maint).
332 * "git status" has trouble showing where it came from by interpreting
333 reflog entries that record certain events, e.g. "checkout @{u}", and
334 gives a hard/fatal error. Even though it inherently is impossible
335 to give a correct answer because the reflog entries lose some
336 information (e.g. "@{u}" does not record what branch the user was
337 on hence which branch 'the upstream' needs to be computed, and even
338 if the record were available, the relationship between branches may
339 have changed), at least hide the error to allow "status" show its
342 * "git status --short" quoted a path with SP in it when tracked, but
343 not those that are untracked, ignored or unmerged. They are all
344 shown quoted consistently.
346 * "git diff/show" on a change that involves a submodule used to read
347 the information on commits in the submodule from a wrong repository
348 and gave a wrong information when the commit-graph is involved.
349 (merge 85a1ec2c32 mf/submodule-summary-with-correct-repository later to maint).
351 * Unlike "git config --local", "git config --worktree" did not fail
352 early and cleanly when started outside a git repository.
353 (merge 378fe5fc3d mt/config-fail-nongit-early later to maint).
355 * There is a logic to estimate how many objects are in the
356 repository, which is mean to run once per process invocation, but
357 it ran every time the estimated value was requested.
358 (merge 67bb65de5d jk/dont-count-existing-objects-twice later to maint).
360 * "git remote set-head" that failed still said something that hints
361 the operation went through, which was misleading.
362 (merge 5a07c6c3c2 cs/don-t-pretend-a-failed-remote-set-head-succeeded later to maint).
364 * "git fetch --all --ipv4/--ipv6" forgot to pass the protocol options
365 to instances of the "git fetch" that talk to individual remotes,
366 which has been corrected.
367 (merge 4e735c1326 ar/fetch-ipversion-in-all later to maint).
369 * Other code cleanup, docfix, build fix, etc.
370 (merge 84544f2ea3 sk/typofixes later to maint).
371 (merge b17f411ab5 ar/help-guides-doc later to maint).
372 (merge 98c6871fad rs/grep-simpler-parse-object-or-die-call later to maint).
373 (merge 861c4ce141 en/typofixes later to maint).
374 (merge 60e47f6773 sg/ci-git-path-fix-with-pyenv later to maint).
375 (merge e2bfa50ac3 jb/doc-packfile-name later to maint).
376 (merge 918d8ff780 es/worktree-cleanup later to maint).
377 (merge dc156bc31f ma/t1450-quotefix later to maint).
378 (merge 56e743426b en/merge-recursive-comment-fixes later to maint).
379 (merge 7d23ff818f rs/bisect-oid-to-hex-fix later to maint).
380 (merge de20baf2c9 ny/notes-doc-sample-update later to maint).
381 (merge f649aaaf82 so/rev-parser-errormessage-fix later to maint).
382 (merge 6103d58b7f bc/sha-256-cvs-svn-updates later to maint).
383 (merge ac900fddb7 ma/stop-progress-null-fix later to maint).
384 (merge e767963ab6 rs/upload-pack-sigchain-fix later to maint).
385 (merge a831908599 rs/preserve-merges-unused-code-removal later to maint).
386 (merge 6dfefe70a9 jb/commit-graph-doc-fix later to maint).
387 (merge 847b37271e pb/set-url-docfix later to maint).
388 (merge 748f733d54 mt/checkout-entry-dead-code-removal later to maint).
389 (merge ce820cbd58 dl/subtree-docs later to maint).
390 (merge 55fe225dde jk/leakfix later to maint).
391 (merge ee22a29215 so/pretty-abbrev-doc later to maint).
392 (merge 3100fd5588 jc/post-checkout-doc later to maint).
393 (merge 17bae89476 pb/doc-external-diff-env later to maint).
394 (merge 27ed6ccc12 jk/worktree-check-clean-leakfix later to maint).
395 (merge 1302badd16 ea/blame-use-oideq later to maint).
396 (merge e6d5a11fed al/t3200-back-on-a-branch later to maint).
397 (merge 324efcf6b6 pw/add-p-leakfix later to maint).
398 (merge 1c6ffb546b jk/add-i-fixes later to maint).
399 (merge e40e936551 cd/commit-graph-doc later to maint).
projects / git / blob