4 Backward compatibility notes
5 ----------------------------
7 When "git push [$there]" does not say what to push, we have used the
8 traditional "matching" semantics so far (all your branches were sent
9 to the remote as long as there already are branches of the same name
10 over there). In Git 2.0, the default is now the "simple" semantics,
13 - only the current branch to the branch with the same name, and only
14 when the current branch is set to integrate with that remote
15 branch, if you are pushing to the same remote as you fetch from; or
17 - only the current branch to the branch with the same name, if you
18 are pushing to a remote that is not where you usually fetch from.
20 You can use the configuration variable "push.default" to change
21 this. If you are an old-timer who wants to keep using the
22 "matching" semantics, you can set the variable to "matching", for
23 example. Read the documentation for other possibilities.
25 When "git add -u" and "git add -A" are run inside a subdirectory
26 without specifying which paths to add on the command line, they
27 operate on the entire tree for consistency with "git commit -a" and
28 other commands (these commands used to operate only on the current
29 subdirectory). Say "git add -u ." or "git add -A ." if you want to
30 limit the operation to the current directory.
32 "git add <path>" is the same as "git add -A <path>" now, so that
33 "git add dir/" will notice paths you removed from the directory and
34 record the removal. In older versions of Git, "git add <path>" used
35 to ignore removals. You can say "git add --ignore-removal <path>" to
36 add only added or modified paths in <path>, if you really want to.
38 The "-q" option to "git diff-files", which does *NOT* mean "quiet",
39 has been removed (it told Git to ignore deletion, which you can do
40 with "git diff-files --diff-filter=d").
42 "git request-pull" lost a few "heuristics" that often led to mistakes.
44 The default prefix for "git svn" has changed in Git 2.0. For a long
45 time, "git svn" created its remote-tracking branches directly under
46 refs/remotes, but it now places them under refs/remotes/origin/ unless
47 it is told otherwise with its "--prefix" option.
50 Updates since v1.9 series
51 -------------------------
53 UI, Workflows & Features
55 * The "multi-mail" post-receive hook (in contrib/) has been updated
56 to a more recent version from upstream.
58 * The "remote-hg/bzr" remote-helper interfaces (in contrib/) are
59 now maintained separately as a third-party plug-in.
61 * "git gc --aggressive" learned "--depth" option and
62 "gc.aggressiveDepth" configuration variable to allow use of a less
63 insane depth than the built-in default value of 250.
65 * "git log" learned the "--show-linear-break" option to show where a
66 single strand-of-pearls is broken in its output.
68 * The "rev-parse --parseopt" mechanism used by scripted Porcelains to
69 parse command-line options and to give help text learned to take
70 the argv-help (the placeholder string for an option parameter,
71 e.g. "key-id" in "--gpg-sign=<key-id>").
73 * The pattern to find where the function begins in C/C++ used in
74 "diff" and "grep -p" has been updated to improve viewing C++
77 * "git rebase" learned to interpret a lone "-" as "@{-1}", the
78 branch that we were previously on.
80 * "git commit --cleanup=<mode>" learned a new mode, scissors.
82 * "git tag --list" output can be sorted using "version sort" with
83 "--sort=version:refname".
85 * Discard the accumulated "heuristics" to guess from which branch the
86 result wants to be pulled from and make sure that what the end user
87 specified is not second-guessed by "git request-pull", to avoid
88 mistakes. When you pushed out your 'master' branch to your public
89 repository as 'for-linus', use the new "master:for-linus" syntax to
90 denote the branch to be pulled.
92 * "git grep" learned to behave in a way similar to native grep when
93 "-h" (no header) and "-c" (count) options are given.
95 * "git push" via transport-helper interface has been updated to
96 allow forced ref updates in a way similar to the natively
99 * The "simple" mode is the default for "git push".
101 * "git add -u" and "git add -A", when run without any pathspec, is a
102 tree-wide operation even when run inside a subdirectory of a
105 * "git add <path>" is the same as "git add -A <path>" now.
107 * "core.statinfo" configuration variable, which is a
108 never-advertised synonym to "core.checkstat", has been removed.
110 * The "-q" option to "git diff-files", which does *NOT* mean
111 "quiet", has been removed (it told Git to ignore deletion, which
112 you can do with "git diff-files --diff-filter=d").
114 * Server operators can loosen the "tips of refs only" restriction for
115 the remote archive service with the uploadarchive.allowUnreachable
116 configuration option.
118 * The progress indicators from various time-consuming commands have
119 been marked for i18n/l10n.
121 * "git notes -C <blob>" diagnoses as an error an attempt to use an
122 object that is not a blob.
124 * "git config" learned to read from the standard input when "-" is
125 given as the value to its "--file" parameter (attempting an
126 operation to update the configuration in the standard input is
127 rejected, of course).
129 * Trailing whitespaces in .gitignore files, unless they are quoted
130 for fnmatch(3), e.g. "path\ ", are warned and ignored. Strictly
131 speaking, this is a backward-incompatible change, but very unlikely
132 to bite any sane user and adjusting should be obvious and easy.
134 * Many commands that create commits, e.g. "pull" and "rebase",
135 learned to take the "--gpg-sign" option on the command line.
137 * "git commit" can be told to always GPG sign the resulting commit
138 by setting the "commit.gpgsign" configuration variable to "true"
139 (the command-line option "--no-gpg-sign" should override it).
141 * "git pull" can be told to only accept fast-forward by setting the
142 new "pull.ff" configuration variable.
144 * "git reset" learned the "-N" option, which does not reset the index
145 fully for paths the index knows about but the tree-ish the command
146 resets to does not (these paths are kept as intend-to-add entries).
149 Performance, Internal Implementation, etc.
151 * The compilation options to port to AIX and to MSVC have been
154 * We started using wildmatch() in place of fnmatch(3) a few releases
155 ago; complete the process and stop using fnmatch(3).
157 * Uses of curl's "multi" interface and "easy" interface do not mix
158 well when we attempt to reuse outgoing connections. Teach the RPC
159 over HTTP code, used in the smart HTTP transport, not to use the
162 * The bitmap-index feature from JGit has been ported, which should
163 significantly improve performance when serving objects from a
164 repository that uses it.
166 * The way "git log --cc" shows a combined diff against multiple
167 parents has been optimized.
169 * The prefixcmp() and suffixcmp() functions are gone. Use
170 starts_with() and ends_with(), and also consider if skip_prefix()
171 suits your needs better when using the former.
174 Also contains various documentation updates and code clean-ups. Many
175 of them came from flurry of activities as GSoC candidate microproject
179 Fixes since v1.9 series
180 -----------------------
182 Unless otherwise noted, all the fixes since v1.9 in the maintenance
183 track are contained in this release (see the maintenance releases'
186 * "git p4" was broken in 1.9 release to deal with changes in binary
188 (merge 749b668 cl/p4-use-diff-tree later to maint).
190 * The shell prompt script (in contrib/), when using the PROMPT_COMMAND
191 interface, used an unsafe construct when showing the branch name in
193 (merge 8976500 rh/prompt-pcmode-avoid-eval-on-refname later to maint).
195 * The remote-helper interface to fast-import/fast-export via the
196 transport-helper has been tightened to avoid leaving the import
197 marks file from a failed/crashed run, as such a file that is out-of-
198 sync with reality confuses a later invocation of itself.
200 * "git rebase" used a POSIX shell construct FreeBSD's /bin/sh does not
202 (merge 8cd6596 km/avoid-non-function-return-in-rebase later to maint).
204 * zsh prompt (in contrib/) leaked unnecessary error messages.
206 * Bash completion (in contrib/) did not complete the refs and remotes
207 correctly given "git pu<TAB>" when "pu" is aliased to "push".
209 * Some more Unicode code points, defined in Unicode 6.3 as having zero
210 width, have been taught to our display column counting logic.
211 (merge d813ab9 tb/unicode-6.3-zero-width later to maint).
213 * Some tests used shell constructs that did not work well on FreeBSD
214 (merge ff7a1c6 km/avoid-bs-in-shell-glob later to maint).
215 (merge 00764ca km/avoid-cp-a later to maint).
217 * "git update-ref --stdin" did not fail a request to create a ref
218 when the ref already existed.
219 (merge b9d56b5 mh/update-ref-batch-create-fix later to maint).
221 * "git diff --no-index -Mq a b" fell into an infinite loop.
222 (merge ad1c3fb jc/fix-diff-no-index-diff-opt-parse later to maint).
224 * "git fetch --prune", when the right-hand side of multiple fetch
225 refspecs overlap (e.g. storing "refs/heads/*" to
226 "refs/remotes/origin/*", while storing "refs/frotz/*" to
227 "refs/remotes/origin/fr/*"), aggressively thought that lack of
228 "refs/heads/fr/otz" on the origin site meant we should remove
229 "refs/remotes/origin/fr/otz" from us, without checking their
230 "refs/frotz/otz" first.
232 Note that such a configuration is inherently unsafe (think what
233 should happen when "refs/heads/fr/otz" does appear on the origin
234 site), but that is not a reason not to be extra careful.
235 (merge e6f6371 cn/fetch-prune-overlapping-destination later to maint).
237 * "git status --porcelain --branch" showed its output with labels
238 "ahead/behind/gone" translated to the user's locale.
239 (merge 7a76c28 mm/status-porcelain-format-i18n-fix later to maint).
241 * A stray environment variable $prefix could have leaked into and
242 affected the behaviour of the "subtree" script (in contrib/).
244 * When it is not necessary to edit a commit log message (e.g. "git
245 commit -m" is given a message without specifying "-e"), we used to
246 disable the spawning of the editor by overriding GIT_EDITOR, but
247 this means all the uses of the editor, other than to edit the
248 commit log message, are also affected.
249 (merge b549be0 bp/commit-p-editor later to maint).
251 * "git mv" that moves a submodule forgot to adjust the array that
252 uses to keep track of which submodules were to be moved to update
254 (merge fb8a4e8 jk/mv-submodules-fix later to maint).
256 * Length limit for the pathname used when removing a path in a deep
257 subdirectory has been removed to avoid buffer overflows.
258 (merge 2f29e0c mh/remove-subtree-long-pathname-fix later to maint).
260 * The test helper lib-terminal always run an actual test_expect_*
261 when included, which screwed up with the use of skil-all that may
262 have to be done later.
263 (merge 7e27173 jk/lib-terminal-lazy later to maint).
265 * "git index-pack" used a wrong variable to name the keep-file in an
266 error message when the file cannot be written or closed.
267 (merge de983a0 nd/index-pack-error-message later to maint).
269 * "rebase -i" produced a broken insn sheet when the title of a commit
270 happened to contain '\n' (or ended with '\c') due to a careless use
272 (merge cb1aefd us/printf-not-echo later to maint).
274 * There were a few instances of 'git-foo' remaining in the
275 documentation that should have been spelled 'git foo'.
276 (merge 3c3e6f5 rr/doc-merge-strategies later to maint).
278 * Serving objects from a shallow repository needs to write a
279 new file to hold the temporary shallow boundaries, but it was not
280 cleaned when we exit due to die() or a signal.
281 (merge 7839632 jk/shallow-update-fix later to maint).
283 * When "git stash pop" stops after failing to apply the stash
284 (e.g. due to conflicting changes), the stash is not dropped. State
285 that explicitly in the output to let the users know.
286 (merge 2d4c993 jc/stash-pop-not-popped later to maint).
288 * The labels in "git status" output that describe the nature of
289 conflicts (e.g. "both deleted") were limited to 20 bytes, which was
290 too short for some l10n (e.g. fr).
291 (merge c7cb333 jn/wt-status later to maint).
293 * "git clean -d pathspec" did not use the given pathspec correctly
294 and ended up cleaning too much.
295 (merge 1f2e108 jk/clean-d-pathspec later to maint).
297 * "git difftool" misbehaved when the repository is bound to the
298 working tree with the ".git file" mechanism, where a textual file
299 ".git" tells us where it is.
300 (merge fcfec8b da/difftool-git-files later to maint).
302 * "git push" did not pay attention to "branch.*.pushremote" if it is
303 defined earlier than "remote.pushdefault"; the order of these two
304 variables in the configuration file should not matter, but it did
306 (merge 98b406f jk/remote-pushremote-config-reading later to maint).
308 * Code paths that parse timestamps in commit objects have been
310 (merge f80d1f9 jk/commit-dates-parsing-fix later to maint).
312 * "git diff --external-diff" incorrectly fed the submodule directory
313 in the working tree to the external diff driver when it knew that it
314 is the same as one of the versions being compared.
315 (merge aba4727 tr/diff-submodule-no-reuse-worktree later to maint).
317 * "git reset" needs to refresh the index when working in a working
318 tree (it can also be used to match the index to the HEAD in an
319 otherwise bare repository), but it failed to set up the working
320 tree properly, causing GIT_WORK_TREE to be ignored.
321 (merge b7756d4 nd/reset-setup-worktree later to maint).
323 * "git check-attr" when working on a repository with a working tree
324 did not work well when the working tree was specified via the
325 "--work-tree" (and obviously with "--git-dir") option.
326 (merge cdbf623 jc/check-attr-honor-working-tree later to maint).
328 * "merge-recursive" was broken in 1.7.7 era and stopped working in
329 an empty (temporary) working tree, when there are renames
330 involved. This has been corrected.
331 (merge 6e2068a bk/refresh-missing-ok-in-merge-recursive later to maint.)
333 * "git rev-parse" was loose in rejecting command-line arguments
334 that do not make sense, e.g. "--default" without the required
335 value for that option.
336 (merge a43219f ds/rev-parse-required-args later to maint.)
338 * "include.path" variable (or any variable that expects a path that
339 can use ~username expansion) in the configuration file is not a
340 boolean, but the code failed to check it.
341 (merge 67beb60 jk/config-path-include-fix later to maint.)
343 * Commands that take pathspecs on the command line misbehaved when
344 the pathspec is given as an absolute pathname (which is a
345 practice not particularly encouraged) that points at a symbolic
346 link in the working tree.
347 (merge later 655ee9e mw/symlinks to maint.)
349 * "git diff --quiet -- pathspec1 pathspec2" sometimes did not return
350 the correct status value.
351 (merge f34b205 nd/diff-quiet-stat-dirty later to maint.)
353 * Attempting to deepen a shallow repository by fetching over smart
354 HTTP transport failed in the protocol exchange, when the no-done
355 extension was used. The fetching side waited for the list of
356 shallow boundary commits after the sending side stopped talking to
358 (merge 0232852 nd/http-fetch-shallow-fix later to maint.)
360 * Allow "git cmd path/", when the 'path' is where a submodule is
361 bound to the top-level working tree, to match 'path', despite the
362 extra and unnecessary trailing slash (such a slash is often
363 given by command-line completion).
364 (merge 2e70c01 nd/submodule-pathspec-ending-with-slash later to maint.)
366 * Documentation and in-code comments had many instances of mistaken
367 use of "nor", which have been corrected.
368 (merge 235e8d5 jl/nor-or-nand-and later to maint).