4 Backward compatibility notes.
6 * Use of an empty string that is used for 'everything matches' is
7 still warned and Git asks users to use a more explicit '.' for that
8 instead. The hope is that existing users will not mind this
9 change, and eventually the warning can be turned into a hard error,
10 upgrading the deprecation into removal of this (mis)feature. That
11 is not scheduled to happen in the upcoming release (yet).
13 * The historical argument order "git merge <msg> HEAD <commit>..."
14 has been deprecated for quite some time, and will be removed in the
21 UI, Workflows & Features
23 * Various updates to "git p4".
25 * "git p4" didn't interact with the internal of .git directory
26 correctly in the modern "git-worktree"-enabled world.
28 * "git branch --list" and friends learned "--ignore-case" option to
29 optionally sort branches and tags case insensitively.
31 * In addition to %(subject), %(body), "log --pretty=format:..."
32 learned a new placeholder %(trailers).
34 * "git rebase" learned "--quit" option, which allows a user to
35 remove the metadata left by an earlier "git rebase" that was
36 manually aborted without using "git rebase --abort".
38 * "git clone --reference $there --recurse-submodules $super" has been
39 taught to guess repositories usable as references for submodules of
40 $super that are embedded in $there while making a clone of the
41 superproject borrow objects from $there; extend the mechanism to
42 also allow submodules of these submodules to borrow repositories
43 embedded in these clones of the submodules embedded in the clone of
46 * Porcelain scripts written in Perl are getting internationalized.
48 * "git merge --continue" has been added as a synonym to "git commit"
49 to conclude a merge that has stopped due to conflicts.
51 * Finer-grained control of what protocols are allowed for transports
52 during clone/fetch/push have been enabled via a new configuration
55 * "git shortlog" learned "--committer" option to group commits by
56 committer, instead of author.
58 * GitLFS integration with "git p4" has been updated.
60 * The isatty() emulation for Windows has been updated to eradicate
61 the previous hack that depended on internals of (older) MSVC
64 * Some platforms no longer understand "latin-1" that is still seen in
65 the wild in e-mail headers; replace them with "iso-8859-1" that is
66 more widely known when conversion fails from/to it.
67 (merge df3755888b jc/latin-1 later to maint).
70 Performance, Internal Implementation, Development Support etc.
72 * Commands that operate on a log message and add lines to the trailer
73 blocks, such as "format-patch -s", "cherry-pick (-x|-s)", and
74 "commit -s", have been taught to use the logic of and share the
75 code with "git interpret-trailer".
77 * The default Travis-CI configuration specifies newer P4 and GitLFS.
78 (merge 5f703e8f02 ls/travis-update-p4-and-lfs later to maint).
80 * The "fast hash" that had disastrous performance issues in some
81 corner cases has been retired from the internal diff.
83 * The character width table has been updated to match Unicode 9.0
84 (merge 9e6e9aefdf bb/unicode-9.0 later to maint).
86 * Update the procedure to generate "tags" for developer support.
87 (merge 046e4c1c09 jk/make-tags-find-sources-tweak later to maint).
89 * The codeflow of setting NOATIME and CLOEXEC on file descriptors Git
90 opens has been simplified.
91 (merge b4d065df03 jc/git-open-cloexec later to maint).
93 * "git diff" and its family had two experimental heuristics to shift
94 the contents of a hunk to make the patch easier to read. One of
95 them turns out to be better than the other, so leave only the
96 "--indent-heuristic" option and remove the other one.
97 (merge 3cde4e02ee jc/retire-compaction-heuristics later to maint).
99 * A new submodule helper "git submodule embedgitdirs" to make it
100 easier to move embedded .git/ directory for submodules in a
101 superproject to .git/modules/ (and point the latter with the former
102 that is turned into a "gitdir:" file) has been added.
105 Also contains various documentation updates and code clean-ups.
110 Unless otherwise noted, all the fixes since v2.9 in the maintenance
111 track are contained in this release (see the maintenance releases'
114 * We often decide if a session is interactive by checking if the
115 standard I/O streams are connected to a TTY, but isatty() that
116 comes with Windows incorrectly returned true if it is used on NUL
117 (i.e. an equivalent to /dev/null). This has been fixed.
118 (merge cbb3f3c9b1 js/mingw-isatty later to maint).
120 * "git svn" did not work well with path components that are "0", and
121 some configuration variable it uses were not documented.
122 (merge ea9a93dcc2 ew/svn-fixes later to maint).
124 * "git rev-parse --symbolic" failed with a more recent notation like
125 "HEAD^-1" and "HEAD^!".
126 (merge a2e7b04c44 jk/rev-parse-symbolic-parents-fix later to maint).
128 * An empty directory in a working tree that can simply be nuked used
129 to interfere while merging or cherry-picking a change to create a
130 submodule directory there, which has been fixed..
131 (merge 5423d2e700 dt/empty-submodule-in-merge later to maint).
133 * The code in "git push" to compute if any commit being pushed in the
134 superproject binds a commit in a submodule that hasn't been pushed
135 out was overly inefficient, making it unusable even for a small
136 project that does not have any submodule but have a reasonable
138 (merge 250ab24ab3 hv/submodule-not-yet-pushed-fix later to maint).
140 * "git push --dry-run --recurse-submodule=on-demand" wasn't
141 "--dry-run" in the submodules.
142 (merge 0301c821c5 bw/push-dry-run later to maint).
144 * The output from "git worktree list" was made in readdir() order,
146 (merge 4df1d4d466 nd/worktree-list-fixup later to maint).
148 * mergetool.<tool>.trustExitCode configuration variable did not apply
149 to built-in tools, but now it does.
150 (merge 2967284456 da/mergetool-trust-exit-code later to maint).
152 * "git p4" LFS support was broken when LFS stores an empty blob.
153 (merge d5eb3cf5e7 ls/p4-empty-file-on-lfs later to maint).
155 * A corner case in merge-recursive regression that crept in
156 during 2.10 development cycle has been fixed.
157 (merge 1c25d2d8ed jc/renormalize-merge-kill-safer-crlf later to maint).
159 * Transport with dumb http can be fooled into following foreign URLs
160 that the end user does not intend to, especially with the server
161 side redirects and http-alternates mechanism, which can lead to
162 security issues. Tighten the redirection and make it more obvious
163 to the end user when it happens.
164 (merge cb4d2d35c4 jk/http-walker-limit-redirect-2.9 later to maint).
166 * Update the error messages from the dumb-http client when it fails
167 to obtain loose objects; we used to give sensible error message
168 only upon 404 but we now forbid unexpected redirects that needs to
169 be reported with something sensible.
170 (merge 3680f16f9d jk/http-walker-limit-redirect later to maint).
172 * When diff.renames configuration is on (and with Git 2.9 and later,
173 it is enabled by default, which made it worse), "git stash"
174 misbehaved if a file is removed and another file with a very
175 similar content is added.
176 (merge 9d4e28ead5 jk/stash-disable-renames-internally later to maint).
178 * "git diff --no-index" did not take "--no-abbrev" option.
179 (merge 43d1948b7b jb/diff-no-index-no-abbrev later to maint).
181 * "git difftool --dir-diff" had a minor regression when started from
182 a subdirectory, which has been fixed.
183 (merge 853e10c197 da/difftool-dir-diff-fix later to maint).
185 * "git commit --allow-empty --only" (no pathspec) with dirty index
186 ought to be an acceptable way to create a new commit that does not
187 change any paths, but it was forbidden, perhaps because nobody
189 (merge beb635ca9c ak/commit-only-allow-empty later to maint).
191 * Git 2.11 had a minor regression in "merge --ff-only" that competed
192 with another process that simultanously attempted to update the
193 index. We used to explain what went wrong with an error message,
194 but the new code silently failed. The error message has been
197 * A pathname that begins with "//" or "\\" on Windows is special but
198 path normalization logic was unaware of it.
199 (merge 7814fbe3f1 js/normalize-path-copy-ceil later to maint).
201 * "git pull --rebase", when there is no new commits on our side since
202 we forked from the upstream, should be able to fast-forward without
203 invoking "git rebase", but it didn't.
204 (merge 33b842a1e9 jc/pull-rebase-ff later to maint).
206 * The way to specify hotkeys to "xxdiff" that is used by "git
207 mergetool" has been modernized to match recent versions of xxdiff.
208 (merge 6cf5f6cef7 da/mergetool-xxdiff-hotkey later to maint).
210 * Unlike "git am --abort", "git cherry-pick --abort" moved HEAD back
211 to where cherry-pick started while picking multiple changes, when
212 the cherry-pick stopped to ask for help from the user, and the user
213 did "git reset --hard" to a different commit in order to re-attempt
215 (merge ce73bb22d8 sb/sequencer-abort-safety later to maint).
217 * Code cleanup in shallow boundary computation.
218 (merge 649b0c316a nd/shallow-fixup later to maint).
220 * A recent update to receive-pack to make it easier to drop garbage
221 objects made it clear that GIT_ALTERNATE_OBJECT_DIRECTORIES cannot
222 have a pathname with a colon in it (no surprise!), and this in turn
223 made it impossible to push into a repository at such a path. This
224 has been fixed by introducing a quoting mechanism used when
225 appending such a path to the colon-separated list.
226 (merge 5e74824fac jk/quote-env-path-list-component later to maint).
228 * The function usage_msg_opt() has been updated to say "fatal:"
229 before the custom message programs give, when they want to die
230 with a message about wrong command line options followed by the
231 standard usage string.
232 (merge 87433261a4 jk/parseopt-usage-msg-opt later to maint).
234 * "git index-pack --stdin" needs an access to an existing repository,
235 but "git index-pack file.pack" to generate an .idx file that
236 corresponds to a packfile does not.
237 (merge 29401e1575 jk/index-pack-wo-repo-from-stdin later to maint).
239 * Fix for NDEBUG builds.
240 (merge 08414938a2 jt/mailinfo-fold-in-body-headers later to maint).
242 * A lazy "git push" without refspec did not internally use a fully
243 specified refspec to perform 'current', 'simple', or 'upstream'
244 push, causing unnecessary "ambiguous ref" errors.
245 (merge b284495e93 jc/push-default-explicit later to maint).
247 * "git p4" misbehaved when swapping a directory and a symbolic link.
248 (merge df8a9e86db ld/p4-compare-dir-vs-symlink later to maint).
250 * Even though an fix was attempted in Git 2.9.3 days, but running
251 "git difftool --dir-diff" from a subdirectory never worked. This
253 (merge ce6926974e jk/difftool-in-subdir later to maint).
255 * "git p4" that tracks multile p4 paths imported a single changelist
256 that touches files in these multiple paths as one commit, followed
257 by many empty commits. This has been fixed.
258 (merge 9943e5b979 gv/p4-multi-path-commit-fix later to maint).
260 * A potential but unlikely buffer overflow in Windows port has been
262 (merge c46458e82f mk/mingw-winansi-ttyname-termination-fix later to maint).
264 * When the http server gives an incomplete response to a smart-http
265 rpc call, it could lead to client waiting for a full response that
266 will never come. Teach the client side to notice this condition
267 and abort the transfer.
268 (merge f8edeaa05d dt/smart-http-detect-server-going-away later to maint).
270 * Compression setting for producing packfiles were spread across
271 three codepaths, one of which did not honor any configuration.
272 Unify these so that all of them honor core.compression and
273 pack.compression variables the same way.
274 (merge 8de7eeb54b jc/compression-config later to maint).
276 * "git fast-import" sometimes mishandled while rebalancing notes
277 tree, which has been fixed.
278 (merge 405d7f4af6 mh/fast-import-notes-fix-new later to maint).
280 * Recent update to the default abbreviation length that auto-scales
281 lacked documentation update, which has been corrected.
282 (merge 48d5014dd4 jc/abbrev-autoscale-config later to maint).
284 * Leakage of lockfiles in the config subsystem has been fixed.
285 (merge c06fa62dfc nd/config-misc-fixes later to maint).
287 * Other minor doc, test and build updates and code cleanups.
288 (merge fa6ca11105 nd/qsort-in-merge-recursive later to maint).
289 (merge fa3142c919 ak/lazy-prereq-mktemp later to maint).
290 (merge 9c48b4fb23 ls/t0021-fixup later to maint).
291 (merge 584f99c87b sb/unpack-trees-grammofix later to maint).
292 (merge 54471fdcc3 jk/readme-gmane-is-no-more later to maint).
293 (merge 9e189f1a5c sb/t3600-cleanup later to maint).
294 (merge e2c20be57c lr/doc-fix-cet later to maint).
295 (merge 47437fd3bd kh/tutorial-grammofix later to maint).
296 (merge f2627d9b19 sb/submodule-config-cleanup later to maint).
297 (merge 7eeda8b821 ls/filter-process later to maint).
298 (merge 6cc823c5c1 jt/fetch-no-redundant-tag-fetch-map later to maint).
299 (merge 235ec24352 mm/push-social-engineering-attack-doc later to maint).
300 (merge f1350d0c12 mm/gc-safety-doc later to maint).