| blob | c617e37dd8a83da8bb3fab922224e5d6d22adb6b |
1 Git 2.16 Release Notes
2 ======================
4 Backward compatibility notes and other notable changes.
6 * Use of an empty string as a pathspec element that is used for
7 'everything matches' is now an error.
10 Updates since v2.15
11 -------------------
13 UI, Workflows & Features
15 * An empty string as a pathspec element that means "everything"
16 i.e. 'git add ""', is now illegal. We started this by first
17 deprecating and warning a pathspec that has such an element in
18 2.11 (Nov 2016).
20 * A hook script that is set unexecutable is simply ignored. Git
21 notifies when such a file is ignored, unless the message is
22 squelched via advice.ignoredHook configuration.
24 * "git pull" has been taught to accept "--[no-]signoff" option and
25 pass it down to "git merge".
27 * The "--push-option=<string>" option to "git push" now defaults to a
28 list of strings configured via push.pushOption variable.
30 * "gitweb" checks if a directory is searchable with Perl's "-x"
31 operator, which can be enhanced by using "filetest 'access'"
32 pragma, which now we do.
34 * "git stash save" has been deprecated in favour of "git stash push".
36 * The set of paths output from "git status --ignored" was tied
37 closely with its "--untracked=<mode>" option, but now it can be
38 controlled more flexibly. Most notably, a directory that is
39 ignored because it is listed to be ignored in the ignore/exclude
40 mechanism can be handled differently from a directory that ends up
41 to be ignored only because all files in it are ignored.
43 * The remote-helper for talking to MediaWiki has been updated to
44 truncate an overlong pagename so that ".mw" suffix can still be
45 added.
47 * The remote-helper for talking to MediaWiki has been updated to
48 work with mediawiki namespaces.
50 * The "--format=..." option "git for-each-ref" takes learned to show
51 the name of the 'remote' repository and the ref at the remote side
52 that is affected for 'upstream' and 'push' via "%(push:remotename)"
53 and friends.
55 * Doc and message updates to teach users "bisect view" is a synonym
56 for "bisect visualize".
58 * "git bisect run" that did not specify any command to run used to go
59 ahead and treated all commits to be tested as 'good'. This has
60 been corrected by making the command error out.
62 * The SubmittingPatches document has been converted to produce an
63 HTML version via AsciiDoc/Asciidoctor.
64 (merge 049e64aa50 bc/submitting-patches-in-asciidoc later to maint).
66 * We learned to talk to watchman to speed up "git status" and other
67 operations that need to see which paths have been modified.
69 * The "diff" family of commands learned to ignore differences in
70 carriage return at the end of line.
72 * Places that know about "sendemail.to", like documentation and shell
73 completion (in contrib/) have been taught about "sendemail.tocmd",
74 too.
76 * "git add --renormalize ." is a new and safer way to record the fact
77 that you are correcting the end-of-line convention and other
78 "convert_to_git()" glitches in the in-repository data.
80 * "git branch" and "git checkout -b" are now forbidden from creating
81 a branch whose name is "HEAD".
83 * "git branch --list" learned to show its output through the pager by
84 default when the output is going to a terminal, which is controlled
85 by the pager.branch configuration variable. This is similar to a
86 recent change to "git tag --list".
88 * "git grep -W", "git diff -W" and their friends learned a heuristic
89 to extend a pre-context beyond the line that matches the "function
90 pattern" (aka "diff.*.xfuncname") to include a comment block, if
91 exists, that immediately precedes it.
94 Performance, Internal Implementation, Development Support etc.
96 * An earlier update made it possible to use an on-stack in-core
97 lockfile structure (as opposed to having to deliberately leak an
98 on-heap one). Many codepaths have been updated to take advantage
99 of this new facility.
101 * Calling cmd_foo() as if it is a general purpose helper function is
102 a no-no. Correct two instances of such to set an example.
104 * We try to see if somebody runs our test suite with a shell that
105 does not support "local" like bash/dash does.
107 * An early part of piece-by-piece rewrite of "git bisect" in C.
109 * GSoC to piece-by-piece rewrite "git submodule" in C.
111 * Optimize the code to find shortest unique prefix of object names.
113 * Pathspec-limited revision traversal was taught not to keep finding
114 unneeded differences once it knows two trees are different inside
115 given pathspec.
117 * Conversion from uchar[20] to struct object_id continues.
119 * Code cleanup.
121 * A single-word "unsigned flags" in the diff options is being split
122 into a structure with many bitfields.
124 * TravisCI build updates.
126 * Parts of a test to drive the long-running content filter interface
127 has been split into its own module, hopefully to eventually become
128 reusable.
130 * Drop (perhaps overly cautious) sanity check before using the index
131 read from the filesystem at runtime.
133 Also contains various documentation updates and code clean-ups.
136 Fixes since v2.15
137 -----------------
139 * "auto" as a value for the columnar output configuration ought to
140 judge "is the output consumed by humans?" with the same criteria as
141 "auto" for coloured output configuration, i.e. either the standard
142 output stream is going to tty, or a pager is in use. We forgot the
143 latter, which has been fixed.
145 * The experimental "color moved lines differently in diff output"
146 feature was buggy around "ignore whitespace changes" edges, which
147 has been corrected.
149 * Instead of using custom line comparison and hashing functions to
150 implement "moved lines" coloring in the diff output, use the pair
151 of these functions from lower-layer xdiff/ code.
153 * Some codepaths did not check for errors when asking what branch the
154 HEAD points at, which have been fixed.
156 * "git commit", after making a commit, did not check for errors when
157 asking on what branch it made the commit, which has been corrected.
159 * "git status --ignored -u" did not stop at a working tree of a
160 separate project that is embedded in an ignored directory and
161 listed files in that other project, instead of just showing the
162 directory itself as ignored.
164 * A broken access to object databases in recent update to "git grep
165 --recurse-submodules" has been fixed.
167 * A recent regression in "git rebase -i" that broke execution of git
168 commands from subdirectories via "exec" instruction has been fixed.
170 * A (possibly flakey) test fix.
172 * "git check-ref-format --branch @{-1}" bit a "BUG()" when run
173 outside a repository for obvious reasons; clarify the documentation
174 and make sure we do not even try to expand the at-mark magic in
175 such a case, but still call the validation logic for branch names.
177 * "git fetch --recurse-submodules" now knows that submodules can be
178 moved around in the superproject in addition to getting updated,
179 and finds the ones that need to be fetched accordingly.
181 * Command line completion (in contrib/) update.
183 * Description of blame.{showroot,blankboundary,showemail,date}
184 configuration variables have been added to "git config --help".
186 * After an error from lstat(), diff_populate_filespec() function
187 sometimes still went ahead and used invalid data in struct stat,
188 which has been fixed.
190 * UNC paths are also relevant in Cygwin builds and they are now
191 tested just like Mingw builds.
193 * Correct start-up sequence so that a repository could be placed
194 immediately under the root directory again (which was broken at
195 around Git 2.13).
197 * The credential helper for libsecret (in contrib/) has been improved
198 to allow possibly prompting the end user to unlock secrets that are
199 currently locked (otherwise the secrets may not be loaded).
201 * MinGW updates.
203 * Error checking in "git imap-send" for empty response has been
204 improved.
206 * Recent update to the refs infrastructure implementation started
207 rewriting packed-refs file more often than before; this has been
208 optimized again for most trivial cases.
209 (merge 7c6bd25c7d mh/avoid-rewriting-packed-refs later to maint).
211 * Some error messages did not quote filenames shown in it, which have
212 been fixed.
214 * "git rebase -i" recently started misbehaving when a submodule that
215 is configured with 'submodule.<name>.ignore' is dirty; this has
216 been corrected.
218 * Building with NO_LIBPCRE1_JIT did not disable it, which has been fixed.
220 * We used to add an empty alternate object database to the system
221 that does not help anything; it has been corrected.
223 * Doc update around use of "format-patch --subject-prefix" etc.
225 * A fix for an ancient bug in "git apply --ignore-space-change" codepath.
227 * Clarify and enhance documentation for "merge-base --fork-point", as
228 it was clear what it computed but not why/what for.
229 (merge 6d1700b8af jc/merge-base-fork-point-doc later to maint).
231 * A few scripts (both in production and tests) incorrectly redirected
232 their error output. These have been corrected.
233 (merge eadf1c8f45 tz/redirect-fix later to maint).
235 * "git notes" sent its error message to its standard output stream,
236 which was corrected.
237 (merge 89b9e31dd5 tz/notes-error-to-stderr later to maint).
239 * The three-way merge performed by "git cherry-pick" was confused
240 when a new submodule was added in the meantime, which has been
241 fixed (or "papered over").
242 (merge c641ca6707 sb/test-cherry-pick-submodule-getting-in-a-way later to maint).
244 * The sequencer machinery (used by "git cherry-pick A..B", and "git
245 rebase -i", among other things) would have lost a commit if stopped
246 due to an unlockable index file, which has been fixed.
247 (merge bd58886775 pw/sequencer-recover-from-unlockable-index later to maint).
249 * "git apply --inaccurate-eof" when used with "--ignore-space-change"
250 triggered an internal sanity check, which has been fixed.
251 (merge 4855de1233 rs/apply-inaccurate-eof-with-incomplete-line later to maint).
253 * Command line completion (in contrib/) has been taught about the
254 "--copy" option of "git branch".
255 (merge 41ca0f773e tz/complete-branch-copy later to maint).
257 * When "git rebase" prepared an mailbox of changes and fed it to "git
258 am" to replay them, it was confused when a stray "From " happened
259 to be in the log message of one of the replayed changes. This has
260 been corrected.
261 (merge ae3b2b04bb ew/rebase-mboxrd later to maint).
263 * There was a recent semantic mismerge in the codepath to write out a
264 section of a configuration section, which has been corrected.
266 * Mentions of "git-rebase" and "git-am" (dashed form) still remained
267 in end-user visible strings emitted by the "git rebase" command;
268 they have been corrected.
269 (merge 82cb775c06 ks/rebase-no-git-foo later to maint).
271 * Contrary to the documentation, "git pull -4/-6 other-args" did not
272 ask the underlying "git fetch" to go over IPv4/IPv6, which has been
273 corrected.
274 (merge ffb4568afe sw/pull-ipv46-passthru later to maint).
276 * Other minor doc, test and build updates and code cleanups.
277 (merge c5e3bc6ec4 sd/branch-copy later to maint).