| blob | 3b65461c0b5a4384fbe28dfbd0e3a30e9a9a533a |
1 Git v2.1 Release Notes
2 ======================
4 Backward compatibility notes
5 ----------------------------
7 * The default value we give to the environment variable LESS has been
8 changed from "FRSX" to "FRX", losing "S" (chop long lines instead
9 of wrapping). Existing users who prefer not to see line-wrapped
10 output may want to set
12 $ git config core.pager "less -S"
14 to restore the traditional behaviour. It is expected that people
15 find output from the most subcommands easier to read with the new
16 default, except for "blame" which tends to produce really long
17 lines. To override the new default only for "git blame", you can
18 do this:
20 $ git config pager.blame "less -S"
22 * A few disused directories in contrib/ have been retired.
25 Updates since v2.0
26 ------------------
28 UI, Workflows & Features
30 * Since the very beginning of Git, we gave the LESS environment a
31 default value "FRSX" when we spawn "less" as the pager. "S" (chop
32 long lines instead of wrapping) has been removed from this default
33 set of options, because it is more or less a personal taste thing,
34 as opposed to others that have good justifications (i.e. "R" is
35 very much justified because many kinds of output we produce are
36 colored and "FX" is justified because output we produce is often
37 shorter than a page).
39 * The logic and data used to compute the display width needed for
40 UTF-8 strings have been updated to match Unicode 7.0 better.
42 * HTTP-based transports learned to propagate the error messages from
43 the webserver better to the client coming over the HTTP transport.
45 * The completion script for bash (in contrib/) has been updated to
46 handle aliases that define complex sequence of commands better.
48 * The "core.preloadindex" configuration variable is by default
49 enabled, allowing modern platforms to take advantage of the
50 multiple cores they have.
52 * "git commit --date=<date>" option learned to read from more
53 timestamp formats, including "--date=now".
55 * The `core.commentChar` configuration variable is used to specify a
56 custom comment character other than the default "#" to be used in
57 the commit log editor. This can be set to `auto` to attempt to
58 choose a different character that does not conflict with what
59 already starts a line in the message being edited for cases like
60 "git commit --amend".
62 * "git format-patch" learned --signature-file=<file> to take the mail
63 signature from.
65 * "git grep" learned grep.fullname configuration variable to force
66 "--full-name" to be default. This may cause regressions on
67 scripted users that do not expect this new behaviour.
69 * "git imap-send" learned to ask the credential helper for auth
70 material.
72 * "git log" and friends now understand the value "auto" set to the
73 "log.decorate" configuration variable to enable the "--decorate"
74 option automatically when the output is sent to tty.
76 * "git merge" without argument, even when there is an upstream
77 defined for the current branch, refused to run until
78 merge.defaultToUpstream is set to true. Flip the default of that
79 configuration variable to true.
81 * "git mergetool" learned to drive the vimdiff3 backend.
83 * mergetool.prompt used to default to 'true', always asking "do you
84 really want to run the tool on this path?". Among the two
85 purposes this prompt serves, ignore the use case to confirm that
86 the user wants to view particular path with the named tool, and
87 redefine the meaning of the prompt only to confirm the choice of
88 the tool made by the autodetection (for those who configured the
89 tool explicitly, the prompt shown for the latter purpose is
90 simply annoying).
92 Strictly speaking, this is a backward incompatible change and the
93 users need to explicitly set the variable to 'true' if they want
94 to resurrect the now-ignored use case.
96 * "git replace" learned the "--edit" subcommand.
98 * "git send-email" learned "--to-cover" and "--cc-cover" options, to
99 tell it to copy To: and Cc: headers found in the first input file
100 when emitting later input files.
102 * "git svn" learned to cope with malformed timestamps with only one
103 digit in the hour part, e.g. 2014-01-07T5:01:02.048176Z, emitted
104 by some broken subversion server implementations.
106 * "git tag" when editing the tag message shows the name of the tag
107 being edited as a comment in the editor.
109 * "git verify-commit" command to check GPG signature in signed
110 commits, in a way similar to "git verify-tag" is used to check
111 signed tags, was added.
114 Performance, Internal Implementation, etc.
116 * Build procedure for 'subtree' (in contrib/) has been cleaned up.
118 * Effort to shrink the size of patches Windows folks maintain on top
119 by upstreaming them continues.
121 * Patches maintained by msysgit folks for Windows port are being
122 upstreamed here a bit by bit.
124 * The leaf function to check validity of a refname format has been
125 micro-optimized, using SSE2 instructions when available. A few
126 breakages during its development have been caught and fixed already
127 but there might remain some more still; please test and report if
128 you find any.
130 * The `core.deltabasecachelimit` used to default to 16 MiB , but this
131 proved to be too small, and has been bumped to 96 MiB.
133 * "git blame" has been optimized greatly by reorganising the data
134 structure that is used to keep track of the work to be done.
136 * "git diff" that compares 3-or-more trees (e.g. parents and the
137 result of a merge) have been optimized.
139 * The API to update/delete references are being converted to handle
140 updates to multiple references in a transactional way. As an
141 example, "update-ref --stdin [-z]" has been updated to use this
142 API.
144 * Parts of the test scripts can be skipped by using a range notation,
145 e.g. "sh t1234-test.sh --run='1-4 6 8-'" to omit test piece 5 and 7
146 and run everything else.
149 Also contains various documentation updates and code clean-ups.
152 Fixes since v2.0
153 ----------------
155 Unless otherwise noted, all the fixes since v2.0 in the maintenance
156 track are contained in this release (see the maintenance releases'
157 notes for details).
159 * We used to unconditionally disable the pager in the pager process
160 we spawn to feed out output, but that prevented people who want to
161 run "less" within "less" from doing so.
162 (merge c0459ca je/pager-do-not-recurse later to maint).
164 * Tools that read diagnostic output in our standard error stream do
165 not want to see terminal control sequence (e.g. erase-to-eol).
166 Detect them by checking if the standard error stream is connected
167 to a tty.
168 (merge 38de156 mn/sideband-no-ansi later to maint).
170 * Mishandling of patterns in .gitignore that has trailing SPs quoted
171 with backslashes (e.g. ones that end with "\ ") have been
172 corrected.
173 (merge 97c1364be6b pb/trim-trailing-spaces later to maint).
175 * Reworded the error message given upon a failure to open an existing
176 loose object file due to e.g. permission issues; it was reported as
177 the object being corrupt, but that is not quite true.
178 (merge d6c8a05 jk/report-fail-to-read-objects-better later to maint).
180 * "git log -2master" is a common typo that shows two commits starting
181 from whichever random branch that is not 'master' that happens to
182 be checked out currently.
183 (merge e3fa568 jc/revision-dash-count-parsing later to maint).
185 * Code to avoid adding the same alternate object store twice was
186 subtly broken for a long time, but nobody seems to have noticed.
187 (merge 80b4785 rs/fix-alt-odb-path-comparison later to maint).
189 * The "%<(10,trunc)%s" pretty format specifier in the log family of
190 commands is used to truncate the string to a given length (e.g. 10
191 in the example) with padding to column-align the output, but did
192 not take into account that number of bytes and number of display
193 columns are different.
194 (merge 7d50987 as/pretty-truncate later to maint).
196 * "%G" (nothing after G) is an invalid pretty format specifier, but
197 the parser did not notice it as garbage.
198 (merge 958b2eb jk/pretty-G-format-fixes later to maint).
200 * A handful of code paths had to read the commit object more than
201 once when showing header fields that are usually not parsed. The
202 internal data structure to keep track of the contents of the commit
203 object has been updated to reduce the need for this double-reading,
204 and to allow the caller find the length of the object.
205 (merge 218aa3a jk/commit-buffer-length later to maint).
207 * The "mailmap.file" configuration option did not support the tilde
208 expansion (i.e. ~user/path and ~/path).
209 (merge 9352fd5 ow/config-mailmap-pathname later to maint).
211 * The completion scripts (in contrib/) did not know about quite a few
212 options that are common between "git merge" and "git pull", and a
213 couple of options unique to "git merge".
214 (merge 8fee872 jk/complete-merge-pull later to maint).
216 * "--ignore-space-change" option of "git apply" ignored the spaces
217 at the beginning of line too aggressively, which is inconsistent
218 with the option of the same name "diff" and "git diff" have.
219 (merge 14d3bb4 jc/apply-ignore-whitespace later to maint).
221 * "git blame" miscounted number of columns needed to show localized
222 timestamps, resulting in jaggy left-side-edge of the source code
223 lines in its output.
224 (merge dd75553 jx/blame-align-relative-time later to maint).
226 * "git blame" assigned the blame to the copy in the working-tree if
227 the repository is set to core.autocrlf=input and the file used CRLF
228 line endings.
229 (merge 4d4813a bc/blame-crlf-test later to maint).
231 * "git clone -b brefs/tags/bar" would have mistakenly thought we were
232 following a single tag, even though it was a name of the branch,
233 because it incorrectly used strstr().
234 (merge 60a5f5f jc/fix-clone-single-starting-at-a-tag later to maint).
236 * "git commit --allow-empty-messag -C $commit" did not work when the
237 commit did not have any log message.
238 (merge 076cbd6 jk/commit-C-pick-empty later to maint).
240 * "git diff --find-copies-harder" sometimes pretended as if the mode
241 bits have changed for paths that are marked with assume-unchanged
242 bit.
243 (merge 5304810 jk/diff-files-assume-unchanged later to maint).
245 * "git format-patch" did not enforce the rule that the "--follow"
246 option from the log/diff family of commands must be used with
247 exactly one pathspec.
248 (merge dd63f16 jk/diff-follow-must-take-one-pathspec later to maint).
250 * "git gc --auto" was recently changed to run in the background to
251 give control back early to the end-user sitting in front of the
252 terminal, but it forgot that housekeeping involving reflogs should
253 be done without other processes competing for accesses to the refs.
254 (merge 62aad18 nd/daemonize-gc later to maint).
256 * "git grep -O" to show the lines that hit in the pager did not work
257 well with case insensitive search. We now spawn "less" with its
258 "-I" option when it is used as the pager (which is the default).
259 (merge f7febbe sk/spawn-less-case-insensitively-from-grep-O-i later to maint).
261 * We used to disable threaded "git index-pack" on platforms without
262 thread-safe pread(); use a different workaround for such
263 platforms to allow threaded "git index-pack".
264 (merge 3953949 nd/index-pack-one-fd-per-thread later to maint).
266 * The error reporting from "git index-pack" has been improved to
267 distinguish missing objects from type errors.
268 (merge 77583e7 jk/index-pack-report-missing later to maint).
270 * "git mailinfo" used to read beyond the end of header string while
271 parsing an incoming e-mail message to extract the patch.
272 (merge b1a013d rs/mailinfo-header-cmp later to maint).
274 * On a case insensitive filesystem, merge-recursive incorrectly
275 deleted the file that is to be renamed to a name that is the same
276 except for case differences.
277 (merge baa37bf dt/merge-recursive-case-insensitive later to maint).
279 * "git pack-objects" unnecessarily copied the previous contents when
280 extending the hashtable, even though it will populate the table
281 from scratch anyway.
282 (merge fb79947 rs/pack-objects-no-unnecessary-realloc later to maint).
284 * Recent updates to "git repack" started to duplicate objects that
285 are in packfiles marked with .keep flag into the new packfile by
286 mistake.
287 (merge d078d85 jk/repack-pack-keep-objects later to maint).
289 * "git rerere forget" did not work well when merge.conflictstyle
290 was set to a non-default value.
291 (merge de3d8bb fc/rerere-conflict-style later to maint).
293 * "git remote rm" and "git remote prune" can involve removing many
294 refs at once, which is not a very efficient thing to do when very
295 many refs exist in the packed-refs file.
296 (merge e6bea66 jl/remote-rm-prune later to maint).
298 * "git log --exclude=<glob> --all | git shortlog" worked as expected,
299 but "git shortlog --exclude=<glob> --all", which is supposed to be
300 identical to the above pipeline, was not accepted at the command
301 line argument parser level.
302 (merge eb07774 jc/shortlog-ref-exclude later to maint).
304 * The autostash mode of "git rebase -i" did not restore the dirty
305 working tree state if the user aborted the interactive rebase by
306 emptying the insn sheet.
307 (merge ddb5432 rr/rebase-autostash-fix later to maint).
309 * During "git rebase --merge", a conflicted patch could not be
310 skipped with "--skip" if the next one also conflicted.
311 (merge 95104c7 bc/fix-rebase-merge-skip later to maint).
313 * "git show -s" (i.e. show log message only) used to incorrectly emit
314 an extra blank line after a merge commit.
315 (merge ad2f725 mk/show-s-no-extra-blank-line-for-merges later to maint).
317 * "git status", even though it is a read-only operation, tries to
318 update the index with refreshed lstat(2) info to optimize future
319 accesses to the working tree opportunistically, but this could
320 race with a "read-write" operation that modify the index while it
321 is running. Detect such a race and avoid overwriting the index.
322 (merge 426ddee ym/fix-opportunistic-index-update-race later to maint).
324 * "git status" (and "git commit") behaved as if changes in a modified
325 submodule are not there if submodule.*.ignore configuration is set,
326 which was misleading. The configuration is only to unclutter diff
327 output during the course of development, and should not to hide
328 changes in the "status" output to cause the users forget to commit
329 them.
330 (merge c215d3d jl/status-added-submodule-is-never-ignored later to maint).
332 * Documentation for "git submodule sync" forgot to say that the subcommand
333 can take the "--recursive" option.
334 (merge 9393ae7 mc/doc-submodule-sync-recurse later to maint).
336 * "git update-index --cacheinfo" in 2.0 release crashed on a
337 malformed command line.
338 (merge c8e1ee4 jc/rev-parse-argh-dashed-multi-words later to maint).
340 * The mode to run tests with HTTP server tests disabled was broken.
341 (merge afa53fe na/no-http-test-in-the-middle later to maint).