]>
Commit | Line | Data |
---|---|---|
048d4d98 JH |
1 | Git v1.8.3 Release Notes |
2 | ======================== | |
3 | ||
4 | Backward compatibility notes (for Git 2.0) | |
5 | ------------------------------------------ | |
6 | ||
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 will change to the "simple" | |
e7a3c902 MB |
11 | semantics that pushes only the current branch to the branch with the same |
12 | name, and only when the current branch is set to integrate with that | |
13 | remote branch. Use the user preference configuration variable | |
048d4d98 | 14 | "push.default" to change this. If you are an old-timer who is used |
e7a3c902 MB |
15 | to the "matching" semantics, you can set the variable to "matching" |
16 | to keep the traditional behaviour. If you want to live in the future | |
17 | early, you can set it to "simple" today without waiting for Git 2.0. | |
18 | ||
19 | When "git add -u" (and "git add -A") is run inside a subdirectory and | |
20 | does not specify which paths to add on the command line, it | |
21 | will operate on the entire tree in Git 2.0 for consistency | |
22 | with "git commit -a" and other commands. There will be no | |
23 | mechanism to make plain "git add -u" behave like "git add -u .". | |
24 | Current users of "git add -u" (without a pathspec) should start | |
25 | training their fingers to explicitly say "git add -u ." | |
26 | before Git 2.0 comes. A warning is issued when these commands are | |
d2949c7b JH |
27 | run without a pathspec and when you have local changes outside the |
28 | current directory, because the behaviour in Git 2.0 will be different | |
29 | from today's version in such a situation. | |
048d4d98 | 30 | |
b75cdfaa JH |
31 | In Git 2.0, "git add <path>" will behave as "git add -A <path>", so |
32 | that "git add dir/" will notice paths you removed from the directory | |
33 | and record the removal. Versions before Git 2.0, including this | |
34 | release, will keep ignoring removals, but the users who rely on this | |
e7a3c902 MB |
35 | behaviour are encouraged to start using "git add --ignore-removal <path>" |
36 | now before 2.0 is released. | |
b75cdfaa | 37 | |
048d4d98 JH |
38 | |
39 | Updates since v1.8.2 | |
40 | -------------------- | |
41 | ||
3d88f83d JH |
42 | Foreign interface |
43 | ||
ab846217 | 44 | * remote-hg and remote-bzr helpers (in contrib/ since v1.8.2) have |
edca4152 JH |
45 | been updated; especially, the latter has been done in an |
46 | accelerated schedule (read: we may not have merged to this release | |
47 | if we were following the usual "cook sufficiently in next before | |
48 | unleashing it to the world" workflow) in order to help Emacs folks, | |
49 | whose primary SCM seems to be stagnating. | |
ab846217 | 50 | |
3d88f83d | 51 | |
048d4d98 JH |
52 | UI, Workflows & Features |
53 | ||
ab846217 JH |
54 | * A handful of updates applied to gitk, including an addition of |
55 | "revert" action, showing dates in tags in a nicer way, making | |
56 | colors configurable, and support for -G'pickaxe' search. | |
57 | ||
de0977d5 JH |
58 | * The prompt string generator (in contrib/completion/) learned to |
59 | show how many changes there are in total and how many have been | |
60 | replayed during a "git rebase" session. | |
61 | ||
d2949c7b JH |
62 | * "git branch --vv" learned to paint the name of the branch it |
63 | integrates with in a different color (color.branch.upstream, | |
64 | which defaults to blue). | |
65 | ||
3d88f83d JH |
66 | * In a sparsely populated working tree, "git checkout <pathspec>" no |
67 | longer unmarks paths that match the given pathspec that were | |
68 | originally ignored with "--sparse" (use --ignore-skip-worktree-bits | |
69 | option to resurrect these paths out of the index if you really want | |
70 | to). | |
71 | ||
562af5b0 JH |
72 | * "git log --format" specifier learned %C(auto) token that tells Git |
73 | to use color when interpolating %d (decoration), %h (short commit | |
74 | object name), etc. for terminal output. | |
75 | ||
3d88f83d JH |
76 | * "git bisect" leaves the final outcome as a comment in its bisect |
77 | log file. | |
78 | ||
79 | * "git clone --reference" can now refer to a gitfile "textual symlink" | |
80 | that points at the real location of the repository. | |
81 | ||
d2949c7b JH |
82 | * "git count-objects" learned "--human-readable" aka "-H" option to |
83 | show various large numbers in Ki/Mi/GiB scaled as necessary. | |
84 | ||
85 | * "git cherry-pick $blob" and "git cherry-pick $tree" are nonsense, | |
86 | and a more readable error message e.g. "can't cherry-pick a tree" | |
87 | is given (we used to say "expected exactly one commit"). | |
88 | ||
1468a583 JH |
89 | * The "--annotate" option to "git send-email" can be turned on (or |
90 | off) by default with sendemail.annotate configuration variable (you | |
91 | can use --no-annotate from the command line to override it). | |
92 | ||
93 | * The "--cover-letter" option to "git format-patch" can be turned on | |
94 | (or off) by default with format.coverLetter configuration | |
95 | variable. By setting it to 'auto', you can turn it on only for a | |
96 | series with two or more patches. | |
97 | ||
98 | * The bash completion support (in contrib/) learned that cherry-pick | |
99 | takes a few more options than it already knew about. | |
100 | ||
7ece7ee6 JH |
101 | * "git help" learned "-g" option to show the list of guides just like |
102 | list of commands are given with "-a". | |
103 | ||
c17b651f JH |
104 | * A triangular "pull from one place, push to another place" workflow |
105 | is supported better by new remote.pushdefault (overrides the | |
106 | "origin" thing) and branch.*.pushremote (overrides the | |
107 | branch.*.remote) configuration variables. | |
108 | ||
109 | * "git status" learned to report that you are in the middle of a | |
110 | revert session, just like it does for a cherry-pick and a bisect | |
111 | session. | |
112 | ||
7ece7ee6 JH |
113 | * The handling by "git branch --set-upstream-to" against various forms |
114 | of erroneous inputs was suboptimal and has been improved. | |
c17b651f | 115 | |
7b592fad | 116 | * When the interactive access to git-shell is not enabled, it issues |
edca4152 JH |
117 | a message meant to help the system administrator to enable it. An |
118 | explicit way has been added to issue custom messages to refuse an | |
119 | access over the network to help the end users who connect to the | |
120 | service expecting an interactive shell. | |
7b592fad | 121 | |
cc3e4eba JH |
122 | * In addition to the case where the user edits the log message with |
123 | the "e)dit" option of "am -i", replace the "Applying: this patch" | |
124 | message with the final log message contents after applymsg hook | |
125 | munges it. | |
126 | ||
127 | * "git status" suggests users to look into using --untracked=no option | |
7b592fad JH |
128 | when it takes too long. |
129 | ||
edca4152 JH |
130 | * "git status" shows a bit more information during a rebase/bisect |
131 | session. | |
cc3e4eba | 132 | |
7b592fad JH |
133 | * "git fetch" learned to fetch a commit at the tip of an unadvertised |
134 | ref by specifying a raw object name from the command line when the | |
135 | server side supports this feature. | |
136 | ||
aec3f779 JH |
137 | * Output from "git log --graph" works better with submodule log |
138 | output now. | |
139 | ||
7b592fad JH |
140 | * "git count-objects -v" learned to report leftover temporary |
141 | packfiles and other garbage in the object store. | |
142 | ||
143 | * A new read-only credential helper (in contrib/) to interact with | |
144 | the .netrc/.authinfo files has been added. | |
145 | ||
146 | * "git send-email" can be used with the credential helper system. | |
048d4d98 | 147 | |
7632cd27 JH |
148 | * There was no Porcelain way to say "I no longer am interested in |
149 | this submodule", once you express your interest in a submodule with | |
150 | "submodule init". "submodule deinit" is the way to do so. | |
151 | ||
152 | * "git pull --rebase" learned to pass "-v/-q" options to underlying | |
153 | "git rebase". | |
154 | ||
155 | * The new "--follow-tags" option tells "git push" to push relevant | |
156 | annotated tags when pushing branches out. | |
157 | ||
21ccebec JH |
158 | * "git merge" and "git pull" can optionally be told to inspect and |
159 | reject when merging a commit that does not carry a trusted GPG | |
160 | signature. | |
161 | ||
2bba2f0e JH |
162 | * "git mergetool" now feeds files to the "p4merge" backend in the |
163 | order that matches the p4 convention, where "theirs" is usually | |
e7a3c902 MB |
164 | shown on the left side, which is the opposite from what other backends |
165 | expect. | |
048d4d98 | 166 | |
961c5129 JH |
167 | * "show/log" now honors gpg.program configuration just like other |
168 | parts of the code that use GnuPG. | |
169 | ||
170 | * "git log" that shows the difference between the parent and the | |
171 | child has been optimized somewhat. | |
172 | ||
8d994db4 JH |
173 | * "git difftool" allows the user to write into the temporary files |
174 | being shown; if the user makes changes to the working tree at the | |
de0977d5 JH |
175 | same time, it now refrains from overwriting the copy in the working |
176 | tree and leaves the temporary file so that changes can be merged | |
177 | manually. | |
8d994db4 JH |
178 | |
179 | * There was no good way to ask "I have a random string that came from | |
180 | outside world. I want to turn it into a 40-hex object name while | |
181 | making sure such an object exists". A new peeling suffix ^{object} | |
182 | can be used for that purpose, together with "rev-parse --verify". | |
183 | ||
048d4d98 JH |
184 | |
185 | Performance, Internal Implementation, etc. | |
186 | ||
7b9a4198 | 187 | * Updates for building under msvc. |
048d4d98 | 188 | |
e7a3c902 | 189 | * A handful of issues in the code that traverses the working tree to find |
562af5b0 JH |
190 | untracked and/or ignored files have been fixed, and the general |
191 | codepath involved in "status -u" and "clean" have been cleaned up | |
192 | and optimized. | |
193 | ||
1468a583 JH |
194 | * The stack footprint of some codepaths that access an object from a |
195 | pack has been shrunk. | |
196 | ||
7ece7ee6 | 197 | * The logic to coalesce the same lines removed from the parents in |
e7a3c902 | 198 | the output from "diff -c/--cc" has been updated, but with O(n^2) |
7ece7ee6 JH |
199 | complexity, so this might turn out to be undesirable. |
200 | ||
c17b651f | 201 | * The code to enforce permission bits on files in $GIT_DIR/ for |
e7a3c902 | 202 | shared repositories has been simplified. |
c17b651f | 203 | |
e7a3c902 MB |
204 | * A few codepaths know how much data they need to put in the |
205 | hashtables they use when they start, but still began with small tables | |
206 | and repeatedly grew and rehashed them. | |
7b592fad | 207 | |
2bba2f0e JH |
208 | * The API to walk reflog entries from the latest to older, which was |
209 | necessary for operations such as "git checkout -", was cumbersome | |
210 | to use correctly and also inefficient. | |
211 | ||
cc3e4eba JH |
212 | * Codepaths that inspect log-message-to-be and decide when to add a |
213 | new Signed-off-by line in various commands have been consolidated. | |
214 | ||
215 | * The pkt-line API, implementation and its callers have been cleaned | |
216 | up to make them more robust. | |
217 | ||
e7a3c902 | 218 | * The Cygwin port has a faster-but-lying lstat(2) emulation whose |
961c5129 | 219 | incorrectness does not matter in practice except for a few |
e7a3c902 | 220 | codepaths, and setting permission bits on directories is a codepath |
961c5129 JH |
221 | that needs to use a more correct one. |
222 | ||
8d994db4 JH |
223 | * "git checkout" had repeated pathspec matches on the same paths, |
224 | which have been consolidated. Also a bug in "git checkout dir/" | |
225 | that is started from an unmerged index has been fixed. | |
226 | ||
7ece7ee6 JH |
227 | * A few bugfixes to "git rerere" working on corner case merge |
228 | conflicts have been applied. | |
229 | ||
048d4d98 | 230 | |
1468a583 | 231 | Also contains various documentation updates and code clean-ups. |
048d4d98 JH |
232 | |
233 | ||
234 | Fixes since v1.8.2 | |
235 | ------------------ | |
236 | ||
237 | Unless otherwise noted, all the fixes since v1.8.2 in the maintenance | |
238 | track are contained in this release (see release notes to them for | |
239 | details). | |
7b9a4198 | 240 | |
de0977d5 JH |
241 | * Recent versions of File::Temp (used by "git svn") started blowing |
242 | up when its tempfile sub is called as a class method; updated the | |
243 | callsite to call it as a plain vanilla function to fix it. | |
244 | (merge eafc2dd hb/git-pm-tempfile later to maint). | |
245 | ||
246 | * Various subcommands of "git remote" simply ignored extraneous | |
247 | command line arguments instead of diagnosing them as errors. | |
de0977d5 | 248 | |
e7a3c902 | 249 | * When receive-pack detects an error in the pack header it received in |
562af5b0 | 250 | order to decide which of unpack-objects or index-pack to run, it |
e7a3c902 | 251 | returned without closing the error stream, which led to a hung |
562af5b0 | 252 | sideband thread. |
562af5b0 | 253 | |
e7a3c902 | 254 | * Zsh completion forgot that the '%' character used to signal untracked |
562af5b0 | 255 | files needs to be escaped with another '%'. |
562af5b0 | 256 | |
3d88f83d | 257 | * A commit object whose author or committer ident are malformed |
e7a3c902 | 258 | crashed some code that trusted that a name, an email and a |
3d88f83d | 259 | timestamp can always be found in it. |
3d88f83d | 260 | |
d2949c7b | 261 | * When "upload-pack" fails while generating a pack in response to |
e7a3c902 MB |
262 | "git fetch" (or "git clone"), the receiving side had |
263 | a programming error that triggered the die handler | |
d2949c7b | 264 | recursively. |
d2949c7b | 265 | |
e7a3c902 MB |
266 | * "rev-list --stdin" and friends kept bogus pointers into the input |
267 | buffer around as human readable object names. This was not a huge | |
d2949c7b JH |
268 | problem but was exposed by a new change that uses these names in |
269 | error output. | |
d2949c7b | 270 | |
1468a583 | 271 | * Smart-capable HTTP servers were not restricted via the |
e7a3c902 MB |
272 | GIT_NAMESPACE mechanism when talking with commit-walking clients, |
273 | like they are when talking with smart HTTP clients. | |
1468a583 JH |
274 | (merge 6130f86 jk/http-dumb-namespaces later to maint). |
275 | ||
276 | * "git merge-tree" did not omit a merge result that is identical to | |
e7a3c902 | 277 | the "our" side in certain cases. |
1468a583 JH |
278 | (merge aacecc3 jk/merge-tree-added-identically later to maint). |
279 | ||
e7a3c902 | 280 | * Perl scripts like "git-svn" closed (instead of redirecting to /dev/null) |
aec3f779 | 281 | the standard error stream, which is not a very smart thing to do. |
e7a3c902 MB |
282 | A later open may return file descriptor #2 for an unrelated purpose, and |
283 | error reporting code may write into it. | |
aec3f779 JH |
284 | |
285 | * "git show-branch" was not prepared to show a very long run of | |
286 | ancestor operators e.g. foobar^2~2^2^2^2...^2~4 correctly. | |
aec3f779 JH |
287 | |
288 | * "git diff --diff-algorithm algo" is also understood as "git diff | |
289 | --diff-algorithm=algo". | |
aec3f779 | 290 | |
e7a3c902 | 291 | * The new core.commentchar configuration was not applied in a few |
aec3f779 | 292 | places. |
aec3f779 | 293 | |
e7a3c902 MB |
294 | * "git bundle" erroneously bailed out when parsing a valid bundle |
295 | containing a prerequisite commit without a commit message. | |
aec3f779 | 296 | |
7ece7ee6 | 297 | * "git log -S/-G" started paying attention to textconv filter, but |
e7a3c902 | 298 | there was no way to disable this. Make it honor the --no-textconv |
7ece7ee6 | 299 | option. |
7ece7ee6 | 300 | |
e7a3c902 | 301 | * When used with the "-d temporary-directory" option, "git filter-branch" |
c17b651f JH |
302 | failed to come back to the original working tree to perform the |
303 | final clean-up procedure. | |
c17b651f | 304 | |
21ccebec JH |
305 | * "git merge $(git rev-parse v1.8.2)" behaved quite differently from |
306 | "git merge v1.8.2", as if v1.8.2 were written as v1.8.2^0 and did | |
307 | not pay much attention to the annotated tag payload. Make the code | |
308 | notice the type of the tag object, in addition to the dwim_ref() | |
309 | based classification the current code uses (i.e. the name appears | |
e7a3c902 | 310 | in refs/tags/) to decide when to special-case tag merging. |
21ccebec | 311 | |
e7a3c902 | 312 | * Fix a 1.8.1.x regression that stopped matching "dir" (without a |
8d994db4 | 313 | trailing slash) to a directory "dir". |
8d994db4 JH |
314 | |
315 | * "git apply --whitespace=fix" was not prepared to see a line getting | |
316 | longer after fixing whitespaces (e.g. tab-in-indent aka Python). | |
8d994db4 JH |
317 | |
318 | * The prompt string generator (in contrib/completion/) did not notice | |
319 | when we are in a middle of a "git revert" session. | |
8d994db4 | 320 | |
e7a3c902 | 321 | * "submodule summary --summary-limit" option did not support the |
8d994db4 | 322 | "--option=value" form. |
8d994db4 | 323 | |
7ece7ee6 | 324 | * "index-pack --fix-thin" used an uninitialized value to compute |
e7a3c902 | 325 | the delta depths of objects it appends to the resulting pack. |
cc3e4eba | 326 | |
e7a3c902 MB |
327 | * "index-pack --verify-stat" used a few counters outside the protection |
328 | of a mutex, possibly showing incorrect numbers. | |
cc3e4eba JH |
329 | |
330 | * The code to keep track of what directory names are known to Git on | |
e7a3c902 MB |
331 | platforms with case insensitive filesystems could get confused upon a |
332 | hash collision between these pathnames and would loop forever. | |
cc3e4eba | 333 | |
e7a3c902 MB |
334 | * Annotated tags outside the refs/tags/ hierarchy were not advertised |
335 | correctly to ls-remote and fetch with recent versions of Git. | |
7632cd27 | 336 | |
e7a3c902 | 337 | * Recent optimizations broke shallow clones. |
7632cd27 JH |
338 | |
339 | * "git cmd -- ':(top'" was not diagnosed as an invalid syntax, and | |
340 | instead the parser kept reading beyond the end of the string. | |
7632cd27 JH |
341 | |
342 | * "git tag -f <tag>" always said "Updated tag '<tag>'" even when | |
e7a3c902 | 343 | creating a new tag (i.e. neither overwriting nor updating). |
7632cd27 JH |
344 | |
345 | * "git p4" did not behave well when the path to the root of the P4 | |
346 | client was not its real path. | |
347 | (merge bbd8486 pw/p4-symlinked-root later to maint). | |
348 | ||
e7a3c902 MB |
349 | * "git archive" reported a failure when asked to create an archive out |
350 | of an empty tree. It is more intuitive to give an empty | |
7632cd27 | 351 | archive back in such a case. |
7632cd27 | 352 | |
e7a3c902 | 353 | * When "format-patch" quoted a non-ascii string in header files, |
7632cd27 | 354 | it incorrectly applied rfc2047 and chopped a single character in |
e7a3c902 | 355 | the middle of the string. |
7632cd27 JH |
356 | |
357 | * An aliased command spawned from a bare repository that does not say | |
e7a3c902 | 358 | it is bare with "core.bare = yes" was treated as non-bare by mistake. |
7632cd27 | 359 | |
e7a3c902 | 360 | * In "git reflog expire", the REACHABLE bit was not cleared from the |
7632cd27 | 361 | correct objects. |
7632cd27 JH |
362 | |
363 | * The logic used by "git diff -M --stat" to shorten the names of | |
364 | files before and after a rename did not work correctly when the | |
365 | common prefix and suffix between the two filenames overlapped. | |
7632cd27 JH |
366 | |
367 | * The "--match=<pattern>" option of "git describe", when used with | |
e7a3c902 | 368 | "--all" to allow refs that are not annotated tags to be a |
7632cd27 | 369 | base of description, did not restrict the output from the command |
e7a3c902 | 370 | to those refs that match the given pattern. |
7632cd27 | 371 | |
7b592fad JH |
372 | * Clarify in the documentation "what" gets pushed to "where" when the |
373 | command line to "git push" does not say these explicitly. | |
7b592fad | 374 | |
7b9a4198 JH |
375 | * The "--color=<when>" argument to the commands in the diff family |
376 | was described poorly. | |
7b9a4198 | 377 | |
e7a3c902 | 378 | * The arguments given to the pre-rebase hook were not documented. |
7b9a4198 JH |
379 | |
380 | * The v4 index format was not documented. | |
7b9a4198 JH |
381 | |
382 | * The "--match=<pattern>" argument "git describe" takes uses glob | |
383 | pattern but it wasn't obvious from the documentation. | |
7b9a4198 JH |
384 | |
385 | * Some sources failed to compile on systems that lack NI_MAXHOST in | |
386 | their system header (e.g. z/OS). | |
7b9a4198 JH |
387 | |
388 | * Add an example use of "--env-filter" in "filter-branch" | |
389 | documentation. | |
7b9a4198 JH |
390 | |
391 | * "git bundle verify" did not say "records a complete history" for a | |
392 | bundle that does not have any prerequisites. | |
7b9a4198 JH |
393 | |
394 | * In the v1.8.0 era, we changed symbols that do not have to be global | |
395 | to file scope static, but a few functions in graph.c were used by | |
e7a3c902 | 396 | CGit sideways, bypassing the entry points of the API the |
7b9a4198 | 397 | in-tree users use. |
7b592fad JH |
398 | |
399 | * "git update-index -h" did not do the usual "-h(elp)" thing. | |
400 | ||
401 | * "git index-pack" had a buffer-overflow while preparing an | |
402 | informational message when the translated version of it was too | |
403 | long. | |
7b592fad JH |
404 | |
405 | * 'git commit -m "$msg"' used to add an extra newline even when | |
406 | $msg already ended with one. | |
7b592fad JH |
407 | |
408 | * The SSL peer verification done by "git imap-send" did not ask for | |
e7a3c902 | 409 | Server Name Indication (RFC 4366), failing to connect to SSL/TLS |
7b592fad | 410 | sites that serve multiple hostnames on a single IP. |
7b592fad JH |
411 | |
412 | * perl/Git.pm::cat_blob slurped everything in core only to write it | |
413 | out to a file descriptor, which was not a very smart thing to do. | |
7b592fad JH |
414 | |
415 | * "git branch" did not bother to check nonsense command line | |
e7a3c902 | 416 | parameters. It now issues errors in many cases. |
7b592fad | 417 | |
e7a3c902 | 418 | * Verification of signed tags was not done correctly when not in C |
7b592fad | 419 | or en/US locale. |
7b592fad JH |
420 | |
421 | * Some platforms and users spell UTF-8 differently; retry with the | |
422 | most official "UTF-8" when the system does not understand the | |
e7a3c902 MB |
423 | user-supplied encoding name that is a common alternative |
424 | spelling of UTF-8. | |
7b592fad | 425 | |
e7a3c902 | 426 | * When export-subst is used, "zip" output recorded an incorrect |
7b592fad | 427 | size of the file. |
7b592fad JH |
428 | |
429 | * "git am $maildir/" applied messages in an unexpected order; sort | |
430 | filenames read from the maildir/ in a way that is more likely to | |
e7a3c902 MB |
431 | sort the messages in the order the writing MUA meant to, by sorting |
432 | numeric segments in numeric order and non-numeric segments in | |
7b592fad | 433 | alphabetical order. |
7b592fad JH |
434 | |
435 | * "git submodule update", when recursed into sub-submodules, did not | |
7ece7ee6 | 436 | accumulate the prefix paths. |